d/freebsd-skq
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
f4cf3aa68b
freebsd-skq/usr.bin/kdump/kdump.c

2220 lines
48 KiB
C
Raw Normal View History

BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
/*-
General further adoption of SPDX licensing ID tags. Mainly focus on files that use BSD 3-Clause license. The Software Package Data Exchange (SPDX) group provides a specification to make it easier for automated tools to detect and summarize well known opensource licenses. We are gradually adopting the specification, noting that the tags are considered only advisory and do not, in any way, superceed or replace the license texts. Special thanks to Wind River for providing access to "The Duke of Highlander" tool: an older (2014) run over FreeBSD tree was useful as a starting point.
2017-11-20 19:49:47 +00:00
* SPDX-License-Identifier: BSD-3-Clause
*
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
* Copyright (c) 1988, 1993
* The Regents of the University of California. All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
Renumber copyright clause 4 Renumber cluase 4 to 3, per what everybody else did when BSD granted them permission to remove clause 3. My insistance on keeping the same numbering for legal reasons is too pedantic, so give up on that point. Submitted by: Jan Schaumann <jschauma@stevens.edu> Pull Request: https://github.com/freebsd/freebsd/pull/96
2017-02-28 23:42:47 +00:00
* 3. Neither the name of the University nor the names of its contributors
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
* may be used to endorse or promote products derived from this software
* without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*/
#ifndef lint
Use err(3). Correct incomplete man page.
1997-07-16 06:49:49 +00:00
static const char copyright[] =
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
"@(#) Copyright (c) 1988, 1993\n\
The Regents of the University of California. All rights reserved.\n";
#endif /* not lint */
#ifndef lint
Use err(3). Correct incomplete man page.
1997-07-16 06:49:49 +00:00
#if 0
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
static char sccsid[] = "@(#)kdump.c 8.1 (Berkeley) 6/6/93";
Use err(3). Correct incomplete man page.
1997-07-16 06:49:49 +00:00
#endif
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
#endif /* not lint */
Consistently use FBSDID
2002-06-30 05:25:07 +00:00
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
Expose kernel-only errno values if _WANT_KERNEL_ERRNO is defined. The kernel uses a few negative errno values for internal conditions such as requesting a system call restart. Normally these errno values are not exposed to userland. However, kdump needs access to these values as some of then can be present in a ktrace system call return record. Previously kdump was defining _KERNEL to gain access to ehse values, but was then having to manually declare 'errno' (and doing it incorrectly). Now, kdump uses _WANT_KERNEL_ERRNO instead of _KERNEL and uses the system-provided declaration of errno.
2016-10-01 22:08:07 +00:00
#define _WANT_KERNEL_ERRNO
Decode kevent structures logged via ktrace(2) in kdump. - Add a new KTR_STRUCT_ARRAY ktrace record type which dumps an array of structures. The structure name in the record payload is preceded by a size_t containing the size of the individual structures. Use this to replace the previous code that dumped the kevent arrays dumped for kevent(). kdump is now able to decode the kevent structures rather than dumping their contents via a hexdump. One change from before is that the 'changes' and 'events' arrays are not marked with separate 'read' and 'write' annotations in kdump output. Instead, the first array is the 'changes' array, and the second array (only present if kevent doesn't fail with an error) is the 'events' array. For kevent(), empty arrays are denoted by an entry with an array containing zero entries rather than no record. - Move kevent decoding tables from truss to libsysdecode. This adds three new functions to decode members of struct kevent: sysdecode_kevent_filter, sysdecode_kevent_flags, and sysdecode_kevent_fflags. kdump uses these helper functions to pretty-print kevent fields. - Move structure definitions for freebsd11 and freebsd32 kevent structures to <sys/event.h> so that they can be shared with userland. The 32-bit structures are only exposed if _WANT_KEVENT32 is defined. The freebsd11 structures are only exposed if _WANT_FREEBSD11_KEVENT is defined. The 32-bit freebsd11 structure requires both. - Decode freebsd11 kevent structures in truss for the compat11.kevent() system call. - Log 32-bit kevent structures via ktrace for 32-bit compat kevent() system calls. - While here, constify the 'void *data' argument to ktrstruct(). Reviewed by: kib (earlier version) MFC after: 1 month Differential Revision: https://reviews.freebsd.org/D12470
2017-11-25 04:49:12 +00:00
#ifdef __LP64__
#define _WANT_KEVENT32
#endif
#define _WANT_FREEBSD11_KEVENT
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
#include <sys/param.h>
Update most userspace consumers of capability.h to use capsicum.h instead. auditdistd is not updated as I will make the change upstream and then do a vendor import sometime in the next week or two. MFC after: 3 weeks
2014-03-16 11:04:44 +00:00
#include <sys/capsicum.h>
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
#include <sys/errno.h>
#include <sys/time.h>
#include <sys/uio.h>
Decode kevent structures logged via ktrace(2) in kdump. - Add a new KTR_STRUCT_ARRAY ktrace record type which dumps an array of structures. The structure name in the record payload is preceded by a size_t containing the size of the individual structures. Use this to replace the previous code that dumped the kevent arrays dumped for kevent(). kdump is now able to decode the kevent structures rather than dumping their contents via a hexdump. One change from before is that the 'changes' and 'events' arrays are not marked with separate 'read' and 'write' annotations in kdump output. Instead, the first array is the 'changes' array, and the second array (only present if kevent doesn't fail with an error) is the 'events' array. For kevent(), empty arrays are denoted by an entry with an array containing zero entries rather than no record. - Move kevent decoding tables from truss to libsysdecode. This adds three new functions to decode members of struct kevent: sysdecode_kevent_filter, sysdecode_kevent_flags, and sysdecode_kevent_fflags. kdump uses these helper functions to pretty-print kevent fields. - Move structure definitions for freebsd11 and freebsd32 kevent structures to <sys/event.h> so that they can be shared with userland. The 32-bit structures are only exposed if _WANT_KEVENT32 is defined. The freebsd11 structures are only exposed if _WANT_FREEBSD11_KEVENT is defined. The 32-bit freebsd11 structure requires both. - Decode freebsd11 kevent structures in truss for the compat11.kevent() system call. - Log 32-bit kevent structures via ktrace for 32-bit compat kevent() system calls. - While here, constify the 'void *data' argument to ktrstruct(). Reviewed by: kib (earlier version) MFC after: 1 month Differential Revision: https://reviews.freebsd.org/D12470
2017-11-25 04:49:12 +00:00
#include <sys/event.h>
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
#include <sys/ktrace.h>
kdump: decode SHM_ANON as first arg to legacy shm_open(2) The first argument to shm_open(2) as well as shm_open2(2) may be a path or SHM_ANON. Decode SHM_ANON, at least- paths will show up as namei results in kdump output, which may be sufficient; in those cases, we'll have printed an address. Future commits will add support for shm_open2() to libsysdecode/truss/kdump. Reported by: kaktus MFC after: 3 days
2020-02-19 14:32:55 +00:00
#include <sys/mman.h>
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
#include <sys/ioctl.h>
Add sockipprotoname() function. Decode the third parameter (protocol) of a socket() call with sockipprotoname() if the first parameter (domain) is PF_INET or PF_INET6. Old parsing behavior before this change: ping6 CALL socket(PF_INET6,SOCK_RAW,0x3a) New behavior after this change: ping6 CALL socket(PF_INET6,SOCK_RAW,IPPROTO_ICMPV6)
2007-01-04 04:46:59 +00:00
#include <sys/socket.h>
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
#include <sys/stat.h>
Teach kdump to understand sv_flags records in the trace files. MFC after: 1 Month.
2011-02-25 22:07:23 +00:00
#include <sys/sysent.h>
Decode the arguments passed to _umtx_op(). In particular, decode the opcode. MFC after: 1 week Sponsored by: Norse
2014-10-13 16:37:06 +00:00
#include <sys/umtx.h>
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
#include <sys/un.h>
Teach kdump to understand sv_flags records in the trace files. MFC after: 1 Month.
2011-02-25 22:07:23 +00:00
#include <sys/queue.h>
- Decode the idtype argument passed to wait6() in kdump and truss. - Don't treat an options argument of 0 to wait4() as an error in kdump. - Decode the wait options passed to wait4() and wait6() in truss and decode the returned rusage and exit status. Approved by: re (kib) MFC after: 1 week
2013-09-12 18:08:25 +00:00
#include <sys/wait.h>
We use a few different ifdef's names to check if we are using Casper or not, let's standardize this. Now we are always use WITH_CASPER name. Discussed with: emaste@ MFC after: 1 month
2017-09-21 14:41:41 +00:00
#ifdef WITH_CASPER
Let the nv.h and dnv.h includes be only in sys directory. Change consumers to include those files from sys. Add duplicated files to ObsoleteFiles. Approved by: pjd (mentor)
2015-07-02 21:58:10 +00:00
#include <sys/nv.h>
#endif
Add two missing include files and prototype for sockfamilyname() which is generated by mksubr.
2009-03-20 18:29:52 +00:00
#include <arpa/inet.h>
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
#include <netinet/in.h>
Add two missing include files and prototype for sockfamilyname() which is generated by mksubr.
2009-03-20 18:29:52 +00:00
#include <ctype.h>
Remove the duplicated code using Capsicum helpers. Reviewed by: cem, ed, bapt, emaste Differential Revision https://reviews.freebsd.org/D8140
2016-10-07 17:56:57 +00:00
#include <capsicum_helpers.h>
Use err(3). Correct incomplete man page.
1997-07-16 06:49:49 +00:00
#include <err.h>
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
#include <grp.h>
#include <inttypes.h>
Use err(3). Correct incomplete man page.
1997-07-16 06:49:49 +00:00
#include <locale.h>
Use getnameinfo(3) instead of inet_ntop(3) to make printable versions of sockaddr_in6 structures. getnameinfo(3) does the same thing, but it is also able to represent a scope zone id as described in the RFC 4007. MFC after: 2 weeks
2013-06-07 08:06:48 +00:00
#include <netdb.h>
MFp4 @229086: Make use of Capsicum to protect kdump(1), as it might be used to parse data from untrusted sources: - Sandbox kdump(1) using capability mode. - Limit stdin descriptor (where opened file is moved to) to only CAP_READ and CAP_FSTAT rights. - Limit stdout descriptor to only CAP_WRITE, CAP_FSTAT and CAP_IOCTL. Plus limit allowed ioctls to TIOCGETA only, which is needed for isatty() to work. - Limit stderr descriptor to only CAP_WRITE and CAP_FSTAT. In addition if the -s option is not given, grant CAP_IOCTL right, but allow for TIOCGWINSZ ioctl only, as we need screen width to dump the data. - Before entering capability mode call catopen("libc", NL_CAT_LOCALE), which opens message catalogs and caches data, so that strerror(3) and strsignal(3) can work in a sandbox. Sponsored by: The FreeBSD Foundation Discussed with: rwatson
2013-05-28 21:25:28 +00:00
#include <nl_types.h>
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
#include <pwd.h>
Handle 64-bit system call arguments (off_t, id_t). In particular, 64-bit system call arguments use up two register_t arguments for 32-bit processes. They must also be aligned on a 64-bit boundary on 32-bit powerpc processes. This fixes the decoding of lseek(), procctl(), and wait6() arguments for 32-bit processes (both native and via freebsd32). Note that the ktrace system call return record only returns a single register, so the return value of lseek is always truncated to the low 32-bits for 32-bit processes.
2016-10-01 22:01:41 +00:00
#include <stddef.h>
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
Start on a new library (libsysdecode) that provides routines for decoding system call information such as system call arguments. Initially this will consist of pulling duplicated code out of truss and kdump though it may prove useful for other utilities in the future. This commit moves the shared utrace(2) record parser out of kdump into the library and updates kdump and truss to use it. One difference from the previous version is that the library version treats unknown events that start with the "RTLD" signature as unknown events. This simplifies the interface and allows the consumer to decide how to handle all non-recognized events. Instead, this function only generates a string description for known malloc() and RTLD records. Reviewed by: bdrewery Differential Revision: https://reviews.freebsd.org/D4537
2015-12-15 00:05:07 +00:00
#include <sysdecode.h>
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
#include <time.h>
Use err(3). Correct incomplete man page.
1997-07-16 06:49:49 +00:00
#include <unistd.h>
#include <vis.h>
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
#include "ktrace.h"
We use a few different ifdef's names to check if we are using Casper or not, let's standardize this. Now we are always use WITH_CASPER name. Discussed with: emaste@ MFC after: 1 month
2017-09-21 14:41:41 +00:00
#ifdef WITH_CASPER
Convert casperd(8) daemon to the libcasper. After calling the cap_init(3) function Casper will fork from it's original process, using pdfork(2). Forking from a process has a lot of advantages: 1. We have the same cwd as the original process. 2. The same uid, gid and groups. 3. The same MAC labels. 4. The same descriptor table. 5. The same routing table. 6. The same umask. 7. The same cpuset(1). From now services are also in form of libraries. We also removed libcapsicum at all and converts existing program using Casper to new architecture. Discussed with: pjd, jonathan, ed, drysdale@google.com, emaste Partially reviewed by: drysdale@google.com, bdrewery Approved by: pjd (mentor) Differential Revision: https://reviews.freebsd.org/D4277
2016-02-25 18:23:40 +00:00
#include <libcasper.h>
#include <casper/cap_grp.h>
#include <casper/cap_pwd.h>
#endif
Teach kdump to understand sv_flags records in the trace files. MFC after: 1 Month.
2011-02-25 22:07:23 +00:00
int fetchprocinfo(struct ktr_header *, u_int *);
Remove spurious ABI tags from kdump output. The abidump routine output an ABI tag when -A was specified for records that were not displayed due to type or pid filtering. To fix, split the code to lookup the ABI from the code to display the ABI, move the code to display the ABI into dumpheader(), and move dumpheader() later in the main loop as a simplification. Previously dumpheader() was called under a condition that repeated conditions made later in the main loop. Reviewed by: kib MFC after: 1 month Sponsored by: DARPA / AFRL Differential Revision: https://reviews.freebsd.org/D16608
2018-08-07 00:10:58 +00:00
u_int findabi(struct ktr_header *);
Improve WARNS situation for kdump: 1) Define _KERNEL while including sys/time.h to get some function prototypes. 2) Add prototypes and ANSIify definitions. 3) Constness changes. 4) Remove register keyword. 5) Actually return a sensible value from main. 6) Make fread_tail take a void * instead of a char *. 7) Avoid a signedness warning by casting to a size_t. Should be safe enough 'cos we also check for nonnegativity. 8) Be extra chummy with sigset_t rather than passing a struct to printf and pretending it is an int.
2002-07-28 16:05:07 +00:00
int fread_tail(void *, int, int);
Remove spurious ABI tags from kdump output. The abidump routine output an ABI tag when -A was specified for records that were not displayed due to type or pid filtering. To fix, split the code to lookup the ABI from the code to display the ABI, move the code to display the ABI into dumpheader(), and move dumpheader() later in the main loop as a simplification. Previously dumpheader() was called under a condition that repeated conditions made later in the main loop. Reviewed by: kib MFC after: 1 month Sponsored by: DARPA / AFRL Differential Revision: https://reviews.freebsd.org/D16608
2018-08-07 00:10:58 +00:00
void dumpheader(struct ktr_header *, u_int);
Teach kdump to understand sv_flags records in the trace files. MFC after: 1 Month.
2011-02-25 22:07:23 +00:00
void ktrsyscall(struct ktr_syscall *, u_int);
void ktrsysret(struct ktr_sysret *, u_int);
Improve WARNS situation for kdump: 1) Define _KERNEL while including sys/time.h to get some function prototypes. 2) Add prototypes and ANSIify definitions. 3) Constness changes. 4) Remove register keyword. 5) Actually return a sensible value from main. 6) Make fread_tail take a void * instead of a char *. 7) Avoid a signedness warning by casting to a size_t. Should be safe enough 'cos we also check for nonnegativity. 8) Be extra chummy with sigset_t rather than passing a struct to printf and pretending it is an int.
2002-07-28 16:05:07 +00:00
void ktrnamei(char *, int);
Two enhancements for kdump. 1) add a "-p pid", which is rather useful for selecting a single pid in a combined trace file (eg: with ktrace -i). 2) display binary genio data in a more precise format.
2003-06-03 01:44:43 +00:00
void hexdump(char *, int, int);
void visdump(char *, int, int);
Improve WARNS situation for kdump: 1) Define _KERNEL while including sys/time.h to get some function prototypes. 2) Add prototypes and ANSIify definitions. 3) Constness changes. 4) Remove register keyword. 5) Actually return a sensible value from main. 6) Make fread_tail take a void * instead of a char *. 7) Avoid a signedness warning by casting to a size_t. Should be safe enough 'cos we also check for nonnegativity. 8) Be extra chummy with sigset_t rather than passing a struct to printf and pretending it is an int.
2002-07-28 16:05:07 +00:00
void ktrgenio(struct ktr_genio *, int);
Teach kdump to decode linux syscalls names too. Fix bug introduced in my previous commit: the kernel always dump native signal numbers, so no need to check the ABI in ktrpsig(). Suggested by: jhb MFC after: 1 Month.
2011-03-01 16:42:28 +00:00
void ktrpsig(struct ktr_psig *);
Improve WARNS situation for kdump: 1) Define _KERNEL while including sys/time.h to get some function prototypes. 2) Add prototypes and ANSIify definitions. 3) Constness changes. 4) Remove register keyword. 5) Actually return a sensible value from main. 6) Make fread_tail take a void * instead of a char *. 7) Avoid a signedness warning by casting to a size_t. Should be safe enough 'cos we also check for nonnegativity. 8) Be extra chummy with sigset_t rather than passing a struct to printf and pretending it is an int.
2002-07-28 16:05:07 +00:00
void ktrcsw(struct ktr_csw *);
Include the associated wait channel message for context switch ktrace records. kdump supports both the old and new messages. Submitted by: Andrey Zonov andrey zonov org MFC after: 1 week
2012-04-20 15:32:36 +00:00
void ktrcsw_old(struct ktr_csw_old *);
Fix most of the warnings in kdump(1). Sponsored by: Norse
2014-10-13 16:17:42 +00:00
void ktruser(int, void *);
Change the cap_rights_t type from uint64_t to a structure that we can extend in the future in a backward compatible (API and ABI) way. The cap_rights_t represents capability rights. We used to use one bit to represent one right, but we are running out of spare bits. Currently the new structure provides place for 114 rights (so 50 more than the previous cap_rights_t), but it is possible to grow the structure to hold at least 285 rights, although we can make it even larger if 285 rights won't be enough. The structure definition looks like this: struct cap_rights { uint64_t cr_rights[CAP_RIGHTS_VERSION + 2]; }; The initial CAP_RIGHTS_VERSION is 0. The top two bits in the first element of the cr_rights[] array contain total number of elements in the array - 2. This means if those two bits are equal to 0, we have 2 array elements. The top two bits in all remaining array elements should be 0. The next five bits in all array elements contain array index. Only one bit is used and bit position in this five-bits range defines array index. This means there can be at most five array elements in the future. To define new right the CAPRIGHT() macro must be used. The macro takes two arguments - an array index and a bit to set, eg. #define CAP_PDKILL CAPRIGHT(1, 0x0000000000000800ULL) We still support aliases that combine few rights, but the rights have to belong to the same array element, eg: #define CAP_LOOKUP CAPRIGHT(0, 0x0000000000000400ULL) #define CAP_FCHMOD CAPRIGHT(0, 0x0000000000002000ULL) #define CAP_FCHMODAT (CAP_FCHMOD | CAP_LOOKUP) There is new API to manage the new cap_rights_t structure: cap_rights_t *cap_rights_init(cap_rights_t *rights, ...); void cap_rights_set(cap_rights_t *rights, ...); void cap_rights_clear(cap_rights_t *rights, ...); bool cap_rights_is_set(const cap_rights_t *rights, ...); bool cap_rights_is_valid(const cap_rights_t *rights); void cap_rights_merge(cap_rights_t *dst, const cap_rights_t *src); void cap_rights_remove(cap_rights_t *dst, const cap_rights_t *src); bool cap_rights_contains(const cap_rights_t *big, const cap_rights_t *little); Capability rights to the cap_rights_init(), cap_rights_set(), cap_rights_clear() and cap_rights_is_set() functions are provided by separating them with commas, eg: cap_rights_t rights; cap_rights_init(&rights, CAP_READ, CAP_WRITE, CAP_FSTAT); There is no need to terminate the list of rights, as those functions are actually macros that take care of the termination, eg: #define cap_rights_set(rights, ...) \ __cap_rights_set((rights), __VA_ARGS__, 0ULL) void __cap_rights_set(cap_rights_t *rights, ...); Thanks to using one bit as an array index we can assert in those functions that there are no two rights belonging to different array elements provided together. For example this is illegal and will be detected, because CAP_LOOKUP belongs to element 0 and CAP_PDKILL to element 1: cap_rights_init(&rights, CAP_LOOKUP | CAP_PDKILL); Providing several rights that belongs to the same array's element this way is correct, but is not advised. It should only be used for aliases definition. This commit also breaks compatibility with some existing Capsicum system calls, but I see no other way to do that. This should be fine as Capsicum is still experimental and this change is not going to 9.x. Sponsored by: The FreeBSD Foundation
2013-09-05 00:09:56 +00:00
void ktrcaprights(cap_rights_t *);
Trace timeval parameters to the getitimer(2) and setitimer(2) syscalls. Reviewed by: jhb Tested by: pho Sponsored by: The FreeBSD Foundation MFC after: 1 week Differential revision: https://reviews.freebsd.org/D7158
2016-07-13 14:37:58 +00:00
void ktritimerval(struct itimerval *it);
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
void ktrsockaddr(struct sockaddr *);
void ktrstat(struct stat *);
void ktrstruct(char *, size_t);
Add a new trace point, KTRFAC_CAPFAIL, which traces capability check failures. It is included in the default set for ktrace(1) and kdump(1).
2011-10-11 20:37:10 +00:00
void ktrcapfail(struct ktr_cap_fail *);
Add new ktrace records for the start and end of VM faults. This gives a pair of records similar to syscall entry and return that a user can use to determine how long page faults take. The new ktrace records are enabled via the 'p' trace type, and are enabled in the default set of trace points. Reviewed by: kib MFC after: 2 weeks
2012-04-05 17:13:14 +00:00
void ktrfault(struct ktr_fault *);
void ktrfaultend(struct ktr_faultend *);
Decode kevent structures logged via ktrace(2) in kdump. - Add a new KTR_STRUCT_ARRAY ktrace record type which dumps an array of structures. The structure name in the record payload is preceded by a size_t containing the size of the individual structures. Use this to replace the previous code that dumped the kevent arrays dumped for kevent(). kdump is now able to decode the kevent structures rather than dumping their contents via a hexdump. One change from before is that the 'changes' and 'events' arrays are not marked with separate 'read' and 'write' annotations in kdump output. Instead, the first array is the 'changes' array, and the second array (only present if kevent doesn't fail with an error) is the 'events' array. For kevent(), empty arrays are denoted by an entry with an array containing zero entries rather than no record. - Move kevent decoding tables from truss to libsysdecode. This adds three new functions to decode members of struct kevent: sysdecode_kevent_filter, sysdecode_kevent_flags, and sysdecode_kevent_fflags. kdump uses these helper functions to pretty-print kevent fields. - Move structure definitions for freebsd11 and freebsd32 kevent structures to <sys/event.h> so that they can be shared with userland. The 32-bit structures are only exposed if _WANT_KEVENT32 is defined. The freebsd11 structures are only exposed if _WANT_FREEBSD11_KEVENT is defined. The 32-bit freebsd11 structure requires both. - Decode freebsd11 kevent structures in truss for the compat11.kevent() system call. - Log 32-bit kevent structures via ktrace for 32-bit compat kevent() system calls. - While here, constify the 'void *data' argument to ktrstruct(). Reviewed by: kib (earlier version) MFC after: 1 month Differential Revision: https://reviews.freebsd.org/D12470
2017-11-25 04:49:12 +00:00
void ktrkevent(struct kevent *);
void ktrstructarray(struct ktr_struct_array *, size_t);
Improve WARNS situation for kdump: 1) Define _KERNEL while including sys/time.h to get some function prototypes. 2) Add prototypes and ANSIify definitions. 3) Constness changes. 4) Remove register keyword. 5) Actually return a sensible value from main. 6) Make fread_tail take a void * instead of a char *. 7) Avoid a signedness warning by casting to a size_t. Should be safe enough 'cos we also check for nonnegativity. 8) Be extra chummy with sigset_t rather than passing a struct to printf and pretending it is an int.
2002-07-28 16:05:07 +00:00
void usage(void);
Make `-R', `-T' and `-E' options mutially non-exclusive. It is often useful to see two or three types at the same time when inspecting the dump. MFC after: 1 month Sponsored by: Sippy Software, Inc.
2014-03-25 23:37:57 +00:00
#define TIMESTAMP_NONE 0x0
#define TIMESTAMP_ABSOLUTE 0x1
#define TIMESTAMP_ELAPSED 0x2
#define TIMESTAMP_RELATIVE 0x4
Use 'bool' instead of 'int' for various boolean flags. Reviewed by: kib MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D16611
2018-08-18 20:23:53 +00:00
static bool abiflag, decimal, fancy = true, resolv, suppressdata, syscallno,
tail, threads;
static int timestamp, maxdata;
Fix most of the warnings in kdump(1). Sponsored by: Norse
2014-10-13 16:17:42 +00:00
static const char *tracefile = DEF_TRACEFILE;
static struct ktr_header ktr_header;
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
#define TIME_FORMAT "%b %e %T %Y"
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
#define eqs(s1, s2) (strcmp((s1), (s2)) == 0)
Handle 64-bit system call arguments (off_t, id_t). In particular, 64-bit system call arguments use up two register_t arguments for 32-bit processes. They must also be aligned on a 64-bit boundary on 32-bit powerpc processes. This fixes the decoding of lseek(), procctl(), and wait6() arguments for 32-bit processes (both native and via freebsd32). Note that the ktrace system call return record only returns a single register, so the return value of lseek is always truncated to the low 32-bits for 32-bit processes.
2016-10-01 22:01:41 +00:00
#define print_number64(first,i,n,c) do { \
uint64_t __v; \
\
if (quad_align && (((ptrdiff_t)((i) - (first))) & 1) == 1) { \
(i)++; \
(n)--; \
} \
if (quad_slots == 2) \
__v = (uint64_t)(uint32_t)(i)[0] | \
((uint64_t)(uint32_t)(i)[1]) << 32; \
else \
__v = (uint64_t)*(i); \
if (decimal) \
printf("%c%jd", (c), (intmax_t)__v); \
else \
printf("%c%#jx", (c), (uintmax_t)__v); \
(i) += quad_slots; \
(n) -= quad_slots; \
(c) = ','; \
} while (0)
The previous commit did not fix the issue since it did not prevent sign extension. Cast to u_register_t first, then to uintmax_t. Submitted by: bde@
2011-10-11 15:04:58 +00:00
#define print_number(i,n,c) do { \
if (decimal) \
printf("%c%jd", c, (intmax_t)*i); \
else \
printf("%c%#jx", c, (uintmax_t)(u_register_t)*i); \
i++; \
n--; \
c = ','; \
Address some of bde@'s concerns with the new code.
2011-10-08 21:02:08 +00:00
} while (0)
Teach kdump to decode linux syscalls names too. Fix bug introduced in my previous commit: the kernel always dump native signal numbers, so no need to check the ABI in ktrpsig(). Suggested by: jhb MFC after: 1 Month.
2011-03-01 16:42:28 +00:00
Teach kdump to understand sv_flags records in the trace files. MFC after: 1 Month.
2011-02-25 22:07:23 +00:00
struct proc_info
{
TAILQ_ENTRY(proc_info) info;
u_int sv_flags;
pid_t pid;
};
Fix most of the warnings in kdump(1). Sponsored by: Norse
2014-10-13 16:17:42 +00:00
static TAILQ_HEAD(trace_procs, proc_info) trace_procs;
Teach kdump to understand sv_flags records in the trace files. MFC after: 1 Month.
2011-02-25 22:07:23 +00:00
We use a few different ifdef's names to check if we are using Casper or not, let's standardize this. Now we are always use WITH_CASPER name. Discussed with: emaste@ MFC after: 1 month
2017-09-21 14:41:41 +00:00
#ifdef WITH_CASPER
Make use of Casper's system.pwd and system.grp services when the -r option is given to convert uids and gids to user names and group names even when running in capability mode sandbox. While here log on stderr when we successfully enter the sandbox. Sponsored by: The FreeBSD Foundation
2013-12-15 23:09:05 +00:00
static cap_channel_t *cappwd, *capgrp;
- Make localtime(3) to work in sandbox. - Move strerror(3) initialization to its own function.
2013-07-18 21:56:10 +00:00
Make use of Casper's system.pwd and system.grp services when the -r option is given to convert uids and gids to user names and group names even when running in capability mode sandbox. While here log on stderr when we successfully enter the sandbox. Sponsored by: The FreeBSD Foundation
2013-12-15 23:09:05 +00:00
static int
cappwdgrp_setup(cap_channel_t **cappwdp, cap_channel_t **capgrpp)
{
cap_channel_t *capcas, *cappwdloc, *capgrploc;
const char *cmds[1], *fields[1];
capcas = cap_init();
if (capcas == NULL) {
Convert casperd(8) daemon to the libcasper. After calling the cap_init(3) function Casper will fork from it's original process, using pdfork(2). Forking from a process has a lot of advantages: 1. We have the same cwd as the original process. 2. The same uid, gid and groups. 3. The same MAC labels. 4. The same descriptor table. 5. The same routing table. 6. The same umask. 7. The same cpuset(1). From now services are also in form of libraries. We also removed libcapsicum at all and converts existing program using Casper to new architecture. Discussed with: pjd, jonathan, ed, drysdale@google.com, emaste Partially reviewed by: drysdale@google.com, bdrewery Approved by: pjd (mentor) Differential Revision: https://reviews.freebsd.org/D4277
2016-02-25 18:23:40 +00:00
err(1, "unable to create casper process");
exit(1);
Make use of Casper's system.pwd and system.grp services when the -r option is given to convert uids and gids to user names and group names even when running in capability mode sandbox. While here log on stderr when we successfully enter the sandbox. Sponsored by: The FreeBSD Foundation
2013-12-15 23:09:05 +00:00
}
cappwdloc = cap_service_open(capcas, "system.pwd");
capgrploc = cap_service_open(capcas, "system.grp");
/* Casper capability no longer needed. */
cap_close(capcas);
if (cappwdloc == NULL || capgrploc == NULL) {
if (cappwdloc == NULL)
warn("unable to open system.pwd service");
if (capgrploc == NULL)
warn("unable to open system.grp service");
If we cannot connect to casperd we don't enter sandbox, but if we can connect to casperd, but we cannot access the service we need we exit with an error. This should not happen and just indicates some configuration error which should be fixed, so we force the user to do it by failing. Discussed with: emaste
2013-12-19 00:51:48 +00:00
exit(1);
Make use of Casper's system.pwd and system.grp services when the -r option is given to convert uids and gids to user names and group names even when running in capability mode sandbox. While here log on stderr when we successfully enter the sandbox. Sponsored by: The FreeBSD Foundation
2013-12-15 23:09:05 +00:00
}
/* Limit system.pwd to only getpwuid() function and pw_name field. */
cmds[0] = "getpwuid";
If we cannot connect to casperd we don't enter sandbox, but if we can connect to casperd, but we cannot access the service we need we exit with an error. This should not happen and just indicates some configuration error which should be fixed, so we force the user to do it by failing. Discussed with: emaste
2013-12-19 00:51:48 +00:00
if (cap_pwd_limit_cmds(cappwdloc, cmds, 1) < 0)
err(1, "unable to limit system.pwd service");
Make use of Casper's system.pwd and system.grp services when the -r option is given to convert uids and gids to user names and group names even when running in capability mode sandbox. While here log on stderr when we successfully enter the sandbox. Sponsored by: The FreeBSD Foundation
2013-12-15 23:09:05 +00:00
fields[0] = "pw_name";
If we cannot connect to casperd we don't enter sandbox, but if we can connect to casperd, but we cannot access the service we need we exit with an error. This should not happen and just indicates some configuration error which should be fixed, so we force the user to do it by failing. Discussed with: emaste
2013-12-19 00:51:48 +00:00
if (cap_pwd_limit_fields(cappwdloc, fields, 1) < 0)
err(1, "unable to limit system.pwd service");
Make use of Casper's system.pwd and system.grp services when the -r option is given to convert uids and gids to user names and group names even when running in capability mode sandbox. While here log on stderr when we successfully enter the sandbox. Sponsored by: The FreeBSD Foundation
2013-12-15 23:09:05 +00:00
/* Limit system.grp to only getgrgid() function and gr_name field. */
cmds[0] = "getgrgid";
If we cannot connect to casperd we don't enter sandbox, but if we can connect to casperd, but we cannot access the service we need we exit with an error. This should not happen and just indicates some configuration error which should be fixed, so we force the user to do it by failing. Discussed with: emaste
2013-12-19 00:51:48 +00:00
if (cap_grp_limit_cmds(capgrploc, cmds, 1) < 0)
err(1, "unable to limit system.grp service");
Make use of Casper's system.pwd and system.grp services when the -r option is given to convert uids and gids to user names and group names even when running in capability mode sandbox. While here log on stderr when we successfully enter the sandbox. Sponsored by: The FreeBSD Foundation
2013-12-15 23:09:05 +00:00
fields[0] = "gr_name";
If we cannot connect to casperd we don't enter sandbox, but if we can connect to casperd, but we cannot access the service we need we exit with an error. This should not happen and just indicates some configuration error which should be fixed, so we force the user to do it by failing. Discussed with: emaste
2013-12-19 00:51:48 +00:00
if (cap_grp_limit_fields(capgrploc, fields, 1) < 0)
err(1, "unable to limit system.grp service");
Make use of Casper's system.pwd and system.grp services when the -r option is given to convert uids and gids to user names and group names even when running in capability mode sandbox. While here log on stderr when we successfully enter the sandbox. Sponsored by: The FreeBSD Foundation
2013-12-15 23:09:05 +00:00
*cappwdp = cappwdloc;
*capgrpp = capgrploc;
return (0);
}
We use a few different ifdef's names to check if we are using Casper or not, let's standardize this. Now we are always use WITH_CASPER name. Discussed with: emaste@ MFC after: 1 month
2017-09-21 14:41:41 +00:00
#endif /* WITH_CASPER */
Make use of Casper's system.pwd and system.grp services when the -r option is given to convert uids and gids to user names and group names even when running in capability mode sandbox. While here log on stderr when we successfully enter the sandbox. Sponsored by: The FreeBSD Foundation
2013-12-15 23:09:05 +00:00
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
static void
print_integer_arg(const char *(*decoder)(int), int value)
{
const char *str;
str = decoder(value);
if (str != NULL)
printf("%s", str);
else {
if (decimal)
printf("<invalid=%d>", value);
else
printf("<invalid=%#x>", value);
}
}
/* Like print_integer_arg but unknown values are treated as valid. */
static void
print_integer_arg_valid(const char *(*decoder)(int), int value)
{
const char *str;
str = decoder(value);
if (str != NULL)
printf("%s", str);
else {
if (decimal)
printf("%d", value);
else
printf("%#x", value);
}
}
kdump/truss: decode new _umtx_op flags In both cases, print the flag bits first followed by the command. Output now looks something like this: (ktrace) _umtx_op(0x8605f7008,0xf<UMTX_OP_WAIT_UINT_PRIVATE>,0,0,0) _umtx_op(0x9fffdce8,0x80000003<UMTX_OP__32BIT|UMTX_OP_WAKE>,0x1,0,0) (truss) _umtx_op(0x7fffffffda50,UMTX_OP_WAKE,0x1,0x0,0x0) = 0 (0x0) _umtx_op(0x9fffdd08,UMTX_OP__32BIT|UMTX_OP_WAKE,0x1,0x0,0x0) = 0 (0x0) Reviewed by: kib Differential Revision: https://reviews.freebsd.org/D27325
2020-12-09 03:24:09 +00:00
static bool
print_mask_arg_part(bool (*decoder)(FILE *, int, int *), int value, int *rem)
{
printf("%#x<", value);
return (decoder(stdout, value, rem));
}
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
static void
print_mask_arg(bool (*decoder)(FILE *, int, int *), int value)
{
bool invalid;
int rem;
kdump/truss: decode new _umtx_op flags In both cases, print the flag bits first followed by the command. Output now looks something like this: (ktrace) _umtx_op(0x8605f7008,0xf<UMTX_OP_WAIT_UINT_PRIVATE>,0,0,0) _umtx_op(0x9fffdce8,0x80000003<UMTX_OP__32BIT|UMTX_OP_WAKE>,0x1,0,0) (truss) _umtx_op(0x7fffffffda50,UMTX_OP_WAKE,0x1,0x0,0x0) = 0 (0x0) _umtx_op(0x9fffdd08,UMTX_OP__32BIT|UMTX_OP_WAKE,0x1,0x0,0x0) = 0 (0x0) Reviewed by: kib Differential Revision: https://reviews.freebsd.org/D27325
2020-12-09 03:24:09 +00:00
invalid = !print_mask_arg_part(decoder, value, &rem);
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
printf(">");
if (invalid)
printf("<invalid>%u", rem);
}
static void
print_mask_arg0(bool (*decoder)(FILE *, int, int *), int value)
{
bool invalid;
int rem;
if (value == 0) {
printf("0");
return;
}
printf("%#x<", value);
invalid = !decoder(stdout, value, &rem);
printf(">");
if (invalid)
printf("<invalid>%u", rem);
}
static void
decode_fileflags(fflags_t value)
{
bool invalid;
fflags_t rem;
if (value == 0) {
printf("0");
return;
}
printf("%#x<", value);
invalid = !sysdecode_fileflags(stdout, value, &rem);
printf(">");
if (invalid)
printf("<invalid>%u", rem);
}
static void
decode_filemode(int value)
{
bool invalid;
int rem;
if (value == 0) {
printf("0");
return;
}
printf("%#o<", value);
invalid = !sysdecode_filemode(stdout, value, &rem);
printf(">");
if (invalid)
printf("<invalid>%u", rem);
}
static void
print_mask_arg32(bool (*decoder)(FILE *, uint32_t, uint32_t *), uint32_t value)
{
bool invalid;
uint32_t rem;
printf("%#x<", value);
invalid = !decoder(stdout, value, &rem);
printf(">");
if (invalid)
printf("<invalid>%u", rem);
}
static void
print_mask_argul(bool (*decoder)(FILE *, u_long, u_long *), u_long value)
{
bool invalid;
u_long rem;
if (value == 0) {
printf("0");
return;
}
printf("%#lx<", value);
invalid = !decoder(stdout, value, &rem);
printf(">");
if (invalid)
printf("<invalid>%lu", rem);
}
Improve WARNS situation for kdump: 1) Define _KERNEL while including sys/time.h to get some function prototypes. 2) Add prototypes and ANSIify definitions. 3) Constness changes. 4) Remove register keyword. 5) Actually return a sensible value from main. 6) Make fread_tail take a void * instead of a char *. 7) Avoid a signedness warning by casting to a size_t. Should be safe enough 'cos we also check for nonnegativity. 8) Be extra chummy with sigset_t rather than passing a struct to printf and pretending it is an int.
2002-07-28 16:05:07 +00:00
int
main(int argc, char *argv[])
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
{
int ch, ktrlen, size;
Improve WARNS situation for kdump: 1) Define _KERNEL while including sys/time.h to get some function prototypes. 2) Add prototypes and ANSIify definitions. 3) Constness changes. 4) Remove register keyword. 5) Actually return a sensible value from main. 6) Make fread_tail take a void * instead of a char *. 7) Avoid a signedness warning by casting to a size_t. Should be safe enough 'cos we also check for nonnegativity. 8) Be extra chummy with sigset_t rather than passing a struct to printf and pretending it is an int.
2002-07-28 16:05:07 +00:00
void *m;
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
int trpoints = ALL_POINTS;
Teach kdump(8) to handle events marked with KTR_DROP. If a record has KTR_DROP set in its header, then we output an extra line to stdout to indicate that events were dropped between the previous record and this record. It is a bit trickier because we need to always notify the user if events are dropped even if KTR_DROP is set on a record of a type that we aren't interested in since kdump(8) doesn't know if the dropped events were of the types that the user has requested. To avoid outputting multiple events dropped notices in between actual event logs, a state variable is set whenever a drop is logged and cleared whenever an actual record is output. Requested by: phk
2003-03-13 18:46:35 +00:00
int drop_logged;
Two enhancements for kdump. 1) add a "-p pid", which is rather useful for selecting a single pid in a combined trace file (eg: with ktrace -i). 2) display binary genio data in a more precise format.
2003-06-03 01:44:43 +00:00
pid_t pid = 0;
Teach kdump to understand sv_flags records in the trace files. MFC after: 1 Month.
2011-02-25 22:07:23 +00:00
u_int sv_flags;
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
setlocale(LC_CTYPE, "");
Add setlocale LC_CTYPE
1995-10-26 22:16:45 +00:00
Make `-R', `-T' and `-E' options mutially non-exclusive. It is often useful to see two or three types at the same time when inspecting the dump. MFC after: 1 month Sponsored by: Sippy Software, Inc.
2014-03-25 23:37:57 +00:00
timestamp = TIMESTAMP_NONE;
Add `-S' to display syscall numbers in the output as well. This is useful for debugging compat modules. Sponsored by: EMC / Isilon Storage Division Obtained from: Isilon OneFS (based on work by Jeff Hughes) MFC after: 2 weeks
2014-03-28 16:11:20 +00:00
while ((ch = getopt(argc,argv,"f:dElm:np:AHRrSsTt:")) != -1)
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
switch (ch) {
Teach kdump to understand sv_flags records in the trace files. MFC after: 1 Month.
2011-02-25 22:07:23 +00:00
case 'A':
Use 'bool' instead of 'int' for various boolean flags. Reviewed by: kib MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D16611
2018-08-18 20:23:53 +00:00
abiflag = true;
Teach kdump to understand sv_flags records in the trace files. MFC after: 1 Month.
2011-02-25 22:07:23 +00:00
break;
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
case 'f':
tracefile = optarg;
break;
case 'd':
Use 'bool' instead of 'int' for various boolean flags. Reviewed by: kib MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D16611
2018-08-18 20:23:53 +00:00
decimal = true;
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
break;
case 'l':
Use 'bool' instead of 'int' for various boolean flags. Reviewed by: kib MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D16611
2018-08-18 20:23:53 +00:00
tail = true;
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
break;
case 'm':
maxdata = atoi(optarg);
break;
case 'n':
Use 'bool' instead of 'int' for various boolean flags. Reviewed by: kib MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D16611
2018-08-18 20:23:53 +00:00
fancy = false;
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
break;
Two enhancements for kdump. 1) add a "-p pid", which is rather useful for selecting a single pid in a combined trace file (eg: with ktrace -i). 2) display binary genio data in a more precise format.
2003-06-03 01:44:43 +00:00
case 'p':
pid = atoi(optarg);
break;
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
case 'r':
Use 'bool' instead of 'int' for various boolean flags. Reviewed by: kib MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D16611
2018-08-18 20:23:53 +00:00
resolv = true;
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
break;
Add `-S' to display syscall numbers in the output as well. This is useful for debugging compat modules. Sponsored by: EMC / Isilon Storage Division Obtained from: Isilon OneFS (based on work by Jeff Hughes) MFC after: 2 weeks
2014-03-28 16:11:20 +00:00
case 'S':
Use 'bool' instead of 'int' for various boolean flags. Reviewed by: kib MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D16611
2018-08-18 20:23:53 +00:00
syscallno = true;
Add `-S' to display syscall numbers in the output as well. This is useful for debugging compat modules. Sponsored by: EMC / Isilon Storage Division Obtained from: Isilon OneFS (based on work by Jeff Hughes) MFC after: 2 weeks
2014-03-28 16:11:20 +00:00
break;
Add "-s" argument to kdump to suppress the display of I/O data. MFC after: 1 week
2005-11-12 14:21:48 +00:00
case 's':
Use 'bool' instead of 'int' for various boolean flags. Reviewed by: kib MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D16611
2018-08-18 20:23:53 +00:00
suppressdata = true;
Add "-s" argument to kdump to suppress the display of I/O data. MFC after: 1 week
2005-11-12 14:21:48 +00:00
break;
Add a -E (elapsed time) flag to kdump. This is like -T, except it is relative to start of the dump. Approved by: re (scottl)
2003-12-07 01:06:32 +00:00
case 'E':
Make `-R', `-T' and `-E' options mutially non-exclusive. It is often useful to see two or three types at the same time when inspecting the dump. MFC after: 1 month Sponsored by: Sippy Software, Inc.
2014-03-25 23:37:57 +00:00
timestamp |= TIMESTAMP_ELAPSED;
Add a -E (elapsed time) flag to kdump. This is like -T, except it is relative to start of the dump. Approved by: re (scottl)
2003-12-07 01:06:32 +00:00
break;
Add a "-H" argument to kdump, which causes kdump to print an additional field holding the threadid. This is more useful for libthr than libpthread, but still quite useful in libpthread as it can be used to process interlaced records from multiple threads over the course of a system call. Detect old ktr_buffer values using the heuristic "if it's negative, then it must not be a valid threadid". This may leave something to be desired. MFC after: 1 month Reviewed by: davidxu
2005-11-01 14:48:23 +00:00
case 'H':
Use 'bool' instead of 'int' for various boolean flags. Reviewed by: kib MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D16611
2018-08-18 20:23:53 +00:00
threads = true;
Add a "-H" argument to kdump, which causes kdump to print an additional field holding the threadid. This is more useful for libthr than libpthread, but still quite useful in libpthread as it can be used to process interlaced records from multiple threads over the course of a system call. Detect old ktr_buffer values using the heuristic "if it's negative, then it must not be a valid threadid". This may leave something to be desired. MFC after: 1 month Reviewed by: davidxu
2005-11-01 14:48:23 +00:00
break;
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
case 'R':
Make `-R', `-T' and `-E' options mutially non-exclusive. It is often useful to see two or three types at the same time when inspecting the dump. MFC after: 1 month Sponsored by: Sippy Software, Inc.
2014-03-25 23:37:57 +00:00
timestamp |= TIMESTAMP_RELATIVE;
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
break;
case 'T':
Make `-R', `-T' and `-E' options mutially non-exclusive. It is often useful to see two or three types at the same time when inspecting the dump. MFC after: 1 month Sponsored by: Sippy Software, Inc.
2014-03-25 23:37:57 +00:00
timestamp |= TIMESTAMP_ABSOLUTE;
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
break;
case 't':
trpoints = getpoints(optarg);
Use err(3). Correct incomplete man page.
1997-07-16 06:49:49 +00:00
if (trpoints < 0)
errx(1, "unknown trace point in %s", optarg);
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
break;
default:
usage();
}
Fix arg parsing. kdump used to allow a single argument, which it silently ignored.
1996-11-18 19:37:52 +00:00
if (argc > optind)
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
usage();
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
m = malloc(size = 1025);
Use err(3). Correct incomplete man page.
1997-07-16 06:49:49 +00:00
if (m == NULL)
errx(1, "%s", strerror(ENOMEM));
Use != 0 to be clear
2016-05-17 00:27:18 +00:00
if (strcmp(tracefile, "-") != 0)
Allow -f - to read from stdin.
2016-05-17 00:08:43 +00:00
if (!freopen(tracefile, "r", stdin))
err(1, "%s", tracefile);
MFp4 @229086: Make use of Capsicum to protect kdump(1), as it might be used to parse data from untrusted sources: - Sandbox kdump(1) using capability mode. - Limit stdin descriptor (where opened file is moved to) to only CAP_READ and CAP_FSTAT rights. - Limit stdout descriptor to only CAP_WRITE, CAP_FSTAT and CAP_IOCTL. Plus limit allowed ioctls to TIOCGETA only, which is needed for isatty() to work. - Limit stderr descriptor to only CAP_WRITE and CAP_FSTAT. In addition if the -s option is not given, grant CAP_IOCTL right, but allow for TIOCGWINSZ ioctl only, as we need screen width to dump the data. - Before entering capability mode call catopen("libc", NL_CAT_LOCALE), which opens message catalogs and caches data, so that strerror(3) and strsignal(3) can work in a sandbox. Sponsored by: The FreeBSD Foundation Discussed with: rwatson
2013-05-28 21:25:28 +00:00
NLS data and tz can be cached using caspicum helpers. Using them we simplifies and unifies sandboxed code.
2018-01-03 07:43:04 +00:00
caph_cache_catpages();
caph_cache_tzdata();
We use a few different ifdef's names to check if we are using Casper or not, let's standardize this. Now we are always use WITH_CASPER name. Discussed with: emaste@ MFC after: 1 month
2017-09-21 14:41:41 +00:00
#ifdef WITH_CASPER
Use 'bool' instead of 'int' for various boolean flags. Reviewed by: kib MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D16611
2018-08-18 20:23:53 +00:00
if (resolv) {
Make use of Casper's system.pwd and system.grp services when the -r option is given to convert uids and gids to user names and group names even when running in capability mode sandbox. While here log on stderr when we successfully enter the sandbox. Sponsored by: The FreeBSD Foundation
2013-12-15 23:09:05 +00:00
if (cappwdgrp_setup(&cappwd, &capgrp) < 0) {
cappwd = NULL;
capgrp = NULL;
}
}
Use 'bool' instead of 'int' for various boolean flags. Reviewed by: kib MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D16611
2018-08-18 20:23:53 +00:00
if (!resolv || (cappwd != NULL && capgrp != NULL)) {
Convert `cap_enter() < 0 && errno != ENOSYS` to `caph_enter() < 0`. No functional change intended.
2018-06-19 23:43:14 +00:00
if (caph_enter() < 0)
Make use of Casper's system.pwd and system.grp services when the -r option is given to convert uids and gids to user names and group names even when running in capability mode sandbox. While here log on stderr when we successfully enter the sandbox. Sponsored by: The FreeBSD Foundation
2013-12-15 23:09:05 +00:00
err(1, "unable to enter capability mode");
}
#else
Use 'bool' instead of 'int' for various boolean flags. Reviewed by: kib MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D16611
2018-08-18 20:23:53 +00:00
if (!resolv) {
Convert `cap_enter() < 0 && errno != ENOSYS` to `caph_enter() < 0`. No functional change intended.
2018-06-19 23:43:14 +00:00
if (caph_enter() < 0)
If the -r option is given we cannot enter capability mode. The option tells kdump to convert numeric UIDs and GIDs into user and group names plus to convert times and dates into locallized versions. This all needs opening various files at various occasions.
2013-05-30 21:59:29 +00:00
err(1, "unable to enter capability mode");
}
Make use of Casper's system.pwd and system.grp services when the -r option is given to convert uids and gids to user names and group names even when running in capability mode sandbox. While here log on stderr when we successfully enter the sandbox. Sponsored by: The FreeBSD Foundation
2013-12-15 23:09:05 +00:00
#endif
Remove the duplicated code using Capsicum helpers. Reviewed by: cem, ed, bapt, emaste Differential Revision https://reviews.freebsd.org/D8140
2016-10-07 17:56:57 +00:00
if (caph_limit_stdio() == -1)
err(1, "unable to limit stdio");
MFp4 @229086: Make use of Capsicum to protect kdump(1), as it might be used to parse data from untrusted sources: - Sandbox kdump(1) using capability mode. - Limit stdin descriptor (where opened file is moved to) to only CAP_READ and CAP_FSTAT rights. - Limit stdout descriptor to only CAP_WRITE, CAP_FSTAT and CAP_IOCTL. Plus limit allowed ioctls to TIOCGETA only, which is needed for isatty() to work. - Limit stderr descriptor to only CAP_WRITE and CAP_FSTAT. In addition if the -s option is not given, grant CAP_IOCTL right, but allow for TIOCGWINSZ ioctl only, as we need screen width to dump the data. - Before entering capability mode call catopen("libc", NL_CAT_LOCALE), which opens message catalogs and caches data, so that strerror(3) and strsignal(3) can work in a sandbox. Sponsored by: The FreeBSD Foundation Discussed with: rwatson
2013-05-28 21:25:28 +00:00
Teach kdump to understand sv_flags records in the trace files. MFC after: 1 Month.
2011-02-25 22:07:23 +00:00
TAILQ_INIT(&trace_procs);
Teach kdump(8) to handle events marked with KTR_DROP. If a record has KTR_DROP set in its header, then we output an extra line to stdout to indicate that events were dropped between the previous record and this record. It is a bit trickier because we need to always notify the user if events are dropped even if KTR_DROP is set on a record of a type that we aren't interested in since kdump(8) doesn't know if the dropped events were of the types that the user has requested. To avoid outputting multiple events dropped notices in between actual event logs, a state variable is set whenever a drop is logged and cleared whenever an actual record is output. Requested by: phk
2003-03-13 18:46:35 +00:00
drop_logged = 0;
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
while (fread_tail(&ktr_header, sizeof(struct ktr_header), 1)) {
Teach kdump(8) to handle events marked with KTR_DROP. If a record has KTR_DROP set in its header, then we output an extra line to stdout to indicate that events were dropped between the previous record and this record. It is a bit trickier because we need to always notify the user if events are dropped even if KTR_DROP is set on a record of a type that we aren't interested in since kdump(8) doesn't know if the dropped events were of the types that the user has requested. To avoid outputting multiple events dropped notices in between actual event logs, a state variable is set whenever a drop is logged and cleared whenever an actual record is output. Requested by: phk
2003-03-13 18:46:35 +00:00
if (ktr_header.ktr_type & KTR_DROP) {
ktr_header.ktr_type &= ~KTR_DROP;
Add a "-H" argument to kdump, which causes kdump to print an additional field holding the threadid. This is more useful for libthr than libpthread, but still quite useful in libpthread as it can be used to process interlaced records from multiple threads over the course of a system call. Detect old ktr_buffer values using the heuristic "if it's negative, then it must not be a valid threadid". This may leave something to be desired. MFC after: 1 month Reviewed by: davidxu
2005-11-01 14:48:23 +00:00
if (!drop_logged && threads) {
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
printf(
- Cast intptr_t, pid_t and time_t values to intmax_t and use %jd with printf. - Cast the system call return value to long and use %ld in a printf in ktrsysret(). PR: bin/123774 MFC after: 2 weeks
2010-02-06 16:01:38 +00:00
"%6jd %6jd %-8.*s Events dropped.\n",
(intmax_t)ktr_header.ktr_pid,
ktr_header.ktr_tid > 0 ?
(intmax_t)ktr_header.ktr_tid : 0,
MAXCOMLEN, ktr_header.ktr_comm);
Add a "-H" argument to kdump, which causes kdump to print an additional field holding the threadid. This is more useful for libthr than libpthread, but still quite useful in libpthread as it can be used to process interlaced records from multiple threads over the course of a system call. Detect old ktr_buffer values using the heuristic "if it's negative, then it must not be a valid threadid". This may leave something to be desired. MFC after: 1 month Reviewed by: davidxu
2005-11-01 14:48:23 +00:00
drop_logged = 1;
} else if (!drop_logged) {
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
printf("%6jd %-8.*s Events dropped.\n",
- Cast intptr_t, pid_t and time_t values to intmax_t and use %jd with printf. - Cast the system call return value to long and use %ld in a printf in ktrsysret(). PR: bin/123774 MFC after: 2 weeks
2010-02-06 16:01:38 +00:00
(intmax_t)ktr_header.ktr_pid, MAXCOMLEN,
Teach kdump(8) to handle events marked with KTR_DROP. If a record has KTR_DROP set in its header, then we output an extra line to stdout to indicate that events were dropped between the previous record and this record. It is a bit trickier because we need to always notify the user if events are dropped even if KTR_DROP is set on a record of a type that we aren't interested in since kdump(8) doesn't know if the dropped events were of the types that the user has requested. To avoid outputting multiple events dropped notices in between actual event logs, a state variable is set whenever a drop is logged and cleared whenever an actual record is output. Requested by: phk
2003-03-13 18:46:35 +00:00
ktr_header.ktr_comm);
drop_logged = 1;
}
}
Use err(3). Correct incomplete man page.
1997-07-16 06:49:49 +00:00
if ((ktrlen = ktr_header.ktr_len) < 0)
errx(1, "bogus length 0x%x", ktrlen);
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
if (ktrlen > size) {
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
m = realloc(m, ktrlen+1);
Use err(3). Correct incomplete man page.
1997-07-16 06:49:49 +00:00
if (m == NULL)
errx(1, "%s", strerror(ENOMEM));
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
size = ktrlen;
}
Use err(3). Correct incomplete man page.
1997-07-16 06:49:49 +00:00
if (ktrlen && fread_tail(m, ktrlen, 1) == 0)
errx(1, "data too short");
Teach kdump to understand sv_flags records in the trace files. MFC after: 1 Month.
2011-02-25 22:07:23 +00:00
if (fetchprocinfo(&ktr_header, (u_int *)m) != 0)
continue;
Allow the -p argument to kdump to accept either a PID or a thread ID. Submitted by: Dmitry Banschikov d.banschikov hostcomm ru MFC after: 1 week
2012-06-04 19:09:14 +00:00
if (pid && ktr_header.ktr_pid != pid &&
ktr_header.ktr_tid != pid)
Two enhancements for kdump. 1) add a "-p pid", which is rather useful for selecting a single pid in a combined trace file (eg: with ktrace -i). 2) display binary genio data in a more precise format.
2003-06-03 01:44:43 +00:00
continue;
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
if ((trpoints & (1<<ktr_header.ktr_type)) == 0)
continue;
Remove spurious ABI tags from kdump output. The abidump routine output an ABI tag when -A was specified for records that were not displayed due to type or pid filtering. To fix, split the code to lookup the ABI from the code to display the ABI, move the code to display the ABI into dumpheader(), and move dumpheader() later in the main loop as a simplification. Previously dumpheader() was called under a condition that repeated conditions made later in the main loop. Reviewed by: kib MFC after: 1 month Sponsored by: DARPA / AFRL Differential Revision: https://reviews.freebsd.org/D16608
2018-08-07 00:10:58 +00:00
sv_flags = findabi(&ktr_header);
dumpheader(&ktr_header, sv_flags);
Teach kdump(8) to handle events marked with KTR_DROP. If a record has KTR_DROP set in its header, then we output an extra line to stdout to indicate that events were dropped between the previous record and this record. It is a bit trickier because we need to always notify the user if events are dropped even if KTR_DROP is set on a record of a type that we aren't interested in since kdump(8) doesn't know if the dropped events were of the types that the user has requested. To avoid outputting multiple events dropped notices in between actual event logs, a state variable is set whenever a drop is logged and cleared whenever an actual record is output. Requested by: phk
2003-03-13 18:46:35 +00:00
drop_logged = 0;
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
switch (ktr_header.ktr_type) {
case KTR_SYSCALL:
Add support to libsysdecode for decoding system call names. A new sysdecode_syscallname() function accepts a system call code and returns a string of the corresponding name (or NULL if the code is unknown). To support different process ABIs, the new function accepts a value from a new sysdecode_abi enum as its first argument to select the ABI in use. Current ABIs supported include FREEBSD (native binaries), FREEBSD32, LINUX, LINUX32, and CLOUDABI64. Note that not all ABIs are supported by all platforms. In general, a given ABI is only supported if a platform can execute binaries for that ABI. To simplify the implementation, libsysdecode's build reuses the existing pre-generated files from the kernel source tree rather than duplicating new copies of said files during the build. kdump(1) and truss(1) now use these functions to map system call identifiers to names. For kdump(1), a new 'syscallname()' function consolidates duplicated code from ktrsyscall() and ktrsyscallret(). The Linux ABI no longer requires custom handling for ktrsyscall() and linux_ktrsyscall() has been removed as a result. Reviewed by: bdrewery Differential Revision: https://reviews.freebsd.org/D4823
2016-01-26 19:07:09 +00:00
ktrsyscall((struct ktr_syscall *)m, sv_flags);
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
break;
case KTR_SYSRET:
Add handling for non-native error values to libsysdecode. Add two new functions, sysdecode_abi_to_freebsd_errno() and sysdecode_freebsd_to_abi_errno(), which convert errno values between the native FreeBSD ABI and other supported ABIs. Note that the mappings are not necessarily perfect meaning in some cases multiple errors in one ABI might map to a single error in another ABI. In that case, the reverse mapping will return one of the errors that maps, but which error is non-deterministic. Change truss to always report the raw error value to the user but use libsysdecode to map it to a native errno value that can be used with strerror() to generate a description. Previously truss reported the "converted" error value. Now the user will always see the exact error value that the application sees. Change kdump to report the truly raw error value to the user. Previously kdump would report the absolute value of the raw error value (so for Linux binaries it didn't output the FreeBSD error value, but the positive value of the Linux error). Now it reports the real (i.e. negative) error value for Linux binaries. Also, use libsysdecode to convert the native FreeBSD error reported in the ktrace record to the raw error used by the ABI. This means that the Linux ABI can now be handled directly in ktrsysret() and removes the need for linux_ktrsysret(). Reviewed by: bdrewery, kib Helpful notes: wblock (manpage) Differential Revision: https://reviews.freebsd.org/D5314
2016-02-23 20:00:55 +00:00
ktrsysret((struct ktr_sysret *)m, sv_flags);
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
break;
case KTR_NAMEI:
Add a new type of KTRACE record for sysctl(3) invocations. It uses the internal sysctl_sysctl_name() handler to map the MIB array to a string name and logs this name in the trace log. This can be useful to see exactly which sysctls a thread is invoking. MFC after: 1 month
2009-03-11 21:48:36 +00:00
case KTR_SYSCTL:
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
ktrnamei(m, ktrlen);
break;
case KTR_GENIO:
ktrgenio((struct ktr_genio *)m, ktrlen);
break;
case KTR_PSIG:
Teach kdump to decode linux syscalls names too. Fix bug introduced in my previous commit: the kernel always dump native signal numbers, so no need to check the ABI in ktrpsig(). Suggested by: jhb MFC after: 1 Month.
2011-03-01 16:42:28 +00:00
ktrpsig((struct ktr_psig *)m);
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
break;
case KTR_CSW:
Include the associated wait channel message for context switch ktrace records. kdump supports both the old and new messages. Submitted by: Andrey Zonov andrey zonov org MFC after: 1 week
2012-04-20 15:32:36 +00:00
if (ktrlen == sizeof(struct ktr_csw_old))
ktrcsw_old((struct ktr_csw_old *)m);
else
ktrcsw((struct ktr_csw *)m);
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
break;
For now we just hexdump the stuff in USER records.
1996-09-19 19:50:45 +00:00
case KTR_USER:
Remove length field from utrace entries.
1996-09-22 18:18:20 +00:00
ktruser(ktrlen, m);
For now we just hexdump the stuff in USER records.
1996-09-19 19:50:45 +00:00
break;
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
case KTR_STRUCT:
ktrstruct(m, ktrlen);
break;
Add a new trace point, KTRFAC_CAPFAIL, which traces capability check failures. It is included in the default set for ktrace(1) and kdump(1).
2011-10-11 20:37:10 +00:00
case KTR_CAPFAIL:
ktrcapfail((struct ktr_cap_fail *)m);
Add new ktrace records for the start and end of VM faults. This gives a pair of records similar to syscall entry and return that a user can use to determine how long page faults take. The new ktrace records are enabled via the 'p' trace type, and are enabled in the default set of trace points. Reviewed by: kib MFC after: 2 weeks
2012-04-05 17:13:14 +00:00
break;
case KTR_FAULT:
ktrfault((struct ktr_fault *)m);
break;
case KTR_FAULTEND:
ktrfaultend((struct ktr_faultend *)m);
break;
Decode kevent structures logged via ktrace(2) in kdump. - Add a new KTR_STRUCT_ARRAY ktrace record type which dumps an array of structures. The structure name in the record payload is preceded by a size_t containing the size of the individual structures. Use this to replace the previous code that dumped the kevent arrays dumped for kevent(). kdump is now able to decode the kevent structures rather than dumping their contents via a hexdump. One change from before is that the 'changes' and 'events' arrays are not marked with separate 'read' and 'write' annotations in kdump output. Instead, the first array is the 'changes' array, and the second array (only present if kevent doesn't fail with an error) is the 'events' array. For kevent(), empty arrays are denoted by an entry with an array containing zero entries rather than no record. - Move kevent decoding tables from truss to libsysdecode. This adds three new functions to decode members of struct kevent: sysdecode_kevent_filter, sysdecode_kevent_flags, and sysdecode_kevent_fflags. kdump uses these helper functions to pretty-print kevent fields. - Move structure definitions for freebsd11 and freebsd32 kevent structures to <sys/event.h> so that they can be shared with userland. The 32-bit structures are only exposed if _WANT_KEVENT32 is defined. The freebsd11 structures are only exposed if _WANT_FREEBSD11_KEVENT is defined. The 32-bit freebsd11 structure requires both. - Decode freebsd11 kevent structures in truss for the compat11.kevent() system call. - Log 32-bit kevent structures via ktrace for 32-bit compat kevent() system calls. - While here, constify the 'void *data' argument to ktrstruct(). Reviewed by: kib (earlier version) MFC after: 1 month Differential Revision: https://reviews.freebsd.org/D12470
2017-11-25 04:49:12 +00:00
case KTR_STRUCT_ARRAY:
ktrstructarray((struct ktr_struct_array *)m, ktrlen);
break;
Add a default case that just outputs a new line for the case of an unknown header type.
2003-03-13 18:58:39 +00:00
default:
printf("\n");
break;
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
}
if (tail)
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
fflush(stdout);
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
}
Improve WARNS situation for kdump: 1) Define _KERNEL while including sys/time.h to get some function prototypes. 2) Add prototypes and ANSIify definitions. 3) Constness changes. 4) Remove register keyword. 5) Actually return a sensible value from main. 6) Make fread_tail take a void * instead of a char *. 7) Avoid a signedness warning by casting to a size_t. Should be safe enough 'cos we also check for nonnegativity. 8) Be extra chummy with sigset_t rather than passing a struct to printf and pretending it is an int.
2002-07-28 16:05:07 +00:00
return 0;
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
}
Improve WARNS situation for kdump: 1) Define _KERNEL while including sys/time.h to get some function prototypes. 2) Add prototypes and ANSIify definitions. 3) Constness changes. 4) Remove register keyword. 5) Actually return a sensible value from main. 6) Make fread_tail take a void * instead of a char *. 7) Avoid a signedness warning by casting to a size_t. Should be safe enough 'cos we also check for nonnegativity. 8) Be extra chummy with sigset_t rather than passing a struct to printf and pretending it is an int.
2002-07-28 16:05:07 +00:00
int
fread_tail(void *buf, int size, int num)
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
{
int i;
while ((i = fread(buf, size, num, stdin)) == 0 && tail) {
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
sleep(1);
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
clearerr(stdin);
}
return (i);
}
Teach kdump to understand sv_flags records in the trace files. MFC after: 1 Month.
2011-02-25 22:07:23 +00:00
int
fetchprocinfo(struct ktr_header *kth, u_int *flags)
{
struct proc_info *pi;
switch (kth->ktr_type) {
case KTR_PROCCTOR:
TAILQ_FOREACH(pi, &trace_procs, info) {
if (pi->pid == kth->ktr_pid) {
TAILQ_REMOVE(&trace_procs, pi, info);
break;
}
}
pi = malloc(sizeof(struct proc_info));
if (pi == NULL)
errx(1, "%s", strerror(ENOMEM));
pi->sv_flags = *flags;
pi->pid = kth->ktr_pid;
TAILQ_INSERT_TAIL(&trace_procs, pi, info);
return (1);
case KTR_PROCDTOR:
TAILQ_FOREACH(pi, &trace_procs, info) {
if (pi->pid == kth->ktr_pid) {
TAILQ_REMOVE(&trace_procs, pi, info);
free(pi);
break;
}
}
return (1);
}
return (0);
}
u_int
Remove spurious ABI tags from kdump output. The abidump routine output an ABI tag when -A was specified for records that were not displayed due to type or pid filtering. To fix, split the code to lookup the ABI from the code to display the ABI, move the code to display the ABI into dumpheader(), and move dumpheader() later in the main loop as a simplification. Previously dumpheader() was called under a condition that repeated conditions made later in the main loop. Reviewed by: kib MFC after: 1 month Sponsored by: DARPA / AFRL Differential Revision: https://reviews.freebsd.org/D16608
2018-08-07 00:10:58 +00:00
findabi(struct ktr_header *kth)
Teach kdump to understand sv_flags records in the trace files. MFC after: 1 Month.
2011-02-25 22:07:23 +00:00
{
struct proc_info *pi;
TAILQ_FOREACH(pi, &trace_procs, info) {
if (pi->pid == kth->ktr_pid) {
Remove spurious ABI tags from kdump output. The abidump routine output an ABI tag when -A was specified for records that were not displayed due to type or pid filtering. To fix, split the code to lookup the ABI from the code to display the ABI, move the code to display the ABI into dumpheader(), and move dumpheader() later in the main loop as a simplification. Previously dumpheader() was called under a condition that repeated conditions made later in the main loop. Reviewed by: kib MFC after: 1 month Sponsored by: DARPA / AFRL Differential Revision: https://reviews.freebsd.org/D16608
2018-08-07 00:10:58 +00:00
return (pi->sv_flags);
Teach kdump to understand sv_flags records in the trace files. MFC after: 1 Month.
2011-02-25 22:07:23 +00:00
}
}
Remove spurious ABI tags from kdump output. The abidump routine output an ABI tag when -A was specified for records that were not displayed due to type or pid filtering. To fix, split the code to lookup the ABI from the code to display the ABI, move the code to display the ABI into dumpheader(), and move dumpheader() later in the main loop as a simplification. Previously dumpheader() was called under a condition that repeated conditions made later in the main loop. Reviewed by: kib MFC after: 1 month Sponsored by: DARPA / AFRL Differential Revision: https://reviews.freebsd.org/D16608
2018-08-07 00:10:58 +00:00
return (0);
Teach kdump to understand sv_flags records in the trace files. MFC after: 1 Month.
2011-02-25 22:07:23 +00:00
}
Improve WARNS situation for kdump: 1) Define _KERNEL while including sys/time.h to get some function prototypes. 2) Add prototypes and ANSIify definitions. 3) Constness changes. 4) Remove register keyword. 5) Actually return a sensible value from main. 6) Make fread_tail take a void * instead of a char *. 7) Avoid a signedness warning by casting to a size_t. Should be safe enough 'cos we also check for nonnegativity. 8) Be extra chummy with sigset_t rather than passing a struct to printf and pretending it is an int.
2002-07-28 16:05:07 +00:00
void
Remove spurious ABI tags from kdump output. The abidump routine output an ABI tag when -A was specified for records that were not displayed due to type or pid filtering. To fix, split the code to lookup the ABI from the code to display the ABI, move the code to display the ABI into dumpheader(), and move dumpheader() later in the main loop as a simplification. Previously dumpheader() was called under a condition that repeated conditions made later in the main loop. Reviewed by: kib MFC after: 1 month Sponsored by: DARPA / AFRL Differential Revision: https://reviews.freebsd.org/D16608
2018-08-07 00:10:58 +00:00
dumpheader(struct ktr_header *kth, u_int sv_flags)
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
{
static char unknown[64];
Don't declare the 'temp' timeval as static.
2016-10-01 22:17:40 +00:00
static struct timeval prevtime, prevtime_e;
struct timeval temp;
Remove spurious ABI tags from kdump output. The abidump routine output an ABI tag when -A was specified for records that were not displayed due to type or pid filtering. To fix, split the code to lookup the ABI from the code to display the ABI, move the code to display the ABI into dumpheader(), and move dumpheader() later in the main loop as a simplification. Previously dumpheader() was called under a condition that repeated conditions made later in the main loop. Reviewed by: kib MFC after: 1 month Sponsored by: DARPA / AFRL Differential Revision: https://reviews.freebsd.org/D16608
2018-08-07 00:10:58 +00:00
const char *abi;
const char *arch;
Improve WARNS situation for kdump: 1) Define _KERNEL while including sys/time.h to get some function prototypes. 2) Add prototypes and ANSIify definitions. 3) Constness changes. 4) Remove register keyword. 5) Actually return a sensible value from main. 6) Make fread_tail take a void * instead of a char *. 7) Avoid a signedness warning by casting to a size_t. Should be safe enough 'cos we also check for nonnegativity. 8) Be extra chummy with sigset_t rather than passing a struct to printf and pretending it is an int.
2002-07-28 16:05:07 +00:00
const char *type;
o Properly init prevtime, so that we don't print bogus value in the first entry reported by the relative mode (-R). o Properly print negative offsets, which I guess may happen if records get re-ordered somehow, possibly due to the locking. Right now we report huge bogus diff (i.e. 2 seconds or so).
2015-04-25 04:58:08 +00:00
const char *sign;
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
switch (kth->ktr_type) {
case KTR_SYSCALL:
type = "CALL";
break;
case KTR_SYSRET:
type = "RET ";
break;
case KTR_NAMEI:
type = "NAMI";
break;
case KTR_GENIO:
type = "GIO ";
break;
case KTR_PSIG:
type = "PSIG";
break;
case KTR_CSW:
Fix alignment of context switch traces. MFC after: 1 week Approved by: re (rwatson: "I like simple patches.")
2007-07-10 00:01:30 +00:00
type = "CSW ";
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
break;
For now we just hexdump the stuff in USER records.
1996-09-19 19:50:45 +00:00
case KTR_USER:
type = "USER";
break;
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
case KTR_STRUCT:
Decode kevent structures logged via ktrace(2) in kdump. - Add a new KTR_STRUCT_ARRAY ktrace record type which dumps an array of structures. The structure name in the record payload is preceded by a size_t containing the size of the individual structures. Use this to replace the previous code that dumped the kevent arrays dumped for kevent(). kdump is now able to decode the kevent structures rather than dumping their contents via a hexdump. One change from before is that the 'changes' and 'events' arrays are not marked with separate 'read' and 'write' annotations in kdump output. Instead, the first array is the 'changes' array, and the second array (only present if kevent doesn't fail with an error) is the 'events' array. For kevent(), empty arrays are denoted by an entry with an array containing zero entries rather than no record. - Move kevent decoding tables from truss to libsysdecode. This adds three new functions to decode members of struct kevent: sysdecode_kevent_filter, sysdecode_kevent_flags, and sysdecode_kevent_fflags. kdump uses these helper functions to pretty-print kevent fields. - Move structure definitions for freebsd11 and freebsd32 kevent structures to <sys/event.h> so that they can be shared with userland. The 32-bit structures are only exposed if _WANT_KEVENT32 is defined. The freebsd11 structures are only exposed if _WANT_FREEBSD11_KEVENT is defined. The 32-bit freebsd11 structure requires both. - Decode freebsd11 kevent structures in truss for the compat11.kevent() system call. - Log 32-bit kevent structures via ktrace for 32-bit compat kevent() system calls. - While here, constify the 'void *data' argument to ktrstruct(). Reviewed by: kib (earlier version) MFC after: 1 month Differential Revision: https://reviews.freebsd.org/D12470
2017-11-25 04:49:12 +00:00
case KTR_STRUCT_ARRAY:
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
type = "STRU";
break;
Add a new type of KTRACE record for sysctl(3) invocations. It uses the internal sysctl_sysctl_name() handler to map the MIB array to a string name and logs this name in the trace log. This can be useful to see exactly which sysctls a thread is invoking. MFC after: 1 month
2009-03-11 21:48:36 +00:00
case KTR_SYSCTL:
type = "SCTL";
break;
Add a new trace point, KTRFAC_CAPFAIL, which traces capability check failures. It is included in the default set for ktrace(1) and kdump(1).
2011-10-11 20:37:10 +00:00
case KTR_CAPFAIL:
type = "CAP ";
break;
Add new ktrace records for the start and end of VM faults. This gives a pair of records similar to syscall entry and return that a user can use to determine how long page faults take. The new ktrace records are enabled via the 'p' trace type, and are enabled in the default set of trace points. Reviewed by: kib MFC after: 2 weeks
2012-04-05 17:13:14 +00:00
case KTR_FAULT:
type = "PFLT";
break;
case KTR_FAULTEND:
type = "PRET";
break;
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
default:
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
sprintf(unknown, "UNKNOWN(%d)", kth->ktr_type);
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
type = unknown;
}
Add a "-H" argument to kdump, which causes kdump to print an additional field holding the threadid. This is more useful for libthr than libpthread, but still quite useful in libpthread as it can be used to process interlaced records from multiple threads over the course of a system call. Detect old ktr_buffer values using the heuristic "if it's negative, then it must not be a valid threadid". This may leave something to be desired. MFC after: 1 month Reviewed by: davidxu
2005-11-01 14:48:23 +00:00
/*
* The ktr_tid field was previously the ktr_buffer field, which held
* the kernel pointer value for the buffer associated with data
* following the record header. It now holds a threadid, but only
* for trace files after the change. Older trace files still contain
* kernel pointers. Detect this and suppress the results by printing
* negative tid's as 0.
*/
if (threads)
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
printf("%6jd %6jd %-8.*s ", (intmax_t)kth->ktr_pid,
- Cast intptr_t, pid_t and time_t values to intmax_t and use %jd with printf. - Cast the system call return value to long and use %ld in a printf in ktrsysret(). PR: bin/123774 MFC after: 2 weeks
2010-02-06 16:01:38 +00:00
kth->ktr_tid > 0 ? (intmax_t)kth->ktr_tid : 0,
MAXCOMLEN, kth->ktr_comm);
Add a "-H" argument to kdump, which causes kdump to print an additional field holding the threadid. This is more useful for libthr than libpthread, but still quite useful in libpthread as it can be used to process interlaced records from multiple threads over the course of a system call. Detect old ktr_buffer values using the heuristic "if it's negative, then it must not be a valid threadid". This may leave something to be desired. MFC after: 1 month Reviewed by: davidxu
2005-11-01 14:48:23 +00:00
else
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
printf("%6jd %-8.*s ", (intmax_t)kth->ktr_pid, MAXCOMLEN,
Add a "-H" argument to kdump, which causes kdump to print an additional field holding the threadid. This is more useful for libthr than libpthread, but still quite useful in libpthread as it can be used to process interlaced records from multiple threads over the course of a system call. Detect old ktr_buffer values using the heuristic "if it's negative, then it must not be a valid threadid". This may leave something to be desired. MFC after: 1 month Reviewed by: davidxu
2005-11-01 14:48:23 +00:00
kth->ktr_comm);
Make `-R', `-T' and `-E' options mutially non-exclusive. It is often useful to see two or three types at the same time when inspecting the dump. MFC after: 1 month Sponsored by: Sippy Software, Inc.
2014-03-25 23:37:57 +00:00
if (timestamp) {
if (timestamp & TIMESTAMP_ABSOLUTE) {
printf("%jd.%06ld ", (intmax_t)kth->ktr_time.tv_sec,
kth->ktr_time.tv_usec);
}
if (timestamp & TIMESTAMP_ELAPSED) {
if (prevtime_e.tv_sec == 0)
prevtime_e = kth->ktr_time;
Use timercmp() and timersub() in kdump. Previously, kdump used the kernel-only timervalsub() macro which required defining _KERNEL when including <sys/time.h>. Now, kdump uses the existing userland API. The timercmp() usage to check for a backwards timestamp is also clearer and simpler than the previous code which checked the result of the subtraction for a negative value. While here, take advantage of the 3-arg timersub() to store the subtraction results in a tempory timeval instead of overwriting the timestamp in the ktrace record and then having to restore it.
2016-10-01 22:12:33 +00:00
timersub(&kth->ktr_time, &prevtime_e, &temp);
printf("%jd.%06ld ", (intmax_t)temp.tv_sec,
temp.tv_usec);
Add a -E (elapsed time) flag to kdump. This is like -T, except it is relative to start of the dump. Approved by: re (scottl)
2003-12-07 01:06:32 +00:00
}
Make `-R', `-T' and `-E' options mutially non-exclusive. It is often useful to see two or three types at the same time when inspecting the dump. MFC after: 1 month Sponsored by: Sippy Software, Inc.
2014-03-25 23:37:57 +00:00
if (timestamp & TIMESTAMP_RELATIVE) {
o Properly init prevtime, so that we don't print bogus value in the first entry reported by the relative mode (-R). o Properly print negative offsets, which I guess may happen if records get re-ordered somehow, possibly due to the locking. Right now we report huge bogus diff (i.e. 2 seconds or so).
2015-04-25 04:58:08 +00:00
if (prevtime.tv_sec == 0)
prevtime = kth->ktr_time;
Use timercmp() and timersub() in kdump. Previously, kdump used the kernel-only timervalsub() macro which required defining _KERNEL when including <sys/time.h>. Now, kdump uses the existing userland API. The timercmp() usage to check for a backwards timestamp is also clearer and simpler than the previous code which checked the result of the subtraction for a negative value. While here, take advantage of the 3-arg timersub() to store the subtraction results in a tempory timeval instead of overwriting the timestamp in the ktrace record and then having to restore it.
2016-10-01 22:12:33 +00:00
if (timercmp(&kth->ktr_time, &prevtime, <)) {
timersub(&prevtime, &kth->ktr_time, &temp);
o Properly init prevtime, so that we don't print bogus value in the first entry reported by the relative mode (-R). o Properly print negative offsets, which I guess may happen if records get re-ordered somehow, possibly due to the locking. Right now we report huge bogus diff (i.e. 2 seconds or so).
2015-04-25 04:58:08 +00:00
sign = "-";
} else {
Use timercmp() and timersub() in kdump. Previously, kdump used the kernel-only timervalsub() macro which required defining _KERNEL when including <sys/time.h>. Now, kdump uses the existing userland API. The timercmp() usage to check for a backwards timestamp is also clearer and simpler than the previous code which checked the result of the subtraction for a negative value. While here, take advantage of the 3-arg timersub() to store the subtraction results in a tempory timeval instead of overwriting the timestamp in the ktrace record and then having to restore it.
2016-10-01 22:12:33 +00:00
timersub(&kth->ktr_time, &prevtime, &temp);
o Properly init prevtime, so that we don't print bogus value in the first entry reported by the relative mode (-R). o Properly print negative offsets, which I guess may happen if records get re-ordered somehow, possibly due to the locking. Right now we report huge bogus diff (i.e. 2 seconds or so).
2015-04-25 04:58:08 +00:00
sign = "";
}
Use timercmp() and timersub() in kdump. Previously, kdump used the kernel-only timervalsub() macro which required defining _KERNEL when including <sys/time.h>. Now, kdump uses the existing userland API. The timercmp() usage to check for a backwards timestamp is also clearer and simpler than the previous code which checked the result of the subtraction for a negative value. While here, take advantage of the 3-arg timersub() to store the subtraction results in a tempory timeval instead of overwriting the timestamp in the ktrace record and then having to restore it.
2016-10-01 22:12:33 +00:00
prevtime = kth->ktr_time;
printf("%s%jd.%06ld ", sign, (intmax_t)temp.tv_sec,
temp.tv_usec);
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
}
}
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
printf("%s ", type);
Remove spurious ABI tags from kdump output. The abidump routine output an ABI tag when -A was specified for records that were not displayed due to type or pid filtering. To fix, split the code to lookup the ABI from the code to display the ABI, move the code to display the ABI into dumpheader(), and move dumpheader() later in the main loop as a simplification. Previously dumpheader() was called under a condition that repeated conditions made later in the main loop. Reviewed by: kib MFC after: 1 month Sponsored by: DARPA / AFRL Differential Revision: https://reviews.freebsd.org/D16608
2018-08-07 00:10:58 +00:00
if (abiflag != 0) {
switch (sv_flags & SV_ABI_MASK) {
case SV_ABI_LINUX:
abi = "L";
break;
case SV_ABI_FREEBSD:
abi = "F";
break;
case SV_ABI_CLOUDABI:
abi = "C";
break;
default:
abi = "U";
break;
}
if ((sv_flags & SV_LP64) != 0)
arch = "64";
else if ((sv_flags & SV_ILP32) != 0)
arch = "32";
else
arch = "00";
printf("%s%s ", abi, arch);
}
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
}
#include <sys/syscall.h>
Move the mkioctls script to libsysdecode and use it to generate a sysdecode_ioctlname() function. This function matches the behavior of the truss variant in that it returns a pointer to a string description for known ioctls. The caller is responsible for displaying unknown ioctl requests. For kdump this meant moving the logic to handle unknown ioctl requests out of the generated function and into an ioctlname() function in kdump.c instead. Differential Revision: https://reviews.freebsd.org/D4610
2015-12-22 20:33:49 +00:00
static void
ioctlname(unsigned long val)
{
const char *str;
str = sysdecode_ioctlname(val);
if (str != NULL)
printf("%s", str);
else if (decimal)
printf("%lu", val);
else
printf("%#lx", val);
}
Add support to libsysdecode for decoding system call names. A new sysdecode_syscallname() function accepts a system call code and returns a string of the corresponding name (or NULL if the code is unknown). To support different process ABIs, the new function accepts a value from a new sysdecode_abi enum as its first argument to select the ABI in use. Current ABIs supported include FREEBSD (native binaries), FREEBSD32, LINUX, LINUX32, and CLOUDABI64. Note that not all ABIs are supported by all platforms. In general, a given ABI is only supported if a platform can execute binaries for that ABI. To simplify the implementation, libsysdecode's build reuses the existing pre-generated files from the kernel source tree rather than duplicating new copies of said files during the build. kdump(1) and truss(1) now use these functions to map system call identifiers to names. For kdump(1), a new 'syscallname()' function consolidates duplicated code from ktrsyscall() and ktrsyscallret(). The Linux ABI no longer requires custom handling for ktrsyscall() and linux_ktrsyscall() has been removed as a result. Reviewed by: bdrewery Differential Revision: https://reviews.freebsd.org/D4823
2016-01-26 19:07:09 +00:00
static enum sysdecode_abi
syscallabi(u_int sv_flags)
{
if (sv_flags == 0)
Add a SYSDECODE_ABI_ prefix to the ABI enums to avoid potential collisions. Suggested by: jmallett Reviewed by: bdrewery, jmallett Differential Revision: https://reviews.freebsd.org/D5123
2016-01-30 01:00:54 +00:00
return (SYSDECODE_ABI_FREEBSD);
Add support to libsysdecode for decoding system call names. A new sysdecode_syscallname() function accepts a system call code and returns a string of the corresponding name (or NULL if the code is unknown). To support different process ABIs, the new function accepts a value from a new sysdecode_abi enum as its first argument to select the ABI in use. Current ABIs supported include FREEBSD (native binaries), FREEBSD32, LINUX, LINUX32, and CLOUDABI64. Note that not all ABIs are supported by all platforms. In general, a given ABI is only supported if a platform can execute binaries for that ABI. To simplify the implementation, libsysdecode's build reuses the existing pre-generated files from the kernel source tree rather than duplicating new copies of said files during the build. kdump(1) and truss(1) now use these functions to map system call identifiers to names. For kdump(1), a new 'syscallname()' function consolidates duplicated code from ktrsyscall() and ktrsyscallret(). The Linux ABI no longer requires custom handling for ktrsyscall() and linux_ktrsyscall() has been removed as a result. Reviewed by: bdrewery Differential Revision: https://reviews.freebsd.org/D4823
2016-01-26 19:07:09 +00:00
switch (sv_flags & SV_ABI_MASK) {
case SV_ABI_FREEBSD:
Add a SYSDECODE_ABI_ prefix to the ABI enums to avoid potential collisions. Suggested by: jmallett Reviewed by: bdrewery, jmallett Differential Revision: https://reviews.freebsd.org/D5123
2016-01-30 01:00:54 +00:00
return (SYSDECODE_ABI_FREEBSD);
Add support to libsysdecode for decoding system call names. A new sysdecode_syscallname() function accepts a system call code and returns a string of the corresponding name (or NULL if the code is unknown). To support different process ABIs, the new function accepts a value from a new sysdecode_abi enum as its first argument to select the ABI in use. Current ABIs supported include FREEBSD (native binaries), FREEBSD32, LINUX, LINUX32, and CLOUDABI64. Note that not all ABIs are supported by all platforms. In general, a given ABI is only supported if a platform can execute binaries for that ABI. To simplify the implementation, libsysdecode's build reuses the existing pre-generated files from the kernel source tree rather than duplicating new copies of said files during the build. kdump(1) and truss(1) now use these functions to map system call identifiers to names. For kdump(1), a new 'syscallname()' function consolidates duplicated code from ktrsyscall() and ktrsyscallret(). The Linux ABI no longer requires custom handling for ktrsyscall() and linux_ktrsyscall() has been removed as a result. Reviewed by: bdrewery Differential Revision: https://reviews.freebsd.org/D4823
2016-01-26 19:07:09 +00:00
case SV_ABI_LINUX:
kdump: simplify/remove per-arch #ifdefs It is acceptable for syscallabi to map SV_ABI to SYSDECODE_ABI on all architectures; libsysdecode will return not-found sentinel values if it does not have a syscall name or errno mapping for a given architecture. Also, use __LP64__ for the SV_ILP32 -> SYSDECODE_ABI_LINUX32 mapping, for any future 32- on 64-bit linuxulator implementation. Reviewed by: jhb Sponsored by: Turing Robotic Industries Inc.
2018-04-24 19:23:26 +00:00
#ifdef __LP64__
Add support to libsysdecode for decoding system call names. A new sysdecode_syscallname() function accepts a system call code and returns a string of the corresponding name (or NULL if the code is unknown). To support different process ABIs, the new function accepts a value from a new sysdecode_abi enum as its first argument to select the ABI in use. Current ABIs supported include FREEBSD (native binaries), FREEBSD32, LINUX, LINUX32, and CLOUDABI64. Note that not all ABIs are supported by all platforms. In general, a given ABI is only supported if a platform can execute binaries for that ABI. To simplify the implementation, libsysdecode's build reuses the existing pre-generated files from the kernel source tree rather than duplicating new copies of said files during the build. kdump(1) and truss(1) now use these functions to map system call identifiers to names. For kdump(1), a new 'syscallname()' function consolidates duplicated code from ktrsyscall() and ktrsyscallret(). The Linux ABI no longer requires custom handling for ktrsyscall() and linux_ktrsyscall() has been removed as a result. Reviewed by: bdrewery Differential Revision: https://reviews.freebsd.org/D4823
2016-01-26 19:07:09 +00:00
if (sv_flags & SV_ILP32)
Add a SYSDECODE_ABI_ prefix to the ABI enums to avoid potential collisions. Suggested by: jmallett Reviewed by: bdrewery, jmallett Differential Revision: https://reviews.freebsd.org/D5123
2016-01-30 01:00:54 +00:00
return (SYSDECODE_ABI_LINUX32);
Add support to libsysdecode for decoding system call names. A new sysdecode_syscallname() function accepts a system call code and returns a string of the corresponding name (or NULL if the code is unknown). To support different process ABIs, the new function accepts a value from a new sysdecode_abi enum as its first argument to select the ABI in use. Current ABIs supported include FREEBSD (native binaries), FREEBSD32, LINUX, LINUX32, and CLOUDABI64. Note that not all ABIs are supported by all platforms. In general, a given ABI is only supported if a platform can execute binaries for that ABI. To simplify the implementation, libsysdecode's build reuses the existing pre-generated files from the kernel source tree rather than duplicating new copies of said files during the build. kdump(1) and truss(1) now use these functions to map system call identifiers to names. For kdump(1), a new 'syscallname()' function consolidates duplicated code from ktrsyscall() and ktrsyscallret(). The Linux ABI no longer requires custom handling for ktrsyscall() and linux_ktrsyscall() has been removed as a result. Reviewed by: bdrewery Differential Revision: https://reviews.freebsd.org/D4823
2016-01-26 19:07:09 +00:00
#endif
Add a SYSDECODE_ABI_ prefix to the ABI enums to avoid potential collisions. Suggested by: jmallett Reviewed by: bdrewery, jmallett Differential Revision: https://reviews.freebsd.org/D5123
2016-01-30 01:00:54 +00:00
return (SYSDECODE_ABI_LINUX);
Add simple support for CloudABI processes to kdump(1). This does not decode arguments to system calls but should properly decode system call names and error return values. Reviewed by: ed Differential Revision: https://reviews.freebsd.org/D5412
2016-02-25 17:43:32 +00:00
case SV_ABI_CLOUDABI:
return (SYSDECODE_ABI_CLOUDABI64);
Add support to libsysdecode for decoding system call names. A new sysdecode_syscallname() function accepts a system call code and returns a string of the corresponding name (or NULL if the code is unknown). To support different process ABIs, the new function accepts a value from a new sysdecode_abi enum as its first argument to select the ABI in use. Current ABIs supported include FREEBSD (native binaries), FREEBSD32, LINUX, LINUX32, and CLOUDABI64. Note that not all ABIs are supported by all platforms. In general, a given ABI is only supported if a platform can execute binaries for that ABI. To simplify the implementation, libsysdecode's build reuses the existing pre-generated files from the kernel source tree rather than duplicating new copies of said files during the build. kdump(1) and truss(1) now use these functions to map system call identifiers to names. For kdump(1), a new 'syscallname()' function consolidates duplicated code from ktrsyscall() and ktrsyscallret(). The Linux ABI no longer requires custom handling for ktrsyscall() and linux_ktrsyscall() has been removed as a result. Reviewed by: bdrewery Differential Revision: https://reviews.freebsd.org/D4823
2016-01-26 19:07:09 +00:00
default:
Add a SYSDECODE_ABI_ prefix to the ABI enums to avoid potential collisions. Suggested by: jmallett Reviewed by: bdrewery, jmallett Differential Revision: https://reviews.freebsd.org/D5123
2016-01-30 01:00:54 +00:00
return (SYSDECODE_ABI_UNKNOWN);
Add support to libsysdecode for decoding system call names. A new sysdecode_syscallname() function accepts a system call code and returns a string of the corresponding name (or NULL if the code is unknown). To support different process ABIs, the new function accepts a value from a new sysdecode_abi enum as its first argument to select the ABI in use. Current ABIs supported include FREEBSD (native binaries), FREEBSD32, LINUX, LINUX32, and CLOUDABI64. Note that not all ABIs are supported by all platforms. In general, a given ABI is only supported if a platform can execute binaries for that ABI. To simplify the implementation, libsysdecode's build reuses the existing pre-generated files from the kernel source tree rather than duplicating new copies of said files during the build. kdump(1) and truss(1) now use these functions to map system call identifiers to names. For kdump(1), a new 'syscallname()' function consolidates duplicated code from ktrsyscall() and ktrsyscallret(). The Linux ABI no longer requires custom handling for ktrsyscall() and linux_ktrsyscall() has been removed as a result. Reviewed by: bdrewery Differential Revision: https://reviews.freebsd.org/D4823
2016-01-26 19:07:09 +00:00
}
}
static void
syscallname(u_int code, u_int sv_flags)
{
const char *name;
name = sysdecode_syscallname(syscallabi(sv_flags), code);
if (name == NULL)
printf("[%d]", code);
else {
printf("%s", name);
if (syscallno)
printf("[%d]", code);
}
}
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
static void
print_signal(int signo)
{
const char *signame;
signame = sysdecode_signal(signo);
if (signame != NULL)
printf("%s", signame);
else
printf("SIG %d", signo);
}
Improve WARNS situation for kdump: 1) Define _KERNEL while including sys/time.h to get some function prototypes. 2) Add prototypes and ANSIify definitions. 3) Constness changes. 4) Remove register keyword. 5) Actually return a sensible value from main. 6) Make fread_tail take a void * instead of a char *. 7) Avoid a signedness warning by casting to a size_t. Should be safe enough 'cos we also check for nonnegativity. 8) Be extra chummy with sigset_t rather than passing a struct to printf and pretending it is an int.
2002-07-28 16:05:07 +00:00
void
Add support to libsysdecode for decoding system call names. A new sysdecode_syscallname() function accepts a system call code and returns a string of the corresponding name (or NULL if the code is unknown). To support different process ABIs, the new function accepts a value from a new sysdecode_abi enum as its first argument to select the ABI in use. Current ABIs supported include FREEBSD (native binaries), FREEBSD32, LINUX, LINUX32, and CLOUDABI64. Note that not all ABIs are supported by all platforms. In general, a given ABI is only supported if a platform can execute binaries for that ABI. To simplify the implementation, libsysdecode's build reuses the existing pre-generated files from the kernel source tree rather than duplicating new copies of said files during the build. kdump(1) and truss(1) now use these functions to map system call identifiers to names. For kdump(1), a new 'syscallname()' function consolidates duplicated code from ktrsyscall() and ktrsyscallret(). The Linux ABI no longer requires custom handling for ktrsyscall() and linux_ktrsyscall() has been removed as a result. Reviewed by: bdrewery Differential Revision: https://reviews.freebsd.org/D4823
2016-01-26 19:07:09 +00:00
ktrsyscall(struct ktr_syscall *ktr, u_int sv_flags)
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
{
Improve WARNS situation for kdump: 1) Define _KERNEL while including sys/time.h to get some function prototypes. 2) Add prototypes and ANSIify definitions. 3) Constness changes. 4) Remove register keyword. 5) Actually return a sensible value from main. 6) Make fread_tail take a void * instead of a char *. 7) Avoid a signedness warning by casting to a size_t. Should be safe enough 'cos we also check for nonnegativity. 8) Be extra chummy with sigset_t rather than passing a struct to printf and pretending it is an int.
2002-07-28 16:05:07 +00:00
int narg = ktr->ktr_narg;
Handle 64-bit system call arguments (off_t, id_t). In particular, 64-bit system call arguments use up two register_t arguments for 32-bit processes. They must also be aligned on a 64-bit boundary on 32-bit powerpc processes. This fixes the decoding of lseek(), procctl(), and wait6() arguments for 32-bit processes (both native and via freebsd32). Note that the ktrace system call return record only returns a single register, so the return value of lseek is always truncated to the low 32-bits for 32-bit processes.
2016-10-01 22:01:41 +00:00
register_t *ip, *first;
Add a new trace point, KTRFAC_CAPFAIL, which traces capability check failures. It is included in the default set for ktrace(1) and kdump(1).
2011-10-11 20:37:10 +00:00
intmax_t arg;
Handle 64-bit system call arguments (off_t, id_t). In particular, 64-bit system call arguments use up two register_t arguments for 32-bit processes. They must also be aligned on a 64-bit boundary on 32-bit powerpc processes. This fixes the decoding of lseek(), procctl(), and wait6() arguments for 32-bit processes (both native and via freebsd32). Note that the ktrace system call return record only returns a single register, so the return value of lseek is always truncated to the low 32-bits for 32-bit processes.
2016-10-01 22:01:41 +00:00
int quad_align, quad_slots;
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
Add support to libsysdecode for decoding system call names. A new sysdecode_syscallname() function accepts a system call code and returns a string of the corresponding name (or NULL if the code is unknown). To support different process ABIs, the new function accepts a value from a new sysdecode_abi enum as its first argument to select the ABI in use. Current ABIs supported include FREEBSD (native binaries), FREEBSD32, LINUX, LINUX32, and CLOUDABI64. Note that not all ABIs are supported by all platforms. In general, a given ABI is only supported if a platform can execute binaries for that ABI. To simplify the implementation, libsysdecode's build reuses the existing pre-generated files from the kernel source tree rather than duplicating new copies of said files during the build. kdump(1) and truss(1) now use these functions to map system call identifiers to names. For kdump(1), a new 'syscallname()' function consolidates duplicated code from ktrsyscall() and ktrsyscallret(). The Linux ABI no longer requires custom handling for ktrsyscall() and linux_ktrsyscall() has been removed as a result. Reviewed by: bdrewery Differential Revision: https://reviews.freebsd.org/D4823
2016-01-26 19:07:09 +00:00
syscallname(ktr->ktr_code, sv_flags);
Handle 64-bit system call arguments (off_t, id_t). In particular, 64-bit system call arguments use up two register_t arguments for 32-bit processes. They must also be aligned on a 64-bit boundary on 32-bit powerpc processes. This fixes the decoding of lseek(), procctl(), and wait6() arguments for 32-bit processes (both native and via freebsd32). Note that the ktrace system call return record only returns a single register, so the return value of lseek is always truncated to the low 32-bits for 32-bit processes.
2016-10-01 22:01:41 +00:00
ip = first = &ktr->ktr_args[0];
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
if (narg) {
char c = '(';
Teach kdump to understand sv_flags records in the trace files. MFC after: 1 Month.
2011-02-25 22:07:23 +00:00
if (fancy &&
Add support to libsysdecode for decoding system call names. A new sysdecode_syscallname() function accepts a system call code and returns a string of the corresponding name (or NULL if the code is unknown). To support different process ABIs, the new function accepts a value from a new sysdecode_abi enum as its first argument to select the ABI in use. Current ABIs supported include FREEBSD (native binaries), FREEBSD32, LINUX, LINUX32, and CLOUDABI64. Note that not all ABIs are supported by all platforms. In general, a given ABI is only supported if a platform can execute binaries for that ABI. To simplify the implementation, libsysdecode's build reuses the existing pre-generated files from the kernel source tree rather than duplicating new copies of said files during the build. kdump(1) and truss(1) now use these functions to map system call identifiers to names. For kdump(1), a new 'syscallname()' function consolidates duplicated code from ktrsyscall() and ktrsyscallret(). The Linux ABI no longer requires custom handling for ktrsyscall() and linux_ktrsyscall() has been removed as a result. Reviewed by: bdrewery Differential Revision: https://reviews.freebsd.org/D4823
2016-01-26 19:07:09 +00:00
(sv_flags == 0 ||
(sv_flags & SV_ABI_MASK) == SV_ABI_FREEBSD)) {
Handle 64-bit system call arguments (off_t, id_t). In particular, 64-bit system call arguments use up two register_t arguments for 32-bit processes. They must also be aligned on a 64-bit boundary on 32-bit powerpc processes. This fixes the decoding of lseek(), procctl(), and wait6() arguments for 32-bit processes (both native and via freebsd32). Note that the ktrace system call return record only returns a single register, so the return value of lseek is always truncated to the low 32-bits for 32-bit processes.
2016-10-01 22:01:41 +00:00
quad_align = 0;
if (sv_flags & SV_ILP32) {
#ifdef __powerpc__
quad_align = 1;
#endif
quad_slots = 2;
} else
quad_slots = 1;
kdump: Improve decoding of various *at calls: * Write AT_FDCWD where appropriate. * Decode the remaining arguments of openat() etc like open() etc.
2013-08-13 19:57:35 +00:00
switch (ktr->ktr_code) {
kdump: Decode AT_FDCWD in first argument of bindat() and connectat().
2013-08-13 20:33:50 +00:00
case SYS_bindat:
Decode arguments to chflagsat().
2017-03-15 22:36:26 +00:00
case SYS_chflagsat:
kdump: Decode AT_FDCWD in first argument of bindat() and connectat().
2013-08-13 20:33:50 +00:00
case SYS_connectat:
kdump: Improve decoding of various *at calls: * Write AT_FDCWD where appropriate. * Decode the remaining arguments of openat() etc like open() etc.
2013-08-13 19:57:35 +00:00
case SYS_faccessat:
case SYS_fchmodat:
case SYS_fchownat:
case SYS_fstatat:
case SYS_futimesat:
case SYS_linkat:
case SYS_mkdirat:
case SYS_mkfifoat:
case SYS_mknodat:
case SYS_openat:
case SYS_readlinkat:
case SYS_renameat:
case SYS_unlinkat:
Add futimens and utimensat system calls. The core kernel part is patch file utimes.2008.4.diff from pluknet@FreeBSD.org. I updated the code for API changes, added the manual page and added compatibility code for old kernels. There is also audit and Capsicum support. A new UTIME_* constant might allow setting birthtimes in future. Differential Revision: https://reviews.freebsd.org/D1426 Submitted by: pluknet (partially) Reviewed by: delphij, pluknet, rwatson Relnotes: yes
2015-01-23 21:07:08 +00:00
case SYS_utimensat:
kdump: Improve decoding of various *at calls: * Write AT_FDCWD where appropriate. * Decode the remaining arguments of openat() etc like open() etc.
2013-08-13 19:57:35 +00:00
putchar('(');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_integer_arg_valid(sysdecode_atfd, *ip);
kdump: Improve decoding of various *at calls: * Write AT_FDCWD where appropriate. * Decode the remaining arguments of openat() etc like open() etc.
2013-08-13 19:57:35 +00:00
c = ',';
ip++;
narg--;
break;
}
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
switch (ktr->ktr_code) {
case SYS_ioctl: {
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
Bring ioctlname() in line with all the other *name() functions, which actually print the name (or the numeric value, if they can't figure out the correct name) instead of just returning a pointer to it. Also, since ioctl numbers are not and probably never will be unique, drop support for using a switch statement instead of an if/else chain.
2011-10-08 12:47:00 +00:00
putchar(c);
Move the mkioctls script to libsysdecode and use it to generate a sysdecode_ioctlname() function. This function matches the behavior of the truss variant in that it returns a pointer to a string description for known ioctls. The caller is responsible for displaying unknown ioctl requests. For kdump this meant moving the logic to handle unknown ioctl requests out of the generated function and into an ioctlname() function in kdump.c instead. Differential Revision: https://reviews.freebsd.org/D4610
2015-12-22 20:33:49 +00:00
ioctlname(*ip);
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
c = ',';
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
}
case SYS_ptrace:
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar('(');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_integer_arg(sysdecode_ptrace_request, *ip);
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
c = ',';
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_access:
case SYS_eaccess:
kdump: Improve decoding of various *at calls: * Write AT_FDCWD where appropriate. * Decode the remaining arguments of openat() etc like open() etc.
2013-08-13 19:57:35 +00:00
case SYS_faccessat:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_mask_arg(sysdecode_access_mode, *ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_open:
kdump: Improve decoding of various *at calls: * Write AT_FDCWD where appropriate. * Decode the remaining arguments of openat() etc like open() etc.
2013-08-13 19:57:35 +00:00
case SYS_openat:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_mask_arg(sysdecode_open_flags, ip[0]);
if ((ip[0] & O_CREAT) == O_CREAT) {
putchar(',');
decode_filemode(ip[1]);
}
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
ip += 2;
narg -= 2;
break;
case SYS_wait4:
- Decode the idtype argument passed to wait6() in kdump and truss. - Don't treat an options argument of 0 to wait4() as an error in kdump. - Decode the wait options passed to wait4() and wait6() in truss and decode the returned rusage and exit status. Approved by: re (kib) MFC after: 1 week
2013-09-12 18:08:25 +00:00
print_number(ip, narg, c);
print_number(ip, narg, c);
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
putchar(',');
print_mask_arg0(sysdecode_wait4_options, *ip);
ip++;
narg--;
- Decode the idtype argument passed to wait6() in kdump and truss. - Don't treat an options argument of 0 to wait4() as an error in kdump. - Decode the wait options passed to wait4() and wait6() in truss and decode the returned rusage and exit status. Approved by: re (kib) MFC after: 1 week
2013-09-12 18:08:25 +00:00
break;
case SYS_wait6:
putchar('(');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_integer_arg(sysdecode_idtype, *ip);
- Decode the idtype argument passed to wait6() in kdump and truss. - Don't treat an options argument of 0 to wait4() as an error in kdump. - Decode the wait options passed to wait4() and wait6() in truss and decode the returned rusage and exit status. Approved by: re (kib) MFC after: 1 week
2013-09-12 18:08:25 +00:00
c = ',';
ip++;
narg--;
Handle 64-bit system call arguments (off_t, id_t). In particular, 64-bit system call arguments use up two register_t arguments for 32-bit processes. They must also be aligned on a 64-bit boundary on 32-bit powerpc processes. This fixes the decoding of lseek(), procctl(), and wait6() arguments for 32-bit processes (both native and via freebsd32). Note that the ktrace system call return record only returns a single register, so the return value of lseek is always truncated to the low 32-bits for 32-bit processes.
2016-10-01 22:01:41 +00:00
print_number64(first, ip, narg, c);
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_mask_arg(sysdecode_wait6_options, *ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_chmod:
case SYS_fchmod:
case SYS_lchmod:
Print the fchmodat mode in human readable fashion MFC after: 1 week
2016-05-03 21:27:17 +00:00
case SYS_fchmodat:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
decode_filemode(*ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
kdump: Improve decoding of various *at calls: * Write AT_FDCWD where appropriate. * Decode the remaining arguments of openat() etc like open() etc.
2013-08-13 19:57:35 +00:00
case SYS_mknodat:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
decode_filemode(*ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_getfsstat:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Update libsysdecode for getfsstat() 'flags' argument changing to 'mode'. As a followup to r310638, update libsysdecode (and kdump) to decode the 'mode' argument to getfsstat(). sysdecode_getfsstat_flags() has been renamed to sysdecode_getfsstat_mode() and now treats the argument as an enumerated value rather than a mask of flags.
2017-01-03 01:39:05 +00:00
print_integer_arg(sysdecode_getfsstat_mode, *ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_mount:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_mask_arg(sysdecode_mount_flags, *ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_unmount:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_mask_arg(sysdecode_mount_flags, *ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_recvmsg:
case SYS_sendmsg:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_mask_arg0(sysdecode_msg_flags, *ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_recvfrom:
case SYS_sendto:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
print_number(ip, narg, c);
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_mask_arg0(sysdecode_msg_flags, *ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_chflags:
Decode arguments to chflagsat().
2017-03-15 22:36:26 +00:00
case SYS_chflagsat:
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
case SYS_fchflags:
case SYS_lchflags:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
decode_fileflags(*ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_kill:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_signal(*ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_reboot:
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar('(');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_mask_arg(sysdecode_reboot_howto, *ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_umask:
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar('(');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
decode_filemode(*ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_msync:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_mask_arg(sysdecode_msync_flags, *ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
kdump has knowledge of lseek() and mmap() arg decoding. Teach it about the new mmap and lseek syscalls. Approved by: re (kensmith)
2007-07-04 23:32:49 +00:00
#ifdef SYS_freebsd6_mmap
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
case SYS_freebsd6_mmap:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_mask_arg(sysdecode_mmap_prot, *ip);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
kdump has knowledge of lseek() and mmap() arg decoding. Teach it about the new mmap and lseek syscalls. Approved by: re (kensmith)
2007-07-04 23:32:49 +00:00
ip++;
narg--;
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_mask_arg(sysdecode_mmap_flags, *ip);
kdump has knowledge of lseek() and mmap() arg decoding. Teach it about the new mmap and lseek syscalls. Approved by: re (kensmith)
2007-07-04 23:32:49 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
kdump has knowledge of lseek() and mmap() arg decoding. Teach it about the new mmap and lseek syscalls. Approved by: re (kensmith)
2007-07-04 23:32:49 +00:00
#endif
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
case SYS_mmap:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_mask_arg(sysdecode_mmap_prot, *ip);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_mask_arg(sysdecode_mmap_flags, *ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_mprotect:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_mask_arg(sysdecode_mmap_prot, *ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_madvise:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_integer_arg(sysdecode_madvice, *ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
Decode pathconf() names, *at() flags, and sysarch() numbers in libsysdecode. Move tables that were previously in truss over to libsysdecode. truss output is unchanged, but kdump has been updated to decode these fields. In addition, sysdecode_sysarch_number() should support all platforms whereas the old table in truss only supported x86.
2017-09-04 05:34:36 +00:00
case SYS_pathconf:
case SYS_lpathconf:
case SYS_fpathconf:
print_number(ip, narg, c);
putchar(',');
print_integer_arg(sysdecode_pathconf_name, *ip);
ip++;
narg--;
break;
Fix decoding of setpriority() arguments. The PRIO_* 'which' value is stored in the first argument to setpriority(2), not the last. While here, decode the arguments to getpriority(2).
2017-06-10 00:35:45 +00:00
case SYS_getpriority:
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
case SYS_setpriority:
Fix decoding of setpriority() arguments. The PRIO_* 'which' value is stored in the first argument to setpriority(2), not the last. While here, decode the arguments to getpriority(2).
2017-06-10 00:35:45 +00:00
putchar('(');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_integer_arg(sysdecode_prio_which, *ip);
Fix decoding of setpriority() arguments. The PRIO_* 'which' value is stored in the first argument to setpriority(2), not the last. While here, decode the arguments to getpriority(2).
2017-06-10 00:35:45 +00:00
c = ',';
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_fcntl:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_integer_arg(sysdecode_fcntl_cmd, ip[0]);
if (sysdecode_fcntl_arg_p(ip[0])) {
putchar(',');
if (ip[0] == F_SETFL)
print_mask_arg(
sysdecode_fcntl_fileflags,
ip[1]);
else
sysdecode_fcntl_arg(stdout,
ip[0], ip[1],
decimal ? 10 : 16);
}
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
ip += 2;
narg -= 2;
break;
case SYS_socket: {
Add sockipprotoname() function. Decode the third parameter (protocol) of a socket() call with sockipprotoname() if the first parameter (domain) is PF_INET or PF_INET6. Old parsing behavior before this change: ping6 CALL socket(PF_INET6,SOCK_RAW,0x3a) New behavior after this change: ping6 CALL socket(PF_INET6,SOCK_RAW,IPPROTO_ICMPV6)
2007-01-04 04:46:59 +00:00
int sockdomain;
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar('(');
Address some of bde@'s concerns with the new code.
2011-10-08 21:02:08 +00:00
sockdomain = *ip;
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_integer_arg(sysdecode_socketdomain,
sockdomain);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_mask_arg(sysdecode_socket_type, *ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
Add sockipprotoname() function. Decode the third parameter (protocol) of a socket() call with sockipprotoname() if the first parameter (domain) is PF_INET or PF_INET6. Old parsing behavior before this change: ping6 CALL socket(PF_INET6,SOCK_RAW,0x3a) New behavior after this change: ping6 CALL socket(PF_INET6,SOCK_RAW,IPPROTO_ICMPV6)
2007-01-04 04:46:59 +00:00
if (sockdomain == PF_INET ||
sockdomain == PF_INET6) {
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_integer_arg(sysdecode_ipproto,
*ip);
Add sockipprotoname() function. Decode the third parameter (protocol) of a socket() call with sockipprotoname() if the first parameter (domain) is PF_INET or PF_INET6. Old parsing behavior before this change: ping6 CALL socket(PF_INET6,SOCK_RAW,0x3a) New behavior after this change: ping6 CALL socket(PF_INET6,SOCK_RAW,IPPROTO_ICMPV6)
2007-01-04 04:46:59 +00:00
ip++;
narg--;
}
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
c = ',';
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
}
case SYS_setsockopt:
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
case SYS_getsockopt: {
const char *str;
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_integer_arg_valid(sysdecode_sockopt_level,
*ip);
str = sysdecode_sockopt_name(ip[0], ip[1]);
if (str != NULL) {
printf(",%s", str);
Only use sockoptname() to parse socket option names for SOL_SOCKET requests. MFC after: 3 days Reported by: Michiel Boland michiel boland.org
2008-01-07 18:50:25 +00:00
ip++;
narg--;
}
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
}
kdump has knowledge of lseek() and mmap() arg decoding. Teach it about the new mmap and lseek syscalls. Approved by: re (kensmith)
2007-07-04 23:32:49 +00:00
#ifdef SYS_freebsd6_lseek
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
case SYS_freebsd6_lseek:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
/* Hidden 'pad' argument, not in lseek(2) */
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
Handle 64-bit system call arguments (off_t, id_t). In particular, 64-bit system call arguments use up two register_t arguments for 32-bit processes. They must also be aligned on a 64-bit boundary on 32-bit powerpc processes. This fixes the decoding of lseek(), procctl(), and wait6() arguments for 32-bit processes (both native and via freebsd32). Note that the ktrace system call return record only returns a single register, so the return value of lseek is always truncated to the low 32-bits for 32-bit processes.
2016-10-01 22:01:41 +00:00
print_number64(first, ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_integer_arg(sysdecode_whence, *ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
kdump has knowledge of lseek() and mmap() arg decoding. Teach it about the new mmap and lseek syscalls. Approved by: re (kensmith)
2007-07-04 23:32:49 +00:00
#endif
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
case SYS_lseek:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
Handle 64-bit system call arguments (off_t, id_t). In particular, 64-bit system call arguments use up two register_t arguments for 32-bit processes. They must also be aligned on a 64-bit boundary on 32-bit powerpc processes. This fixes the decoding of lseek(), procctl(), and wait6() arguments for 32-bit processes (both native and via freebsd32). Note that the ktrace system call return record only returns a single register, so the return value of lseek is always truncated to the low 32-bits for 32-bit processes.
2016-10-01 22:01:41 +00:00
print_number64(first, ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_integer_arg(sysdecode_whence, *ip);
kdump has knowledge of lseek() and mmap() arg decoding. Teach it about the new mmap and lseek syscalls. Approved by: re (kensmith)
2007-07-04 23:32:49 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_flock:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_mask_arg(sysdecode_flock_operation, *ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_mkfifo:
kdump: Improve decoding of various *at calls: * Write AT_FDCWD where appropriate. * Decode the remaining arguments of openat() etc like open() etc.
2013-08-13 19:57:35 +00:00
case SYS_mkfifoat:
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
case SYS_mkdir:
kdump: Improve decoding of various *at calls: * Write AT_FDCWD where appropriate. * Decode the remaining arguments of openat() etc like open() etc.
2013-08-13 19:57:35 +00:00
case SYS_mkdirat:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
decode_filemode(*ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_shutdown:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_integer_arg(sysdecode_shutdown_how, *ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_socketpair:
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar('(');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_integer_arg(sysdecode_socketdomain, *ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_mask_arg(sysdecode_socket_type, *ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
c = ',';
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_getrlimit:
case SYS_setrlimit:
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar('(');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_integer_arg(sysdecode_rlimit, *ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
c = ',';
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
Decode the 'who' argument passed to getrusage(). Add a new sysdecode_getrusage_who() which decodes the RUSAGE_* constant passed as the first argument to getrusage(). Use this function in both kdump and truss to decode the first argument to getrusage(). PR: 215448 Submitted by: Anton Yuzhaninov <citrin+pr@citrin.ru> MFC after: 1 month
2017-06-03 14:22:15 +00:00
case SYS_getrusage:
putchar('(');
print_integer_arg(sysdecode_getrusage_who, *ip);
ip++;
narg--;
c = ',';
break;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
case SYS_quotactl:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
if (!sysdecode_quotactl_cmd(stdout, *ip)) {
if (decimal)
printf("<invalid=%d>", (int)*ip);
else
printf("<invalid=%#x>",
(int)*ip);
}
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
c = ',';
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_nfssvc:
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar('(');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_integer_arg(sysdecode_nfssvc_flags, *ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
c = ',';
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_rtprio:
Decode arguments to rtprio_thread() (same as rtprio()).
2017-06-10 01:32:18 +00:00
case SYS_rtprio_thread:
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar('(');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_integer_arg(sysdecode_rtprio_function,
*ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
c = ',';
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS___semctl:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_integer_arg(sysdecode_semctl_cmd, *ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_semget:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_mask_arg(sysdecode_semget_flags, *ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_msgctl:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_integer_arg(sysdecode_msgctl_cmd, *ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_shmat:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_mask_arg(sysdecode_shmat_flags, *ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_shmctl:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_integer_arg(sysdecode_shmctl_cmd, *ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
sysent: regenerate after r352705 This also implements it, fixes kdump, and removes no longer needed bits from lib/libc/sys/shm_open.c for the interim.
2019-09-25 18:09:19 +00:00
#ifdef SYS_freebsd12_shm_open
case SYS_freebsd12_shm_open:
kdump: decode SHM_ANON as first arg to legacy shm_open(2) The first argument to shm_open(2) as well as shm_open2(2) may be a path or SHM_ANON. Decode SHM_ANON, at least- paths will show up as namei results in kdump output, which may be sufficient; in those cases, we'll have printed an address. Future commits will add support for shm_open2() to libsysdecode/truss/kdump. Reported by: kaktus MFC after: 3 days
2020-02-19 14:32:55 +00:00
if (ip[0] == (uintptr_t)SHM_ANON) {
printf("(SHM_ANON");
ip++;
} else {
print_number(ip, narg, c);
}
Add kdump support for shm_open(). MFC after: 1 week
2014-08-01 23:28:21 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_mask_arg(sysdecode_open_flags, ip[0]);
putchar(',');
decode_filemode(ip[1]);
ip += 2;
narg -= 2;
Add kdump support for shm_open(). MFC after: 1 week
2014-08-01 23:28:21 +00:00
break;
sysent: regenerate after r352705 This also implements it, fixes kdump, and removes no longer needed bits from lib/libc/sys/shm_open.c for the interim.
2019-09-25 18:09:19 +00:00
#endif
kdump: decode shm_open2 This is the kdump counterpart of the truss support added in r358116, and also a part of D23733. shm_open2 is the successor to shm_open. Reviewed by: kaktus
2020-02-19 14:55:59 +00:00
case SYS_shm_open2:
if (ip[0] == (uintptr_t)SHM_ANON) {
printf("(SHM_ANON");
ip++;
} else {
print_number(ip, narg, c);
}
putchar(',');
print_mask_arg(sysdecode_open_flags, ip[0]);
putchar(',');
decode_filemode(ip[1]);
putchar(',');
print_mask_arg(sysdecode_shmflags, ip[2]);
ip += 3;
narg -= 3;
break;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
case SYS_minherit:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_integer_arg(sysdecode_minherit_inherit,
*ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_rfork:
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar('(');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_mask_arg(sysdecode_rfork_flags, *ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
c = ',';
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_lio_listio:
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar('(');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_integer_arg(sysdecode_lio_listio_mode,
*ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
c = ',';
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_mlockall:
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar('(');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_mask_arg(sysdecode_mlockall_flags, *ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_sched_setscheduler:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_integer_arg(sysdecode_scheduler_policy,
*ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_sched_get_priority_max:
case SYS_sched_get_priority_min:
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar('(');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_integer_arg(sysdecode_scheduler_policy,
*ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_sendfile:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
print_number(ip, narg, c);
print_number(ip, narg, c);
print_number(ip, narg, c);
print_number(ip, narg, c);
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_mask_arg(sysdecode_sendfile_flags, *ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_kldsym:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_integer_arg(sysdecode_kldsym_cmd, *ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_sigprocmask:
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar('(');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_integer_arg(sysdecode_sigprocmask_how,
*ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
c = ',';
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS___acl_get_file:
case SYS___acl_set_file:
case SYS___acl_get_fd:
case SYS___acl_set_fd:
case SYS___acl_delete_file:
case SYS___acl_delete_fd:
case SYS___acl_aclcheck_file:
case SYS___acl_aclcheck_fd:
case SYS___acl_get_link:
case SYS___acl_set_link:
case SYS___acl_delete_link:
case SYS___acl_aclcheck_link:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_integer_arg(sysdecode_acltype, *ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_sigaction:
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar('(');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_signal(*ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
c = ',';
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_extattrctl:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_integer_arg(sysdecode_extattrnamespace,
*ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_nmount:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_mask_arg(sysdecode_mount_flags, *ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_thr_create:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_mask_arg(sysdecode_thr_create_flags, *ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_thr_kill:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_signal(*ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
case SYS_kldunloadf:
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_integer_arg(sysdecode_kldunload_flags,
*ip);
Change kdump to print more useful information, i.e. it changes from 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,0,0x1b6) 32229 telnet CALL socket(0x2,0x2,0) to 32229 telnet CALL mmap(0,0x8000,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0,0,0) 32229 telnet CALL open(0x2807bc28,O_RDONLY,<unused>0x1b6) 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) David wanted to implement the suggestions which came up at the review from arch@ too, but real life rejected this proposal. So I commit what we already got and let another volunteer pick the remaining work from the ideas list. Submitted by: "David Kirchner" <dpk@dpk.net> Suggested by: FreeBSD ideas list page Reviewed by: arch
2006-05-20 14:27:22 +00:00
ip++;
narg--;
C has had swicth statements for 40 years or so. It's about time we started using them.
2011-10-08 11:51:48 +00:00
break;
kdump: Improve decoding of various *at calls: * Write AT_FDCWD where appropriate. * Decode the remaining arguments of openat() etc like open() etc.
2013-08-13 19:57:35 +00:00
case SYS_linkat:
case SYS_renameat:
case SYS_symlinkat:
print_number(ip, narg, c);
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_integer_arg_valid(sysdecode_atfd, *ip);
kdump: Improve decoding of various *at calls: * Write AT_FDCWD where appropriate. * Decode the remaining arguments of openat() etc like open() etc.
2013-08-13 19:57:35 +00:00
ip++;
narg--;
Decode pathconf() names, *at() flags, and sysarch() numbers in libsysdecode. Move tables that were previously in truss over to libsysdecode. truss output is unchanged, but kdump has been updated to decode these fields. In addition, sysdecode_sysarch_number() should support all platforms whereas the old table in truss only supported x86.
2017-09-04 05:34:36 +00:00
print_number(ip, narg, c);
kdump: Improve decoding of various *at calls: * Write AT_FDCWD where appropriate. * Decode the remaining arguments of openat() etc like open() etc.
2013-08-13 19:57:35 +00:00
break;
Merge Capsicum overhaul: - Capability is no longer separate descriptor type. Now every descriptor has set of its own capability rights. - The cap_new(2) system call is left, but it is no longer documented and should not be used in new code. - The new syscall cap_rights_limit(2) should be used instead of cap_new(2), which limits capability rights of the given descriptor without creating a new one. - The cap_getrights(2) syscall is renamed to cap_rights_get(2). - If CAP_IOCTL capability right is present we can further reduce allowed ioctls list with the new cap_ioctls_limit(2) syscall. List of allowed ioctls can be retrived with cap_ioctls_get(2) syscall. - If CAP_FCNTL capability right is present we can further reduce fcntls that can be used with the new cap_fcntls_limit(2) syscall and retrive them with cap_fcntls_get(2). - To support ioctl and fcntl white-listing the filedesc structure was heavly modified. - The audit subsystem, kdump and procstat tools were updated to recognize new syscalls. - Capability rights were revised and eventhough I tried hard to provide backward API and ABI compatibility there are some incompatible changes that are described in detail below: CAP_CREATE old behaviour: - Allow for openat(2)+O_CREAT. - Allow for linkat(2). - Allow for symlinkat(2). CAP_CREATE new behaviour: - Allow for openat(2)+O_CREAT. Added CAP_LINKAT: - Allow for linkat(2). ABI: Reuses CAP_RMDIR bit. - Allow to be target for renameat(2). Added CAP_SYMLINKAT: - Allow for symlinkat(2). Removed CAP_DELETE. Old behaviour: - Allow for unlinkat(2) when removing non-directory object. - Allow to be source for renameat(2). Removed CAP_RMDIR. Old behaviour: - Allow for unlinkat(2) when removing directory. Added CAP_RENAMEAT: - Required for source directory for the renameat(2) syscall. Added CAP_UNLINKAT (effectively it replaces CAP_DELETE and CAP_RMDIR): - Allow for unlinkat(2) on any object. - Required if target of renameat(2) exists and will be removed by this call. Removed CAP_MAPEXEC. CAP_MMAP old behaviour: - Allow for mmap(2) with any combination of PROT_NONE, PROT_READ and PROT_WRITE. CAP_MMAP new behaviour: - Allow for mmap(2)+PROT_NONE. Added CAP_MMAP_R: - Allow for mmap(PROT_READ). Added CAP_MMAP_W: - Allow for mmap(PROT_WRITE). Added CAP_MMAP_X: - Allow for mmap(PROT_EXEC). Added CAP_MMAP_RW: - Allow for mmap(PROT_READ | PROT_WRITE). Added CAP_MMAP_RX: - Allow for mmap(PROT_READ | PROT_EXEC). Added CAP_MMAP_WX: - Allow for mmap(PROT_WRITE | PROT_EXEC). Added CAP_MMAP_RWX: - Allow for mmap(PROT_READ | PROT_WRITE | PROT_EXEC). Renamed CAP_MKDIR to CAP_MKDIRAT. Renamed CAP_MKFIFO to CAP_MKFIFOAT. Renamed CAP_MKNODE to CAP_MKNODEAT. CAP_READ old behaviour: - Allow pread(2). - Disallow read(2), readv(2) (if there is no CAP_SEEK). CAP_READ new behaviour: - Allow read(2), readv(2). - Disallow pread(2) (CAP_SEEK was also required). CAP_WRITE old behaviour: - Allow pwrite(2). - Disallow write(2), writev(2) (if there is no CAP_SEEK). CAP_WRITE new behaviour: - Allow write(2), writev(2). - Disallow pwrite(2) (CAP_SEEK was also required). Added convinient defines: #define CAP_PREAD (CAP_SEEK | CAP_READ) #define CAP_PWRITE (CAP_SEEK | CAP_WRITE) #define CAP_MMAP_R (CAP_MMAP | CAP_SEEK | CAP_READ) #define CAP_MMAP_W (CAP_MMAP | CAP_SEEK | CAP_WRITE) #define CAP_MMAP_X (CAP_MMAP | CAP_SEEK | 0x0000000000000008ULL) #define CAP_MMAP_RW (CAP_MMAP_R | CAP_MMAP_W) #define CAP_MMAP_RX (CAP_MMAP_R | CAP_MMAP_X) #define CAP_MMAP_WX (CAP_MMAP_W | CAP_MMAP_X) #define CAP_MMAP_RWX (CAP_MMAP_R | CAP_MMAP_W | CAP_MMAP_X) #define CAP_RECV CAP_READ #define CAP_SEND CAP_WRITE #define CAP_SOCK_CLIENT \ (CAP_CONNECT | CAP_GETPEERNAME | CAP_GETSOCKNAME | CAP_GETSOCKOPT | \ CAP_PEELOFF | CAP_RECV | CAP_SEND | CAP_SETSOCKOPT | CAP_SHUTDOWN) #define CAP_SOCK_SERVER \ (CAP_ACCEPT | CAP_BIND | CAP_GETPEERNAME | CAP_GETSOCKNAME | \ CAP_GETSOCKOPT | CAP_LISTEN | CAP_PEELOFF | CAP_RECV | CAP_SEND | \ CAP_SETSOCKOPT | CAP_SHUTDOWN) Added defines for backward API compatibility: #define CAP_MAPEXEC CAP_MMAP_X #define CAP_DELETE CAP_UNLINKAT #define CAP_MKDIR CAP_MKDIRAT #define CAP_RMDIR CAP_UNLINKAT #define CAP_MKFIFO CAP_MKFIFOAT #define CAP_MKNOD CAP_MKNODAT #define CAP_SOCK_ALL (CAP_SOCK_CLIENT | CAP_SOCK_SERVER) Sponsored by: The FreeBSD Foundation Reviewed by: Christoph Mallon <christoph.mallon@gmx.de> Many aspects discussed with: rwatson, benl, jonathan ABI compatibility discussed with: kib
2013-03-02 00:53:12 +00:00
case SYS_cap_fcntls_limit:
print_number(ip, narg, c);
putchar(',');
arg = *ip;
ip++;
narg--;
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_mask_arg32(sysdecode_cap_fcntlrights, arg);
Merge Capsicum overhaul: - Capability is no longer separate descriptor type. Now every descriptor has set of its own capability rights. - The cap_new(2) system call is left, but it is no longer documented and should not be used in new code. - The new syscall cap_rights_limit(2) should be used instead of cap_new(2), which limits capability rights of the given descriptor without creating a new one. - The cap_getrights(2) syscall is renamed to cap_rights_get(2). - If CAP_IOCTL capability right is present we can further reduce allowed ioctls list with the new cap_ioctls_limit(2) syscall. List of allowed ioctls can be retrived with cap_ioctls_get(2) syscall. - If CAP_FCNTL capability right is present we can further reduce fcntls that can be used with the new cap_fcntls_limit(2) syscall and retrive them with cap_fcntls_get(2). - To support ioctl and fcntl white-listing the filedesc structure was heavly modified. - The audit subsystem, kdump and procstat tools were updated to recognize new syscalls. - Capability rights were revised and eventhough I tried hard to provide backward API and ABI compatibility there are some incompatible changes that are described in detail below: CAP_CREATE old behaviour: - Allow for openat(2)+O_CREAT. - Allow for linkat(2). - Allow for symlinkat(2). CAP_CREATE new behaviour: - Allow for openat(2)+O_CREAT. Added CAP_LINKAT: - Allow for linkat(2). ABI: Reuses CAP_RMDIR bit. - Allow to be target for renameat(2). Added CAP_SYMLINKAT: - Allow for symlinkat(2). Removed CAP_DELETE. Old behaviour: - Allow for unlinkat(2) when removing non-directory object. - Allow to be source for renameat(2). Removed CAP_RMDIR. Old behaviour: - Allow for unlinkat(2) when removing directory. Added CAP_RENAMEAT: - Required for source directory for the renameat(2) syscall. Added CAP_UNLINKAT (effectively it replaces CAP_DELETE and CAP_RMDIR): - Allow for unlinkat(2) on any object. - Required if target of renameat(2) exists and will be removed by this call. Removed CAP_MAPEXEC. CAP_MMAP old behaviour: - Allow for mmap(2) with any combination of PROT_NONE, PROT_READ and PROT_WRITE. CAP_MMAP new behaviour: - Allow for mmap(2)+PROT_NONE. Added CAP_MMAP_R: - Allow for mmap(PROT_READ). Added CAP_MMAP_W: - Allow for mmap(PROT_WRITE). Added CAP_MMAP_X: - Allow for mmap(PROT_EXEC). Added CAP_MMAP_RW: - Allow for mmap(PROT_READ | PROT_WRITE). Added CAP_MMAP_RX: - Allow for mmap(PROT_READ | PROT_EXEC). Added CAP_MMAP_WX: - Allow for mmap(PROT_WRITE | PROT_EXEC). Added CAP_MMAP_RWX: - Allow for mmap(PROT_READ | PROT_WRITE | PROT_EXEC). Renamed CAP_MKDIR to CAP_MKDIRAT. Renamed CAP_MKFIFO to CAP_MKFIFOAT. Renamed CAP_MKNODE to CAP_MKNODEAT. CAP_READ old behaviour: - Allow pread(2). - Disallow read(2), readv(2) (if there is no CAP_SEEK). CAP_READ new behaviour: - Allow read(2), readv(2). - Disallow pread(2) (CAP_SEEK was also required). CAP_WRITE old behaviour: - Allow pwrite(2). - Disallow write(2), writev(2) (if there is no CAP_SEEK). CAP_WRITE new behaviour: - Allow write(2), writev(2). - Disallow pwrite(2) (CAP_SEEK was also required). Added convinient defines: #define CAP_PREAD (CAP_SEEK | CAP_READ) #define CAP_PWRITE (CAP_SEEK | CAP_WRITE) #define CAP_MMAP_R (CAP_MMAP | CAP_SEEK | CAP_READ) #define CAP_MMAP_W (CAP_MMAP | CAP_SEEK | CAP_WRITE) #define CAP_MMAP_X (CAP_MMAP | CAP_SEEK | 0x0000000000000008ULL) #define CAP_MMAP_RW (CAP_MMAP_R | CAP_MMAP_W) #define CAP_MMAP_RX (CAP_MMAP_R | CAP_MMAP_X) #define CAP_MMAP_WX (CAP_MMAP_W | CAP_MMAP_X) #define CAP_MMAP_RWX (CAP_MMAP_R | CAP_MMAP_W | CAP_MMAP_X) #define CAP_RECV CAP_READ #define CAP_SEND CAP_WRITE #define CAP_SOCK_CLIENT \ (CAP_CONNECT | CAP_GETPEERNAME | CAP_GETSOCKNAME | CAP_GETSOCKOPT | \ CAP_PEELOFF | CAP_RECV | CAP_SEND | CAP_SETSOCKOPT | CAP_SHUTDOWN) #define CAP_SOCK_SERVER \ (CAP_ACCEPT | CAP_BIND | CAP_GETPEERNAME | CAP_GETSOCKNAME | \ CAP_GETSOCKOPT | CAP_LISTEN | CAP_PEELOFF | CAP_RECV | CAP_SEND | \ CAP_SETSOCKOPT | CAP_SHUTDOWN) Added defines for backward API compatibility: #define CAP_MAPEXEC CAP_MMAP_X #define CAP_DELETE CAP_UNLINKAT #define CAP_MKDIR CAP_MKDIRAT #define CAP_RMDIR CAP_UNLINKAT #define CAP_MKFIFO CAP_MKFIFOAT #define CAP_MKNOD CAP_MKNODAT #define CAP_SOCK_ALL (CAP_SOCK_CLIENT | CAP_SOCK_SERVER) Sponsored by: The FreeBSD Foundation Reviewed by: Christoph Mallon <christoph.mallon@gmx.de> Many aspects discussed with: rwatson, benl, jonathan ABI compatibility discussed with: kib
2013-03-02 00:53:12 +00:00
break;
Pretty-print the advice constants passed to posix_fadvise(2). MFC after: 2 weeks
2012-02-23 21:08:21 +00:00
case SYS_posix_fadvise:
Fix style in previous commit. Submitted by: bde
2012-02-24 20:12:53 +00:00
print_number(ip, narg, c);
print_number(ip, narg, c);
print_number(ip, narg, c);
Pretty-print the advice constants passed to posix_fadvise(2). MFC after: 2 weeks
2012-02-23 21:08:21 +00:00
(void)putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_integer_arg(sysdecode_fadvice, *ip);
Pretty-print the advice constants passed to posix_fadvise(2). MFC after: 2 weeks
2012-02-23 21:08:21 +00:00
ip++;
narg--;
break;
Extend the support for exempting processes from being killed when swap is exhausted. - Add a new protect(1) command that can be used to set or revoke protection from arbitrary processes. Similar to ktrace it can apply a change to all existing descendants of a process as well as future descendants. - Add a new procctl(2) system call that provides a generic interface for control operations on processes (as opposed to the debugger-specific operations provided by ptrace(2)). procctl(2) uses a combination of idtype_t and an id to identify the set of processes on which to operate similar to wait6(). - Add a PROC_SPROTECT control operation to manage the protection status of a set of processes. MADV_PROTECT still works for backwards compatability. - Add a p_flag2 to struct proc (and a corresponding ki_flag2 to kinfo_proc) the first bit of which is used to track if P_PROTECT should be inherited by new child processes. Reviewed by: kib, jilles (earlier version) Approved by: re (delphij) MFC after: 1 month
2013-09-19 18:53:42 +00:00
case SYS_procctl:
putchar('(');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_integer_arg(sysdecode_idtype, *ip);
Extend the support for exempting processes from being killed when swap is exhausted. - Add a new protect(1) command that can be used to set or revoke protection from arbitrary processes. Similar to ktrace it can apply a change to all existing descendants of a process as well as future descendants. - Add a new procctl(2) system call that provides a generic interface for control operations on processes (as opposed to the debugger-specific operations provided by ptrace(2)). procctl(2) uses a combination of idtype_t and an id to identify the set of processes on which to operate similar to wait6(). - Add a PROC_SPROTECT control operation to manage the protection status of a set of processes. MADV_PROTECT still works for backwards compatability. - Add a p_flag2 to struct proc (and a corresponding ki_flag2 to kinfo_proc) the first bit of which is used to track if P_PROTECT should be inherited by new child processes. Reviewed by: kib, jilles (earlier version) Approved by: re (delphij) MFC after: 1 month
2013-09-19 18:53:42 +00:00
c = ',';
ip++;
narg--;
Handle 64-bit system call arguments (off_t, id_t). In particular, 64-bit system call arguments use up two register_t arguments for 32-bit processes. They must also be aligned on a 64-bit boundary on 32-bit powerpc processes. This fixes the decoding of lseek(), procctl(), and wait6() arguments for 32-bit processes (both native and via freebsd32). Note that the ktrace system call return record only returns a single register, so the return value of lseek is always truncated to the low 32-bits for 32-bit processes.
2016-10-01 22:01:41 +00:00
print_number64(first, ip, narg, c);
Extend the support for exempting processes from being killed when swap is exhausted. - Add a new protect(1) command that can be used to set or revoke protection from arbitrary processes. Similar to ktrace it can apply a change to all existing descendants of a process as well as future descendants. - Add a new procctl(2) system call that provides a generic interface for control operations on processes (as opposed to the debugger-specific operations provided by ptrace(2)). procctl(2) uses a combination of idtype_t and an id to identify the set of processes on which to operate similar to wait6(). - Add a PROC_SPROTECT control operation to manage the protection status of a set of processes. MADV_PROTECT still works for backwards compatability. - Add a p_flag2 to struct proc (and a corresponding ki_flag2 to kinfo_proc) the first bit of which is used to track if P_PROTECT should be inherited by new child processes. Reviewed by: kib, jilles (earlier version) Approved by: re (delphij) MFC after: 1 month
2013-09-19 18:53:42 +00:00
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_integer_arg(sysdecode_procctl_cmd, *ip);
Extend the support for exempting processes from being killed when swap is exhausted. - Add a new protect(1) command that can be used to set or revoke protection from arbitrary processes. Similar to ktrace it can apply a change to all existing descendants of a process as well as future descendants. - Add a new procctl(2) system call that provides a generic interface for control operations on processes (as opposed to the debugger-specific operations provided by ptrace(2)). procctl(2) uses a combination of idtype_t and an id to identify the set of processes on which to operate similar to wait6(). - Add a PROC_SPROTECT control operation to manage the protection status of a set of processes. MADV_PROTECT still works for backwards compatability. - Add a p_flag2 to struct proc (and a corresponding ki_flag2 to kinfo_proc) the first bit of which is used to track if P_PROTECT should be inherited by new child processes. Reviewed by: kib, jilles (earlier version) Approved by: re (delphij) MFC after: 1 month
2013-09-19 18:53:42 +00:00
ip++;
narg--;
break;
kdump/truss: decode new _umtx_op flags In both cases, print the flag bits first followed by the command. Output now looks something like this: (ktrace) _umtx_op(0x8605f7008,0xf<UMTX_OP_WAIT_UINT_PRIVATE>,0,0,0) _umtx_op(0x9fffdce8,0x80000003<UMTX_OP__32BIT|UMTX_OP_WAKE>,0x1,0,0) (truss) _umtx_op(0x7fffffffda50,UMTX_OP_WAKE,0x1,0x0,0x0) = 0 (0x0) _umtx_op(0x9fffdd08,UMTX_OP__32BIT|UMTX_OP_WAKE,0x1,0x0,0x0) = 0 (0x0) Reviewed by: kib Differential Revision: https://reviews.freebsd.org/D27325
2020-12-09 03:24:09 +00:00
case SYS__umtx_op: {
int op;
Decode the arguments passed to _umtx_op(). In particular, decode the opcode. MFC after: 1 week Sponsored by: Norse
2014-10-13 16:37:06 +00:00
print_number(ip, narg, c);
putchar(',');
kdump/truss: decode new _umtx_op flags In both cases, print the flag bits first followed by the command. Output now looks something like this: (ktrace) _umtx_op(0x8605f7008,0xf<UMTX_OP_WAIT_UINT_PRIVATE>,0,0,0) _umtx_op(0x9fffdce8,0x80000003<UMTX_OP__32BIT|UMTX_OP_WAKE>,0x1,0,0) (truss) _umtx_op(0x7fffffffda50,UMTX_OP_WAKE,0x1,0x0,0x0) = 0 (0x0) _umtx_op(0x9fffdd08,UMTX_OP__32BIT|UMTX_OP_WAKE,0x1,0x0,0x0) = 0 (0x0) Reviewed by: kib Differential Revision: https://reviews.freebsd.org/D27325
2020-12-09 03:24:09 +00:00
if (print_mask_arg_part(sysdecode_umtx_op_flags,
*ip, &op))
putchar('|');
print_integer_arg(sysdecode_umtx_op, op);
putchar('>');
Decode the arguments passed to _umtx_op(). In particular, decode the opcode. MFC after: 1 week Sponsored by: Norse
2014-10-13 16:37:06 +00:00
switch (*ip) {
case UMTX_OP_CV_WAIT:
ip++;
narg--;
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_mask_argul(
sysdecode_umtx_cvwait_flags, *ip);
Decode the arguments passed to _umtx_op(). In particular, decode the opcode. MFC after: 1 week Sponsored by: Norse
2014-10-13 16:37:06 +00:00
break;
case UMTX_OP_RW_RDLOCK:
ip++;
narg--;
putchar(',');
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_mask_argul(
sysdecode_umtx_rwlock_flags, *ip);
Decode the arguments passed to _umtx_op(). In particular, decode the opcode. MFC after: 1 week Sponsored by: Norse
2014-10-13 16:37:06 +00:00
break;
}
ip++;
narg--;
Decode arguments to truncate and ftruncate. In particular, decode the off_t argument as a 64-bit argument to fix decoding for 32-bit processes.
2016-10-01 22:03:41 +00:00
break;
kdump/truss: decode new _umtx_op flags In both cases, print the flag bits first followed by the command. Output now looks something like this: (ktrace) _umtx_op(0x8605f7008,0xf<UMTX_OP_WAIT_UINT_PRIVATE>,0,0,0) _umtx_op(0x9fffdce8,0x80000003<UMTX_OP__32BIT|UMTX_OP_WAKE>,0x1,0,0) (truss) _umtx_op(0x7fffffffda50,UMTX_OP_WAKE,0x1,0x0,0x0) = 0 (0x0) _umtx_op(0x9fffdd08,UMTX_OP__32BIT|UMTX_OP_WAKE,0x1,0x0,0x0) = 0 (0x0) Reviewed by: kib Differential Revision: https://reviews.freebsd.org/D27325
2020-12-09 03:24:09 +00:00
}
Decode arguments to truncate and ftruncate. In particular, decode the off_t argument as a 64-bit argument to fix decoding for 32-bit processes.
2016-10-01 22:03:41 +00:00
case SYS_ftruncate:
case SYS_truncate:
print_number(ip, narg, c);
print_number64(first, ip, narg, c);
break;
Decode pathconf() names, *at() flags, and sysarch() numbers in libsysdecode. Move tables that were previously in truss over to libsysdecode. truss output is unchanged, but kdump has been updated to decode these fields. In addition, sysdecode_sysarch_number() should support all platforms whereas the old table in truss only supported x86.
2017-09-04 05:34:36 +00:00
case SYS_fchownat:
print_number(ip, narg, c);
print_number(ip, narg, c);
print_number(ip, narg, c);
break;
case SYS_fstatat:
case SYS_utimensat:
print_number(ip, narg, c);
print_number(ip, narg, c);
break;
case SYS_unlinkat:
print_number(ip, narg, c);
break;
case SYS_sysarch:
putchar('(');
print_integer_arg(sysdecode_sysarch_number, *ip);
ip++;
narg--;
c = ',';
break;
}
switch (ktr->ktr_code) {
case SYS_chflagsat:
case SYS_fchownat:
case SYS_faccessat:
case SYS_fchmodat:
case SYS_fstatat:
case SYS_linkat:
case SYS_unlinkat:
case SYS_utimensat:
putchar(',');
print_mask_arg0(sysdecode_atflags, *ip);
ip++;
narg--;
break;
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
}
}
Use a safety belt for cases where corrupted narg can be passed to the ktrsyscall(). print_number() does decrement the number of arguments, leading to infinite loops for negative values. Reported by: Patrick Lamaiziere <patpr at davenulle dot org>, Jonathan Pascal <jkpyvxmzsa at mailinator dot com> Submitted by: jh PR: bin/120055, kern/119564 MFC: 1 week
2009-11-07 21:46:34 +00:00
while (narg > 0) {
Whitespace.
2011-10-08 12:10:16 +00:00
print_number(ip, narg, c);
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
}
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(')');
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
}
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar('\n');
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
}
Improve WARNS situation for kdump: 1) Define _KERNEL while including sys/time.h to get some function prototypes. 2) Add prototypes and ANSIify definitions. 3) Constness changes. 4) Remove register keyword. 5) Actually return a sensible value from main. 6) Make fread_tail take a void * instead of a char *. 7) Avoid a signedness warning by casting to a size_t. Should be safe enough 'cos we also check for nonnegativity. 8) Be extra chummy with sigset_t rather than passing a struct to printf and pretending it is an int.
2002-07-28 16:05:07 +00:00
void
Add support to libsysdecode for decoding system call names. A new sysdecode_syscallname() function accepts a system call code and returns a string of the corresponding name (or NULL if the code is unknown). To support different process ABIs, the new function accepts a value from a new sysdecode_abi enum as its first argument to select the ABI in use. Current ABIs supported include FREEBSD (native binaries), FREEBSD32, LINUX, LINUX32, and CLOUDABI64. Note that not all ABIs are supported by all platforms. In general, a given ABI is only supported if a platform can execute binaries for that ABI. To simplify the implementation, libsysdecode's build reuses the existing pre-generated files from the kernel source tree rather than duplicating new copies of said files during the build. kdump(1) and truss(1) now use these functions to map system call identifiers to names. For kdump(1), a new 'syscallname()' function consolidates duplicated code from ktrsyscall() and ktrsyscallret(). The Linux ABI no longer requires custom handling for ktrsyscall() and linux_ktrsyscall() has been removed as a result. Reviewed by: bdrewery Differential Revision: https://reviews.freebsd.org/D4823
2016-01-26 19:07:09 +00:00
ktrsysret(struct ktr_sysret *ktr, u_int sv_flags)
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
{
Improve WARNS situation for kdump: 1) Define _KERNEL while including sys/time.h to get some function prototypes. 2) Add prototypes and ANSIify definitions. 3) Constness changes. 4) Remove register keyword. 5) Actually return a sensible value from main. 6) Make fread_tail take a void * instead of a char *. 7) Avoid a signedness warning by casting to a size_t. Should be safe enough 'cos we also check for nonnegativity. 8) Be extra chummy with sigset_t rather than passing a struct to printf and pretending it is an int.
2002-07-28 16:05:07 +00:00
register_t ret = ktr->ktr_retval;
int error = ktr->ktr_error;
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
Add support to libsysdecode for decoding system call names. A new sysdecode_syscallname() function accepts a system call code and returns a string of the corresponding name (or NULL if the code is unknown). To support different process ABIs, the new function accepts a value from a new sysdecode_abi enum as its first argument to select the ABI in use. Current ABIs supported include FREEBSD (native binaries), FREEBSD32, LINUX, LINUX32, and CLOUDABI64. Note that not all ABIs are supported by all platforms. In general, a given ABI is only supported if a platform can execute binaries for that ABI. To simplify the implementation, libsysdecode's build reuses the existing pre-generated files from the kernel source tree rather than duplicating new copies of said files during the build. kdump(1) and truss(1) now use these functions to map system call identifiers to names. For kdump(1), a new 'syscallname()' function consolidates duplicated code from ktrsyscall() and ktrsyscallret(). The Linux ABI no longer requires custom handling for ktrsyscall() and linux_ktrsyscall() has been removed as a result. Reviewed by: bdrewery Differential Revision: https://reviews.freebsd.org/D4823
2016-01-26 19:07:09 +00:00
syscallname(ktr->ktr_code, sv_flags);
printf(" ");
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
if (error == 0) {
if (fancy) {
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
printf("%ld", (long)ret);
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
if (ret < 0 || ret > 9)
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
printf("/%#lx", (unsigned long)ret);
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
} else {
if (decimal)
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
printf("%ld", (long)ret);
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
else
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
printf("%#lx", (unsigned long)ret);
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
}
} else if (error == ERESTART)
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
printf("RESTART");
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
else if (error == EJUSTRETURN)
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
printf("JUSTRETURN");
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
else {
Add handling for non-native error values to libsysdecode. Add two new functions, sysdecode_abi_to_freebsd_errno() and sysdecode_freebsd_to_abi_errno(), which convert errno values between the native FreeBSD ABI and other supported ABIs. Note that the mappings are not necessarily perfect meaning in some cases multiple errors in one ABI might map to a single error in another ABI. In that case, the reverse mapping will return one of the errors that maps, but which error is non-deterministic. Change truss to always report the raw error value to the user but use libsysdecode to map it to a native errno value that can be used with strerror() to generate a description. Previously truss reported the "converted" error value. Now the user will always see the exact error value that the application sees. Change kdump to report the truly raw error value to the user. Previously kdump would report the absolute value of the raw error value (so for Linux binaries it didn't output the FreeBSD error value, but the positive value of the Linux error). Now it reports the real (i.e. negative) error value for Linux binaries. Also, use libsysdecode to convert the native FreeBSD error reported in the ktrace record to the raw error used by the ABI. This means that the Linux ABI can now be handled directly in ktrsysret() and removes the need for linux_ktrsysret(). Reviewed by: bdrewery, kib Helpful notes: wblock (manpage) Differential Revision: https://reviews.freebsd.org/D5314
2016-02-23 20:00:55 +00:00
printf("-1 errno %d", sysdecode_freebsd_to_abi_errno(
syscallabi(sv_flags), error));
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
if (fancy)
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
printf(" %s", strerror(ktr->ktr_error));
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
}
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar('\n');
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
}
Improve WARNS situation for kdump: 1) Define _KERNEL while including sys/time.h to get some function prototypes. 2) Add prototypes and ANSIify definitions. 3) Constness changes. 4) Remove register keyword. 5) Actually return a sensible value from main. 6) Make fread_tail take a void * instead of a char *. 7) Avoid a signedness warning by casting to a size_t. Should be safe enough 'cos we also check for nonnegativity. 8) Be extra chummy with sigset_t rather than passing a struct to printf and pretending it is an int.
2002-07-28 16:05:07 +00:00
void
ktrnamei(char *cp, int len)
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
{
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
printf("\"%.*s\"\n", len, cp);
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
}
Improve WARNS situation for kdump: 1) Define _KERNEL while including sys/time.h to get some function prototypes. 2) Add prototypes and ANSIify definitions. 3) Constness changes. 4) Remove register keyword. 5) Actually return a sensible value from main. 6) Make fread_tail take a void * instead of a char *. 7) Avoid a signedness warning by casting to a size_t. Should be safe enough 'cos we also check for nonnegativity. 8) Be extra chummy with sigset_t rather than passing a struct to printf and pretending it is an int.
2002-07-28 16:05:07 +00:00
void
Two enhancements for kdump. 1) add a "-p pid", which is rather useful for selecting a single pid in a combined trace file (eg: with ktrace -i). 2) display binary genio data in a more precise format.
2003-06-03 01:44:43 +00:00
hexdump(char *p, int len, int screenwidth)
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
{
Two enhancements for kdump. 1) add a "-p pid", which is rather useful for selecting a single pid in a combined trace file (eg: with ktrace -i). 2) display binary genio data in a more precise format.
2003-06-03 01:44:43 +00:00
int n, i;
Improve WARNS situation for kdump: 1) Define _KERNEL while including sys/time.h to get some function prototypes. 2) Add prototypes and ANSIify definitions. 3) Constness changes. 4) Remove register keyword. 5) Actually return a sensible value from main. 6) Make fread_tail take a void * instead of a char *. 7) Avoid a signedness warning by casting to a size_t. Should be safe enough 'cos we also check for nonnegativity. 8) Be extra chummy with sigset_t rather than passing a struct to printf and pretending it is an int.
2002-07-28 16:05:07 +00:00
int width;
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
Two enhancements for kdump. 1) add a "-p pid", which is rather useful for selecting a single pid in a combined trace file (eg: with ktrace -i). 2) display binary genio data in a more precise format.
2003-06-03 01:44:43 +00:00
width = 0;
do {
width += 2;
i = 13; /* base offset */
i += (width / 2) + 1; /* spaces every second byte */
i += (width * 2); /* width of bytes */
i += 3; /* " |" */
i += width; /* each byte */
i += 1; /* "|" */
} while (i < screenwidth);
width -= 2;
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
Two enhancements for kdump. 1) add a "-p pid", which is rather useful for selecting a single pid in a combined trace file (eg: with ktrace -i). 2) display binary genio data in a more precise format.
2003-06-03 01:44:43 +00:00
for (n = 0; n < len; n += width) {
for (i = n; i < n + width; i++) {
if ((i % width) == 0) { /* beginning of line */
printf(" 0x%04x", i);
}
if ((i % 2) == 0) {
printf(" ");
}
if (i < len)
printf("%02x", p[i] & 0xff);
else
printf(" ");
}
printf(" |");
for (i = n; i < n + width; i++) {
if (i >= len)
break;
if (p[i] >= ' ' && p[i] <= '~')
printf("%c", p[i]);
else
printf(".");
}
printf("|\n");
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
}
Two enhancements for kdump. 1) add a "-p pid", which is rather useful for selecting a single pid in a combined trace file (eg: with ktrace -i). 2) display binary genio data in a more precise format.
2003-06-03 01:44:43 +00:00
if ((i % width) != 0)
printf("\n");
}
void
visdump(char *dp, int datalen, int screenwidth)
{
int col = 0;
char *cp;
int width;
char visbuf[5];
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
printf(" \"");
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
col = 8;
for (;datalen > 0; datalen--, dp++) {
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
vis(visbuf, *dp, VIS_CSTYLE, *(dp+1));
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
cp = visbuf;
/*
* Keep track of printables and
* space chars (like fold(1)).
*/
if (col == 0) {
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar('\t');
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
col = 8;
}
switch(*cp) {
case '\n':
col = 0;
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar('\n');
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
continue;
case '\t':
width = 8 - (col&07);
break;
default:
width = strlen(cp);
}
if (col + width > (screenwidth-2)) {
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
printf("\\\n\t");
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
col = 8;
}
col += width;
do {
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
putchar(*cp++);
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
} while (*cp);
}
if (col == 0)
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
printf(" ");
printf("\"\n");
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
}
Two enhancements for kdump. 1) add a "-p pid", which is rather useful for selecting a single pid in a combined trace file (eg: with ktrace -i). 2) display binary genio data in a more precise format.
2003-06-03 01:44:43 +00:00
void
ktrgenio(struct ktr_genio *ktr, int len)
{
int datalen = len - sizeof (struct ktr_genio);
char *dp = (char *)ktr + sizeof (struct ktr_genio);
static int screenwidth = 0;
int i, binary;
MFp4 @229085: Rearrange the code so we don't call ioctl(TIOCGWINSZ) if the -s option is given, as the result won't be used then. Sponsored by: The FreeBSD Foundation
2013-05-28 21:21:46 +00:00
printf("fd %d %s %d byte%s\n", ktr->ktr_fd,
ktr->ktr_rw == UIO_READ ? "read" : "wrote", datalen,
datalen == 1 ? "" : "s");
if (suppressdata)
return;
Two enhancements for kdump. 1) add a "-p pid", which is rather useful for selecting a single pid in a combined trace file (eg: with ktrace -i). 2) display binary genio data in a more precise format.
2003-06-03 01:44:43 +00:00
if (screenwidth == 0) {
struct winsize ws;
if (fancy && ioctl(fileno(stderr), TIOCGWINSZ, &ws) != -1 &&
ws.ws_col > 8)
screenwidth = ws.ws_col;
else
screenwidth = 80;
}
if (maxdata && datalen > maxdata)
datalen = maxdata;
for (i = 0, binary = 0; i < datalen && binary == 0; i++) {
if (dp[i] >= 32 && dp[i] < 127)
continue;
if (dp[i] == 10 || dp[i] == 13 || dp[i] == 0 || dp[i] == 9)
continue;
binary = 1;
}
if (binary)
hexdump(dp, datalen, screenwidth);
else
visdump(dp, datalen, screenwidth);
}
Improve WARNS situation for kdump: 1) Define _KERNEL while including sys/time.h to get some function prototypes. 2) Add prototypes and ANSIify definitions. 3) Constness changes. 4) Remove register keyword. 5) Actually return a sensible value from main. 6) Make fread_tail take a void * instead of a char *. 7) Avoid a signedness warning by casting to a size_t. Should be safe enough 'cos we also check for nonnegativity. 8) Be extra chummy with sigset_t rather than passing a struct to printf and pretending it is an int.
2002-07-28 16:05:07 +00:00
void
Teach kdump to decode linux syscalls names too. Fix bug introduced in my previous commit: the kernel always dump native signal numbers, so no need to check the ABI in ktrpsig(). Suggested by: jhb MFC after: 1 Month.
2011-03-01 16:42:28 +00:00
ktrpsig(struct ktr_psig *psig)
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
{
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
const char *str;
print_signal(psig->signo);
kdump: Pretty-print signal codes. MFC after: 1 week
2012-09-22 12:40:00 +00:00
if (psig->action == SIG_DFL) {
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
printf(" SIG_DFL");
kdump: Pretty-print signal codes. MFC after: 1 week
2012-09-22 12:40:00 +00:00
} else {
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
printf(" caught handler=0x%lx mask=0x%x",
kdump: Pretty-print signal codes. MFC after: 1 week
2012-09-22 12:40:00 +00:00
(u_long)psig->action, psig->mask.__bits[0]);
Improve WARNS situation for kdump: 1) Define _KERNEL while including sys/time.h to get some function prototypes. 2) Add prototypes and ANSIify definitions. 3) Constness changes. 4) Remove register keyword. 5) Actually return a sensible value from main. 6) Make fread_tail take a void * instead of a char *. 7) Avoid a signedness warning by casting to a size_t. Should be safe enough 'cos we also check for nonnegativity. 8) Be extra chummy with sigset_t rather than passing a struct to printf and pretending it is an int.
2002-07-28 16:05:07 +00:00
}
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
printf(" code=");
str = sysdecode_sigcode(psig->signo, psig->code);
if (str != NULL)
printf("%s", str);
else
printf("<invalid=%#x>", psig->code);
putchar('\n');
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
}
Improve WARNS situation for kdump: 1) Define _KERNEL while including sys/time.h to get some function prototypes. 2) Add prototypes and ANSIify definitions. 3) Constness changes. 4) Remove register keyword. 5) Actually return a sensible value from main. 6) Make fread_tail take a void * instead of a char *. 7) Avoid a signedness warning by casting to a size_t. Should be safe enough 'cos we also check for nonnegativity. 8) Be extra chummy with sigset_t rather than passing a struct to printf and pretending it is an int.
2002-07-28 16:05:07 +00:00
void
Include the associated wait channel message for context switch ktrace records. kdump supports both the old and new messages. Submitted by: Andrey Zonov andrey zonov org MFC after: 1 week
2012-04-20 15:32:36 +00:00
ktrcsw_old(struct ktr_csw_old *cs)
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
{
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
printf("%s %s\n", cs->out ? "stop" : "resume",
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
cs->user ? "user" : "kernel");
}
Include the associated wait channel message for context switch ktrace records. kdump supports both the old and new messages. Submitted by: Andrey Zonov andrey zonov org MFC after: 1 week
2012-04-20 15:32:36 +00:00
void
ktrcsw(struct ktr_csw *cs)
{
printf("%s %s \"%s\"\n", cs->out ? "stop" : "resume",
cs->user ? "user" : "kernel", cs->wmesg);
}
Improve WARNS situation for kdump: 1) Define _KERNEL while including sys/time.h to get some function prototypes. 2) Add prototypes and ANSIify definitions. 3) Constness changes. 4) Remove register keyword. 5) Actually return a sensible value from main. 6) Make fread_tail take a void * instead of a char *. 7) Avoid a signedness warning by casting to a size_t. Should be safe enough 'cos we also check for nonnegativity. 8) Be extra chummy with sigset_t rather than passing a struct to printf and pretending it is an int.
2002-07-28 16:05:07 +00:00
void
Fix most of the warnings in kdump(1). Sponsored by: Norse
2014-10-13 16:17:42 +00:00
ktruser(int len, void *p)
For now we just hexdump the stuff in USER records.
1996-09-19 19:50:45 +00:00
{
Fix most of the warnings in kdump(1). Sponsored by: Norse
2014-10-13 16:17:42 +00:00
unsigned char *cp;
Add code to parse the utrace(2) entries generated by malloc(3) in a more human-readable format. Note that we report 'realloc(p, 0)' as 'free(p)' since both cases are encoded the same way and 'free()' is more common than a realloc() to 0. MFC after: 1 week
2007-01-05 21:04:37 +00:00
Start on a new library (libsysdecode) that provides routines for decoding system call information such as system call arguments. Initially this will consist of pulling duplicated code out of truss and kdump though it may prove useful for other utilities in the future. This commit moves the shared utrace(2) record parser out of kdump into the library and updates kdump and truss to use it. One difference from the previous version is that the library version treats unknown events that start with the "RTLD" signature as unknown events. This simplifies the interface and allows the consumer to decide how to handle all non-recognized events. Instead, this function only generates a string description for known malloc() and RTLD records. Reviewed by: bdrewery Differential Revision: https://reviews.freebsd.org/D4537
2015-12-15 00:05:07 +00:00
if (sysdecode_utrace(stdout, p, len)) {
truss: Add support for utrace(2). This uses the kdump(1) utrace support code directly until a common library is created. This allows malloc(3) tracing with MALLOC_CONF=utrace:true and rtld tracing with LD_UTRACE=1. Unknown utrace(2) data is just printed as hex. PR: 43819 [inspired by] Reviewed by: jhb MFC after: 2 weeks Relnotes: yes Differential Revision: https://reviews.freebsd.org/D3819
2015-10-06 21:58:38 +00:00
printf("\n");
Add code to parse the utrace(2) entries generated by malloc(3) in a more human-readable format. Note that we report 'realloc(p, 0)' as 'free(p)' since both cases are encoded the same way and 'free()' is more common than a realloc() to 0. MFC after: 1 week
2007-01-05 21:04:37 +00:00
return;
}
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
printf("%d ", len);
Fix most of the warnings in kdump(1). Sponsored by: Norse
2014-10-13 16:17:42 +00:00
cp = p;
Remove length field from utrace entries.
1996-09-22 18:18:20 +00:00
while (len--)
Respect decimal flag when dumping USER type records.
2004-03-25 12:33:55 +00:00
if (decimal)
Fix most of the warnings in kdump(1). Sponsored by: Norse
2014-10-13 16:17:42 +00:00
printf(" %d", *cp++);
Respect decimal flag when dumping USER type records.
2004-03-25 12:33:55 +00:00
else
Fix most of the warnings in kdump(1). Sponsored by: Norse
2014-10-13 16:17:42 +00:00
printf(" %02x", *cp++);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
printf("\n");
For now we just hexdump the stuff in USER records.
1996-09-19 19:50:45 +00:00
}
Change the cap_rights_t type from uint64_t to a structure that we can extend in the future in a backward compatible (API and ABI) way. The cap_rights_t represents capability rights. We used to use one bit to represent one right, but we are running out of spare bits. Currently the new structure provides place for 114 rights (so 50 more than the previous cap_rights_t), but it is possible to grow the structure to hold at least 285 rights, although we can make it even larger if 285 rights won't be enough. The structure definition looks like this: struct cap_rights { uint64_t cr_rights[CAP_RIGHTS_VERSION + 2]; }; The initial CAP_RIGHTS_VERSION is 0. The top two bits in the first element of the cr_rights[] array contain total number of elements in the array - 2. This means if those two bits are equal to 0, we have 2 array elements. The top two bits in all remaining array elements should be 0. The next five bits in all array elements contain array index. Only one bit is used and bit position in this five-bits range defines array index. This means there can be at most five array elements in the future. To define new right the CAPRIGHT() macro must be used. The macro takes two arguments - an array index and a bit to set, eg. #define CAP_PDKILL CAPRIGHT(1, 0x0000000000000800ULL) We still support aliases that combine few rights, but the rights have to belong to the same array element, eg: #define CAP_LOOKUP CAPRIGHT(0, 0x0000000000000400ULL) #define CAP_FCHMOD CAPRIGHT(0, 0x0000000000002000ULL) #define CAP_FCHMODAT (CAP_FCHMOD | CAP_LOOKUP) There is new API to manage the new cap_rights_t structure: cap_rights_t *cap_rights_init(cap_rights_t *rights, ...); void cap_rights_set(cap_rights_t *rights, ...); void cap_rights_clear(cap_rights_t *rights, ...); bool cap_rights_is_set(const cap_rights_t *rights, ...); bool cap_rights_is_valid(const cap_rights_t *rights); void cap_rights_merge(cap_rights_t *dst, const cap_rights_t *src); void cap_rights_remove(cap_rights_t *dst, const cap_rights_t *src); bool cap_rights_contains(const cap_rights_t *big, const cap_rights_t *little); Capability rights to the cap_rights_init(), cap_rights_set(), cap_rights_clear() and cap_rights_is_set() functions are provided by separating them with commas, eg: cap_rights_t rights; cap_rights_init(&rights, CAP_READ, CAP_WRITE, CAP_FSTAT); There is no need to terminate the list of rights, as those functions are actually macros that take care of the termination, eg: #define cap_rights_set(rights, ...) \ __cap_rights_set((rights), __VA_ARGS__, 0ULL) void __cap_rights_set(cap_rights_t *rights, ...); Thanks to using one bit as an array index we can assert in those functions that there are no two rights belonging to different array elements provided together. For example this is illegal and will be detected, because CAP_LOOKUP belongs to element 0 and CAP_PDKILL to element 1: cap_rights_init(&rights, CAP_LOOKUP | CAP_PDKILL); Providing several rights that belongs to the same array's element this way is correct, but is not advised. It should only be used for aliases definition. This commit also breaks compatibility with some existing Capsicum system calls, but I see no other way to do that. This should be fine as Capsicum is still experimental and this change is not going to 9.x. Sponsored by: The FreeBSD Foundation
2013-09-05 00:09:56 +00:00
void
ktrcaprights(cap_rights_t *rightsp)
{
printf("cap_rights_t ");
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
sysdecode_cap_rights(stdout, rightsp);
Change the cap_rights_t type from uint64_t to a structure that we can extend in the future in a backward compatible (API and ABI) way. The cap_rights_t represents capability rights. We used to use one bit to represent one right, but we are running out of spare bits. Currently the new structure provides place for 114 rights (so 50 more than the previous cap_rights_t), but it is possible to grow the structure to hold at least 285 rights, although we can make it even larger if 285 rights won't be enough. The structure definition looks like this: struct cap_rights { uint64_t cr_rights[CAP_RIGHTS_VERSION + 2]; }; The initial CAP_RIGHTS_VERSION is 0. The top two bits in the first element of the cr_rights[] array contain total number of elements in the array - 2. This means if those two bits are equal to 0, we have 2 array elements. The top two bits in all remaining array elements should be 0. The next five bits in all array elements contain array index. Only one bit is used and bit position in this five-bits range defines array index. This means there can be at most five array elements in the future. To define new right the CAPRIGHT() macro must be used. The macro takes two arguments - an array index and a bit to set, eg. #define CAP_PDKILL CAPRIGHT(1, 0x0000000000000800ULL) We still support aliases that combine few rights, but the rights have to belong to the same array element, eg: #define CAP_LOOKUP CAPRIGHT(0, 0x0000000000000400ULL) #define CAP_FCHMOD CAPRIGHT(0, 0x0000000000002000ULL) #define CAP_FCHMODAT (CAP_FCHMOD | CAP_LOOKUP) There is new API to manage the new cap_rights_t structure: cap_rights_t *cap_rights_init(cap_rights_t *rights, ...); void cap_rights_set(cap_rights_t *rights, ...); void cap_rights_clear(cap_rights_t *rights, ...); bool cap_rights_is_set(const cap_rights_t *rights, ...); bool cap_rights_is_valid(const cap_rights_t *rights); void cap_rights_merge(cap_rights_t *dst, const cap_rights_t *src); void cap_rights_remove(cap_rights_t *dst, const cap_rights_t *src); bool cap_rights_contains(const cap_rights_t *big, const cap_rights_t *little); Capability rights to the cap_rights_init(), cap_rights_set(), cap_rights_clear() and cap_rights_is_set() functions are provided by separating them with commas, eg: cap_rights_t rights; cap_rights_init(&rights, CAP_READ, CAP_WRITE, CAP_FSTAT); There is no need to terminate the list of rights, as those functions are actually macros that take care of the termination, eg: #define cap_rights_set(rights, ...) \ __cap_rights_set((rights), __VA_ARGS__, 0ULL) void __cap_rights_set(cap_rights_t *rights, ...); Thanks to using one bit as an array index we can assert in those functions that there are no two rights belonging to different array elements provided together. For example this is illegal and will be detected, because CAP_LOOKUP belongs to element 0 and CAP_PDKILL to element 1: cap_rights_init(&rights, CAP_LOOKUP | CAP_PDKILL); Providing several rights that belongs to the same array's element this way is correct, but is not advised. It should only be used for aliases definition. This commit also breaks compatibility with some existing Capsicum system calls, but I see no other way to do that. This should be fine as Capsicum is still experimental and this change is not going to 9.x. Sponsored by: The FreeBSD Foundation
2013-09-05 00:09:56 +00:00
printf("\n");
}
Trace timeval parameters to the getitimer(2) and setitimer(2) syscalls. Reviewed by: jhb Tested by: pho Sponsored by: The FreeBSD Foundation MFC after: 1 week Differential revision: https://reviews.freebsd.org/D7158
2016-07-13 14:37:58 +00:00
static void
ktrtimeval(struct timeval *tv)
{
printf("{%ld, %ld}", (long)tv->tv_sec, tv->tv_usec);
}
void
ktritimerval(struct itimerval *it)
{
printf("itimerval { .interval = ");
ktrtimeval(&it->it_interval);
printf(", .value = ");
ktrtimeval(&it->it_value);
printf(" }\n");
}
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
void
ktrsockaddr(struct sockaddr *sa)
{
/*
TODO: Support additional address families
- Trim an unused and bogus Makefile for mount_smbfs. - Reconnect with some minor modifications, in particular now selsocket() internals are adapted to use sbintime units after recent'ish calloutng switch.
2013-06-28 21:00:08 +00:00
#include <netsmb/netbios.h>
struct sockaddr_nb *nb;
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
*/
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
const char *str;
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
char addr[64];
/*
* note: ktrstruct() has already verified that sa points to a
* buffer at least sizeof(struct sockaddr) bytes long and exactly
* sa->sa_len bytes long.
*/
printf("struct sockaddr { ");
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
str = sysdecode_sockaddr_family(sa->sa_family);
if (str != NULL)
printf("%s", str);
else
printf("<invalid=%d>", sa->sa_family);
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
printf(", ");
#define check_sockaddr_len(n) \
Make kdump compile cleanly at WARNS level 6, with one exception: the ipfilter headers contain a duplicated function declaration. Turn off -Werror to allow kdump to compile in spite of this. It would be neat to be able to turn off -Werror on a file-by-file basis... PR: bin/161478 Submitted by: Garrett Cooper <yanegomi@gmail.com>
2011-10-12 21:45:12 +00:00
if (sa_##n.s##n##_len < sizeof(struct sockaddr_##n)) { \
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
printf("invalid"); \
break; \
}
switch(sa->sa_family) {
case AF_INET: {
Make kdump compile cleanly at WARNS level 6, with one exception: the ipfilter headers contain a duplicated function declaration. Turn off -Werror to allow kdump to compile in spite of this. It would be neat to be able to turn off -Werror on a file-by-file basis... PR: bin/161478 Submitted by: Garrett Cooper <yanegomi@gmail.com>
2011-10-12 21:45:12 +00:00
struct sockaddr_in sa_in;
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
Make kdump compile cleanly at WARNS level 6, with one exception: the ipfilter headers contain a duplicated function declaration. Turn off -Werror to allow kdump to compile in spite of this. It would be neat to be able to turn off -Werror on a file-by-file basis... PR: bin/161478 Submitted by: Garrett Cooper <yanegomi@gmail.com>
2011-10-12 21:45:12 +00:00
memset(&sa_in, 0, sizeof(sa_in));
- Use correct size of copying different socket structures. MFC after: 1 week
2013-02-12 18:48:11 +00:00
memcpy(&sa_in, sa, sa->sa_len);
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
check_sockaddr_len(in);
Make kdump compile cleanly at WARNS level 6, with one exception: the ipfilter headers contain a duplicated function declaration. Turn off -Werror to allow kdump to compile in spite of this. It would be neat to be able to turn off -Werror on a file-by-file basis... PR: bin/161478 Submitted by: Garrett Cooper <yanegomi@gmail.com>
2011-10-12 21:45:12 +00:00
inet_ntop(AF_INET, &sa_in.sin_addr, addr, sizeof addr);
printf("%s:%u", addr, ntohs(sa_in.sin_port));
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
break;
}
case AF_INET6: {
Make kdump compile cleanly at WARNS level 6, with one exception: the ipfilter headers contain a duplicated function declaration. Turn off -Werror to allow kdump to compile in spite of this. It would be neat to be able to turn off -Werror on a file-by-file basis... PR: bin/161478 Submitted by: Garrett Cooper <yanegomi@gmail.com>
2011-10-12 21:45:12 +00:00
struct sockaddr_in6 sa_in6;
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
Make kdump compile cleanly at WARNS level 6, with one exception: the ipfilter headers contain a duplicated function declaration. Turn off -Werror to allow kdump to compile in spite of this. It would be neat to be able to turn off -Werror on a file-by-file basis... PR: bin/161478 Submitted by: Garrett Cooper <yanegomi@gmail.com>
2011-10-12 21:45:12 +00:00
memset(&sa_in6, 0, sizeof(sa_in6));
- Use correct size of copying different socket structures. MFC after: 1 week
2013-02-12 18:48:11 +00:00
memcpy(&sa_in6, sa, sa->sa_len);
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
check_sockaddr_len(in6);
Use getnameinfo(3) instead of inet_ntop(3) to make printable versions of sockaddr_in6 structures. getnameinfo(3) does the same thing, but it is also able to represent a scope zone id as described in the RFC 4007. MFC after: 2 weeks
2013-06-07 08:06:48 +00:00
getnameinfo((struct sockaddr *)&sa_in6, sizeof(sa_in6),
addr, sizeof(addr), NULL, 0, NI_NUMERICHOST);
Make kdump compile cleanly at WARNS level 6, with one exception: the ipfilter headers contain a duplicated function declaration. Turn off -Werror to allow kdump to compile in spite of this. It would be neat to be able to turn off -Werror on a file-by-file basis... PR: bin/161478 Submitted by: Garrett Cooper <yanegomi@gmail.com>
2011-10-12 21:45:12 +00:00
printf("[%s]:%u", addr, htons(sa_in6.sin6_port));
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
break;
}
case AF_UNIX: {
Make kdump compile cleanly at WARNS level 6, with one exception: the ipfilter headers contain a duplicated function declaration. Turn off -Werror to allow kdump to compile in spite of this. It would be neat to be able to turn off -Werror on a file-by-file basis... PR: bin/161478 Submitted by: Garrett Cooper <yanegomi@gmail.com>
2011-10-12 21:45:12 +00:00
struct sockaddr_un sa_un;
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
Make kdump compile cleanly at WARNS level 6, with one exception: the ipfilter headers contain a duplicated function declaration. Turn off -Werror to allow kdump to compile in spite of this. It would be neat to be able to turn off -Werror on a file-by-file basis... PR: bin/161478 Submitted by: Garrett Cooper <yanegomi@gmail.com>
2011-10-12 21:45:12 +00:00
memset(&sa_un, 0, sizeof(sa_un));
- Use correct size of copying different socket structures. MFC after: 1 week
2013-02-12 18:48:11 +00:00
memcpy(&sa_un, sa, sa->sa_len);
Make kdump compile cleanly at WARNS level 6, with one exception: the ipfilter headers contain a duplicated function declaration. Turn off -Werror to allow kdump to compile in spite of this. It would be neat to be able to turn off -Werror on a file-by-file basis... PR: bin/161478 Submitted by: Garrett Cooper <yanegomi@gmail.com>
2011-10-12 21:45:12 +00:00
printf("%.*s", (int)sizeof(sa_un.sun_path), sa_un.sun_path);
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
break;
}
default:
printf("unknown address family");
}
printf(" }\n");
}
void
ktrstat(struct stat *statp)
{
char mode[12], timestr[PATH_MAX + 4];
struct passwd *pwd;
struct group *grp;
struct tm *tm;
/*
* note: ktrstruct() has already verified that statp points to a
* buffer exactly sizeof(struct stat) bytes long.
*/
printf("struct stat {");
When displaying a struct stat, if the -r option was not specified, display the numeric rather than symbolic representation of st_mode. Approved by: re (glebius) MFC after: 1 week
2013-10-07 11:23:01 +00:00
printf("dev=%ju, ino=%ju, ",
(uintmax_t)statp->st_dev, (uintmax_t)statp->st_ino);
Use 'bool' instead of 'int' for various boolean flags. Reviewed by: kib MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D16611
2018-08-18 20:23:53 +00:00
if (!resolv)
When displaying a struct stat, if the -r option was not specified, display the numeric rather than symbolic representation of st_mode. Approved by: re (glebius) MFC after: 1 week
2013-10-07 11:23:01 +00:00
printf("mode=0%jo, ", (uintmax_t)statp->st_mode);
else {
strmode(statp->st_mode, mode);
printf("mode=%s, ", mode);
}
printf("nlink=%ju, ", (uintmax_t)statp->st_nlink);
Use 'bool' instead of 'int' for various boolean flags. Reviewed by: kib MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D16611
2018-08-18 20:23:53 +00:00
if (!resolv) {
Make use of Casper's system.pwd and system.grp services when the -r option is given to convert uids and gids to user names and group names even when running in capability mode sandbox. While here log on stderr when we successfully enter the sandbox. Sponsored by: The FreeBSD Foundation
2013-12-15 23:09:05 +00:00
pwd = NULL;
} else {
We use a few different ifdef's names to check if we are using Casper or not, let's standardize this. Now we are always use WITH_CASPER name. Discussed with: emaste@ MFC after: 1 month
2017-09-21 14:41:41 +00:00
#ifdef WITH_CASPER
Make use of Casper's system.pwd and system.grp services when the -r option is given to convert uids and gids to user names and group names even when running in capability mode sandbox. While here log on stderr when we successfully enter the sandbox. Sponsored by: The FreeBSD Foundation
2013-12-15 23:09:05 +00:00
if (cappwd != NULL)
pwd = cap_getpwuid(cappwd, statp->st_uid);
else
#endif
pwd = getpwuid(statp->st_uid);
}
if (pwd == NULL)
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
printf("uid=%ju, ", (uintmax_t)statp->st_uid);
else
printf("uid=\"%s\", ", pwd->pw_name);
Use 'bool' instead of 'int' for various boolean flags. Reviewed by: kib MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D16611
2018-08-18 20:23:53 +00:00
if (!resolv) {
Make use of Casper's system.pwd and system.grp services when the -r option is given to convert uids and gids to user names and group names even when running in capability mode sandbox. While here log on stderr when we successfully enter the sandbox. Sponsored by: The FreeBSD Foundation
2013-12-15 23:09:05 +00:00
grp = NULL;
} else {
We use a few different ifdef's names to check if we are using Casper or not, let's standardize this. Now we are always use WITH_CASPER name. Discussed with: emaste@ MFC after: 1 month
2017-09-21 14:41:41 +00:00
#ifdef WITH_CASPER
Make use of Casper's system.pwd and system.grp services when the -r option is given to convert uids and gids to user names and group names even when running in capability mode sandbox. While here log on stderr when we successfully enter the sandbox. Sponsored by: The FreeBSD Foundation
2013-12-15 23:09:05 +00:00
if (capgrp != NULL)
grp = cap_getgrgid(capgrp, statp->st_gid);
else
#endif
grp = getgrgid(statp->st_gid);
}
if (grp == NULL)
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
printf("gid=%ju, ", (uintmax_t)statp->st_gid);
else
printf("gid=\"%s\", ", grp->gr_name);
printf("rdev=%ju, ", (uintmax_t)statp->st_rdev);
printf("atime=");
Use 'bool' instead of 'int' for various boolean flags. Reviewed by: kib MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D16611
2018-08-18 20:23:53 +00:00
if (!resolv)
Change all our own code to use st_*tim instead of st_*timespec. Also remove some local patches to diff(1) which are now unneeded.
2010-03-28 13:16:08 +00:00
printf("%jd", (intmax_t)statp->st_atim.tv_sec);
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
else {
Change all our own code to use st_*tim instead of st_*timespec. Also remove some local patches to diff(1) which are now unneeded.
2010-03-28 13:16:08 +00:00
tm = localtime(&statp->st_atim.tv_sec);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
strftime(timestr, sizeof(timestr), TIME_FORMAT, tm);
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
printf("\"%s\"", timestr);
}
Change all our own code to use st_*tim instead of st_*timespec. Also remove some local patches to diff(1) which are now unneeded.
2010-03-28 13:16:08 +00:00
if (statp->st_atim.tv_nsec != 0)
printf(".%09ld, ", statp->st_atim.tv_nsec);
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
else
printf(", ");
Fix humanized decoding of struct stat with respect to .st_mtim st_mtim was being incorrectly described as "stime=", not "mtime=". This was introduced with the original feature commit (r176471). MFC after: 1 week PR: 209699 Submitted by: naddy Sponsored by: EMC / Isilon Storage Division
2016-05-22 18:20:45 +00:00
printf("mtime=");
Use 'bool' instead of 'int' for various boolean flags. Reviewed by: kib MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D16611
2018-08-18 20:23:53 +00:00
if (!resolv)
Change all our own code to use st_*tim instead of st_*timespec. Also remove some local patches to diff(1) which are now unneeded.
2010-03-28 13:16:08 +00:00
printf("%jd", (intmax_t)statp->st_mtim.tv_sec);
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
else {
Change all our own code to use st_*tim instead of st_*timespec. Also remove some local patches to diff(1) which are now unneeded.
2010-03-28 13:16:08 +00:00
tm = localtime(&statp->st_mtim.tv_sec);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
strftime(timestr, sizeof(timestr), TIME_FORMAT, tm);
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
printf("\"%s\"", timestr);
}
Change all our own code to use st_*tim instead of st_*timespec. Also remove some local patches to diff(1) which are now unneeded.
2010-03-28 13:16:08 +00:00
if (statp->st_mtim.tv_nsec != 0)
printf(".%09ld, ", statp->st_mtim.tv_nsec);
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
else
printf(", ");
printf("ctime=");
Use 'bool' instead of 'int' for various boolean flags. Reviewed by: kib MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D16611
2018-08-18 20:23:53 +00:00
if (!resolv)
Change all our own code to use st_*tim instead of st_*timespec. Also remove some local patches to diff(1) which are now unneeded.
2010-03-28 13:16:08 +00:00
printf("%jd", (intmax_t)statp->st_ctim.tv_sec);
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
else {
Change all our own code to use st_*tim instead of st_*timespec. Also remove some local patches to diff(1) which are now unneeded.
2010-03-28 13:16:08 +00:00
tm = localtime(&statp->st_ctim.tv_sec);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
strftime(timestr, sizeof(timestr), TIME_FORMAT, tm);
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
printf("\"%s\"", timestr);
}
Change all our own code to use st_*tim instead of st_*timespec. Also remove some local patches to diff(1) which are now unneeded.
2010-03-28 13:16:08 +00:00
if (statp->st_ctim.tv_nsec != 0)
printf(".%09ld, ", statp->st_ctim.tv_nsec);
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
else
printf(", ");
printf("birthtime=");
Use 'bool' instead of 'int' for various boolean flags. Reviewed by: kib MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D16611
2018-08-18 20:23:53 +00:00
if (!resolv)
Change all our own code to use st_*tim instead of st_*timespec. Also remove some local patches to diff(1) which are now unneeded.
2010-03-28 13:16:08 +00:00
printf("%jd", (intmax_t)statp->st_birthtim.tv_sec);
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
else {
Change all our own code to use st_*tim instead of st_*timespec. Also remove some local patches to diff(1) which are now unneeded.
2010-03-28 13:16:08 +00:00
tm = localtime(&statp->st_birthtim.tv_sec);
I appreciate the logic behind using a (void) cast to indicate that the return value is intentionally ignored, but frankly, all it does is get in the way of the code. Also fix a few other incorrect casts, such as (void *)malloc(foo) and passing signed values to %x.
2011-10-08 12:28:06 +00:00
strftime(timestr, sizeof(timestr), TIME_FORMAT, tm);
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
printf("\"%s\"", timestr);
}
Change all our own code to use st_*tim instead of st_*timespec. Also remove some local patches to diff(1) which are now unneeded.
2010-03-28 13:16:08 +00:00
if (statp->st_birthtim.tv_nsec != 0)
printf(".%09ld, ", statp->st_birthtim.tv_nsec);
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
else
printf(", ");
printf("size=%jd, blksize=%ju, blocks=%jd, flags=0x%x",
(uintmax_t)statp->st_size, (uintmax_t)statp->st_blksize,
(intmax_t)statp->st_blocks, statp->st_flags);
printf(" }\n");
}
void
ktrstruct(char *buf, size_t buflen)
{
char *name, *data;
size_t namelen, datalen;
int i;
Change the cap_rights_t type from uint64_t to a structure that we can extend in the future in a backward compatible (API and ABI) way. The cap_rights_t represents capability rights. We used to use one bit to represent one right, but we are running out of spare bits. Currently the new structure provides place for 114 rights (so 50 more than the previous cap_rights_t), but it is possible to grow the structure to hold at least 285 rights, although we can make it even larger if 285 rights won't be enough. The structure definition looks like this: struct cap_rights { uint64_t cr_rights[CAP_RIGHTS_VERSION + 2]; }; The initial CAP_RIGHTS_VERSION is 0. The top two bits in the first element of the cr_rights[] array contain total number of elements in the array - 2. This means if those two bits are equal to 0, we have 2 array elements. The top two bits in all remaining array elements should be 0. The next five bits in all array elements contain array index. Only one bit is used and bit position in this five-bits range defines array index. This means there can be at most five array elements in the future. To define new right the CAPRIGHT() macro must be used. The macro takes two arguments - an array index and a bit to set, eg. #define CAP_PDKILL CAPRIGHT(1, 0x0000000000000800ULL) We still support aliases that combine few rights, but the rights have to belong to the same array element, eg: #define CAP_LOOKUP CAPRIGHT(0, 0x0000000000000400ULL) #define CAP_FCHMOD CAPRIGHT(0, 0x0000000000002000ULL) #define CAP_FCHMODAT (CAP_FCHMOD | CAP_LOOKUP) There is new API to manage the new cap_rights_t structure: cap_rights_t *cap_rights_init(cap_rights_t *rights, ...); void cap_rights_set(cap_rights_t *rights, ...); void cap_rights_clear(cap_rights_t *rights, ...); bool cap_rights_is_set(const cap_rights_t *rights, ...); bool cap_rights_is_valid(const cap_rights_t *rights); void cap_rights_merge(cap_rights_t *dst, const cap_rights_t *src); void cap_rights_remove(cap_rights_t *dst, const cap_rights_t *src); bool cap_rights_contains(const cap_rights_t *big, const cap_rights_t *little); Capability rights to the cap_rights_init(), cap_rights_set(), cap_rights_clear() and cap_rights_is_set() functions are provided by separating them with commas, eg: cap_rights_t rights; cap_rights_init(&rights, CAP_READ, CAP_WRITE, CAP_FSTAT); There is no need to terminate the list of rights, as those functions are actually macros that take care of the termination, eg: #define cap_rights_set(rights, ...) \ __cap_rights_set((rights), __VA_ARGS__, 0ULL) void __cap_rights_set(cap_rights_t *rights, ...); Thanks to using one bit as an array index we can assert in those functions that there are no two rights belonging to different array elements provided together. For example this is illegal and will be detected, because CAP_LOOKUP belongs to element 0 and CAP_PDKILL to element 1: cap_rights_init(&rights, CAP_LOOKUP | CAP_PDKILL); Providing several rights that belongs to the same array's element this way is correct, but is not advised. It should only be used for aliases definition. This commit also breaks compatibility with some existing Capsicum system calls, but I see no other way to do that. This should be fine as Capsicum is still experimental and this change is not going to 9.x. Sponsored by: The FreeBSD Foundation
2013-09-05 00:09:56 +00:00
cap_rights_t rights;
Trace timeval parameters to the getitimer(2) and setitimer(2) syscalls. Reviewed by: jhb Tested by: pho Sponsored by: The FreeBSD Foundation MFC after: 1 week Differential revision: https://reviews.freebsd.org/D7158
2016-07-13 14:37:58 +00:00
struct itimerval it;
The kdump data stream is an unaligned data stream for stat and sockaddr structures. As such, we have top copy the data structure into a local buffer before we can reference it, otherwise we have unaligned references (these are fixed up automatically on some CPUs, but not on others). We do this unconditionally to make the code easier to read and understand. Submitted by: Grzegorz Bernacki
2010-02-18 16:05:09 +00:00
struct stat sb;
struct sockaddr_storage ss;
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
for (name = buf, namelen = 0;
namelen < buflen && name[namelen] != '\0';
++namelen)
/* nothing */;
if (namelen == buflen)
goto invalid;
if (name[namelen] != '\0')
goto invalid;
data = buf + namelen + 1;
datalen = buflen - namelen - 1;
if (datalen == 0)
goto invalid;
/* sanity check */
Make kdump compile cleanly at WARNS level 6, with one exception: the ipfilter headers contain a duplicated function declaration. Turn off -Werror to allow kdump to compile in spite of this. It would be neat to be able to turn off -Werror on a file-by-file basis... PR: bin/161478 Submitted by: Garrett Cooper <yanegomi@gmail.com>
2011-10-12 21:45:12 +00:00
for (i = 0; i < (int)namelen; ++i)
if (!isalpha(name[i]))
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
goto invalid;
Change the cap_rights_t type from uint64_t to a structure that we can extend in the future in a backward compatible (API and ABI) way. The cap_rights_t represents capability rights. We used to use one bit to represent one right, but we are running out of spare bits. Currently the new structure provides place for 114 rights (so 50 more than the previous cap_rights_t), but it is possible to grow the structure to hold at least 285 rights, although we can make it even larger if 285 rights won't be enough. The structure definition looks like this: struct cap_rights { uint64_t cr_rights[CAP_RIGHTS_VERSION + 2]; }; The initial CAP_RIGHTS_VERSION is 0. The top two bits in the first element of the cr_rights[] array contain total number of elements in the array - 2. This means if those two bits are equal to 0, we have 2 array elements. The top two bits in all remaining array elements should be 0. The next five bits in all array elements contain array index. Only one bit is used and bit position in this five-bits range defines array index. This means there can be at most five array elements in the future. To define new right the CAPRIGHT() macro must be used. The macro takes two arguments - an array index and a bit to set, eg. #define CAP_PDKILL CAPRIGHT(1, 0x0000000000000800ULL) We still support aliases that combine few rights, but the rights have to belong to the same array element, eg: #define CAP_LOOKUP CAPRIGHT(0, 0x0000000000000400ULL) #define CAP_FCHMOD CAPRIGHT(0, 0x0000000000002000ULL) #define CAP_FCHMODAT (CAP_FCHMOD | CAP_LOOKUP) There is new API to manage the new cap_rights_t structure: cap_rights_t *cap_rights_init(cap_rights_t *rights, ...); void cap_rights_set(cap_rights_t *rights, ...); void cap_rights_clear(cap_rights_t *rights, ...); bool cap_rights_is_set(const cap_rights_t *rights, ...); bool cap_rights_is_valid(const cap_rights_t *rights); void cap_rights_merge(cap_rights_t *dst, const cap_rights_t *src); void cap_rights_remove(cap_rights_t *dst, const cap_rights_t *src); bool cap_rights_contains(const cap_rights_t *big, const cap_rights_t *little); Capability rights to the cap_rights_init(), cap_rights_set(), cap_rights_clear() and cap_rights_is_set() functions are provided by separating them with commas, eg: cap_rights_t rights; cap_rights_init(&rights, CAP_READ, CAP_WRITE, CAP_FSTAT); There is no need to terminate the list of rights, as those functions are actually macros that take care of the termination, eg: #define cap_rights_set(rights, ...) \ __cap_rights_set((rights), __VA_ARGS__, 0ULL) void __cap_rights_set(cap_rights_t *rights, ...); Thanks to using one bit as an array index we can assert in those functions that there are no two rights belonging to different array elements provided together. For example this is illegal and will be detected, because CAP_LOOKUP belongs to element 0 and CAP_PDKILL to element 1: cap_rights_init(&rights, CAP_LOOKUP | CAP_PDKILL); Providing several rights that belongs to the same array's element this way is correct, but is not advised. It should only be used for aliases definition. This commit also breaks compatibility with some existing Capsicum system calls, but I see no other way to do that. This should be fine as Capsicum is still experimental and this change is not going to 9.x. Sponsored by: The FreeBSD Foundation
2013-09-05 00:09:56 +00:00
if (strcmp(name, "caprights") == 0) {
if (datalen != sizeof(cap_rights_t))
goto invalid;
memcpy(&rights, data, datalen);
ktrcaprights(&rights);
Trace timeval parameters to the getitimer(2) and setitimer(2) syscalls. Reviewed by: jhb Tested by: pho Sponsored by: The FreeBSD Foundation MFC after: 1 week Differential revision: https://reviews.freebsd.org/D7158
2016-07-13 14:37:58 +00:00
} else if (strcmp(name, "itimerval") == 0) {
if (datalen != sizeof(struct itimerval))
goto invalid;
memcpy(&it, data, datalen);
ktritimerval(&it);
Change the cap_rights_t type from uint64_t to a structure that we can extend in the future in a backward compatible (API and ABI) way. The cap_rights_t represents capability rights. We used to use one bit to represent one right, but we are running out of spare bits. Currently the new structure provides place for 114 rights (so 50 more than the previous cap_rights_t), but it is possible to grow the structure to hold at least 285 rights, although we can make it even larger if 285 rights won't be enough. The structure definition looks like this: struct cap_rights { uint64_t cr_rights[CAP_RIGHTS_VERSION + 2]; }; The initial CAP_RIGHTS_VERSION is 0. The top two bits in the first element of the cr_rights[] array contain total number of elements in the array - 2. This means if those two bits are equal to 0, we have 2 array elements. The top two bits in all remaining array elements should be 0. The next five bits in all array elements contain array index. Only one bit is used and bit position in this five-bits range defines array index. This means there can be at most five array elements in the future. To define new right the CAPRIGHT() macro must be used. The macro takes two arguments - an array index and a bit to set, eg. #define CAP_PDKILL CAPRIGHT(1, 0x0000000000000800ULL) We still support aliases that combine few rights, but the rights have to belong to the same array element, eg: #define CAP_LOOKUP CAPRIGHT(0, 0x0000000000000400ULL) #define CAP_FCHMOD CAPRIGHT(0, 0x0000000000002000ULL) #define CAP_FCHMODAT (CAP_FCHMOD | CAP_LOOKUP) There is new API to manage the new cap_rights_t structure: cap_rights_t *cap_rights_init(cap_rights_t *rights, ...); void cap_rights_set(cap_rights_t *rights, ...); void cap_rights_clear(cap_rights_t *rights, ...); bool cap_rights_is_set(const cap_rights_t *rights, ...); bool cap_rights_is_valid(const cap_rights_t *rights); void cap_rights_merge(cap_rights_t *dst, const cap_rights_t *src); void cap_rights_remove(cap_rights_t *dst, const cap_rights_t *src); bool cap_rights_contains(const cap_rights_t *big, const cap_rights_t *little); Capability rights to the cap_rights_init(), cap_rights_set(), cap_rights_clear() and cap_rights_is_set() functions are provided by separating them with commas, eg: cap_rights_t rights; cap_rights_init(&rights, CAP_READ, CAP_WRITE, CAP_FSTAT); There is no need to terminate the list of rights, as those functions are actually macros that take care of the termination, eg: #define cap_rights_set(rights, ...) \ __cap_rights_set((rights), __VA_ARGS__, 0ULL) void __cap_rights_set(cap_rights_t *rights, ...); Thanks to using one bit as an array index we can assert in those functions that there are no two rights belonging to different array elements provided together. For example this is illegal and will be detected, because CAP_LOOKUP belongs to element 0 and CAP_PDKILL to element 1: cap_rights_init(&rights, CAP_LOOKUP | CAP_PDKILL); Providing several rights that belongs to the same array's element this way is correct, but is not advised. It should only be used for aliases definition. This commit also breaks compatibility with some existing Capsicum system calls, but I see no other way to do that. This should be fine as Capsicum is still experimental and this change is not going to 9.x. Sponsored by: The FreeBSD Foundation
2013-09-05 00:09:56 +00:00
} else if (strcmp(name, "stat") == 0) {
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
if (datalen != sizeof(struct stat))
goto invalid;
The kdump data stream is an unaligned data stream for stat and sockaddr structures. As such, we have top copy the data structure into a local buffer before we can reference it, otherwise we have unaligned references (these are fixed up automatically on some CPUs, but not on others). We do this unconditionally to make the code easier to read and understand. Submitted by: Grzegorz Bernacki
2010-02-18 16:05:09 +00:00
memcpy(&sb, data, datalen);
ktrstat(&sb);
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
} else if (strcmp(name, "sockaddr") == 0) {
The kdump data stream is an unaligned data stream for stat and sockaddr structures. As such, we have top copy the data structure into a local buffer before we can reference it, otherwise we have unaligned references (these are fixed up automatically on some CPUs, but not on others). We do this unconditionally to make the code easier to read and understand. Submitted by: Grzegorz Bernacki
2010-02-18 16:05:09 +00:00
if (datalen > sizeof(ss))
goto invalid;
memcpy(&ss, data, datalen);
- Make actually printing path of AF_LOCAL socket types. MFC after: 1 week
2013-02-12 19:10:54 +00:00
if (datalen != ss.ss_len)
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
goto invalid;
The kdump data stream is an unaligned data stream for stat and sockaddr structures. As such, we have top copy the data structure into a local buffer before we can reference it, otherwise we have unaligned references (these are fixed up automatically on some CPUs, but not on others). We do this unconditionally to make the code easier to read and understand. Submitted by: Grzegorz Bernacki
2010-02-18 16:05:09 +00:00
ktrsockaddr((struct sockaddr *)&ss);
This patch adds a new ktrace(2) record type, KTR_STRUCT, whose payload consists of the null-terminated name and the contents of any structure you wish to record. A new ktrstruct() function constructs and emits a KTR_STRUCT record. It is accompanied by convenience macros for struct stat and struct sockaddr. In kdump(1), KTR_STRUCT records are handled by a dispatcher function that runs stringent sanity checks on its contents before handing it over to individual decoding funtions for each type of structure. Currently supported structures are struct stat and struct sockaddr for the AF_INET, AF_INET6 and AF_UNIX families; support for AF_APPLETALK and AF_IPX is present but disabled, as I am unable to test it properly. Since 's' was already taken, the letter 't' is used by ktrace(1) to enable KTR_STRUCT trace points, and in kdump(1) to enable their decoding. Derived from patches by Andrew Li <andrew2.li@citi.com>. PR: kern/117836 MFC after: 3 weeks
2008-02-23 01:01:49 +00:00
} else {
printf("unknown structure\n");
}
return;
invalid:
printf("invalid record\n");
}
Add a new trace point, KTRFAC_CAPFAIL, which traces capability check failures. It is included in the default set for ktrace(1) and kdump(1).
2011-10-11 20:37:10 +00:00
void
ktrcapfail(struct ktr_cap_fail *ktr)
{
Revisit the capability failure trace points. The initial implementation only logged instances where an operation on a file descriptor required capabilities which the file descriptor did not have. By adding a type enum to struct ktr_cap_fail, we can catch other types of capability failures as well, such as disallowed system calls or attempts to wrap a file descriptor with more capabilities than it had to begin with.
2011-10-18 07:28:58 +00:00
switch (ktr->cap_type) {
case CAPFAIL_NOTCAPABLE:
/* operation on fd with insufficient capabilities */
printf("operation requires ");
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
sysdecode_cap_rights(stdout, &ktr->cap_needed);
Capability rights are held by descriptors, not processes. Reported by: jonathan
2014-02-23 22:13:16 +00:00
printf(", descriptor holds ");
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
sysdecode_cap_rights(stdout, &ktr->cap_held);
Revisit the capability failure trace points. The initial implementation only logged instances where an operation on a file descriptor required capabilities which the file descriptor did not have. By adding a type enum to struct ktr_cap_fail, we can catch other types of capability failures as well, such as disallowed system calls or attempts to wrap a file descriptor with more capabilities than it had to begin with.
2011-10-18 07:28:58 +00:00
break;
case CAPFAIL_INCREASE:
/* requested more capabilities than fd already has */
printf("attempt to increase capabilities from ");
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
sysdecode_cap_rights(stdout, &ktr->cap_held);
Fix copy-pasto in CAPFAIL_INCREASE case. Noticed by: pjd
2011-10-18 09:39:41 +00:00
printf(" to ");
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
sysdecode_cap_rights(stdout, &ktr->cap_needed);
Revisit the capability failure trace points. The initial implementation only logged instances where an operation on a file descriptor required capabilities which the file descriptor did not have. By adding a type enum to struct ktr_cap_fail, we can catch other types of capability failures as well, such as disallowed system calls or attempts to wrap a file descriptor with more capabilities than it had to begin with.
2011-10-18 07:28:58 +00:00
break;
case CAPFAIL_SYSCALL:
/* called restricted syscall */
printf("disallowed system call");
break;
case CAPFAIL_LOOKUP:
kdump: expand comment on reasons for CAPFAIL_LOOKUP Comment for CAPFAIL_LOOKUP refered only to paths containing ".." but it is returned for other restricted VFS lookup cases, such as absolute paths or openat(AT_FDCWD, ...).
2019-02-18 03:49:16 +00:00
/* absolute or AT_FDCWD path, ".." path, etc. */
Revisit the capability failure trace points. The initial implementation only logged instances where an operation on a file descriptor required capabilities which the file descriptor did not have. By adding a type enum to struct ktr_cap_fail, we can catch other types of capability failures as well, such as disallowed system calls or attempts to wrap a file descriptor with more capabilities than it had to begin with.
2011-10-18 07:28:58 +00:00
printf("restricted VFS lookup");
break;
default:
printf("unknown capability failure: ");
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
sysdecode_cap_rights(stdout, &ktr->cap_needed);
Revisit the capability failure trace points. The initial implementation only logged instances where an operation on a file descriptor required capabilities which the file descriptor did not have. By adding a type enum to struct ktr_cap_fail, we can catch other types of capability failures as well, such as disallowed system calls or attempts to wrap a file descriptor with more capabilities than it had to begin with.
2011-10-18 07:28:58 +00:00
printf(" ");
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
sysdecode_cap_rights(stdout, &ktr->cap_held);
Revisit the capability failure trace points. The initial implementation only logged instances where an operation on a file descriptor required capabilities which the file descriptor did not have. By adding a type enum to struct ktr_cap_fail, we can catch other types of capability failures as well, such as disallowed system calls or attempts to wrap a file descriptor with more capabilities than it had to begin with.
2011-10-18 07:28:58 +00:00
break;
}
Add new ktrace records for the start and end of VM faults. This gives a pair of records similar to syscall entry and return that a user can use to determine how long page faults take. The new ktrace records are enabled via the 'p' trace type, and are enabled in the default set of trace points. Reviewed by: kib MFC after: 2 weeks
2012-04-05 17:13:14 +00:00
printf("\n");
}
void
ktrfault(struct ktr_fault *ktr)
{
kdump: eliminate new clang warnings. MFC after: 1 week
2015-01-06 18:53:09 +00:00
printf("0x%jx ", (uintmax_t)ktr->vaddr);
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
print_mask_arg(sysdecode_vmprot, ktr->type);
Add new ktrace records for the start and end of VM faults. This gives a pair of records similar to syscall entry and return that a user can use to determine how long page faults take. The new ktrace records are enabled via the 'p' trace type, and are enabled in the default set of trace points. Reviewed by: kib MFC after: 2 weeks
2012-04-05 17:13:14 +00:00
printf("\n");
}
void
ktrfaultend(struct ktr_faultend *ktr)
{
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
const char *str;
Add new ktrace records for the start and end of VM faults. This gives a pair of records similar to syscall entry and return that a user can use to determine how long page faults take. The new ktrace records are enabled via the 'p' trace type, and are enabled in the default set of trace points. Reviewed by: kib MFC after: 2 weeks
2012-04-05 17:13:14 +00:00
Move mksubr from kdump into libsysdecode. Restructure this script so that it generates a header of tables instead of a source file. The tables are included in a flags.c source file which provides functions to decode various system call arguments. For functions that decode an enumeration, the function returns a pointer to a string for known values and NULL for unknown values. For functions that do more complex decoding (typically of a bitmask), the function accepts a pointer to a FILE object (open_memstream() can be used as a string builder) to which decoded values are written. If the function operates on a bitmask, the function returns true if any bits were decoded or false if the entire value was valid. Additionally, the third argument accepts a pointer to a value to which any undecoded bits are stored. This pointer can be NULL if the caller doesn't care about remaining bits. Convert kdump over to using decoder functions from libsysdecode instead of mksubr. truss also uses decoders from libsysdecode instead of private lookup tables, though lookup tables for objects not decoded by kdump remain in truss for now. Eventually most of these tables should move into libsysdecode as the automated table generation approach from mksubr is less stale than the static tables in truss. Some changes have been made to truss and kdump output: - The flags passed to open() are now properly decoded in that one of O_RDONLY, O_RDWR, O_WRONLY, or O_EXEC is always included in a decoded mask. - Optional arguments to open(), openat(), and fcntl() are only printed in kdump if they exist (e.g. the mode is only printed for open() if O_CREAT is set in the flags). - Print argument to F_GETLK/SETLK/SETLKW in kdump as a pointer, not int. - Include all procctl() commands. - Correctly decode pipe2() flags in truss by not assuming full open()-like flags with O_RDONLY, etc. - Decode file flags passed to *chflags() as file flags (UF_* and SF_*) rather than as a file mode. - Fix decoding of quotactl() commands by splitting out the two command components instead of assuming the raw command value matches the primary command component. In addition, truss and kdump now build without triggering any warnings. All of the sysdecode manpages now include the required headers in the synopsis. Reviewed by: kib (several older versions), wblock (manpages) MFC after: 2 months Differential Revision: https://reviews.freebsd.org/D7847
2016-10-17 22:37:07 +00:00
str = sysdecode_vmresult(ktr->result);
if (str != NULL)
printf("%s", str);
else
printf("<invalid=%d>", ktr->result);
Add new ktrace records for the start and end of VM faults. This gives a pair of records similar to syscall entry and return that a user can use to determine how long page faults take. The new ktrace records are enabled via the 'p' trace type, and are enabled in the default set of trace points. Reviewed by: kib MFC after: 2 weeks
2012-04-05 17:13:14 +00:00
printf("\n");
Add a new trace point, KTRFAC_CAPFAIL, which traces capability check failures. It is included in the default set for ktrace(1) and kdump(1).
2011-10-11 20:37:10 +00:00
}
Decode kevent structures logged via ktrace(2) in kdump. - Add a new KTR_STRUCT_ARRAY ktrace record type which dumps an array of structures. The structure name in the record payload is preceded by a size_t containing the size of the individual structures. Use this to replace the previous code that dumped the kevent arrays dumped for kevent(). kdump is now able to decode the kevent structures rather than dumping their contents via a hexdump. One change from before is that the 'changes' and 'events' arrays are not marked with separate 'read' and 'write' annotations in kdump output. Instead, the first array is the 'changes' array, and the second array (only present if kevent doesn't fail with an error) is the 'events' array. For kevent(), empty arrays are denoted by an entry with an array containing zero entries rather than no record. - Move kevent decoding tables from truss to libsysdecode. This adds three new functions to decode members of struct kevent: sysdecode_kevent_filter, sysdecode_kevent_flags, and sysdecode_kevent_fflags. kdump uses these helper functions to pretty-print kevent fields. - Move structure definitions for freebsd11 and freebsd32 kevent structures to <sys/event.h> so that they can be shared with userland. The 32-bit structures are only exposed if _WANT_KEVENT32 is defined. The freebsd11 structures are only exposed if _WANT_FREEBSD11_KEVENT is defined. The 32-bit freebsd11 structure requires both. - Decode freebsd11 kevent structures in truss for the compat11.kevent() system call. - Log 32-bit kevent structures via ktrace for 32-bit compat kevent() system calls. - While here, constify the 'void *data' argument to ktrstruct(). Reviewed by: kib (earlier version) MFC after: 1 month Differential Revision: https://reviews.freebsd.org/D12470
2017-11-25 04:49:12 +00:00
void
ktrkevent(struct kevent *kev)
{
printf("{ ident=");
switch (kev->filter) {
case EVFILT_READ:
case EVFILT_WRITE:
case EVFILT_VNODE:
case EVFILT_PROC:
case EVFILT_TIMER:
case EVFILT_PROCDESC:
case EVFILT_EMPTY:
printf("%ju", (uintmax_t)kev->ident);
break;
case EVFILT_SIGNAL:
print_signal(kev->ident);
break;
default:
printf("%p", (void *)kev->ident);
}
printf(", filter=");
print_integer_arg(sysdecode_kevent_filter, kev->filter);
printf(", flags=");
print_mask_arg0(sysdecode_kevent_flags, kev->flags);
printf(", fflags=");
sysdecode_kevent_fflags(stdout, kev->filter, kev->fflags,
decimal ? 10 : 16);
printf(", data=%#jx, udata=%p }", (uintmax_t)kev->data, kev->udata);
}
void
ktrstructarray(struct ktr_struct_array *ksa, size_t buflen)
{
struct kevent kev;
char *name, *data;
size_t namelen, datalen;
int i;
bool first;
buflen -= sizeof(*ksa);
for (name = (char *)(ksa + 1), namelen = 0;
namelen < buflen && name[namelen] != '\0';
++namelen)
/* nothing */;
if (namelen == buflen)
goto invalid;
if (name[namelen] != '\0')
goto invalid;
/* sanity check */
for (i = 0; i < (int)namelen; ++i)
if (!isalnum(name[i]) && name[i] != '_')
goto invalid;
data = name + namelen + 1;
datalen = buflen - namelen - 1;
printf("struct %s[] = { ", name);
first = true;
for (; datalen >= ksa->struct_size;
data += ksa->struct_size, datalen -= ksa->struct_size) {
if (!first)
printf("\n ");
else
first = false;
if (strcmp(name, "kevent") == 0) {
if (ksa->struct_size != sizeof(kev))
goto bad_size;
memcpy(&kev, data, sizeof(kev));
ktrkevent(&kev);
} else if (strcmp(name, "kevent_freebsd11") == 0) {
struct kevent_freebsd11 kev11;
if (ksa->struct_size != sizeof(kev11))
goto bad_size;
memcpy(&kev11, data, sizeof(kev11));
memset(&kev, 0, sizeof(kev));
kev.ident = kev11.ident;
kev.filter = kev11.filter;
kev.flags = kev11.flags;
kev.fflags = kev11.fflags;
kev.data = kev11.data;
kev.udata = kev11.udata;
ktrkevent(&kev);
#ifdef _WANT_KEVENT32
} else if (strcmp(name, "kevent32") == 0) {
struct kevent32 kev32;
if (ksa->struct_size != sizeof(kev32))
goto bad_size;
memcpy(&kev32, data, sizeof(kev32));
memset(&kev, 0, sizeof(kev));
kev.ident = kev32.ident;
kev.filter = kev32.filter;
kev.flags = kev32.flags;
kev.fflags = kev32.fflags;
#if BYTE_ORDER == BIG_ENDIAN
kev.data = kev32.data2 | ((int64_t)kev32.data1 << 32);
#else
kev.data = kev32.data1 | ((int64_t)kev32.data2 << 32);
#endif
kev.udata = (void *)(uintptr_t)kev32.udata;
ktrkevent(&kev);
} else if (strcmp(name, "kevent32_freebsd11") == 0) {
struct kevent32_freebsd11 kev32;
if (ksa->struct_size != sizeof(kev32))
goto bad_size;
memcpy(&kev32, data, sizeof(kev32));
memset(&kev, 0, sizeof(kev));
kev.ident = kev32.ident;
kev.filter = kev32.filter;
kev.flags = kev32.flags;
kev.fflags = kev32.fflags;
kev.data = kev32.data;
kev.udata = (void *)(uintptr_t)kev32.udata;
ktrkevent(&kev);
#endif
} else {
printf("<unknown structure> }\n");
return;
}
}
printf(" }\n");
return;
invalid:
printf("invalid record\n");
return;
bad_size:
printf("<bad size> }\n");
return;
}
Improve WARNS situation for kdump: 1) Define _KERNEL while including sys/time.h to get some function prototypes. 2) Add prototypes and ANSIify definitions. 3) Constness changes. 4) Remove register keyword. 5) Actually return a sensible value from main. 6) Make fread_tail take a void * instead of a char *. 7) Avoid a signedness warning by casting to a size_t. Should be safe enough 'cos we also check for nonnegativity. 8) Be extra chummy with sigset_t rather than passing a struct to printf and pretending it is an int.
2002-07-28 16:05:07 +00:00
void
usage(void)
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
{
Add `-S' to display syscall numbers in the output as well. This is useful for debugging compat modules. Sponsored by: EMC / Isilon Storage Division Obtained from: Isilon OneFS (based on work by Jeff Hughes) MFC after: 2 weeks
2014-03-28 16:11:20 +00:00
fprintf(stderr, "usage: kdump [-dEnlHRrSsTA] [-f trfile] "
Spell -t option's argument by name.
2008-04-02 09:41:29 +00:00
"[-m maxdata] [-p pid] [-t trstr]\n");
BSD 4.4 Lite Usr.bin Sources
1994-05-27 12:33:43 +00:00
exit(1);
}
Reference in New Issue Copy Permalink