Commit Graph

226832 Commits

Author SHA1 Message Date
Fedor Uporov
f0a993bbdb Do not free bufs in case of extents metadata blocks + remove unneeded asserts.
Approved by:    pfg (mentor)
MFC after:      6 months
2017-10-24 19:14:33 +00:00
Alan Cox
3a757e5403 Micro-optimize the handling of fictitious pages in vm_page_free_prep().
A fictitious page is always wired, so there is no point in trying to
remove one from the page queues.

Completely remove one inaccurate comment from vm_page_free_prep() and
correct another.

Reviewed by:	kib, markj
MFC after:	1 week
2017-10-24 17:14:53 +00:00
Ruslan Bukin
d27927f731 Extract a set of pmcstat functions and interfaces to the new internal
library -- libpmcstat.

This includes PMC logging module, symbols lookup functions,
ELF parsing, process management, PMC attachment, etc.

This allows to reuse code while building new hwpmc(4)-based applications.

Also add pmcstat_symbol_search_by_name() function that allows to find
mapped IP range for a given function name.

Reviewed by:	kib
Sponsored by:	DARPA, AFRL
Differential Revision:	https://reviews.freebsd.org/D12718
2017-10-24 16:28:00 +00:00
Michael Tuexen
701492a5f6 Fix a bug in handling special ABORT chunks.
Thanks to Felix Weinrank for finding this issue using libfuzzer with
the userland stack.

MFC after:	3 days
2017-10-24 16:24:12 +00:00
Andriy Gapon
e54fb4ff8c iscsi_shutdown_post: do nothing if panic-ing
There is nothing that that routine should or could really do in that
context.

Reported by:	Ben RUBSON <ben.rubson@gmail.com>
MFC after:	1 week
2017-10-24 14:59:31 +00:00
Alan Somers
42b22a3e74 Bump man page revision dates for r324941
Reported by:	jhb
MFC after:	20 days
X-MFC-with:	324941
Sponsored by:	Spectra Logic Corp
2017-10-24 14:34:25 +00:00
Michael Tuexen
adc59f7f46 Fix a locking issue found by running AFL on the userland stack.
Thanks to Felix Weinrank for reporting the issue.

MFC after:	3 days
2017-10-24 14:28:56 +00:00
Edward Tomasz Napierala
b4ed9a87f9 Remove unneeded calls to access(2) from rtld(1); just call open(2) instead.
The result looks like this:

--- przed       2017-10-21 23:19:21.445034000 +0100
+++ po  2017-10-21 23:18:50.031865000 +0100
@@ -11,7 +11,6 @@ mmap(0x0,102,PROT_READ,MAP_PRIVATE,3,0x0)      = 343665418
 close(3)                                        = 0 (0x0)
 open("/usr/local/etc/libmap.d",O_RDONLY|O_NONBLOCK|O_DIRECTORY|O_CLOEXEC,0165) ERR#2 'No such file or directory'
 munmap(0x80067d000,102)                                 = 0 (0x0)
-access("/usr/local/lib/libintl.so.8",F_OK)      = 0 (0x0)
 openat(AT_FDCWD,"/usr/local/lib/libintl.so.8",O_RDONLY|O_CLOEXEC|O_VERIFY,00) = 3 (0x3)
 fstat(3,{ mode=-rw-r--r-- ,inode=642560,size=55188,blksize=32768 }) = 0 (0x0)
 mmap(0x0,4096,PROT_READ,MAP_PRIVATE|MAP_PREFAULT_READ,3,0x0) = 34366541824 (0x80067d000)
@@ -20,14 +19,13 @@ mmap(0x800877000,40960,PROT_READ|PROT_EXEC,MAP_PRIVATE
 mmap(0x800a81000,4096,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_FIXED|MAP_PREFAULT_READ,3,0xa000) = 34370752512 (0x800a81000)
 munmap(0x80067d000,4096)                        = 0 (0x0)
 close(3)                                        = 0 (0x0)
-access("/usr/local/lib/libc.so.7",F_OK)                 ERR#2 'No such file or directory'
+openat(AT_FDCWD,"/usr/local/lib/libc.so.7",O_RDONLY|O_CLOEXEC|O_VERIFY,00) ERR#2 'No such file or directory'
 openat(AT_FDCWD,"/var/run/ld-elf.so.hints",O_RDONLY|O_CLOEXEC,00) = 3 (0x3)
 read(3,"Ehnt\^A\0\0\0\M^@\0\0\0\M-2\0\0"...,128) = 128 (0x80)
 fstat(3,{ mode=-r--r--r-- ,inode=970684,size=306,blksize=32768 }) = 0 (0x0)
 lseek(3,0x80,SEEK_SET)                          = 128 (0x80)
 read(3,"/lib:/usr/lib:/usr/lib/compat:/u"...,178) = 178 (0xb2)
 close(3)                                        = 0 (0x0)
-access("/lib/libc.so.7",F_OK)                   = 0 (0x0)
 openat(AT_FDCWD,"/lib/libc.so.7",O_RDONLY|O_CLOEXEC|O_VERIFY,00) = 3 (0x3)
 fstat(3,{ mode=-r--r--r-- ,inode=1605239,size=1910320,blksize=32768 }) = 0 (0x0)
 mmap(0x0,4096,PROT_READ,MAP_PRIVATE|MAP_PREFAULT_READ,3,0x0) = 34366541824 (0x80067d000)

Reviewed by:	kib
MFC after:	2 weeks
Sponsored by:	DARPA, AFRL
Differential Revision:	https://reviews.freebsd.org/D12766
2017-10-24 12:56:08 +00:00
Edward Tomasz Napierala
2fe071d971 Replace lseek(2)/read(2) pair with pread(2), removing yet another syscall
from the binary startup code.

MFC after:	2 weeks
Sponsored by:	DARPA, AFRL
2017-10-24 12:04:07 +00:00
Edward Tomasz Napierala
1689a3c4ac Make find_library() conform to style(9). No functional changes.
Suggested by:	kib
MFC after:	2 weeks
Sponsored by:	DARPA, AFRL
2017-10-24 11:24:54 +00:00
Edward Tomasz Napierala
e65ad973b2 Reword the conditional; it was ugly, and adding another parameter,
which I'm going to do in a subsequent commit, would make it even uglier.
No functional changes.

MFC after:	2 weeks
Sponsored by:	DARPA, AFRL
2017-10-24 11:16:38 +00:00
Edward Tomasz Napierala
a8b31c14d2 Use xmalloc and read(2) instead of mmap(2) to read in libmap.conf(5).
This removes the need to call munmap(2) afterwards.

MFC after:	2 weeks
Sponsored by:	DARPA, AFRL
Differential Revision:	https://reviews.freebsd.org/D12767
2017-10-24 10:48:26 +00:00
Andrey V. Elsukov
5c70ebfa57 Add IPv6 support for O_TCPDATALEN opcode.
PR:		222746
MFC after:	1 week
2017-10-24 08:39:05 +00:00
Navdeep Parhar
5bcae8ddfa cxgbe(4): Read the MPS buffer group map from the firmware as it could be
different from hardware defaults.  The congestion channel map, which is
still fixed, needs to be tracked separately now.  Change the congestion
setting for TOE rx queues to match the drivers on other OSes while here.

MFC after:	2 weeks
Sponsored by:	Chelsio Communications
2017-10-24 05:41:48 +00:00
Warner Losh
6ca2fb6623 Treat a 'current' value of 0 as unlimited as a failsfe.
When limiting I/O, a value of 0 makes no sense as a limit. No progress
can be made. Trade the possibility that someone might be doing
something clever to achieve ultra-low I/O limits vs the damage of not
ever making progress on an I/O in favor of making progress. Now the
machine won't be useless if this accidentally gets requested.

Sponsored by: Netflix
2017-10-24 02:25:42 +00:00
Alan Somers
913b932900 Remove artificial restriction on lio_listio's operation count
In r322258 I made p1003_1b.aio_listio_max a tunable. However, further
investigation shows that there was never any good reason for that limit to
exist in the first place. It's used in two completely different ways:

* To size a UMA zone, which globally limits the number of concurrent
  aio_suspend calls.

* To artifically limit the number of operations in a single lio_listio call.
  There doesn't seem to be any memory allocation associated with this limit.

This change does two things:

* Properly names aio_suspend's UMA zone, and sizes it based on a new constant.

* Eliminates the artifical restriction on lio_listio. Instead, lio_listio
  calls will now be limited by the more generous max_aio_queue_per_proc. The
  old p1003_1b.aio_listio_max is now an alias for
  vfs.aio.max_aio_queue_per_proc, so sysconf(3) will still work with
  _SC_AIO_LISTIO_MAX.

Reported by:	bde
Reviewed by:	jhb
MFC after:	3 weeks
Sponsored by:	Spectra Logic Corp
Differential Revision:	https://reviews.freebsd.org/D12120
2017-10-23 23:12:01 +00:00
Alan Somers
659058b06f Fix the error message when creating a zpool on a too-small device
Don't check for SPA_MINDEVSIZE in vdev_geom_attach when opening by path.
It's redundant with the check in vdev_open, and failing to attach here
results in the wrong error message being printed.  However, still check for
it in some other situations:

* When opening by guids, so we don't get bogged down reading from slow
  devices like floppy drives.
* In vdev_geom_read_pool_label for the same reason, because we iterate over
  all providers.
* If the caller requests that we verify the guid, because then we'll have to
  read from the device before vdev_open verifies the size.

PR:		222227
Reported by:	Marie Helene Kvello-Aune <marieheleneka@gmail.com>
Reviewed by:	avg, mav
MFC after:	3 weeks
Sponsored by:	Spectra Logic Corp
Differential Revision:	https://reviews.freebsd.org/D12531
2017-10-23 23:05:29 +00:00
Dimitry Andric
cc718496a9 After jemalloc was updated to version 5.0.0 in r319971, i386 executables
linked with AddressSanitizer (even those linked on earlier versions of
FreeBSD, or with external versions of clang) started failing with errors
similar to:

  ==14688==AddressSanitizer CHECK failed:
  /usr/src/contrib/compiler-rt/lib/asan/asan_poisoning.cc:36
  "((AddrIsAlignedByGranularity(addr))) != (0)" (0x0, 0x0)

This is because AddressSanitizer expects all the TLS data in the program
to be aligned to at least 8 bytes.

Before the jemalloc 5.0.0 update, all the TLS data in the i386 version
of libc.so added up to 80 bytes (a multiple of 8), but 5.0.0 made this
grow to 2404 bytes (not a multiple of 8).  This is due to added caching
data in jemalloc's internal struct tsd_s.

To fix AddressSanitizer, ensure this struct is aligned to at least 16
bytes, which can be done unconditionally for all architectures.  (An
earlier version of the fix aligned the struct to 8 bytes, but only for
ILP32 architectures.  This was deemed unnecessarily complicated.)

PR:		221337
X-MFC-With:	r319971
2017-10-23 21:31:04 +00:00
Stephen Hurd
3429c02f82 Some cache related optimizations
1. prefetch 128 bytes of mbufs.
2. Re-order filling the pkt_info so cache stalls happen at the end
3. Define empty prefetch2cachelines() macro when the function isn't present.

Provides small performance improvments on some hardware

Reviewed by:	sbruno
Approved by:	sbruno (mentor)
Sponsored by:	Limelight Networks
Differential Revision:	https://reviews.freebsd.org/D12447
2017-10-23 20:50:08 +00:00
Ed Maste
aa86fca1d1 uefi.8: mention use on (32-bit) arm
Sponsored by:	The FreeBSD Foundation
2017-10-23 20:32:18 +00:00
Enji Cooper
fcc46733b3 Clean up trailing whitespace
MFC after:	1 week
2017-10-23 16:55:22 +00:00
Enji Cooper
df6a67d624 Remove dead stores
The return value of various snprintf calls was stored in `len` and not used
in many functions.

MFC after:	1 week
Reported by:	clang-scanbuild
2017-10-23 16:54:30 +00:00
Konstantin Belousov
c9b6b5826c Expand explanation of atomicity.
Mention per-location total order, out of thin air, and torn writes
guarantees.  Mention C11 standard' memory model and one most important
FreeBSD additional requirement, that is aligned ordinary loads and
stores are atomic on processors.

The text is introductional and informal.  Reference the C11 and
C++1{1,4,7} standards for authorative description.

In collaboration with:	alc
Sponsored by:	The FreeBSD Foundation (kib)
MFC after:	1 week
2017-10-23 16:14:55 +00:00
Mark Johnston
c9748df8ff Remove resource_set_*() declarations from sys/bus.h.
The corresponding definitions were removed in r78135.

PR:		223189
Submitted by:	marc.priggemeyer@gmail.com
MFC after:	1 week
2017-10-23 16:02:48 +00:00
Matt Joras
ba19246e07 Move clear_unrhdr to tmpfs_free_tmp.
Clearing the unr in tmpfs_unmount is not correct. In the case of
multiple references to the tmpfs mount (e.g. when there are lookup
threads using it) it will not be the one to finish tmpfs_free_tmp. In
those cases tmpfs_free_node_locked will be the final one to execute
tmpfs_free_tmp, and until then the unr must be valid.

Reported by:	pho
Approved/reviewed by:	rstone (mentor)
Sponsored by:	Dell EMC Isilon
Differential Revision:	https://reviews.freebsd.org/D12749
2017-10-23 15:43:38 +00:00
Mark Johnston
5fca1d90c1 Fix the VM_NRESERVLEVEL == 0 build.
Add VM_NRESERVLEVEL guards in the pmaps that implement transparent
superpage promotion using reservations.

Reviewed by:	alc, kib
MFC after:	1 week
Differential Revision:	https://reviews.freebsd.org/D12764
2017-10-23 15:34:05 +00:00
Enji Cooper
0c095a65a1 Fix an omission in a comment
Soft float API support applies to armv7 too after r324340
2017-10-23 07:56:56 +00:00
Mateusz Guzik
5132933a08 Bump WITNESS_PENDLIST to accomodate sleepq chain bump.
Reported by:	ngie
2017-10-23 01:00:35 +00:00
Warner Losh
6c458a6a56 Use preferred defined paths, rather than relative paths in fdt.
Sponsored by: Netflix
2017-10-22 22:52:27 +00:00
Warner Losh
c02a64d8be Create a shell script to build sys/boot on all the architectures.
One could run this from any directory, but it's designed to do
regression testing on sys/boot (it only tests on a subset of
architectures since all of them would take a lot longer and not help).
This will also ensure that future commits to sys/boot compile
everywhere.

Sponsored by: Netflix
2017-10-22 22:52:23 +00:00
Warner Losh
404a0a2f40 Use SYSDIR instead of ${.CURDIR}/../..<etc>/sys.
Sponsored by: Netflix
2017-10-22 22:50:28 +00:00
Warner Losh
b022947e0e Use BOOTSRC here.
sponsored by: Netflix
2017-10-22 22:50:23 +00:00
Warner Losh
45e0c11351 Prefer SRCTOP paths for bits we're grabbing from libc.
Sponsored by: Netflix
2017-10-22 22:50:19 +00:00
Warner Losh
254ba2c985 Make at91 boot loader compile again.
No clue if it actually still works, but it might.
2017-10-22 22:50:15 +00:00
Warner Losh
cb3ee5c694 End source directories with SRC rather than a hodgepodge of names
BOOTDIR->BOOTSRC
FICLDIR->FICLSRC
LDR_MI->LDRSRC

This matches the patterns used in the rest of the system a bit vetter.

Suggested by: rgrimes@
Sponsored by: Netflix
2017-10-22 22:50:08 +00:00
Warner Losh
392c874409 Move fdt and uboot defines into common uboot.mk.
Sponsored by: Netflix
2017-10-22 22:49:51 +00:00
Mateusz Guzik
9e68989764 Make the sleepq chain hash size configurable per-arch and bump on amd64.
While here cache-align chains.

This shortens longest found chain during poudriere -j 80 from 32 to 16.

Pushing this higher up will probably require allocation on boot.
2017-10-22 20:43:50 +00:00
Mateusz Guzik
5a17c5524f sdt: make all sdt probe sites test one variable
This saves on cache misses at the expense of a slight grow of .text.

Note this is a bandaid for lack of hotpatching.

Discussed with:	markj
2017-10-22 20:22:23 +00:00
Mark Johnston
cc1307b67f Delete declarations of struct pfs_bitmap, removed in r143841.
MFC after:	1 week
2017-10-22 20:22:11 +00:00
Mateusz Guzik
4f7663af9f sdt: whack unused SDT_PROBE_ENABLED 2017-10-22 20:14:48 +00:00
Jilles Tjoelker
1f04a45950 libc: Do not refer to _DefaultRuneLocale in ctype inlines
Referring to _DefaultRuneLocale causes this >4KB structure to be copied to
all executables that use <ctype.h> inlines (except PIE executables).

This only affects the case where thread local storage is available.

_CurrentRuneLocale cannot be NULL, so the check can be removed entirely.

_DefaultRuneLocale needs to remain available for now since libc++ uses it.
The __isctype inline in include/_ctype.h also refers to _DefaultRuneLocale
and remains available because it may still be used by third party software.

Reviewed by:	bdrewery, theraven
Differential Revision:	https://reviews.freebsd.org/D10363
2017-10-22 20:01:07 +00:00
Mark Johnston
7421ff0751 Address some miscellaneous issues in the CTF man page.
- Fix a number of typos.
- Replace some illumos-specific references.
- Note that a type definition of kind CTF_K_FUNCTION may be followed by
  a null type identifier in order to provide 4-byte alignment for the
  next type definition.

MFC after:	2 weeks
2017-10-22 19:17:25 +00:00
Mark Johnston
ef36b3f756 MFV r323105 (partial): 8300 fix man page issues found by mandoc 1.14.1
illumos/illumos-gate@72d3dbb9ab
72d3dbb9ab

https://www.illumos.org/issues/8300
  Prior to integrating the mdocml update to 1.14.1, fix issues found by
  new version, especially the "new sentence, new line" style rule.

FreeBSD note: this revision merges only the changes to the CTF manual
page. The changes to the ZFS pages cannot be applied directly.

Reviewed by: Robert Mustacchi <rm@joyent.com>
Reviewed by: Toomas Soome <tsoome@me.com>
Approved by: Gordon Ross <gwr@nexenta.com>
Author: Yuri Pankov <yuri.pankov@nexenta.com>

Discussed with:	avg
MFC after:	2 weeks
2017-10-22 18:32:28 +00:00
Mateusz Guzik
614e1868d6 Change kdb_active type to u_char.
Fixes warnings from gcc and keeps the small size. Perhaps nesting should be moved
to another variablle.

Reported by:	ngie
2017-10-22 13:42:56 +00:00
Enji Cooper
f2374e0cc5 Clean up trailing whitespace in kdb_thr_ctx(..)
MFC after:	1 week
2017-10-22 12:12:52 +00:00
Bruce M Simpson
ab28146b6c Modernise this man page somewhat.
1. Add a reference to a good 3rd party list of compatible cables, but
provide suggestions for 'known good' vendors.

2. Change IP-based USB host-host example to a modern Ethernet one which
works 'out of box' with current Linux systems.

3. Explain that USB 3.0 is host-host, even though point-to-point soft
Ethernet can be achieved.

MFC after:	3 weeks
2017-10-22 11:40:55 +00:00
Bruce M Simpson
1d4c696a71 Add Prolific PL27A1 USB 3.0 Host-Host device to udbp(4).
Tested with a Plugable cable in VirtualBox against Linux 4.11.

MFC after:	2 weeks
2017-10-22 11:15:58 +00:00
Edward Tomasz Napierala
be7d4ac586 Add OID for the vm.overcommit sysctl. This makes it possible to remove
one call to sysctl(2) from jemalloc startup code. (That also requires
changes to jemalloc, but I plan to push those to upstream first.)

Reviewed by:	kib
MFC after:	2 weeks
Sponsored by:	DARPA, AFRL
Differential Revision:	https://reviews.freebsd.org/D12745
2017-10-22 10:35:29 +00:00
Edward Tomasz Napierala
6b61e3e47a Don't call realpath(3) from libmap rtld code. This gets rid of a few calls
to fstatat(2) at binary startup; the difference looks like this:

--- przed       2017-10-14 13:55:49.983528000 +0100
+++ po  2017-10-14 14:10:39.134343000 +0100
@@ -1,15 +1,10 @@
 mmap(0x0,32768,PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANON,-1,0x0) = 34366173184 (0x800623000)
 issetugid()                                     = 0 (0x0)
-fstatat(AT_FDCWD,"/etc",{ mode=drwxr-xr-x ,inode=1364352,size=2560,blksize=32768 },AT_SYMLINK_NOFOLLOW) = 0 (0x0)
-fstatat(AT_FDCWD,"/etc/libmap.conf",{ mode=-rw-r--r-- ,inode=1373288,size=102,blksize=32768 },AT_SYMLINK_NOFOLLOW) = 0 (0x0)
 openat(AT_FDCWD,"/etc/libmap.conf",O_RDONLY|O_CLOEXEC,00) = 3 (0x3)
 fstat(3,{ mode=-rw-r--r-- ,inode=1373288,size=102,blksize=32768 }) = 0 (0x0)
 mmap(0x0,102,PROT_READ,MAP_PRIVATE,3,0x0)       = 34366205952 (0x80062b000)
 close(3)                                        = 0 (0x0)
-fstatat(AT_FDCWD,"/usr",{ mode=drwxr-xr-x ,inode=561792,size=512,blksize=32768 },AT_SYMLINK_NOFOLLOW) = 0 (0x0)
-fstatat(AT_FDCWD,"/usr/local",{ mode=drwxr-xr-x ,inode=561800,size=512,blksize=32768 },AT_SYMLINK_NOFOLLOW) = 0 (0x0)
-fstatat(AT_FDCWD,"/usr/local/etc",{ mode=drwxr-xr-x ,inode=653279,size=1536,blksize=32768 },AT_SYMLINK_NOFOLLOW) = 0 (0x0)
-fstatat(AT_FDCWD,"/usr/local/etc/libmap.d",0x7fffffffcf50,AT_SYMLINK_NOFOLLOW) ERR#2 'No such file or directory'
+open("/usr/local/etc/libmap.d",O_RDONLY|O_NONBLOCK|O_DIRECTORY|O_CLOEXEC,0165) ERR#2 'No such file or directory'
 munmap(0x80062b000,102)                                 = 0 (0x0)
 openat(AT_FDCWD,"/var/run/ld-elf.so.hints",O_RDONLY|O_CLOEXEC,00) = 3 (0x3)
 read(3,"Ehnt\^A\0\0\0\M^@\0\0\0\M-2\0\0"...,128) = 128 (0x80)

Reviewed by:	kib
MFC after:	2 weeks
Sponsored by:	DARPA, AFRL
Differential Revision:	https://reviews.freebsd.org/D12741
2017-10-22 10:32:40 +00:00
Konstantin Belousov
456a73ef01 Remove the support for mknod(S_IFMT), which created dummy vnodes with
VBAD type.

FFS ffs_write() VOP catches such vnodes and panics, other VOPs do not
check for the type and their behaviour is really undefined.  The
comment claims that this support was done for 'badsect' to flag bad
sectors, we do not have such facility in kernel anyway.

Reported by:	Dmitry Vyukov <dvyukov@google.com>
Sponsored by:	The FreeBSD Foundation
MFC after:	1 week
2017-10-22 08:11:45 +00:00