Commit Graph

172 Commits

Author SHA1 Message Date
bcr
6f33a79b0a Fix typos in comments, no functional changes.
Found by:       codespell
Reviewed by:    alfred
MFC after:      1 week
2011-05-19 20:35:40 +00:00
jilles
f046771b04 Simplify various code that allowed for sys_signame being lower case.
This was changed in r218285.
2011-03-06 19:50:47 +00:00
jhb
6190d8e2a6 Properly check for errors from waitpid().
Submitted by:	gcooper
MFC after:	1 week
2011-02-15 12:42:18 +00:00
jilles
dbecc33067 Make sys_signame upper case.
This matches the constants from <signal.h> with 'SIG' removed, which POSIX
requires kill and trap to accept and 'kill -l' to write.

'kill -l', 'trap', 'trap -l' output is now upper case.

In Turkish locales, signal names with an upper case 'I' are now accepted,
while signal names with a lower case 'i' are no longer accepted, and the
output of 'killall -l' now contains proper capital 'I' without dot instead
of a dotted capital 'I'.
2011-02-04 16:40:50 +00:00
jh
6664a6800a Don't write the terminating NUL past end of buffer.
PR:		bin/152345
Submitted by:	Mateusz Guzik
2010-12-06 09:18:11 +00:00
jh
2b1dea9bfc Set FD_CLOEXEC for the output file only when the file has been specified
with the -o option. Setting the flag for stderr (the default) could
cause the traced process to redirect stderr to a random file.

PR:		bin/152151
Submitted by:	ashish
MFC after:	5 days
2010-11-13 09:28:49 +00:00
ed
e24538c770 Remove setpgid() call before executing child process.
Using a separate process group here is bad, since (for example) job
control in the TTY layer prevents interaction with the TTY, causing the
child process to hang.

Mentioned on:	current@
MFC after:	2 weeks
2010-10-20 09:35:20 +00:00
bcr
2ea7523b1b s/sytem/system in comments, no functional changes.
Reviewed by:    alfred@
2010-10-13 20:08:02 +00:00
nwhitehorn
7ce28b7066 Fix breakage introduced in r211725 and improve functionality of truss on
64-bit powerpc by adding 32-bit compatibility features.
2010-08-28 15:04:53 +00:00
imp
c3a399c4ba MFtbemd:
Prefer MACHNE_CPUARCH to MACHINE_ARCH in most contexts where you want
to test of all the CPUs of a given family conform.
2010-08-23 22:24:11 +00:00
nwhitehorn
4652fc6f95 Teach truss and xlint how to operate on 64-bit PowerPC systems. 2010-07-12 00:54:41 +00:00
imp
9563548aa0 Fix copyright spelling.
PR:		139825
Submitted by:	Ruslan Mahmatkhanov
2010-03-10 20:31:30 +00:00
rrs
b3974c3946 This fix makes it so the args are malloc'd before
we use them (always a good idea ;-0)

This was found and fixed by JC

Obtained from:	JC (c.jayachandran@gmail.com)
2010-02-03 13:03:47 +00:00
ed
9b380e30d4 Build usr.bin/ with WARNS=6 by default.
Also add some missing $FreeBSD$ to keep svn happy.
2010-01-02 10:27:05 +00:00
brooks
a5cc24440b The devices that supported EVFILT_NETDEV kqueue filters were removed in
r195175.  Remove all definitions, documentation, and usage.

fifo_misc.c:
	Remove all kqueue tests as fifo_io.c performs all those that
	would have remained.

Reviewed by:	rwatson
MFC after:	3 weeks
X-MFC note:	don't change vlan_link_state() function signature
2009-12-31 20:29:58 +00:00
ed
a0fa5bcb40 Allow proper tracing of posix_openpt(2). 2009-12-23 15:22:50 +00:00
jh
34da792643 Cast time_t values to intmax_t and use %jd with printf.
OK'ed by:	delphij
Approved by:	trasz (mentor)
2009-12-21 17:41:57 +00:00
jh
2eaf7d742e Remove non-working special case for pipe(2) from amd64-fbsd32.c and
i386-fbsd.c. Add pipe(2) to syscall table to decode it's pointer
argument properly and re-add special handling for pipe(2) return value
to print_syscall_ret().

PR:		bin/120870
Approved by:	trasz (mentor)
2009-12-21 17:38:13 +00:00
jh
05e68d005a Avoid sharing the file descriptor of the output file with traced
processes by setting the FD_CLOEXEC flag for the output file.

PR:		bin/140493
Submitted by:	Erik Lax
OK'ed by:	delphij
Approved by:	trasz (mentor)
2009-12-20 11:00:53 +00:00
jh
d51d2e56fc Add fork(2), getegid(2), geteuid(2), getgid(2), getpid(2), getpgid(2),
getpgrp(2), getppid(2), getsid(2) and getuid(2) to syscall table to
decode their arguments correctly.

OK'ed by:	delphij
Approved by:	trasz (mentor)
2009-12-20 10:58:34 +00:00
delphij
0842b7d53f Revert most part of 200420 as requested, as more review and polish is
needed.
2009-12-13 03:14:06 +00:00
delphij
b963db4652 Remove unneeded header includes from usr.bin/ except contributed code.
Tested with:	make universe
2009-12-11 23:35:38 +00:00
delphij
1e64043f43 Correct off-by-one issue in truss(1) which happens when system call number
is nsyscalls.

PR:		bin/134916
Submitted by:	Steven Hartland <steven hartland multiplay co uk>
MFC after:	2 weeks
2009-05-28 00:38:24 +00:00
delphij
09f0a4d580 Reduce diff against my local version: replace malloc+memset() cases to calloc(). 2009-05-15 19:41:10 +00:00
dds
07e8ac4814 Fix print_syscall_ret parameter order. 2009-05-13 13:00:52 +00:00
dds
51f76634a5 Fix compilation error introduced in r192025. 2009-05-13 12:43:37 +00:00
dds
bea63f2169 Add -c option to summarize number of calls, errors, and system time.
Reviewed by:	alfred
2009-05-12 20:42:12 +00:00
delphij
feaa2ee03c Don't exit until all truss children were exit. 2009-04-13 16:23:32 +00:00
delphij
882a2894c3 Correct a bug where pr_data should have been assigned. 2009-04-13 16:22:38 +00:00
imp
442a51fa71 Add preliminary support for truss on MIPS. It compiles, but has not
been extensively tested.  And the ELF64 stuff likely is not quite
right...

# There's a lot of cut-n-paste code here that could easily be
# refactored, at least for FreeBSD syscalls.
2009-02-15 01:26:49 +00:00
trhodes
9e3e63a72b Attaching to the init process returns EINVAL,
so give an example that is more likely to work.
Stolen from the ktrace(1) manual page.

PR:		128222
Submitted by:	Mateusz Guzik <mjguzik@gmail.com>
2009-01-23 00:58:14 +00:00
des
99ec4c843d Try to make this code slightly less painful to read. 2008-07-31 17:15:21 +00:00
jhb
7617274408 Retire some stale alpha references. 2008-05-16 20:09:29 +00:00
jhb
697ba009c1 Teach truss about 32-bit FreeBSD and Linux binaries on amd64. Some
additional work is needed to handle ABI-specific syscall argument parsing,
but this gets the basic tracing working.

MFC after:	1 week
2008-05-16 15:34:06 +00:00
des
f7514f1316 Revert CLEANDEPFILES commit per ru@'s request; it does not really solve
the problem.  The correct fix will follow.
2008-02-05 08:52:36 +00:00
des
0e657b23ab Normally, when a header file is removed from the build (as i4b headers
were recently), a simple 'make cleandepend; make depend' is sufficient
to keep the tree buildable after a cvs update when doing incremental
builds.

However, kdump and truss use a script which searches for header files
that define ioctls, and generates C code that includes them.  This
script will usually not need updating when a header file is removed,
so the normal dependency mechanism will not realize that it needs to
be re-run.  One is therefore left with code that references dead files
but will only be removed by a full 'make clean', which defeats the
purpose of incremental builds.

To work around this, modify the cleandepend target in bsd.dep.mk to
also remove any files listed in a new variable named CLEANDEPFILES,
and modify kdump's and truss's Makefiles accordingly.

MFC after:	2 weeks
2008-02-03 11:34:56 +00:00
jasone
7719fcccbc The break() system call takes a pointer argument, not an integer. This
change fixes output for break() on LP64 systems.
2008-01-03 00:18:03 +00:00
marcel
8be873747a Print integer-typed arguments as integers. This makes sure that
on 64-bit platforms the result is more reliable. For example,
-1 was previously printed as 0xffffffff.

Approved by: re (kensmith)
2007-07-28 23:18:39 +00:00
marcel
aa233d32b1 Fix handling of Quad-type arguments. Previously, syscalls
containing 64-bit arguments would have explicit padding.
On 64-bit platforms there was no padding, so the dummy
argument was not covering anything. On 32-bit platforms
with weak alignment (i.e. i386) the 64-bit argument did
not need to be aligned, so there too an aditional argument
was introduced. On 32-bit platforms with strong alignment
(i.e. PowerPC) the dummy argument in fact cover the padding.
By elimininating the dummy argument, 64-bit platforms now
have 1 argument less. This also applies to 32-bit platforms
with weak alignment. On PowerPC this doesn't matter, because
the padding is still there. We just don't "name" it.
Deal with those 3 cases.

Approved by: re (kensmith)
2007-07-28 23:15:04 +00:00
marcel
0333870131 Syscalls have at most 6 argument, not 5. See mmap(2) for example.
Previously the offset argument to mmap(2) would be bogus as we
weren't reading it in.

Approved by: re (kensmith)
2007-07-28 23:00:42 +00:00
delphij
affd0ee004 MFp4: Bugfixes for truss(1):
- Fix logic handling execve().  We will not be able to
   obtain information otherwise.
 - truss coredump [1].
 - truss does not work against itself [2].

PR:		bin/58970 [1], bin/45193 [2]
Submitted by:	Howard Su
Approved by:	re (kensmith)
2007-06-26 22:42:37 +00:00
ru
05dcdedaca Don't want a stray systrace_args.c in objdir. 2007-05-22 10:36:12 +00:00
keramida
2049c78bad Remove duplicate ' 2' manpage section number.
Submitted by:	 Rui Paulo
MFC after:	 3 days
2007-05-21 20:33:51 +00:00
delphij
70cda62de5 Make use of ptrace(2) instead of procfs in truss(1), eliminating
yet another need of an available /proc/ mount.

Tested with:	make universe
Submitted by:	howardsu
Reviewed by:	alfred
2007-04-10 04:03:34 +00:00
ru
164381741c Add missing markup bits. 2006-12-11 11:34:44 +00:00
ru
33e34aeeb5 Markup fixes. 2006-09-29 15:20:48 +00:00
ru
aa1113efea Remove alpha-specific stuff. 2006-08-23 12:12:56 +00:00
delphij
9328356917 Make sure that the sc pointer gets initialized. No functional change
should happen with this change.
2006-07-26 08:58:00 +00:00
pav
988c4b7d34 - Fix warning when compiling with -DDEBUG
Approved by:	alfred
2006-05-15 22:09:45 +00:00
pav
a233d0a2ec - Add decoding of kse_release, kevent, sigprocmask, unmount, socket, getrusage,
rename, __getcwd, shutdown, getrlimit, setrlimit, _umtx_lock, _umtx_unlock,
  pathconf, truncate, ftruncate, kill

- Decode more arguments of open, mprot, *stat, and fcntl.

- Convert all constant-macro and bitfield decoding to lookup tables; much
  cleaner than previous code.

- Print the timestamp of process exit and signal reception when -d or -D are in
  use

- Try six times with 1/2 second delay to debug the child

PR:		bin/52190 (updated)
Submitted by:	Dan Nelson <dnelson@allantgroup.com>
Approved by:	alfred
2006-05-15 21:18:28 +00:00