| Commit message (Collapse) | Author | Age | Files | Lines |
... | |
|
|
|
|
|
|
|
| |
* procfs.c (_stp_proc_write_cmd): For STP_SYMBOLS,
type field is a long to preserve alignment.
Use STP_ALLOC_FLAGS.
* symbols.c: Use STP_ALLOC_FLAGS.
|
|
|
|
|
|
|
| |
* alloc.c (STP_ALLOC_FLAGS): Define.
Cleanup ifdefs.
* map.c: Use STP_ALLOC_FLAGS.
* stat.c: ditto.
|
|
|
|
| |
bugfix of lket-b2a
|
|
|
|
| |
bugfix of the obsolete buf_info and consumed_info.
|
|
|
|
|
|
|
|
| |
* transport_msgs.h: Change all ints to int32_t. Prefix
all struct names with "_stp".
* transport.c: Use new struct names.
(_stp_handle_start): Send pointer size and endianess.
* procfs.c: Use new struct names.
|
|
|
|
|
|
|
|
|
|
|
| |
* librelay.c: Change all references to transport messages
to use the new names with "_stp" prefix.
(stp_main_loop): For STP_SYMBOLS, check pointer size and
endianess to confirm staprun is compatible with the kernel.
* librelay.h: Move a bunch of common includes here.
* stpd.c: Cleanup includes.
* symbols.c: Ditto.
|
|
|
|
| |
* sym.h: Change int to int32_t.
|
|
|
|
|
|
|
|
| |
make fork/execce trace hooks lket internally and turned on by default.
add stoptrace_fork and stoptrace_execve flags for lket to control
stopping fork/execve in user scripts.
update lket-b2a and lket man page.
|
| |
|
| |
|
| |
|
|
|
|
| |
* symbols.c (_stp_do_module): Fix error message.
|
| |
|
|
|
|
|
| |
* io.c (_stp_vlog, _stp_dbug): Make first parameter const.
* runtime.h: Update prototype for _stp_dbug.
|
|
|
|
|
| |
* runtime/regs.c: patch from Bibo Mao for fixing the return
value on IA64 (bz #3404).
|
| |
|
|
|
|
|
|
| |
Document signal trace hooks into lket man page
bugfix of lket-b2a for event description data insert
|
| |
|
| |
|
|
|
|
|
|
| |
header files
* runtime/lket/b2a/Makefile.am: set compile flags according to the existance of mysql_config
|
|
|
|
| |
* stack-ppc64.c (__stp_stack_sprint): Declare sp before _sp.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* stack.c: Reorganize and split arch-specific functions to
separate files.
(_stp_kta): Better checking.
(_stp_stack_sprint): Better handling of return probes.
* stack-i386.c: New file. Uses 2.6.18 DWARF unwinder if available.
* stack-x86_64.c: New file. Uses 2.6.18 DWARF unwinder if available.
* stack-ppc64.c: New file.
* stack-ia64.c: New file.
* sym.c (_stp_kallsyms_lookup_name): New function. Like
kallsyms_lookup_name() except use our internal lookup table.
|
| |
|
| |
|
|
|
|
| |
clean up the unused database tables and destory GTree on exit.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Adding the support of automatically database creation and
dumping lket trace into mysql database
* configure.ac:
add the checking of mysql client library
* Makefile.in, configure:
regenerated
* runtime/lket/b2a/Makefile.am:
add the checking and optional compiling for mysql client library
* runtime/lket/b2a/Makefile.in:
regenerated
|
|
|
|
|
|
|
| |
From David Wilder <dwilder@us.ibm.com>
* time.c (__stp_estimate_cpufreq): Short-circuit on s390.
(_stp_gettimeofday_ns): Use a fixed formula on s390 to
convert TOD clocks to nanoseconds.
|
|
|
|
|
|
|
|
|
|
| |
* stack.c (_stp_kta): Rewrite. Use the _stap_symbols
struct instead of calling into the kernel.
* sym.c (_stp_kallsyms_lookup): Move here from runtime.h
* runtime.h: Get rid of all the symbol stuff that
did not belong here.
|
|
|
|
|
|
|
|
|
|
| |
* README: Changed 'stpd' references to 'staprun'.
* README.doc: Ditto.
* TODO: Ditto.
* io.c: Ditto.
* print.c: Ditto.
* runtime.h: Ditto.
* bench2/bench.rb: Ditto.
|
|
|
|
|
|
| |
* bench/run_bench: Changed 'stpd' references to 'staprun'.
* bench/trans_bench: Ditto.
* where_func/README: Ditto.
|
|
|
|
|
|
| |
* Makefile: Changed 'stpd' references to 'staprun'.
* librelay.c: Ditto.
* stpd.c: Ditto.
|
|
|
|
|
| |
* transport.c: Changed 'stpd' references to 'staprun'.
* transport.txt: Ditto.
|
|
|
|
|
| |
* io.c: Changed 'stpd' references to 'staprun'.
* print.c: Ditto.
|
|
|
|
|
|
|
| |
* transport.c (_stp_handle_start): Don't initialize timers here.
(_stp_transport_init): Initialize timers here.
(_stp_cleanup_and_exit): Don't kill timers or free print buffers here.
(_stp_transport_close): kill timers and free print buffers here.
|
|
|
|
|
|
|
| |
* time.c (_stp_init_time): Use dynamic percpu allocations
instead of evil static allocations.
(_stp_kill_time): Free percpu allocations and set
_stp_time to NULL so the timers are only deleted once.
|
|
|
|
|
| |
* io.c (_stp_vlog): Remove ifdef for STP_RELAYFS. Relayfs now
behaves the same as procfs.
|
| |
|
|
|
|
|
| |
* print.c: Replace STP_PRINT_BUF_LEN with STP_BUFFER_SIZE.
* string.c: Ditto.
|
| |
|
|
|
|
|
| |
* transport.c (_stp_work_queue): Reenable some cleanup
code.
|
|
|
|
|
| |
* transport.c (_stp_transport_init): Call _stp_print_init().
(_stp_cleanup_and_exit): Call _stp_print_cleanup().
|
|
|
|
|
|
|
|
|
|
|
| |
PR 3232
* print.c (_stp_print_init): New. Alloc per-cpu buffers.
(_stp_print_cleanup): New. Free per-cpu buffers.
(_stp_print_flush): Use per_cpu_ptr().
* string.c (_stp_sprintf): Ditto.
(_stp_vsprintf): Ditto.
(_stp_string_cat_cstr): Ditto.
(_stp_string_cat_char): Ditto.
|
| |
|
|
|
|
|
| |
* lket/b2a/lket_b2a.c (main): Correct arguments to fseek.
(dump_data): Work around a spurious compiler warning.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
PR 3233
./
* stapfuncs.5.in: Document new gettimeofday_ns.
runtime/
* time.c (stp_time_t): Use ns for the base time, and freq is now kHz.
(__stp_estimate_cpufreq): Compute kHz instead of MHz.
(__stp_time_timer_callback, __stp_init_time): Compute base in ns.
(__stp_time_cpufreq_callback): Record kHz instead of MHz.
(_stp_init_time): Record kHz, and disable preemption to avoid a race
in the cpufreq notifier.
(_stp_gettimeofday_ns): Converted from _stp_gettimeofday_us.
tapset/
* timestamp.stp (gettimeofday_ns): New function
(gettimeofday_us, gettimeofday_ms, gettimeofday_s):
Use gettimeofday_ns as the base unit.
testsuite/
* buildok/timestamp.stp: add gettimeofday_ns test.
|
| |
|
|
|
|
|
|
|
|
|
| |
Be sync with the runtime printing changes made by Martin
lket/b2a/lket_b2a.c, lket/b2a/lket_b2a.h:
Be sync with the recent runtime printing changes made by Martin.
Some clean up work before adding the support of dumping data into MySQL
Bug fix of segment fault when an event is not registered.
|
|
|
|
|
|
| |
PR 3220
From Eugeniy Meshcheryakov <eugen@debian.org>:
* lket/b2a/lket_b2a.c: Check that appNameTree exists before destroying.
|
| |
|