Messages in this thread | | | From | Adrian Hunter <> | Subject | [PATCH V6 00/17] perf tools: Introduce an abstraction for AUX Area and Instruction Tracing | Date | Fri, 29 May 2015 16:33:28 +0300 |
| |
Hi
Here is V6 patches for the introduction of an abstraction for using the AUX area and Instruction tracing. The patches for AUX area support have been applied, just leaving patches for Intel PT and Intel BTS.
The patches can also be found here:
http://git.infradead.org/users/ahunter/linux-perf.git
An example (unchanged from V3) perf.data file and build id archive can be found here:
http://git.infradead.org/~ahunter/tfr/
There is also a tar of the 3 most relevant files with debugging symbols. These need to be placed in under the correct paths in /usr/lib/debug to get symbols.
Changes in V6:
Some minor expansion of commit messages.
Patches already applied: perf tools: Disallow PMU events intel_pt and intel_bts until there is support
perf db-export: Fix thread ref-counting New patch
perf tools: Ensure thread-stack is flushed New patch
perf tools: Add Intel PT support Support thread ref-counting
perf tools: Add Intel PT decoder Fix a bug: FUP packet in PSB to update last IP
perf tools: Take Intel PT into use Add Overview and Quickstart sections to intel_pt.txt
perf tools: Add Intel BTS support Add Overview to intel_bts.txt Support thread ref-counting
perf tools: Add example call-graph script Add documentation comments to scripts
Changes in V5:
Patches already applied: perf report: Fix placement of itrace option in documentation perf tools: Add AUX area tracing index perf tools: Hit all build ids when AUX area tracing perf tools: Add build option NO_AUXTRACE to exclude AUX area tracing perf auxtrace: Add option to synthesize events for transactions perf tools: Add support for PERF_RECORD_AUX perf tools: Add support for PERF_RECORD_ITRACE_START perf tools: Add AUX area tracing Snapshot Mode perf record: Add AUX area tracing Snapshot Mode support
perf tools: Disallow PMU events intel_pt and intel_bts until there is support New patch
perf tools: Add Intel PT decoder Style improvements pointed out by Acme: aligning '=', single line initializing Make use of zalloc() not malloc / memset Make use of zfree Map internal error codes to fixed constants for output Change intel_pt_error_message() to intel_pt__strerror()
perf tools: Add Intel PT support Make use of zfree
perf tools: Take Intel PT into use Allow "intel_pt" PMU to be selected as an event
perf tools: Add Intel BTS support Allow "intel_bts" PMU to be selected as an event Make use of zfree Map internal error codes to fixed constants for output Let "intel_bts" show up in 'perf list'
perf tools: Output sample flags and insn_len from intel_bts Map internal error codes to fixed constants for output
Changes on V4:
perf tools: Amend mmap ref counting for the AUX area mmap Dropped because already applied
perf script: Always allow fields 'addr' and 'cpu' for auxtrace Dropped because already applied
perf report: Add Instruction Tracing support Dropped because already applied
perf report: Fix placement of itrace option in documentation New patch
perf tools: Add AUX area tracing index Change size checks for more flexibility i.e. - don't mind if an indexed auxtrace_event is bigger than struct auxtrace_event - don't mind if the auxtrace index does not fill the whole file section Rename 'index' variable to 'ent' to avoid build errors on older gcc
perf tools: Add build option NO_AUXTRACE to exclude AUX area tracing Fix whitespace alignment of NO_AUXTRACE=1 Add NO_AUXTRACE=1 to make_minimal
perf tools: Add support for PERF_RECORD_AUX Expand commit message
perf tools: Add AUX area tracing Snapshot Mode Whitespace fixups
perf record: Add AUX area tracing Snapshot Mode support Whitespace fixups Don't init static variables to 0 or NULL
perf tools: Add Intel PT packet decoder Whitespace fixups
perf tools: Add Intel PT instruction decoder Avoid build error on older (broken) gcc by adding -Wno-override-init Avoid build errors due to funny collate sequences i.e. use LC_COLLATE=C etc
perf tools: Add Intel PT decoder Avoid build errors initializing structures to 0
perf tools: Add Intel PT support Avoid build errors initializing structures to 0 Allow for perf_pmu__config_terms() having an extra parameter now Allow for parse_events() having an extra parameter now Rename 'div' variable to 'd' to avoid build errors Whitespace fixup Remove a couple of unused enums
perf tools: Add Intel BTS support Avoid build errors initializing structures to 0 Allow for parse_events() having an extra parameter now
perf tools: Put itrace options into an asciidoc include New patch
Changes in V3:
New patch: perf tools: Amend mmap ref counting for the AUX area mmap
Move some code under arch: perf tools: Add Intel PT support perf tools: Add Intel BTS support
Updated documentation: perf report: Add Instruction Tracing support perf auxtrace: Add option to synthesize events for transactions perf tools: Take Intel PT into use perf tools: Add Intel BTS support
Patches already applied: perf header: Add AUX area tracing feature perf evlist: Add support for mmapping an AUX area buffer perf tools: Add user events for AUX area tracing perf tools: Add support for AUX area recording perf record: Add basic AUX area tracing support perf record: Extend -m option for AUX area tracing mmap pages perf tools: Add a user event for AUX area tracing errors perf session: Add hooks to allow transparent decoding of AUX area tracing data perf session: Add instruction tracing options perf auxtrace: Add helpers for AUX area tracing errors perf auxtrace: Add helpers for queuing AUX area tracing data perf auxtrace: Add a heap for sorting AUX area tracing queues perf auxtrace: Add processing for AUX area tracing events perf auxtrace: Add a hashtable for caching perf tools: Add member to struct dso for an instruction cache perf script: Add Instruction Tracing support perf inject: Re-pipe AUX area tracing events perf inject: Add Instruction Tracing support perf script: Add field option 'flags' to print sample flags perf tools: Add aux_watermark member of struct perf_event_attr
Changes in V2:
Get rid of MIN() perf auxtrace: Add helpers for AUX area tracing errors perf inject: Re-pipe AUX area tracing events perf tools: Add build option NO_AUXTRACE to exclude AUX area tracing
Intel BTS can be used on most recent Intel CPUs. Intel PT is available on Broadwell.
Examples:
Trace 'ls' with Intel BTS userspace only
perf record --per-thread -e intel_bts//u ls perf report perf script
Trace 'ls' with Intel BTS kernel and userspace
~/libexec/perf-core/perf-with-kcore record bts-ls --per-thread -e intel_bts// -- ls ~/libexec/perf-core/perf-with-kcore report bts-ls ~/libexec/perf-core/perf-with-kcore script bts-ls
Trace 'ls' with Intel PT userspace only
perf record -e intel_pt//u ls perf report perf script
Trace 'ls' with Intel PT kernel and userspace
~/libexec/perf-core/perf-with-kcore record pt-ls -e intel_pt// -- ls ~/libexec/perf-core/perf-with-kcore report pt-ls ~/libexec/perf-core/perf-with-kcore script pt-ls
The abstraction has two separate aspects: 1. recording AUX area data 2. processing AUX area data
Recording consists of mmapping a separate buffer and copying the data into the perf.data file. The buffer is an AUX area buffer. The data is written preceded by a new user event PERF_RECORD_AUXTRACE. The data is too big to fit in the event but follows immediately afterward. Session processing has to skip to get to the next event header in a similar fashion to the existing PERF_RECORD_HEADER_TRACING_DATA event. The main recording patches are:
perf evlist: Add support for mmapping an AUX area buffer perf tools: Add user events for AUX area tracing perf tools: Add support for AUX area recording perf record: Add basic AUX area tracing support
Processing consists of providing hooks in session processing to enable a decoder to see all the events and deliver synthesized events transparently into the event stream. The main processing patch is:
perf session: Add hooks to allow transparent decoding of AUX area tracing data
Adrian Hunter (17): perf db-export: Fix thread ref-counting perf tools: Ensure thread-stack is flushed perf auxtrace: Add Intel PT as an AUX area tracing type perf tools: Add Intel PT packet decoder perf tools: Add Intel PT instruction decoder perf tools: Add Intel PT log perf tools: Add Intel PT decoder perf tools: Add Intel PT support perf tools: Take Intel PT into use perf tools: Allow auxtrace data alignment perf tools: Add Intel BTS support perf tools: Output sample flags and insn_len from intel_pt perf tools: Output sample flags and insn_len from intel_bts perf tools: Intel PT to always update thread stack trace number perf tools: Intel BTS to always update thread stack trace number perf tools: Put itrace options into an asciidoc include perf tools: Add example call-graph script
tools/build/Makefile.build | 2 + tools/perf/.gitignore | 2 + tools/perf/Documentation/intel-bts.txt | 86 + tools/perf/Documentation/intel-pt.txt | 588 ++++++ tools/perf/Documentation/itrace.txt | 22 + tools/perf/Documentation/perf-inject.txt | 23 +- tools/perf/Documentation/perf-report.txt | 23 +- tools/perf/Documentation/perf-script.txt | 23 +- tools/perf/Makefile.perf | 12 +- tools/perf/arch/x86/util/Build | 5 + tools/perf/arch/x86/util/auxtrace.c | 83 + tools/perf/arch/x86/util/intel-bts.c | 458 +++++ tools/perf/arch/x86/util/intel-pt.c | 752 ++++++++ tools/perf/arch/x86/util/pmu.c | 18 + .../scripts/python/call-graph-from-postgresql.py | 327 ++++ tools/perf/scripts/python/export-to-postgresql.py | 47 + tools/perf/util/Build | 3 + tools/perf/util/auxtrace.c | 16 +- tools/perf/util/auxtrace.h | 3 + tools/perf/util/db-export.c | 19 +- tools/perf/util/intel-bts.c | 921 ++++++++++ tools/perf/util/intel-bts.h | 43 + tools/perf/util/intel-pt-decoder/Build | 14 + .../perf/util/intel-pt-decoder/intel-pt-decoder.c | 1759 ++++++++++++++++++ .../perf/util/intel-pt-decoder/intel-pt-decoder.h | 102 ++ .../util/intel-pt-decoder/intel-pt-insn-decoder.c | 246 +++ .../util/intel-pt-decoder/intel-pt-insn-decoder.h | 65 + tools/perf/util/intel-pt-decoder/intel-pt-log.c | 155 ++ tools/perf/util/intel-pt-decoder/intel-pt-log.h | 52 + .../util/intel-pt-decoder/intel-pt-pkt-decoder.c | 400 +++++ .../util/intel-pt-decoder/intel-pt-pkt-decoder.h | 64 + tools/perf/util/intel-pt.c | 1895 ++++++++++++++++++++ tools/perf/util/intel-pt.h | 51 + tools/perf/util/machine.c | 21 + tools/perf/util/machine.h | 3 + tools/perf/util/pmu.c | 4 - tools/perf/util/session.c | 20 + tools/perf/util/thread-stack.c | 18 +- tools/perf/util/thread-stack.h | 1 + 39 files changed, 8259 insertions(+), 87 deletions(-) create mode 100644 tools/perf/Documentation/intel-bts.txt create mode 100644 tools/perf/Documentation/intel-pt.txt create mode 100644 tools/perf/Documentation/itrace.txt create mode 100644 tools/perf/arch/x86/util/auxtrace.c create mode 100644 tools/perf/arch/x86/util/intel-bts.c create mode 100644 tools/perf/arch/x86/util/intel-pt.c create mode 100644 tools/perf/arch/x86/util/pmu.c create mode 100644 tools/perf/scripts/python/call-graph-from-postgresql.py create mode 100644 tools/perf/util/intel-bts.c create mode 100644 tools/perf/util/intel-bts.h create mode 100644 tools/perf/util/intel-pt-decoder/Build create mode 100644 tools/perf/util/intel-pt-decoder/intel-pt-decoder.c create mode 100644 tools/perf/util/intel-pt-decoder/intel-pt-decoder.h create mode 100644 tools/perf/util/intel-pt-decoder/intel-pt-insn-decoder.c create mode 100644 tools/perf/util/intel-pt-decoder/intel-pt-insn-decoder.h create mode 100644 tools/perf/util/intel-pt-decoder/intel-pt-log.c create mode 100644 tools/perf/util/intel-pt-decoder/intel-pt-log.h create mode 100644 tools/perf/util/intel-pt-decoder/intel-pt-pkt-decoder.c create mode 100644 tools/perf/util/intel-pt-decoder/intel-pt-pkt-decoder.h create mode 100644 tools/perf/util/intel-pt.c create mode 100644 tools/perf/util/intel-pt.h
Regards Adrian
| |