summaryrefslogtreecommitdiffstats
path: root/runtime
Commit message (Collapse)AuthorAgeFilesLines
...
* 2006-09-21 Martin Hunt <hunt@redhat.com>hunt2006-09-223-16/+34
| | | | | * transport.c (_stp_transport_init): Call _stp_print_init(). (_stp_cleanup_and_exit): Call _stp_print_cleanup().
* 2006-09-21 Martin Hunt <hunt@redhat.com>hunt2006-09-223-7/+31
| | | | | | | | | | | 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.
* Use semaphore if mutex not availabletrz2006-09-212-2/+27
|
* 2006-09-20 Josh Stone <joshua.i.stone@intel.com>jistone2006-09-202-2/+5
| | | | | * lket/b2a/lket_b2a.c (main): Correct arguments to fseek. (dump_data): Work around a spurious compiler warning.
* 2006-09-20 Josh Stone <joshua.i.stone@intel.com>jistone2006-09-202-32/+63
| | | | | | | | | | | | | | | | | | | | | | | | 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.
* Conversion from relay-on-procfs to relay-on-debugfstrz2006-09-197-164/+142
|
* lket_trace.stp,process.stp,register_event.stp:guanglei2006-09-193-96/+138
| | | | | | | | | 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.
* 2006-09-18 Josh Stone <joshua.i.stone@intel.com>jistone2006-09-182-1/+8
| | | | | | PR 3220 From Eugeniy Meshcheryakov <eugen@debian.org>: * lket/b2a/lket_b2a.c: Check that appNameTree exists before destroying.
* *** empty log message ***hunt2006-09-181-0/+7
|
* 2006-09-18 Martin Hunt <hunt@redhat.com>hunt2006-09-182-5/+48
| | | | | | | * bench2/bench.rb: Remove the deprecated "-m" option to stpd. Also, don't use merge mode for relayfs. * bench2/run_binary_print: New test. Compare various ways of printing binary data.
* 2006-09-18 Martin Hunt <hunt@redhat.com>hunt2006-09-183-1/+14
| | | | | * transport_msgs.h (struct transport_info): Add merge field. * transport.c (_stp_transport_open): Set merge field.
* 2006-09-18 Martin Hunt <hunt@redhat.com>hunt2006-09-183-35/+57
| | | | | | | | | * stpd.c (usage): Remove "-m" option. (main): Print warning if "-m" is used. * librelay.c (merge_output): Rewrite to handle new format that support binary. (stp_main_loop): Read merge option from the transport info message.
* 2006-09-18 Martin Hunt <hunt@redhat.com>hunt2006-09-184-88/+155
| | | | | | | | | | | | | | | | * print.c (_stp_print_flush): Rewrite so one version works for relayfs or procfs. Use proper per-cpu functions. (_stp_reserve_bytes): New function. Reserve bytes in the output buffer. (_stp_print_binary): New function. Write a variable number of 64-bit values directly into the output buffer. * string.c (_stp_sprintf): Rewrite using new per-cpu buffers. (_stp_vsprintf): Ditto. (_stp_string_cat_cstr): Ditto. (_stp_string_cat_char): Ditto. * runtime.h: Set defaults for MAXTRYLOCK and TRYLOCKDELAY to make runtime tests in bench2 happy.
* 2006-09-13 Martin Hunt <hunt@redhat.com>hunt2006-09-134-59/+59
| | | | | | | | | | * librelay.c (init_relayfs): Exec stp_check and find relay_filebase. * librelay.h (stp_main_loop): Fix declaration of init_stp(). * stpd.c (usage): Remove "-r" option. (main): Don't find stpd_filebase and don't send it to init_stp().
* Systemtap perfmon support to access the processors perfmon hardware.wcohen2006-09-123-0/+165
|
* add another two kinds of timing mechanisms for LKET, i.e. get_cycles() andguanglei2006-08-293-6/+89
| | | | sched_clock()
* PR 3037wcohen2006-08-143-2/+6
| | | | | * string.c: * print.c: Remove includes for <linux/config.h>.
* 2006-08-11 David Smith <dsmith@redhat.com>dsmith2006-08-112-0/+6
| | | | | * lket/b2a/.cvsignore: Added file to let cvs ignore generated files.
* Fix for bug #2992trz2006-08-022-2/+13
|
* a bug fix for the zero-length stpd_cpu fileguanglei2006-07-251-2/+2
|
* 2006-07-21 Martin Hunt <hunt@redhat.com>hunt2006-07-212-0/+5
| | | | * runtime.h: Include compat.h.
* 2006-07-20 Martin Hunt <hunt@redhat.com>hunt2006-07-212-5/+12
| | | | * vsprintf.c (_stp_vsnprintf): %p format fixes.
* 2006-07-20 Martin Hunt <hunt@redhat.com>hunt2006-07-202-2/+9
| | | | | * librelay.c (stp_main_loop): If module doesn't start, kill any target command.
* bug fix of lket-b2a: segment fault caused by event registerguanglei2006-07-191-14/+8
| | | | | | change the timestamp printing format of lket-b2a: usec --> sec.usec fix an error in "CUSTOMIZED TRACE DATA" : addevent.netdev --> addevent.netdev.transmit
* 2006-07-14 Josh Stone <joshua.i.stone@intel.com>jistone2006-07-142-12/+41
| | | | | | | | | | | | PR 2922 * time.c (_stp_init_time): call cpufreq_get for each cpu from a non-atomic context (module init time). * time.c (__stp_init_time): remove call to cpufreq_get from IPI. * time.c (__stp_time_timer_callback): disable IRQ's to avoid perturbations in measurements. * time.c (_stp_gettimeofday_us): disable preemption to avoid CPU swaps while we're computing the time.
* 2006-07-12 Martin Hunt <hunt@redhat.com>hunt2006-07-122-0/+9
| | | | | * runtime.h (for_each_cpu): Define for new kernels which no longer define it.
* Minor cleanuphunt2006-07-121-3/+3
|
* 2006-07-11 Martin Hunt <hunt@redhat.com>hunt2006-07-122-8/+14
| | | | * string.c (_stp_text_str): Use __get_user().
* 2006-07-11 Martin Hunt <hunt@redhat.com>hunt2006-07-114-17/+42
| | | | | | * string.c (_stp_text_str): Add a parameter to support userspace strings too. * string.h: (_stp_text_str): Fix proto.
* changes to addevent.process to generate adequate info of process relationshipsguanglei2006-07-112-10/+25
| | | | | | | | use current->parent->tgid as the parent process id for ascii tracing modified lket-b2a to be sync with the new addevent.process event hooks updated lket man pages to be sync with newly changes to addevent.process event hooks
* _lket_trace() won't log ppid and tgid any more. I use some bit-OPs toguanglei2006-06-293-33/+39
| | | | | | make _lket_trace's format string shorter for better performance. I also modified lket-b2a to be sync with _lket_trace()
* .roland2006-06-271-0/+4
|
* 2006-06-27 Roland McGrath <roland@redhat.com>roland2006-06-271-1/+22
| | | | * loc2c-runtime.h: Add some comments.
* aio.stp: new event hooks for AIOguanglei2006-06-262-7/+71
| | | | | register_events.stp, hookid_defs.stp: changes for aio.stp. register_events.stp, process.stp: bugfix for ascii tracing
* 2006-06-23 Josh Stone <joshua.i.stone@intel.com>jistone2006-06-232-6/+23
| | | | * time.c: Do cpufreq notification only if CONFIG_CPU_FREQ defined.
* Part of fix for bug #2532trz2006-06-232-3/+10
|
* add PR 2525 note to ChangeLogsjistone2006-06-222-0/+2
|
* 2006-06-21 Josh Stone <joshua.i.stone@intel.com>jistone2006-06-224-1/+200
| | | | | | | | | | | | | runtime/ * time.c: Time-estimation with minimal dependency on xtime. runtime/transport/ * transport/transport.c (_stp_handle_start): Initialize timer functions. * transport.c (_stp_cleanup_and_exit): Teardown timer functions. tapset/ * timestamp.stp (gettimeofday_us, gettimeofday_ms, gettimeofday_s): Convert to using the runtime-provided _stp_gettimeofday_us().
* .roland2006-06-171-50/+54
|
* 2006-06-16 Roland McGrath <roland@redhat.com>roland2006-06-171-4/+2
| | | | * lket/b2a/Makefile.in: Regenerated with automake-1.9.6-2.
* 2006-06-13 Martin Hunt <hunt@redhat.com>hunt2006-06-132-17/+14
| | | | | * librelay.c (start_cmd): Rewrite using sigwait() to eliminate a race.
* 2006-06-13 Martin Hunt <hunt@redhat.com>hunt2006-06-132-3/+8
| | | | | * transport.c (_stp_transport_close): Destroy workqueue when done.
* 2006-06-09 Martin Hunt <hunt@redhat.com>hunt2006-06-092-2/+16
| | | | | | * transport.c: Using the default workqueue was causing problems because of the long delays on probe_exit(). Created a new systemtap work queue and used that instead.
* Revert my former changes to lket_trace_extra() in translator.guanglei2006-06-093-308/+205
| | | | | a new way of allowing user add arbitrary trace data. And also provide a way of letting user write the trace data format.
* 2006-06-05 Martin Hunt <hunt@redhat.com>hunt2006-06-052-8/+14
| | | | | * procfs.c (_stp_write): Always use spin_lock_irqsave() because this function can sometimes be called with interrupts enabled.
* eliminate lket-b2a compiling warnings on in 64-bit modeguanglei2006-06-022-7/+11
|
* make lket-b2a able to read user added trace data and backtraceguanglei2006-06-013-11/+37
|
* 2006-05-30 Martin Hunt <hunt@redhat.com>hunt2006-05-303-0/+106
| | | | | * string.c (_stp_text_str): New function. * string.h (_stp_text_str): Declare.
* add conditional build of lket-b2a. If glib2-devel is not found, just skipsguanglei2006-05-272-5/+2
| | | | | | and gives a warning deleted runtime/lket/b2a/Makefile since it should be generated by configure
* 2006-05-25 Martin Hunt <hunt@redhat.com>hunt2006-05-252-1/+22
| | | | | * vsprintf.c (_stp_vsnprintf): Change %p to work like libc and automatically insert "0x" before the pointer.
generated by cgit (git 2.34.1) at 2025-08-09 01:04:06 +0000