aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAge
...
* parse-events: Allow expressions in __print_symbolic() fieldsStefan Hajnoczi2012-02-08
| | | | | | | | | | | | | | | | | | The __print_symbolic() function takes a sequence of key-value pairs for pretty-printing a constant. The new kvm:kvm_exit print fmt uses the expression: __print_symbolic(..., { 0x040 + 1, "DB excp" }, ...) Currently only atoms are supported and this print fmt fails to parse. This patch adds support for expressions instead of just atoms so that 0x040 + 1 is parsed successfully. Link: http://lkml.kernel.org/r/1315131959-7452-1-git-send-email-stefanha@linux.vnet.ibm.com Signed-off-by: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com> Cc: Avi Kivity <avi@redhat.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* trace-cmd: fix length calculation for padding eventsDavid Sharp2012-01-27
| | | | | | | | | | | | | Padding events store size in bytes, not words. Usually this ends up pushing the parser off the current page, but occasionally not, and when not, it ends up desynchronizing the parser from the event stream. This would manifest as a lot of "ug!" messages from trace-cmd, and kernelshark crashing before displaying the UI. Link: http://lkml.kernel.org/r/1327633215-4013-1-git-send-email-dhsharp@google.com Signed-off-by: David Sharp <dhsharp@google.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* blk plugin: replace BLK_TC_BARRIER with BLK_TC_FLUSH/BLK_TC_FUAStefan Hajnoczi2012-01-09
| | | | | | | | | | | | | | | | | | | | | The BLK_TC_BARRIER flag was dropped in Linux commit c09c47caedc in August 2011. The blk plugin fails to build against recent kernel headers. Since no flag bits were left, the new BLK_TC_FLUSH flag reused the BLK_TC_BARRIER bit. The new BLK_TC_FUA flag was also added. This patch updates fill_rwbs() to reflect the new BLK_TC_FLUSH/BLK_TC_FUA flags. This allows plugin_blk.c to build successfully on recent kernels. Most of the patch deals with detecting old vs new kernel headers so we can build successfully on both. (Namhyung Kim recommended using the makefile check used by perf and other tools) Link: http://lkml.kernel.org/r/1326108639-13904-1-git-send-email-stefanha@linux.vnet.ibm.com Signed-off-by: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com> Cc: Namhyung Kim <namhyung@gmail.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* parse-events: Fix %pM print format arg handlingSteven Rostedt2012-01-09
| | | | | | | | | | | | When %pM is used, the arg value must be a 6 byte character that will be printed as a 6 byte MAC address. But the code does a break over the main code which updates the current processing arg to point to the next arg. If there are other print arguments after a %pM, they will be off by one. The next arg will still be processing the %pM arg. Reported-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* trace-cmd: Cross-compile fixes for LDFLAGS and include pathDarren Hart2011-11-30
| | | | | | | | | | | Add ability for the Makefile to respect LDFLAGS. Also remove hardcoded /usr/local/include include path. Link: http://lkml.kernel.org/r/4ED6C808.9030003@linux.intel.com Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org> Signed-off-by: Darren Hart <dvhart@linux.intel.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* parse-events: Add support to show migrate disable counterSteven Rostedt2011-11-16
| | | | | | | The RT kernel added a migrate disable counter in all events. Add support to show this in the latency format. Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* parse-events: Add support for "%.*s" in bprintk eventsSteven Rostedt2011-11-15
| | | | | | | | | The arg notation of '*' in bprintks is not handled by the parser. Implement it so that they show up properly in the output and do not kill the tracer from reporting events. Reported-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* parse-event: Let filtering numbers by string use function namesSteven Rostedt2011-11-15
| | | | | | | | | As a pointer can be converted into a function name, let the filters work with the function name as well as with the pointer number. If the comparison expects a string, then convert numbers into functions, but only when the number is the same size as a long. Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* parse-event: Fix memset pointer size bugsSteven Rostedt2011-11-07
| | | | | | Fix memset(ptr, 0, sizeof(ptr)) to memset(ptr, 0, sizeof(*ptr)); Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* trace-cmd: Version 1.2Steven Rostedt2011-11-07
| | | | Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* trace-cmd: Build and install doc by defaultSteven Rostedt2011-11-07
| | | | | | | | I have been informed that documentation (man pages) should be installed by default and not just separate. Reported-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* parse-event: Fix parsing of __print_flags() in TP_printk()Steven Rostedt2011-11-07
| | | | | | | | | | | | A update is made to the sched:sched_switch event that adds some logic to the first parameter of the __print_flags() that shows the state of tasks. This change cause trace-cmd to fail parsing the flags. A simple fix is needed to have the parser be able to process ops within the argument. Reported-by: Andrew Vagin <avagin@openvz.org> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* parse-events: Modify header to work in C++ programsSteven Rostedt2011-10-17
| | | | | | | Replace keywords "private" and "bool" from parse-events.h to allow it to be used in C++ programs. Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* parse-events: Allow '*' and '/' operations in TP_printkSteven Rostedt2011-10-15
| | | | | | Add multiply and divide operations in the printk format. Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* trace-cmd: Do not record all events twice in extractSteven Rostedt2011-10-14
| | | | | | | | If -a is set with trace-cmd extract, simply ignore it, otherwise, all events will be recorded twice. Once at this location, and once during extracting. Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* trace-cmd: Do not use threads for extractSteven Rostedt2011-10-14
| | | | | | | | | | | | | | Currently the trace-cmd extract function uses the forked threads to read the data from the ring buffers. As this is not a live trace and the program does not end till all the buffers are flushed to disk, there is no need to use the threads. Just serially read each buffer into a file and then pull them together normally. This also fixes a bug that was triggered in the kernel where reading the trace_pipe_raw file after EOF will yield the last page again, causing the trace-cmd extract to produce duplicate pages. Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* trace-cmd: Use splice to filter out rest of bufferSteven Rostedt2011-10-14
| | | | | | | | | When the main thread tells the recorders that they are finished, have the recorders still use splice till all data is read that splice can handle (full page sizes only). Then use the read function for the rest. Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* trace-cmd: Do not make trace-cmd have to build with ptraceSteven Rostedt2011-09-28
| | | | | | | | If ptrace is not available on an OS or arch, do not fail the build. Simply do not support the child tracing feature. Reported-by: Shawn Bohrer <sbohrer@rgmadvisors.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* trace-cmd: Allow report to ignore the date optionSteven Rostedt2011-09-02
| | | | | | | | | | | Add --nodate to trace-cmd report to ignore the date offset and give the raw timestamp count of the records. This also adds a flags field to the tracecmd_input struct and is set and cleared with tracecmd_set_flag() and tracecmd_clear_flag() respectively. Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* trace-cmd: Have timestamp offset set at cpu descriptor not recordSteven Rostedt2011-09-02
| | | | | | | | | The cpu descriptor is referenced by many functions, and needs to include the ts_offset, otherwise those functions will be incorrect. The record->ts is based on the cpu descriptor timestamp, so by modifying the cpu timestamp the record->ts is also modified. Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* trace-cmd: Record CPU trace stats before disable allSteven Rostedt2011-08-18
| | | | | | | | | | | | The disable_all() function is called before we print out the per cpu buffer stats. But the disable_all() function also resets those same stats so the output is useless. Instead of printing out the stats while tracing is still active (although tracing_on is off), record the stats into individual per cpu trace_seq buffers and then still print when all has been disabled. Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* trace-cmd: Show better data when reporting offsetsSteven Rostedt2011-08-18
| | | | | | | | | | | | | | | | | | After a record, the offset of the CPU data is printed out for the user to see. This format is not very informative as the data just says "offset=27f000". Change this to a better format of: CPU0 data recorded at offset=0x2d5000 319488 bytes in size CPU1 data recorded at offset=0x323000 258048 bytes in size CPU2 data recorded at offset=0x362000 454656 bytes in size CPU3 data recorded at offset=0x3d1000 90112 bytes in size Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* trace-cmd: Update man pages for 'check-events' optionVaibhav Nagarnaik2011-08-18
| | | | | | | | | | | | | | | There is a new option added to the 'report' target to check event formats of the events recorded in a trace data file. This patch updates the documentation in the man page. This patch also adds a new man page for 'check-events' target which verifies event formats on the local machine. Cc: Michael Rubin <mrubin@google.com> Cc: David Sharp <dhsharp@google.com> Signed-off-by: Vaibhav Nagarnaik <vnagarnaik@google.com> Link: http://lkml.kernel.org/r/1313091459-12049-5-git-send-email-vnagarnaik@google.com Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* trace-cmd: Add --check-events to reportVaibhav Nagarnaik2011-08-18
| | | | | | | | | | | | | Add a new option --check-events to the report target which makes sure that the event formats in the trace data file can be parsed correctly on the machine that trace-cmd is running on. It returns EINVAL(22) for parsing failures and 0 for success. Cc: Michael Rubin <mrubin@google.com> Cc: David Sharp <dhsharp@google.com> Signed-off-by: Vaibhav Nagarnaik <vnagarnaik@google.com> Link: http://lkml.kernel.org/r/1313091459-12049-4-git-send-email-vnagarnaik@google.com Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* trace-cmd: Load plugins before parsing events in check-eventsVaibhav Nagarnaik2011-08-18
| | | | | | | | | | | | | | | The target check-events parses the event format strings on the local machine and returns whether they can be parsed or not. The parsing functionality is extended by loading plugins, if available. This patch loads the plugins before starting parsing of the event formats. Cc: Michael Rubin <mrubin@google.com> Cc: David Sharp <dhsharp@google.com> Signed-off-by: Vaibhav Nagarnaik <vnagarnaik@google.com> Link: http://lkml.kernel.org/r/1313091459-12049-3-git-send-email-vnagarnaik@google.com Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* trace-cmd: Don't call exit from tracecmd_blk_hack()Vaibhav Nagarnaik2011-08-18
| | | | | | | | | | | | | Commit 9e0fd22b14805a3a3219a99bc5eccebf70f5484a adds a hack to parse blktrace events. In tracecmd_blk_hack(), if the block trace events could not be parsed, it calls exit(0). Get rid of it and only return -1 to let the caller know that parsing failed. Cc: Michael Rubin <mrubin@google.com> Cc: David Sharp <dhsharp@google.com> Signed-off-by: Vaibhav Nagarnaik <vnagarnaik@google.com> Link: http://lkml.kernel.org/r/1313091459-12049-2-git-send-email-vnagarnaik@google.com Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* trace-cmd: Add function declaration to fix compile warningVaibhav Nagarnaik2011-08-18
| | | | | | | | | | | The function trace_util_ftrace_options() is not declared and causes a compile warning. This patch declares it in trace-cmd.h. Cc: Michael Rubin <mrubin@google.com> Cc: David Sharp <dhsharp@google.com> Signed-off-by: Vaibhav Nagarnaik <vnagarnaik@google.com> Link: http://lkml.kernel.org/r/1313091459-12049-1-git-send-email-vnagarnaik@google.com Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* trace-cmd: Simplify check-event pevent error handlingSteven Rostedt2011-07-29
| | | | Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* parse-events: Let pevent_free() take a NULL pointerSteven Rostedt2011-07-29
| | | | | | | | The pevent_free() should ack like other free()s and allow a NULL pointer to be passed to it which makes error handling a bit easier for the users of pevent_free(). Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* trace-cmd: Add parse error checking targetVaibhav Nagarnaik2011-07-29
| | | | | | | | | | | | | | Add another target 'check-events' which parses all the event formats and returns whether there are any issues with the print format strings. With an error in the format, the return value is 22 (EINVAL) and for success, it is 0. Cc: Michael Rubin <mrubin@google.com> Cc: David Sharp <dhsharp@google.com> Signed-off-by: Vaibhav Nagarnaik <vnagarnaik@google.com> Link: http://lkml.kernel.org/r/1311619188-4905-1-git-send-email-vnagarnaik@google.com Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* trace-cmd: Close open fds on execNathan Lynch2011-07-29
| | | | | | | | | | | | | | | | | | | | | | | | | trace-cmd's record mode allows child processes to inherit tracing-related file descriptors: $ trace-cmd record -esched: ls -l /proc/self/fd total 0 lrwx------. 1 root root 64 May 20 12:44 0 -> /dev/pts/11 lrwx------. 1 root root 64 May 20 12:44 1 -> /dev/pts/11 lrwx------. 1 root root 64 May 20 12:44 2 -> /dev/pts/11 lrwx------. 1 root root 64 May 20 12:44 3 -> /sys/kernel/debug/tracing/tracing_on l-wx------. 1 root root 64 May 20 12:44 4 -> /sys/kernel/debug/tracing/set_ftrace_pid l-wx------. 1 root root 64 May 20 12:44 5 -> /sys/kernel/debug/tracing/tracing_enabled lr-x------. 1 root root 64 May 20 12:44 6 -> /proc/31036/fd This can be problematic for applications (such as lxc) that interrogate their open file descriptors. Open these with O_CLOEXEC so they are not inherited. Signed-off-by: Nathan Lynch <ntl@pobox.com> Link: http://lkml.kernel.org/r/1305915732.2429.4.camel@orca.stoopid.dyndns.org Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* parse-events: Handle opcode parsing errorVaibhav Nagarnaik2011-07-29
| | | | | | | | | | | | | | | If an invalid opcode is encountered in parsing event print format, the trace-cmd calls exit() without parsing any other events. This patch adds handling for such an error where the get_op_prio() is called. If the return value is -1, then the event print format parsing is skipped and parsing continues. Cc: Michael Rubin <mrubin@google.com> Cc: David Sharp <dhsharp@google.com> Signed-off-by: Vaibhav Nagarnaik <vnagarnaik@google.com> Link: http://lkml.kernel.org/r/1311619257-4970-1-git-send-email-vnagarnaik@google.com Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* parse-events: Support '+' opcode in print formatVaibhav Nagarnaik2011-07-29
| | | | | | | | | | | The '+' opcode is not supported in the arguments for the print format. This patch adds support for it. Cc: Michael Rubin <mrubin@google.com> Cc: David Sharp <dhsharp@google.com> Signed-off-by: Vaibhav Nagarnaik <vnagarnaik@google.com> Link: http://lkml.kernel.org/r/1310785241-3799-4-git-send-email-vnagarnaik@google.com Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* parse-events: Handle invalid opcode parsing gracefullyVaibhav Nagarnaik2011-07-29
| | | | | | | | | | | | | | | | If an invalid opcode is encountered, trace-cmd exits with an error. Instead it can be treated as a soft error where the event's print format is not parsed and its binary data is dumped out. This patch adds a return value to arg_num_eval() function to indicate if the parsing was successful. If not, then the error is considered soft and the parsing of the offending event fails. Cc: Michael Rubin <mrubin@google.com> Cc: David Sharp <dhsharp@google.com> Signed-off-by: Vaibhav Nagarnaik <vnagarnaik@google.com> Link: http://lkml.kernel.org/r/1310785241-3799-2-git-send-email-vnagarnaik@google.com Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* trace-cmd: Fix warning for undeclared function trace_util_ftrace_optionsSteven Rostedt2011-07-29
| | | | Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* trace-cmd: Add check for truncated filesVaibhav Nagarnaik2011-06-29
| | | | | | | | | | | | | | | | If the trace data file that is being read is truncated, mmap'ing its non-existent page does not give any error. When the page is read, the process dies with a SIGBUS. This patch adds a file size check while initializing the trace data and makes sure that the file size matches the expected size; otherwise an error is returned. Cc: Michael Rubin <mrubin@google.com> Cc: David Sharp <dhsharp@google.com> Signed-off-by: Vaibhav Nagarnaik <vnagarnaik@google.com> Link: http://lkml.kernel.org/r/1308615578-556-1-git-send-email-vnagarnaik@google.com Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* trace-cmd: Use size of record to determine kernel stack traceSteven Rostedt2011-06-29
| | | | | | | | | The kernel stack trace may soon become dynamic, and the size will depend on the size of the record and not the size of the stack array. Use the size of the record to determine the end of the stack output instead of the array. Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* trace-cmd: Allow use of enabling the nop tracerSteven Rostedt2011-06-29
| | | | | | | | For those cases that we just want to enable tracing because we want to view trace_printk()s in the kernel or for whatever, allow the nop tracer to be enabled. Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* trace-cmd: Add hack to ignore ARM '$x' functionsSteven Rostedt2011-06-08
| | | | | | | | | | | | | | | | | | | | | | | | | | Some ARM boards add '$x' symbols into the kallsyms file: 00000000bf0007e8 $a [cfg80211] 00000000bf000890 $a [cfg80211] 00000000bf000a30 $d [cfg80211] 00000000bf000a40 $a [cfg80211] 00000000bf000e78 $d [cfg80211] 00000000bf000e88 $a [cfg80211] 00000000bf0011f8 $d [cfg80211] 00000000bf001244 $a [cfg80211] 00000000bf0012a8 $a [cfg80211] 00000000bf00131c $a [cfg80211] 00000000bf001340 $a [cfg80211] 00000000bf00146c $d [cfg80211] Unfortunately, trace-cmd seems to pick these for the function names it reports, making the function names totally useless. Add a hack to not include these names in the function table of trace-cmd. Reported-by: Luciano Coelho <coelho@ti.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* trace-graph: Fix saved pointer corruption in update_last_record()Steven Rostedt2011-04-18
| | | | | | | | The saved pointer was created with tracecmd_peek_data() and then set_cpu_to_time() was called freeing the saved pointer. We could simply reference the pointer if need be, but using tracecmd_read_data() is a better solution. Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* trace-cmd: Release handle in proper location in trace-output.cSteven Rostedt2011-04-18
| | | | | | | | | tracecmd_append_cpu_data() does not create the handle and should not release it. Move the release of the handle to the location that created it: tracecmd_create_file_glob() Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* trace-cmd: Print an error if record exec failsSteven Rostedt2011-03-30
| | | | | | | | Currently if the exec fails in trace-cmd record, it silently exits without letting the user know. Add a big message telling the user if the exec fails. Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* Fix address of the Free Software FoundationUwe Kleine-König2011-03-22
| | | | | Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* trace-cmd: Add ftrace options with fgraph tailprint optionSteven Rostedt2011-03-18
| | | | | | | | | Added internal options for ftrace, and included a fgraph:tailprint option that will print the function names at the function exit. When the option is set, instead of just '}' the name will be added '} /* func_name */' Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* trace-cmd: Fix record of function_graphSteven Rostedt2011-03-18
| | | | | | | | | For safety reasons we always set ftrace_stack_trace to zero when starting function tracer. But we seem to be doing that for function_graph tracer as well, where that option does not exist. Don't fail if it does not exist, quietly move on. Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* parse-events: fold meSteven Rostedt2011-03-18
|
* trace-cmd: Add trace-cmd optionsSteven Rostedt2011-03-18
| | | | | | | Add trace-cmd options that reads all the plugins and lists the available options for trace-cmd report. Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* trace-cmd: Add -O to trace-cmd report to pass options to pluginsSteven Rostedt2011-03-18
| | | | | | | | The -O field now passes options to the plugins. The format is of: -O [plugin:]var[=val] Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* trace-cmd: Add utility to add options to pluginsSteven Rostedt2011-03-18
| | | | | | Added trace_util_add_option() and trace_util_read_plugin_options(); Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* parse-events: Add names for plugin optionsSteven Rostedt2011-03-18
| | | | | | | Add PEVENT_PLUGIN_ALIAS and PEVENT_PLUGIN_OPTIONS for plugins to use. Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
generated by cgit v1.2.2 (git 2.25.0) at 2025-04-04 07:05:52 -0400