IF YOU WOULD LIKE TO GET AN ACCOUNT, please write an
email to Administrator. User accounts are meant only to access repo
and report issues and/or generate pull requests.
This is a purpose-specific Git hosting for
BaseALT
projects. Thank you for your understanding!
Только зарегистрированные пользователи имеют доступ к сервису!
Для получения аккаунта, обратитесь к администратору.
Add support for printing control messages containing software and/or
hardware timestamps, which are typically enabled by NTP and PTP
implementations.
* print_timeval.c (print_struct_timeval_data_size): New mpers printer.
* print_timespec.c (print_struct_timespec_data_size,
print_struct_timespec_array_data_size): Likewise.
* msghdr.c (print_scm_timestamp, print_scm_timestampns,
print_scm_timestamping): New functions.
(cmsg_socket_printers): Add them.
* NEWS: Mention this change.
* tests/msg_control.c: Include "xlat.h" and "xlat/scmvals.h".
(test_scm_timestamp, test_scm_timestampns, test_scm_timestamping):
New functions.
(test_sol_socket): Use them.
Signed-off-by: Miroslav Lichvar <mlichvar@redhat.com>
Signed-off-by: Dmitry V. Levin <ldv@altlinux.org>
This fixes generation of syscallent.i and subsequent scno.h files
on mips where syscallent.h checks macros defined in config.h.
* scno.am (syscallent.i): Add $(top_builddir)/config.h
to prerequisites, include it.
* NEWS: Mention this fix.
Reported-by: David Robins <dbrobins@i4031.net>
On alpha, struct old_sigaction.sa_flags has a 32-bit type.
On mips, only first word of old_sigaction.sa_mask is read.
On all architectures except alpha and mips that have old sigaction
syscall, struct old_sigaction has sa_restorer field that has to be
decoded.
* signal.c (struct old_sigaction) [ALPHA]: Change sa_flags type
to unsigned int, add ATTRIBUTE_PACKED.
[MIPS]: Dhrink sa_mask array to 1 element.
[!ALPHA && !MIPS]: Define sa_restorer field unconditionally.
(struct old_sigaction32): Likewise.
(decode_old_sigaction) [!current_wordsize]: Initialize sa_restorer field
from old_sigaction32.sa_restorer unconditionally.
[!ALPHA && !MIPS]: Print old_sigaction.sa_restorer if SA_RESTORER flag
is set.
* NEWS: Mention this change.
When traced command is terminated by a blocked signal, unblock
that signal to ensure strace termination with the same signal.
* strace.c (main): Unblock the signal raised for strace termination.
* NEWS: Mention this change.
While strace resets SIGCHLD handler to the default action so that
waitpid definitely works without losing track of children, tracees
should not inherit this change.
* strace.c (struct exec_params): Add child_sa field.
(init): When setting SIGCHLD handler to SIG_DFL, save the old handler.
(exec_or_die): Restore SIGCHLD handler if it was different from SIG_DFL
at startup.
* NEWS: Mention this change.
In interactive mode (-I2), those signals that were blocked at startup
will remain blocked for the whole period of strace execution.
* strace.c (empty_set): Rename to start_set. All users changed.
(init): Initialize start_set and blocked_set to the set of currently
blocked signals.
(startup_attach, trace): Use SIG_SETMASK instead of SIG_BLOCK.
* NEWS: Mention this change.
Old ssetmask syscall takes an argument and returns a value of type int.
Old sgetmask syscall may return an error.
* signal.c (SYS_FUNC(ssetmask)): Explicitly convert the argument
and return value to unsigned int, print it using sprint_old_sigmask_val.
(SYS_FUNC(sgetmask)): Do not print the mask in case of syserror.
* tests/sxetmask.c: New file.
* tests/gen_tests.in (sxetmask): New entry.
* tests/pure_executables.list: Add sxetmask.
* tests/.gitignore: Likewise.
* NEWS: Mention this fix.
On some architectures old sigsuspend syscall takes the signal mask from
the 3rd argument, on some from the 1st. And, if it wasn't peculiar
enough, the signal mask is passed by value on all architectures except
mips where it's passed by reference.
* signal.c (SYS_FUNC(sigsuspend)): Take the signal mask from the last
argument.
* linux/alpha/syscallent.h (sigsuspend): Set nargs to 1.
* linux/crisv10/syscallent.h: Likewise.
* linux/mips/syscallent-o32.h: Likewise.
* linux/powerpc/syscallent.h: Likewise.
* linux/powerpc64/syscallent.h: Likewise.
* linux/sh/syscallent.h: Likewise.
* linux/sh64/syscallent.h: Likewise.
* linux/sparc/syscallent.h: Likewise.
* linux/sparc64/syscallent.h: Likewise.
* tests/sigsuspend.c: New file.
* tests/gen_tests.in (sigsuspend): New entry.
* tests/pure_executables.list: Add sigsuspend.
* tests/.gitignore: Likewise.
* NEWS: Mention this fix.
Refactor sprinttime: implement sprinttime_nsec and sprinttime_usec
that handle nanoseconds and microseconds, respectively.
Always print raw values of time data fields, format string
representations of time as comments.
* defs.h (sprinttime): Change argument type from time_t to long long.
(sprinttime_nsec, sprinttime_usec): New prototypes.
* util.c (sprinttime_ex, sprinttime_nsec, sprinttime_usec): New
functions.
(sprinttime): Turn into a thin wrapper around sprinttime_ex.
* stat.h (struct strace_stat): Add has_nsec field.
* fetch_struct_stat.c (HAVE_NSEC): New macro.
(fetch_struct_stat): Initialize has_nsec field with HAVE_NSEC.
* fetch_struct_stat64.c (HAVE_NSEC): New macro.
(fetch_struct_stat64): Initialize has_nsec field with HAVE_NSEC.
* print_struct_stat.c (print_struct_stat) <PRINT_ST_TIME>:
Print raw values of time fields, use sprinttime_nsec to format a string
representation of time, use tprints_comment to print it as a comment.
* statx.c (SYS_FUNC(statx)) <PRINT_FIELD_TIME>: Likewise.
* utime.c (SYS_FUNC(utime)): Print raw values of struct utimbuf.actime
and struct utimbuf.modtime fields, use sprinttime to format a string
representation of time, use tprints_comment to print it as a comment.
* tests/tests.h (print_time_t_nsec): Add int argument.
* tests/print_time.c (print_time_t_ex): New function.
(print_time_t_nsec): Add int argument, turn into a thin wrapper around
print_time_t_ex.
* tests/utime.c (main): Update expected output.
* tests/xstatx.c [!IS_STATX] (HAVE_NSEC): New macro.
[!IS_STATX] (PRINT_ST_TIME), [IS_STATX] (PRINT_FIELD_TIME): Update
expected output.
* NEWS: Mention this timestamps representation improvement.
Co-authored-by: Dmitry V. Levin <ldv@altlinux.org>
* qualify.c: Include <regex.h>.
(qualify_syscall_regex): New function.
(qualify_syscall): Use it.
* strace.1: Document -e trace=/regex option.
* NEWS: Mention -e trace=/regex option.
* tests/regex.test: New test.
* tests/Makefile.am (DECODER_TESTS): Add it.
* tests/options-syntax.test: Add checks for invaild regexp
and for regexp that doesn't match a syscall.
* v4l2.c: Include "xlat/v4l2_vbi_flags.h" and "xlat/v4l2_sliced_flags.h".
(struct_v4l2_clip): New typedef. Mpersify it.
(print_v4l2_clip): New function.
(print_v4l2_format_fmt): Use it. Add struct tcb argument. Implement
decoding of the remaining V4L2_BUF_TYPE_* types.
* xlat/v4l2_vbi_flags.in: Add V4L2_VBI_UNSYNC and V4L2_VBI_INTERLACED
introduced by linux kernel commit v2.5.46~39^2~23^2~4.
Add V4L2_VBI_ITU_525_F1_START, V4L2_VBI_ITU_525_F2_START,
V4L2_VBI_ITU_625_F1_START, and V4L2_VBI_ITU_625_F2_START introduced by
linux kernel commit v3.17-rc1~112^2~217.
* xlat/v4l2_sliced_flags.in: Add V4L2_SLICED_TELETEXT_B, V4L2_SLICED_VPS,
V4L2_SLICED_CAPTION_525, V4L2_SLICED_WSS_625, V4L2_SLICED_VBI_525, and
V4L2_SLICED_VBI_625 introduced by linux kernel commit v2.6.14-rc2~64.
* configure.ac (AC_CHECK_DECLS): Add V4L2_BUF_TYPE_SDR_CAPTURE and
V4L2_BUF_TYPE_SDR_OUTPUT.
(AC_CHECK_MEMBERS): Add struct v4l2_window.global_alpha and
struct v4l2_sdr_format.buffersize.
* NEWS: Mention this change.
Signed-off-by: Edgar Kaziahmedov <edos@linux.com>
This saves up to 25% of ptrace syscalls in case of trace filtering.
* syscall.c (trace_syscall_exiting): Do not call gettimeofday
and get_regs for filtered syscalls.
* NEWS: Mention this change.
* tests/get_regs.test: New test.
* tests/Makefile.am (MISC_TESTS): Add it.
(as suggested in
https://www.mail-archive.com/strace-devel@lists.sourceforge.net/msg05147.html )
In order to move them to a different namespace than syscall names. This is
already a problem in case of ipc class which shadows ipc syscall.
Old behaviour is retained in order to preserve backwards compatibility.
* qualify.c (lookup_class) <syscall_class>: Add %-prefixed class definitions.
* strace.1 (.SS Filtering): Add information regarding %-prefixed class
syntax, declare legacy class syntax deprecated.
* tests/net.test: Update to use %-prefixed syscall class name.
* tests/netlink_protocol.test: Likewise.
* tests/qual_fault-exit_group.test: Likewise.
* tests/qual_syscall.test: Likewise.
* tests/scm_rights-fd.test: Likewise.
* tests/sigreturn.test: Likewise.
* tests/uio.test: Likewise.
* NEWS: Mention this change.
x32 is the only linux architecture where preadv2 and pwritev2 syscalls
take 5 arguments instead of 6.
* io.c (PREADV2_PWRITEV2_FLAGS_ARG_NO): New macro.
(SYS_FUNC(preadv2), SYS_FUNC(pwritev2)): Use it.
* linux/x32/syscallent.h (preadv2, pwritev2): Change nargs to 5.
* NEWS: Mention this fix.
* tests/preadv2-pwritev2.c (main): Fix invocation of preadv2
and pwritev2 syscalls on x32.
Implement read/write semantics of struct sched_attr.size argument
of sched_setattr syscall. Do not print members of struct sched_attr
besides sched_attr.size when the specified structure size is less than
the minimal size allowed by the kernel.
* sched.c (print_sched_attr): Fetch struct sched_attr.size and use it
as the structure size. Print struct sched_attr.size only when the
structure size is less than SCHED_ATTR_MIN_SIZE.
(SYS_FUNC(sched_setattr)): Call print_sched_attr with zero size
argument. Print struct sched_attr.size returned by the kernel
on exiting syscall in case of E2BIG. Print the last syscall argument
on exiting syscall.
(SYS_FUNC(sched_getattr)): Do not call print_sched_attr with zero size
argument.
* NEWS: Mention it.
* tests/sched_xetattr.c (main): Check it.
Introduce -e inject= as a general injection option,
limit -e fault= option to syscall fault injection.
Change default return code of syscall fault injection to ENOSYS.
* qualify.c (parse_inject_token): Add fault_tokens_only argument,
do not accept retval= and signal= tokens when fault_tokens_only
is set to true.
(parse_inject_expression): Add fault_tokens_only argument,
forward it to parse_inject_token.
(qualify_inject_common): New function.
(qualify_fault): Use it.
(qualify_inject): New function.
(qual_options): New entry.
* strace.1: Describe -e inject= option.
* NEWS: Mention -e inject= option.
* tests/qual_fault-syntax.test: Test that -e fault= option does not
support retval=, signal=, and multiple error= tokens.
* tests/qual_fault.c (DEFAULT_ERRNO): Set to ENOSYS unconditionally.
* tests/qual_inject-retval.test: Replace -e fault= option
with -e inject= option.
* tests/qual_inject-syntax.test: New test.
* tests/Makefile.am (MISC_TESTS): Add it.
Introduce new -e fault=EXPR syntax that can be used to specify a subset
of syscalls that are subject of syscall fault injection, an error code
that has to be injected, and a frequency of injection.
The expression specifying syscall fault injection has the following
format: SET[:error=ERRNO][:when=FIRST[+[STEP]]]
where only SET is a required part and all the rest is optional.
The method used to implement syscall fault injection is the following:
on entering syscall the syscall number is substituted by an invalid
syscall number -1, and on exiting syscall the error code returned by
the kernel is substituted with the error code specified in the fault
expression.
This implementaion is based on the prototype developed
by Nahim El Atmani as a part of his GSoC 2016 strace project.
* defs.h (struct fault_opts): New forward declaration.
(struct tcb): Add fault_vec field.
(TCB_FAULT_INJ, QUAL_FAULT): New macros.
* strace.1: Document -e fault expression syntax.
* strace.c (usage): Mention -e fault expression.
(droptcb): Deallocate fault_vec member.
* syscall.c (qual_fault, arch_set_scno, arch_set_error): New prototypes.
(qual_options): Add "fault" option.
(struct fault_opts): New structure.
(num_faults): New variable.
(fault_vec): New array.
(syscall_fault_injected, tcb_fault_opts, reallocate_fault,
find_errno_by_name, qual_syscall_ex, strip_prefix, parse_fault_token,
parse_fault_expression, qual_fault, inject_syscall_fault_entering,
update_syscall_fault_exiting): New functions.
(qual_syscall): Use qual_syscall_ex.
(qualify_one): Add argument: a pointer to struct fault_opts, all callers
changed. Copy struct fault_opts from the pointer to fault_vec.
Use reallocate_fault.
(qualify_scno, qualify_syscall_class, qualify_syscall_name): Add
argument: a pointer to struct fault_opts.
(qualify): Use reallocate_fault. Do not check "all" class for
QUAL_FAULT qualifier.
(lookup_class): Check for "all" class.
(trace_syscall_entering): Use inject_syscall_fault_entering.
(trace_syscall_exiting): Use update_syscall_fault_exiting. Clear
TCB_FAULT_INJ flag along with TCB_INSYSCALL. Print " (INJECTED)" suffix
when the syscall has been injected successfully.
[ARCH_REGS_FOR_GETREGSET && !HAVE_GETREGS_OLD]
(ptrace_setregset): New function.
(ptrace_setregset_or_setregs): Define to ptrace_setregset.
[ARCH_REGS_FOR_GETREGS && !HAVE_GETREGS_OLD]
(ptrace_setregs): New function.
(ptrace_setregset_or_setregs): Define to ptrace_setregs.
[ptrace_setregset_or_setregs] (set_regs): New function.
Include "set_scno.c" and "set_error.c"
* NEWS: Mention this enhancement.
* defs.h (evdev_abs): New xlat prototype.
(evdev_ioctl): Remove.
* evdev.c: Do not include "xlat/evdev_abs.h".
Mpersify struct ff_effect and ioctl numbers.
(evdev_ioctl): Mpersify.
* NEWS: Mention this enhancement.
* io.c (SYS_FUNC(preadv)): Call tprint_iov_upto instead
of tprint_iov and specify syscall return value as a data size limit.
* NEWS: Mention it.
* tests/preadv.c (main): Add a test case for preadv short read.
As -yy parser, compared to -y, needs to do at least 5 extra syscalls
(getxattr, socket, sendmsg, recvmsg, close) to print socket details,
caching results of netlink conversations between strace and kernel
noticeably reduces amount of system time spent by strace.
The caching is safe since sockets do not change their addresses after
successful bind or connect syscall.
* defs.h (string_quote, print_sockaddr_by_inode_cached): New prototypes.
* socketutils.c (cache_entry): New type.
(CACHE_SIZE, CACHE_MASK): New macros.
(cache): New static array.
(cache_and_print_inode_details): New static function.
(print_sockaddr_by_inode_cached): New function.
(inet_parse_response, unix_parse_response): Use
cache_and_print_inode_details.
* util.c (printfd): Use string_quote and print_sockaddr_by_inode_cached.
(string_quote): Remove static qualifier.
* NEWS: Mention this improvement.
* tests/unix-yy.c (main): Update.
* mem.c (SYS_FUNC(mincore)): Print only those bytes of the vector
that were written by the kernel.
* NEWS: Mention this fix.
* tests/mincore.c: New file.
* tests/mincore.test: New test.
* tests/Makefile.am (check_PROGRAMS): Add mincore.
(TESTS): Add mincore.test.
* tests/.gitignore: Add mincore.
* strace.c (init): Allow -p option along with a command.
(startup_child): In -D mode, record the parent of the tracer process
as strace_child.
(startup_attach): Save trace_tracer_pid before -D mode fork.
When tracing a command in -f mode, do not check for the command's
threads as it has no threads at this moment.
Never attach to the tracer process.
In -D mode, never attach to the parent of the tracer process,
terminate that process only once at the end of startup_attach,
and reset strace_child.
* strace.1: Document that -p option can be used along with tracing
of a command.
* NEWS: Mention it.
* tests/attach-p-cmd-cmd.c: New file.
* tests/attach-p-cmd-p.c: Likewise.
* tests/attach-p-cmd.test: New test.
* tests/.gitignore: Add attach-p-cmd-cmd and attach-p-cmd-p.
* tests/Makefile.am (check_PROGRAMS): Likewise.
(TESTS): Add attach-p-cmd.test.
This fixes Debian bug #549942.
* defs.h (dumpiov_upto): New prototype.
(dumpiov): Change to a wrapper around dumpiov_upto.
* util.c (dumpiov): Rename to dumpiov_upto, add and check data_size
argument.
* io.c (SYS_FUNC(readv)): Call tprint_iov_upto instead
of tprint_iov and specify syscall return value as a data size limit.
* syscall.c (dumpio): In case of SEN_readv, call dumpiov_upto instead
of dumpiov and specify syscall return value as a data size limit.
* NEWS: Mention this fix.
* tests/readv.c: New file.
* tests/readv.test: New test.
* tests/Makefile.am (check_PROGRAMS): Add readv.
(TESTS): Add readv.test.
* tests/.gitignore: Add readv.
On s390/s390x, syscalls with NR up to 255 can be implemented
directly using "svc NR", for NR >= 256 "svc 0" with %r1=NR is used.
The latter method is allowed for NR < 256, too.
When the syscall number specified directly or indirectly is recognized
by the kernel, i.e. it is less than its NR_syscalls value, it is stored
in %r2 and is available to arch_get_scno via s390_regset.gprs[2].
For syscall numbers >= NR_syscalls this register is set to 0,
but %r1 remains unchanged and could be used by arch_get_scno
via s390_regset.gprs[1] to decide what the syscall number is.
* linux/s390/get_scno.c (arch_get_scno): If s390_regset.gprs[2] is zero,
take syscall number from s390_regset.gprs[1].
* NEWS: Mention this fix.
This fixes Debian bug #485979 and Fedora bug #1298294.
Print the second return value of getxpid, getxuid, and getxgid syscalls
that return a pair of values using the same mechanism as pipe syscall.
* alpha.c: New file.
* Makefile.am (strace_SOURCES): Add it.
* linux/alpha/syscallent.h [20]: Change SEN(getpid) to SEN(getxpid).
[24]: Change SEN(getuid) to SEN(getxuid).
[47]: Change SEN(getgid) to SEN(getxgid).
* NEWS: Mention this enhancement.
* tests/uid.awk: Update for getxgid output change.
* tests/uid.test: Cleanup.
* tests/getxxid.c: New file.
* tests/getxxid.test: New test.
* tests/Makefile.am (check_PROGRAMS): Add getxxid.
(TESTS): Add getxxid.test.
* tests/.gitignore: Add getxxid.
This change complements commit v4.9-359-gd93d9f8 by fixing
RVAL_UDECIMAL case.
The only syscall that appears to be affected is the times syscall.
* syscall.c (trace_syscall_exiting): In case of RVAL_UDECIMAL,
when current personality is 32-bit, print 32-bit return code.
* NEWS: Mention this fix.
Reported-by: Steve McIntyre <steve@einval.com>
Print cpu_set_t as a set of integers, similar to the way
fd_set is printed as a set of descriptors.
* affinity.c: Include <sched.h>.
(get_cpuset_size): New function.
(print_affinitylist): Rewrite using get_cpuset_size and next_set_bit.
(sched_getaffinity, sched_setaffinity): Print first two args as ints.
* NEWS: Mention this enhancement.
* tests/sched_xetaffinity.c: New file.
* tests/sched_xetaffinity.test: New test.
* tests/Makefile.am (check_PROGRAMS): Add sched_xetaffinity.
(TESTS): Add sched_xetaffinity.test.
* tests/.gitignore: Add sched_xetaffinity.
Reported-by: Joe Korty <joe.korty@ccur.com>
* personality.c (SYS_FUNC(personality)): On entering syscall,
print first argument not as a hex value but as a set of flags.
On exiting syscall, print return code as a set of flags.
* NEWS: Mention this enhancement.
strace -k does not produce a reliable output on all supported
configurations yet, even basic strace-k.test is known to fail
on some of them.
* strace.c (usage): Document -k option as experimental.
* strace.1: Likewise.
* NEWS: Likewise.
Time spent in system time is not useful where a syscall depends on some
non-CPU resource, eg. typically open() or stat() to a network drive.
This patch adds a new flag (-w) to produce a summary of the time
difference between beginning and end of the system call (ie. latency)
This functionality has been useful to profile slow processes that
are not CPU-bound.
Signed-off-by: Mark Hills <mark.hills@framestore.com>
* NEWS (Improvements): Mention recent recvmsg/recvmmsg decoders
enhancements.
(Portability): Add a recommendation for the minimum Linux kernel
version to use.
* strace.c (main) [LINUX]: When attaching for -p, look in
/proc/PID/task for all threads and attach them as presumed
CLONE_THREAD children.
* NEWS: Mention the feature.
* strace.c (usage): Omit -z, since it has never worked properly.
* NEWS: Likewise.
* strace.c (main): Grok new option `-E var=val' or `-E var' to put
var=val in environ or to remove var, respectively.
(usage): Mention it.
* strace.1, NEWS: Document it.
