perf: enable compression of record mode trace to save storage space
Subsystems (2)
v10 cover perf: enable compression of record mode trace to save storage space
The patch set implements runtime trace compression (-z option) in record mode and trace auto decompression in report and inject modes. Streaming Zstd API [1] is used for compression and decompression of data that come from kernel mmaped data buffers.
Usage of implemented -z,--compression_level=n option provides ~3-5x avg. trace file size reduction on variety of tested workloads what saves storage space on larger server systems where trace file size can easily reach several tens or even hundreds of GiBs, especially when profiling with dwarf-based stacks and tracing of context switches. Default option value is 1 (fastest compression).
Implemented --mmap-flush option can be used to specify minimal size of data chunk that is extracted from mmaped kernel buffer to store into a trace. The option is independent from -z setting and doesn't vary with compression level. The default option value is 1 byte what means every time trace writing thread finds some new data in the mmaped buffer the data is extracted, possibly compressed and written to a trace. The option serves two purposes the first one is to increase the compression ratio of trace data and the second one is to avoid live-lock self tool process monitoring in system wide (-a) profiling mode. Profiling in system wide mode with compression (-a -z) can additionally induce data into the kernel buffers along with the data from monitored processes. If performance data rate and volume from the monitored processes is high then trace streaming and compression activity in the tool is also high. It can lead to subtle live-lock effect of endless activity when compression of single new byte from some of mmaped kernel buffer induces the next single byte at some mmaped buffer. So perf tool thread never stops on polling event file descriptors. Varying data chunk size to be extracted from mmap buffers allows avoiding live-locking self monitoring in system wide mode and makes mmap buffers polling loop manageable. Possible usage examples:
$ tools/perf/perf record -z -e cycles -- matrix.gcc $ tools/perf/perf record --aio -z -e cycles -- matrix.gcc $ tools/perf/perf record -z --mmap-flush 1024 -e cycles -- matrix.gcc $ tools/perf/perf record --aio -z --mmap-flush 1K -e cycles -- matrix.gcc
Runtime compression overhead has been measured for serial and AIO trace writing modes when profiling matrix multiplication workload:
-------------------------------------------------------------
| SERIAL | AIO-1 |
----|-----------------------------|-----------------------------| |-z | OVH(x) | ratio(x) size(MiB) | OVH(x) | ratio(x) size(MiB) | |---|--------|--------------------|--------|--------------------| | 0 | 1,00 | 1,000 179,424 | 1,00 | 1,000 187,527 | | 1 | 1,04 | 8,427 181,148 | 1,01 | 8,474 188,562 | | 2 | 1,07 | 8,055 186,953 | 1,03 | 7,912 191,773 | | 3 | 1,04 | 8,283 181,908 | 1,03 | 8,220 191,078 | | 5 | 1,09 | 8,101 187,705 | 1,05 | 7,780 190,065 | | 8 | 1,05 | 9,217 179,191 | 1,12 | 6,111 193,024 |
-----------------------------------------------------------------
OVH = (Execution time with -z N) / (Execution time with -z 0)
ratio - compression ratio
size - number of bytes that was compressed
size ~= trace file x ratio
See complete description of measurement conditions with details below.
Introduced compression functionality can be disabled or configured from the command line using NO_LIBZSTD and LIBZSTD_DIR defines:
$ make -C tools/perf NO_LIBZSTD=1 clean all $ make -C tools/perf LIBZSTD_DIR=/path/to/zstd/sources/ clean all
If your system has some version of the zstd package preinstalled then the build system finds and uses it during the build. Auto detection feature status is reported just before compilation starts, as usual. If you still prefer to compile with some other version of zstd you have capability to refer the compilation to that version using LIBZSTD_DIR define.
See 'perf test' results below for enabled and disabled (NO_LIBZSTD=1) feature configurations.
---
Alexey Budankov (12):
feature: implement libzstd check, LIBZSTD_DIR and NO_LIBZSTD defines perf record: implement --mmap-flush=<number> option
perf session: define bytes_transferred and bytes_compressed metrics perf record: implement COMPRESSED event record and its attributes perf mmap: implement dedicated memory buffer for data compression perf util: introduce Zstd streaming based compression API perf record: implement compression for serial trace streaming perf record: implement compression for AIO trace streaming perf record: implement -z,--compression_level[=<n>] option perf report: implement record trace decompression
perf inject: enable COMPRESSED records decompression
perf tests: implement Zstd comp/decomp integration test
tools/build/Makefile.feature | 6 +- tools/build/feature/Makefile | 6 +- tools/build/feature/test-all.c | 5 + tools/build/feature/test-libzstd.c | 12 + tools/perf/Documentation/perf-record.txt | 17 ++ .../Documentation/perf.data-file-format.txt | 24 ++ tools/perf/Makefile.config | 20 ++ tools/perf/Makefile.perf | 3 + tools/perf/builtin-inject.c | 4 + tools/perf/builtin-record.c | 285 +++++++++++++++--- tools/perf/builtin-report.c | 5 +- tools/perf/builtin-version.c | 2 + tools/perf/perf.h | 2 + .../tests/shell/record+zstd_comp_decomp.sh | 35 +++ tools/perf/util/Build | 2 + tools/perf/util/compress.h | 54 ++++ tools/perf/util/env.h | 11 + tools/perf/util/event.c | 1 + tools/perf/util/event.h | 7 + tools/perf/util/evlist.c | 8 +- tools/perf/util/evlist.h | 3 +- tools/perf/util/header.c | 55 +++- tools/perf/util/header.h | 1 + tools/perf/util/mmap.c | 106 ++----- tools/perf/util/mmap.h | 17 +- tools/perf/util/session.c | 129 +++++++- tools/perf/util/session.h | 14 + tools/perf/util/tool.h | 2 + tools/perf/util/zstd.c | 111 +++++++ 29 files changed, 813 insertions(+), 134 deletions(-) create mode 100644 tools/build/feature/test-libzstd.c create mode 100755 tools/perf/tests/shell/record+zstd_comp_decomp.sh create mode 100644 tools/perf/util/zstd.c
---
Changes in v10:
- separated decomp list deallocation into perf_session__release_decomp_events
- extended the test with suggested decompression validation
Changes in v9:
- fixed issue with improper max COMPRESSED record size calculation
- moved up calculation of ratio variable in 03/12
- made minor corrections in changelogs
- corrected several checkpatch.pl warnings and errors
Changes in v8:
- avoid using -f for --mmap-flush option
- move stubs to compress.h and avoid unconditional compiling of zstd.c
- fixed silent interruption for perf record collection
- implemented -z 1 as default
Changes in v7:
- rebased to Arnaldo's perf/core tip
- implemented B/K/M/G suffixes for -f option
- reworked record__mmap_read_evlist() to replace perf_mmap__aio_push() by perf_mmap__push() in AIO case
- extended "[ perf record: Captured ... ]" message with compression statistics
- extended changelog for v5 06/10
- used PERF_SAMPLE_MAX_SIZE for compressed record size calculations
- renamed record__zstd_compress to zstd_compress and record__process_comp_header to process_comp_header
- separated nr_cblocks_max applying
Changes in v6:
- extended docs with description of PERF_RECORD_COMPRESSED record and HEADER_COMPRESSED feature layouts
Changes in v5:
- implemented perf version --build-options extension for aio and zstd - see TESTING below
- adjusted commit message and perf-record.txt content for -f option
- fixed build errors in case of NO_AIO=1 and NO_LIBZSTD=1
Changes in v4:
- implemented integration tests
- adjusted zstd_ stub functions
- rebased on tip of Arnaldo's perf/core
Changes in v3:
- moved -f,--mmap-flush option implementation into a separate patch
- moved definition and printing of bytes_transferred and bytes_compressed into a separate patch
- moved COMPRESSED feature into a separate patch
- added versioning and stored COMPRESSED feature attributes as u32
- implemented dedicated memory buffer for compression in case of serial streaming
- moved low level Zstd based compression functions into util/{compress.h,zstd.c}
- made compress function to be a param of __push(), __aio_push() functions
- enabled perf inject to decompress COMPRESSED records
- measured compression overhead for serial and AIO streaming using
basic matrix multiplication workload on 8 core skylake
Changes in v2:
- moved compression/decompression code to session layer
- enabled allocation aio data buffers for compression
- enabled trace compression for serial trace streaming
---
[1] https://github.com/facebook/zstd
---
OVERHEAD MEASUREMENTS:
uname -a
Linux localhost 4.20.7-200.fc29.x86_64 #1 SMP Wed Feb 6 19:16:42 UTC 2019 x86_64 x86_64 x86_64 GNU/Linux
cat /proc/cpuinfo
processor : 7vendor_id : GenuineIntel
cpu family : 6
model : 94
model name : Intel(R) Core(TM) i7-6700K CPU @ 4.00GHz
stepping : 3
microcode : 0xc6
cpu MHz : 4021.884
cache size : 8192 KB
physical id : 0
siblings : 8
core id : 3
cpu cores : 4
apicid : 7
initial apicid : 7
fpu : yesfpu_exception : yes
cpuid level : 22
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf tsc_known_freq pni pclmulqdq dtes64 monitor ds_cpl vmx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault invpcid_single pti ssbd ibrs ibpb stibp tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2 erms invpcid rtm mpx rdseed adx smap clflushopt intel_pt xsaveopt xsavec xgetbv1 xsaves dtherm ida arat pln pts hwp hwp_notify hwp_act_window hwp_epp flush_l1d
bugs : cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass l1tf
bogomips : 8016.00
clflush size : 64cache_alignment : 64
address sizes : 39 bits physical, 48 bits virtual
power management:
-----------------------------------------------------------------
#!/bin/bash -xv
echo 0 > /proc/sys/kernel/perf_event_paranoid
+ echo 0
cat /proc/sys/kernel/perf_event_paranoid
+ cat /proc/sys/kernel/perf_event_paranoid
0
echo performance | tee /sys/devices/system/cpu/cpu*/cpufreq/scaling_governor
+ echo performance
+ tee /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor /sys/devices/system/cpu/cpu1/cpufreq/scaling_governor /sys/devices/system/cpu/cpu2/cpufreq/scaling_governor /sys/devices/system/cpu/cpu3/cpufreq/scaling_governor /sys/devices/system/cpu/cpu4/cpufreq/scaling_governor /sys/devices/system/cpu/cpu5/cpufreq/scaling_governor /sys/devices/system/cpu/cpu6/cpufreq/scaling_governor /sys/devices/system/cpu/cpu7/cpufreq/scaling_governor performance
for i in 0 1 2 3 5 8
do
/usr/bin/time tools/perf/perf record -z $i -F 25000 -N -B -T -R -e cycles -- ../../matrix/linux/matrix.gcc
done
+ for i in 0 1 2 3 5 8
+ /usr/bin/time tools/perf/perf record -z 0 -F 25000 -N -B -T -R -e cycles -- ../../matrix/linux/matrix.gcc Addr of buf1 = 0x7fe36de5c010
Offs of buf1 = 0x7fe36de5c180
Addr of buf2 = 0x7fe36be5b010
Offs of buf2 = 0x7fe36be5b1c0
Addr of buf3 = 0x7fe369e5a010
Offs of buf3 = 0x7fe369e5a100
Addr of buf4 = 0x7fe367e59010
Offs of buf4 = 0x7fe367e59140
Threads #: 8 Pthreads
Matrix size: 2048
Using multiply kernel: multiply1
Execution time = 16.949 seconds
[ perf record: Woken up 309 times to write data ]
[ perf record: Captured and wrote 179.424 MB perf.data ]
133.67user 0.35system 0:17.08elapsed 784%CPU (0avgtext+0avgdata 100580maxresident)k 0inputs+367480outputs (0major+34737minor)pagefaults 0swaps
+ for i in 0 1 2 3 5 8
+ /usr/bin/time tools/perf/perf record -z 1 -F 25000 -N -B -T -R -e cycles -- ../../matrix/linux/matrix.gcc Addr of buf1 = 0x7fcaec334010
Offs of buf1 = 0x7fcaec334180
Addr of buf2 = 0x7fcaea333010
Offs of buf2 = 0x7fcaea3331c0
Addr of buf3 = 0x7fcae8332010
Offs of buf3 = 0x7fcae8332100
Addr of buf4 = 0x7fcae6331010
Offs of buf4 = 0x7fcae6331140
Threads #: 8 Pthreads
Matrix size: 2048
Using multiply kernel: multiply1
Execution time = 17.608 seconds
[ perf record: Woken up 595 times to write data ]
[ perf record: Compressed 181.148 MB to 21.497 MB, ratio is 8.427 ]
[ perf record: Captured and wrote 21.527 MB perf.data ]
135.69user 0.24system 0:17.73elapsed 766%CPU (0avgtext+0avgdata 100500maxresident)k 0inputs+44112outputs (0major+35033minor)pagefaults 0swaps
+ for i in 0 1 2 3 5 8
+ /usr/bin/time tools/perf/perf record -z 2 -F 25000 -N -B -T -R -e cycles -- ../../matrix/linux/matrix.gcc Addr of buf1 = 0x7f1336f8d010
Offs of buf1 = 0x7f1336f8d180
Addr of buf2 = 0x7f1334f8c010
Offs of buf2 = 0x7f1334f8c1c0
Addr of buf3 = 0x7f1332f8b010
Offs of buf3 = 0x7f1332f8b100
Addr of buf4 = 0x7f1330f8a010
Offs of buf4 = 0x7f1330f8a140
Threads #: 8 Pthreads
Matrix size: 2048
Using multiply kernel: multiply1
Execution time = 18.175 seconds
[ perf record: Woken up 521 times to write data ]
[ perf record: Compressed 186.953 MB to 23.210 MB, ratio is 8.055 ]
[ perf record: Captured and wrote 23.239 MB perf.data ]
140.21user 0.25system 0:18.32elapsed 766%CPU (0avgtext+0avgdata 100560maxresident)k 0inputs+47608outputs (0major+35263minor)pagefaults 0swaps
+ for i in 0 1 2 3 5 8
+ /usr/bin/time tools/perf/perf record -z 3 -F 25000 -N -B -T -R -e cycles -- ../../matrix/linux/matrix.gcc Addr of buf1 = 0x7f97060e3010
Offs of buf1 = 0x7f97060e3180
Addr of buf2 = 0x7f97040e2010
Offs of buf2 = 0x7f97040e21c0
Addr of buf3 = 0x7f97020e1010
Offs of buf3 = 0x7f97020e1100
Addr of buf4 = 0x7f97000e0010
Offs of buf4 = 0x7f97000e0140
Threads #: 8 Pthreads
Matrix size: 2048
Using multiply kernel: multiply1
Execution time = 17.688 seconds
[ perf record: Woken up 485 times to write data ]
[ perf record: Compressed 181.908 MB to 21.962 MB, ratio is 8.283 ]
[ perf record: Captured and wrote 21.991 MB perf.data ]
136.87user 0.23system 0:17.81elapsed 769%CPU (0avgtext+0avgdata 100616maxresident)k 0inputs+45064outputs (0major+35773minor)pagefaults 0swaps
+ for i in 0 1 2 3 5 8
+ /usr/bin/time tools/perf/perf record -z 5 -F 25000 -N -B -T -R -e cycles -- ../../matrix/linux/matrix.gcc Addr of buf1 = 0x7f477b444010
Offs of buf1 = 0x7f477b444180
Addr of buf2 = 0x7f4779443010
Offs of buf2 = 0x7f47794431c0
Addr of buf3 = 0x7f4777442010
Offs of buf3 = 0x7f4777442100
Addr of buf4 = 0x7f4775441010
Offs of buf4 = 0x7f4775441140
Threads #: 8 Pthreads
Matrix size: 2048
Using multiply kernel: multiply1
Execution time = 18.406 seconds
[ perf record: Woken up 416 times to write data ]
[ perf record: Compressed 187.705 MB to 23.170 MB, ratio is 8.101 ]
[ perf record: Captured and wrote 23.200 MB perf.data ]
142.72user 0.26system 0:18.53elapsed 771%CPU (0avgtext+0avgdata 100520maxresident)k 0inputs+47528outputs (0major+36928minor)pagefaults 0swaps
+ for i in 0 1 2 3 5 8
+ /usr/bin/time tools/perf/perf record -z 8 -F 25000 -N -B -T -R -e cycles -- ../../matrix/linux/matrix.gcc Addr of buf1 = 0x7fb5bf032010
Offs of buf1 = 0x7fb5bf032180
Addr of buf2 = 0x7fb5bd031010
Offs of buf2 = 0x7fb5bd0311c0
Addr of buf3 = 0x7fb5bb030010
Offs of buf3 = 0x7fb5bb030100
Addr of buf4 = 0x7fb5b902f010
Offs of buf4 = 0x7fb5b902f140
Threads #: 8 Pthreads
Matrix size: 2048
Using multiply kernel: multiply1
Execution time = 17.751 seconds
[ perf record: Woken up 391 times to write data ]
[ perf record: Compressed 179.191 MB to 19.441 MB, ratio is 9.217 ]
[ perf record: Captured and wrote 19.502 MB perf.data ]
138.90user 0.29system 0:17.88elapsed 778%CPU (0avgtext+0avgdata 100612maxresident)k 0inputs+39968outputs (0major+37436minor)pagefaults 0swaps
for i in 0 1 2 3 5 8
do
/usr/bin/time tools/perf/perf record --aio=1 -z $i -F 25000 -N -B -T -R -e cycles -- ../../matrix/linux/matrix.gcc
done
+ for i in 0 1 2 3 5 8
+ /usr/bin/time tools/perf/perf record --aio=1 -z 0 -F 25000 -N -B -T -R -e cycles -- ../../matrix/linux/matrix.gcc Addr of buf1 = 0x7feee4519010
Offs of buf1 = 0x7feee4519180
Addr of buf2 = 0x7feee2518010
Offs of buf2 = 0x7feee25181c0
Addr of buf3 = 0x7feee0517010
Offs of buf3 = 0x7feee0517100
Addr of buf4 = 0x7feede516010
Offs of buf4 = 0x7feede516140
Threads #: 8 Pthreads
Matrix size: 2048
Using multiply kernel: multiply1
Execution time = 17.912 seconds
[ perf record: Woken up 390 times to write data ]
[ perf record: Captured and wrote 187.527 MB perf.data ]
139.70user 0.39system 0:18.04elapsed 776%CPU (0avgtext+0avgdata 100624maxresident)k 0inputs+384072outputs (0major+35257minor)pagefaults 0swaps
+ for i in 0 1 2 3 5 8
+ /usr/bin/time tools/perf/perf record --aio=1 -z 1 -F 25000 -N -B -T -R -e cycles -- ../../matrix/linux/matrix.gcc Addr of buf1 = 0x7f72b93ac010
Offs of buf1 = 0x7f72b93ac180
Addr of buf2 = 0x7f72b73ab010
Offs of buf2 = 0x7f72b73ab1c0
Addr of buf3 = 0x7f72b53aa010
Offs of buf3 = 0x7f72b53aa100
Addr of buf4 = 0x7f72b33a9010
Offs of buf4 = 0x7f72b33a9140
Threads #: 8 Pthreads
Matrix size: 2048
Using multiply kernel: multiply1
Execution time = 18.198 seconds
[ perf record: Woken up 416 times to write data ]
[ perf record: Compressed 188.562 MB to 22.252 MB, ratio is 8.474 ]
[ perf record: Captured and wrote 22.284 MB perf.data ]
141.12user 0.32system 0:18.32elapsed 771%CPU (0avgtext+0avgdata 100576maxresident)k 0inputs+45664outputs (0major+35040minor)pagefaults 0swaps
+ for i in 0 1 2 3 5 8
+ /usr/bin/time tools/perf/perf record --aio=1 -z 2 -F 25000 -N -B -T -R -e cycles -- ../../matrix/linux/matrix.gcc Addr of buf1 = 0x7ffb9caf3010
Offs of buf1 = 0x7ffb9caf3180
Addr of buf2 = 0x7ffb9aaf2010
Offs of buf2 = 0x7ffb9aaf21c0
Addr of buf3 = 0x7ffb98af1010
Offs of buf3 = 0x7ffb98af1100
Addr of buf4 = 0x7ffb96af0010
Offs of buf4 = 0x7ffb96af0140
Threads #: 8 Pthreads
Matrix size: 2048
Using multiply kernel: multiply1
Execution time = 18.360 seconds
[ perf record: Woken up 442 times to write data ]
[ perf record: Compressed 191.773 MB to 24.238 MB, ratio is 7.912 ]
[ perf record: Captured and wrote 24.290 MB perf.data ]
143.76user 0.49system 0:18.50elapsed 779%CPU (0avgtext+0avgdata 100596maxresident)k 0inputs+49760outputs (0major+35276minor)pagefaults 0swaps
+ for i in 0 1 2 3 5 8
+ /usr/bin/time tools/perf/perf record --aio=1 -z 3 -F 25000 -N -B -T -R -e cycles -- ../../matrix/linux/matrix.gcc Addr of buf1 = 0x7f13f2df2010
Offs of buf1 = 0x7f13f2df2180
Addr of buf2 = 0x7f13f0df1010
Offs of buf2 = 0x7f13f0df11c0
Addr of buf3 = 0x7f13eedf0010
Offs of buf3 = 0x7f13eedf0100
Addr of buf4 = 0x7f13ecdef010
Offs of buf4 = 0x7f13ecdef140
Threads #: 8 Pthreads
Matrix size: 2048
Using multiply kernel: multiply1
Execution time = 18.383 seconds
[ perf record: Woken up 499 times to write data ]
[ perf record: Compressed 191.078 MB to 23.246 MB, ratio is 8.220 ]
[ perf record: Captured and wrote 23.282 MB perf.data ]
143.72user 0.34system 0:18.51elapsed 778%CPU (0avgtext+0avgdata 100616maxresident)k 0inputs+47704outputs (0major+35783minor)pagefaults 0swaps
+ for i in 0 1 2 3 5 8
+ /usr/bin/time tools/perf/perf record --aio=1 -z 5 -F 25000 -N -B -T -R -e cycles -- ../../matrix/linux/matrix.gcc Addr of buf1 = 0x7fca0d091010
Offs of buf1 = 0x7fca0d091180
Addr of buf2 = 0x7fca0b090010
Offs of buf2 = 0x7fca0b0901c0
Addr of buf3 = 0x7fca0908f010
Offs of buf3 = 0x7fca0908f100
Addr of buf4 = 0x7fca0708e010
Offs of buf4 = 0x7fca0708e140
Threads #: 8 Pthreads
Matrix size: 2048
Using multiply kernel: multiply1
Execution time = 18.758 seconds
[ perf record: Woken up 535 times to write data ]
[ perf record: Compressed 190.065 MB to 24.430 MB, ratio is 7.780 ]
[ perf record: Captured and wrote 24.519 MB perf.data ]
144.62user 0.66system 0:18.88elapsed 769%CPU (0avgtext+0avgdata 100528maxresident)k 0inputs+50232outputs (0major+36942minor)pagefaults 0swaps
+ for i in 0 1 2 3 5 8
+ /usr/bin/time tools/perf/perf record --aio=1 -z 8 -F 25000 -N -B -T -R -e cycles -- ../../matrix/linux/matrix.gcc Addr of buf1 = 0x7f7e1f449010
Offs of buf1 = 0x7f7e1f449180
Addr of buf2 = 0x7f7e1d448010
Offs of buf2 = 0x7f7e1d4481c0
Addr of buf3 = 0x7f7e1b447010
Offs of buf3 = 0x7f7e1b447100
Addr of buf4 = 0x7f7e19446010
Offs of buf4 = 0x7f7e19446140
Threads #: 8 Pthreads
Matrix size: 2048
Using multiply kernel: multiply1
Execution time = 20.103 seconds
[ perf record: Woken up 260 times to write data ]
[ perf record: Compressed 193.024 MB to 31.588 MB, ratio is 6.111 ]
[ perf record: Captured and wrote 32.139 MB perf.data ]
151.73user 4.21system 0:20.23elapsed 770%CPU (0avgtext+0avgdata 100616maxresident)k 0inputs+65848outputs (0major+37431minor)pagefaults 0swaps
---
TESTING:
$ tools/perf/perf version --build-options
perf version 4.13.rc5.gd8d056b
dwarf: [ on ] # HAVE_DWARF_SUPPORT
dwarf_getlocations: [ on ] # HAVE_DWARF_GETLOCATIONS_SUPPORT
glibc: [ on ] # HAVE_GLIBC_SUPPORT
gtk2: [ on ] # HAVE_GTK2_SUPPORT
syscall_table: [ on ] # HAVE_SYSCALL_TABLE_SUPPORT
libbfd: [ on ] # HAVE_LIBBFD_SUPPORT
libelf: [ on ] # HAVE_LIBELF_SUPPORT
libnuma: [ on ] # HAVE_LIBNUMA_SUPPORT
numa_num_possible_cpus: [ on ] # HAVE_LIBNUMA_SUPPORT
libperl: [ on ] # HAVE_LIBPERL_SUPPORT
libpython: [ on ] # HAVE_LIBPYTHON_SUPPORT
libslang: [ on ] # HAVE_SLANG_SUPPORT
libcrypto: [ on ] # HAVE_LIBCRYPTO_SUPPORT
libunwind: [ on ] # HAVE_LIBUNWIND_SUPPORT
libdw-dwarf-unwind: [ on ] # HAVE_DWARF_SUPPORT
zlib: [ on ] # HAVE_ZLIB_SUPPORT
lzma: [ on ] # HAVE_LZMA_SUPPORT
get_cpuid: [ on ] # HAVE_AUXTRACE_SUPPORT
bpf: [ on ] # HAVE_LIBBPF_SUPPORT
aio: [ OFF ] # HAVE_AIO_SUPPORT
zstd: [ OFF ] # HAVE_ZSTD_SUPPORT $ tools/perf/perf version --build-options
perf version 4.13.rc5.gd8d056b
dwarf: [ on ] # HAVE_DWARF_SUPPORT
dwarf_getlocations: [ on ] # HAVE_DWARF_GETLOCATIONS_SUPPORT
glibc: [ on ] # HAVE_GLIBC_SUPPORT
gtk2: [ on ] # HAVE_GTK2_SUPPORT
syscall_table: [ on ] # HAVE_SYSCALL_TABLE_SUPPORT
libbfd: [ on ] # HAVE_LIBBFD_SUPPORT
libelf: [ on ] # HAVE_LIBELF_SUPPORT
libnuma: [ on ] # HAVE_LIBNUMA_SUPPORT
numa_num_possible_cpus: [ on ] # HAVE_LIBNUMA_SUPPORT
libperl: [ on ] # HAVE_LIBPERL_SUPPORT
libpython: [ on ] # HAVE_LIBPYTHON_SUPPORT
libslang: [ on ] # HAVE_SLANG_SUPPORT
libcrypto: [ on ] # HAVE_LIBCRYPTO_SUPPORT
libunwind: [ on ] # HAVE_LIBUNWIND_SUPPORT
libdw-dwarf-unwind: [ on ] # HAVE_DWARF_SUPPORT
zlib: [ on ] # HAVE_ZLIB_SUPPORT
lzma: [ on ] # HAVE_LZMA_SUPPORT
get_cpuid: [ on ] # HAVE_AUXTRACE_SUPPORT
bpf: [ on ] # HAVE_LIBBPF_SUPPORT
aio: [ on ] # HAVE_AIO_SUPPORT
zstd: [ on ] # HAVE_ZSTD_SUPPORT$ make -C tools/perf clean all ... $ pushd tools/perf/ && ./perf test && popd ~/abudanko/kernel/acme/tools/perf ~/abudanko/kernel/acme 1: vmlinux symtab matches kallsyms : Skip 2: Detect openat syscall event : Ok 3: Detect openat syscall event on all cpus : Ok 4: Read samples using the mmap interface : Ok 5: Test data source output : Ok 6: Parse event definition strings : Ok 7: Simple expression parser : Ok 8: PERF_RECORD_* events & perf_sample fields : Ok 9: Parse perf pmu format : Ok 10: DSO data read : Ok 11: DSO data cache : Ok 12: DSO data reopen : Ok 13: Roundtrip evsel->name : Ok 14: Parse sched tracepoints fields : Ok 15: syscalls:sys_enter_openat event fields : Ok 16: Setup struct perf_event_attr : Ok 17: Match and link multiple hists : Ok 18: 'import perf' in python : Ok 19: Breakpoint overflow signal handler : Ok 20: Breakpoint overflow sampling : Ok 21: Breakpoint accounting : Ok 22: Watchpoint : 22.1: Read Only Watchpoint : Skip 22.2: Write Only Watchpoint : Ok 22.3: Read / Write Watchpoint : Ok 22.4: Modify Watchpoint : Ok 23: Number of exit events of a simple workload : Ok 24: Software clock events period values : Ok 25: Object code reading : Ok 26: Sample parsing : Ok 27: Use a dummy software event to keep tracking : Ok 28: Parse with no sample_id_all bit set : Ok 29: Filter hist entries : Ok 30: Lookup mmap thread : Ok 31: Share thread mg : Ok 32: Sort output of hist entries : Ok 33: Cumulate child hist entries : Ok 34: Track with sched_switch : Ok 35: Filter fds with revents mask in a fdarray : Ok 36: Add fd to a fdarray, making it autogrow : Ok 37: kmod_path__parse : Ok 38: Thread map : Ok 39: LLVM search and compile : 39.1: Basic BPF llvm compile : Skip 39.2: kbuild searching : Skip 39.3: Compile source for BPF prologue generation : Skip 39.4: Compile source for BPF relocation : Skip 40: Session topology : Ok 41: BPF filter : 41.1: Basic BPF filtering : Skip 41.2: BPF pinning : Skip 41.3: BPF prologue generation : Skip 41.4: BPF relocation checker : Skip 42: Synthesize thread map : Ok 43: Remove thread map : Ok 44: Synthesize cpu map : Ok 45: Synthesize stat config : Ok 46: Synthesize stat : Ok 47: Synthesize stat round : Ok 48: Synthesize attr update : Ok 49: Event times : Ok 50: Read backward ring buffer : Ok 51: Print cpu map : Ok 52: Probe SDT events : Ok 53: is_printable_array : Ok 54: Print bitmap : Ok 55: perf hooks : Ok 56: builtin clang support : Skip (not compiled in) 57: unit_number__scnprintf : Ok 58: mem2node : Ok 59: x86 rdpmc : Ok 60: Convert perf time to TSC : Ok 61: DWARF unwind : Ok 62: x86 instruction decoder - new instructions : Ok 63: x86 bp modify : Ok 64: Check open filename arg using perf trace + vfs_getname: Skip 65: Add vfs_getname probe to get syscall args filenames : Skip 66: probe libc's inet_pton & backtrace it with ping : Ok 67: Use vfs_getname probe to get syscall args filenames : Skip 68: record trace Zstd compression/decompression : Ok ~/abudanko/kernel/acme
$ make -C tools/perf NO_LIBZSTD=1 clean all ... $ pushd tools/perf/ && ./perf test && popd ~/abudanko/kernel/acme/tools/perf ~/abudanko/kernel/acme ... 68: record trace Zstd compression/decompression : Skip ~/abudanko/kernel/acme
PATCH v10 1/12 feature: implement libzstd check, LIBZSTD_DIR and NO_LIBZSTD defines
Implement libzstd feature check, NO_LIBZSTD and LIBZSTD_DIR defines to override Zstd library sources or disable the feature from the command line:
$ make -C tools/perf LIBZSTD_DIR=/path/to/zstd/sources/ clean all $ make -C tools/perf NO_LIBZSTD=1 clean all
Auto detection feature status is reported just before compilation starts. If your system has some version of the zstd library preinstalled then the build system finds and uses it during the build.
If you still prefer to compile with some other version of zstd library you have capability to refer the compilation to that version using LIBZSTD_DIR define.
--- tools/build/Makefile.feature | 6 ++++-- tools/build/feature/Makefile | 6 +++++- tools/build/feature/test-all.c | 5 +++++ tools/build/feature/test-libzstd.c | 12 ++++++++++++ tools/perf/Makefile.config | 20 ++++++++++++++++++++ tools/perf/Makefile.perf | 3 +++ tools/perf/builtin-version.c | 2 ++ 7 files changed, 51 insertions(+), 3 deletions(-) create mode 100644 tools/build/feature/test-libzstd.c
tools/build/Makefile.feature +4 -2
@@ -66,7 +66,8 @@ FEATURE_TESTS_BASIC := \ sched_getcpu \ sdt \ setns \ - libaio + libaio \ + libzstd # FEATURE_TESTS_BASIC + FEATURE_TESTS_EXTRA is the complete list # of all feature tests
@@ -118,7 +119,8 @@ FEATURE_DISPLAY ?= \ lzma \ get_cpuid \ bpf \ - libaio + libaio \ + libzstd # Set FEATURE_CHECK_(C|LD)FLAGS-all for all FEATURE_TESTS features. # If in the future we need per-feature checks/flags for features not
tools/build/feature/Makefile +5 -1
@@ -62,7 +62,8 @@ FILES= \ test-clang.bin \ test-llvm.bin \ test-llvm-version.bin \ - test-libaio.bin + test-libaio.bin \ + test-libzstd.bin FILES := $(addprefix $(OUTPUT),$(FILES))
@@ -301,6 +302,9 @@ $(OUTPUT)test-clang.bin: $(OUTPUT)test-libaio.bin: $(BUILD) -lrt +$(OUTPUT)test-libzstd.bin: + $(BUILD) -lzstd + ############################### clean:
tools/build/feature/test-all.c +5 -0
@@ -178,6 +178,10 @@# include "test-reallocarray.c" #undef main +#define main main_test_zstd +# include "test-libzstd.c" +#undef main +int main (int argc,char *argv[]) {main_test_libpython ();
@@ -219,6 +223,7 @@ int main(int argc, char *argv[])main_test_setns ();main_test_libaio ();main_test_reallocarray (); +main_test_libzstd ();return 0 ; }
tools/build/feature/test-libzstd.c new file +12 -0
@@ -0,0 +1,12 @@ +// SPDX-License-Identifier: GPL-2.0 +#include <zstd.h> + +int main (void ) +{ +ZSTD_CStream *cstream; + + cstream =ZSTD_createCStream (); +ZSTD_freeCStream (cstream); + +return 0 ; +}
tools/perf/Makefile.config +20 -0
@@ -152,6 +152,13 @@ endif FEATURE_CHECK_CFLAGS-libbabeltrace := $(LIBBABELTRACE_CFLAGS) FEATURE_CHECK_LDFLAGS-libbabeltrace := $(LIBBABELTRACE_LDFLAGS) -lbabeltrace-ctf +ifdef LIBZSTD_DIR + LIBZSTD_CFLAGS := -I$(LIBZSTD_DIR)/lib + LIBZSTD_LDFLAGS := -L$(LIBZSTD_DIR)/lib +endif +FEATURE_CHECK_CFLAGS-libzstd := $(LIBZSTD_CFLAGS) +FEATURE_CHECK_LDFLAGS-libzstd := $(LIBZSTD_LDFLAGS) + FEATURE_CHECK_CFLAGS-bpf = -I. -I$(srctree)/tools/include -I$(srctree)/tools/arch/$(SRCARCH)/include/uapi -I$(srctree)/tools/include/uapi # include ARCH specific config -include $(src-perf)/arch/$(SRCARCH)/Makefile
@@ -782,6 +789,19 @@ ifndef NO_LZMA endif endif +ifndef NO_LIBZSTD + ifeq ($(feature-libzstd), 1) + CFLAGS += -DHAVE_ZSTD_SUPPORT + CFLAGS += $(LIBZSTD_CFLAGS) + LDFLAGS += $(LIBZSTD_LDFLAGS) + EXTLIBS += -lzstd + $(call detected,CONFIG_ZSTD) + else + msg := $(warning No libzstd found, disables trace compression, please install libzstd-dev[el] and/or set LIBZSTD_DIR); + NO_LIBZSTD := 1 + endif +endif + ifndef NO_BACKTRACE ifeq ($(feature-backtrace), 1) CFLAGS += -DHAVE_BACKTRACE_SUPPORT
tools/perf/Makefile.perf +3 -0
@@ -108,6 +108,9 @@ include ../scripts/utilities.mak # streaming for record mode. Currently Posix AIO trace streaming is # supported only when linking with glibc. # +# Define NO_LIBZSTD if you do not want support of Zstandard based runtime +# trace compression in record mode. +# # As per kernel Makefile, avoid funny character set dependencies unexport LC_ALL
tools/perf/builtin-version.c +2 -0
@@ -78,6 +78,8 @@ static void library_status(void)STATUS (HAVE_LZMA_SUPPORT , lzma);STATUS (HAVE_AUXTRACE_SUPPORT , get_cpuid);STATUS (HAVE_LIBBPF_SUPPORT , bpf); +STATUS (HAVE_AIO_SUPPORT , aio); +STATUS (HAVE_ZSTD_SUPPORT , zstd); }int cmd_version (int argc,const char **argv)
Signature · 1 line
2.20.1
tip-bot for Alexey Budankov
merge bot[tip:perf/core] tools build: Implement libzstd feature check, LIBZSTD_DIR and NO_LIBZSTD defines
tools build: Implement libzstd feature check, LIBZSTD_DIR and NO_LIBZSTD defines
Commit-ID: 3b1c5d9659718263c7f9c93af82f98221c58f171 Gitweb: https://git.kernel.org/tip/3b1c5d9659718263c7f9c93af82f98221c58f171 Author: Alexey Budankov <[email protected]> AuthorDate: Mon, 18 Mar 2019 20:39:49 +0300 Committer: Arnaldo Carvalho de Melo <[email protected]> CommitDate: Mon, 1 Apr 2019 15:18:10 -0300
tools build: Implement libzstd feature check, LIBZSTD_DIR and NO_LIBZSTD defines
Implement libzstd feature check, NO_LIBZSTD and LIBZSTD_DIR defines to override Zstd library sources or disable the feature from the command line:
$ make -C tools/perf LIBZSTD_DIR=/path/to/zstd/sources/ clean all $ make -C tools/perf NO_LIBZSTD=1 clean all
Auto detection feature status is reported just before compilation starts. If your system has some version of the zstd library preinstalled then the build system finds and uses it during the build.
If you still prefer to compile with some other version of zstd library you have capability to refer the compilation to that version using LIBZSTD_DIR define.
--- tools/build/Makefile.feature | 2 ++ tools/build/feature/Makefile | 6 +++++- tools/build/feature/test-all.c | 5 +++++ tools/build/feature/test-libzstd.c | 12 ++++++++++++ tools/perf/Makefile.config | 20 ++++++++++++++++++++ tools/perf/Makefile.perf | 3 +++ tools/perf/builtin-version.c | 2 ++ 7 files changed, 49 insertions(+), 1 deletion(-)
tools/build/Makefile.feature +2 -0
@@ -67,6 +67,7 @@ FEATURE_TESTS_BASIC := \ sdt \ setns \ libaio \ + libzstd \ disassembler-four-args # FEATURE_TESTS_BASIC + FEATURE_TESTS_EXTRA is the complete list
@@ -120,6 +121,7 @@ FEATURE_DISPLAY ?= \ get_cpuid \ bpf \ libaio \ + libzstd \ disassembler-four-args # Set FEATURE_CHECK_(C|LD)FLAGS-all for all FEATURE_TESTS features.
tools/build/feature/Makefile +5 -1
@@ -62,7 +62,8 @@ FILES= \ test-clang.bin \ test-llvm.bin \ test-llvm-version.bin \ - test-libaio.bin + test-libaio.bin \ + test-libzstd.bin FILES := $(addprefix $(OUTPUT),$(FILES))
@@ -301,6 +302,9 @@ $(OUTPUT)test-clang.bin: $(OUTPUT)test-libaio.bin: $(BUILD) -lrt +$(OUTPUT)test-libzstd.bin: + $(BUILD) -lzstd + ############################### clean:
tools/build/feature/test-all.c +5 -0
@@ -182,6 +182,10 @@# include "test-disassembler-four-args.c" #undef main +#define main main_test_zstd +# include "test-libzstd.c" +#undef main +int main (int argc,char *argv[]) {main_test_libpython ();
@@ -224,6 +228,7 @@ int main(int argc, char *argv[])main_test_libaio ();main_test_reallocarray ();main_test_disassembler_four_args (); +main_test_libzstd ();return 0 ; }
tools/build/feature/test-libzstd.c new file +12 -0
@@ -0,0 +1,12 @@ +// SPDX-License-Identifier: GPL-2.0 +#include <zstd.h> + +int main (void ) +{ +ZSTD_CStream *cstream; + + cstream =ZSTD_createCStream (); +ZSTD_freeCStream (cstream); + +return 0 ; +}
tools/perf/Makefile.config +20 -0
@@ -152,6 +152,13 @@ endif FEATURE_CHECK_CFLAGS-libbabeltrace := $(LIBBABELTRACE_CFLAGS) FEATURE_CHECK_LDFLAGS-libbabeltrace := $(LIBBABELTRACE_LDFLAGS) -lbabeltrace-ctf +ifdef LIBZSTD_DIR + LIBZSTD_CFLAGS := -I$(LIBZSTD_DIR)/lib + LIBZSTD_LDFLAGS := -L$(LIBZSTD_DIR)/lib +endif +FEATURE_CHECK_CFLAGS-libzstd := $(LIBZSTD_CFLAGS) +FEATURE_CHECK_LDFLAGS-libzstd := $(LIBZSTD_LDFLAGS) + FEATURE_CHECK_CFLAGS-bpf = -I. -I$(srctree)/tools/include -I$(srctree)/tools/arch/$(SRCARCH)/include/uapi -I$(srctree)/tools/include/uapi # include ARCH specific config -include $(src-perf)/arch/$(SRCARCH)/Makefile
@@ -787,6 +794,19 @@ ifndef NO_LZMA endif endif +ifndef NO_LIBZSTD + ifeq ($(feature-libzstd), 1) + CFLAGS += -DHAVE_ZSTD_SUPPORT + CFLAGS += $(LIBZSTD_CFLAGS) + LDFLAGS += $(LIBZSTD_LDFLAGS) + EXTLIBS += -lzstd + $(call detected,CONFIG_ZSTD) + else + msg := $(warning No libzstd found, disables trace compression, please install libzstd-dev[el] and/or set LIBZSTD_DIR); + NO_LIBZSTD := 1 + endif +endif + ifndef NO_BACKTRACE ifeq ($(feature-backtrace), 1) CFLAGS += -DHAVE_BACKTRACE_SUPPORT
tools/perf/Makefile.perf +3 -0
@@ -108,6 +108,9 @@ include ../scripts/utilities.mak # streaming for record mode. Currently Posix AIO trace streaming is # supported only when linking with glibc. # +# Define NO_LIBZSTD if you do not want support of Zstandard based runtime +# trace compression in record mode. +# # As per kernel Makefile, avoid funny character set dependencies unexport LC_ALL
tools/perf/builtin-version.c +2 -0
@@ -78,6 +78,8 @@ static void library_status(void)STATUS (HAVE_LZMA_SUPPORT , lzma);STATUS (HAVE_AUXTRACE_SUPPORT , get_cpuid);STATUS (HAVE_LIBBPF_SUPPORT , bpf); +STATUS (HAVE_AIO_SUPPORT , aio); +STATUS (HAVE_ZSTD_SUPPORT , zstd); }int cmd_version (int argc,const char **argv)
PATCH v10 2/12 perf record: implement --mmap-flush=<number> option
In reply to Alexey BudankovImplemented --mmap-flush option that specifies minimal number of bytes that is extracted from mmaped kernel buffer to store into a trace. The default option value is 1 byte what means every time trace writing thread finds some new data in the mmaped buffer the data is extracted, possibly compressed and written to a trace.
$ tools/perf/perf record --mmap-flush 1024 -e cycles -- matrix.gcc $ tools/perf/perf record --aio --mmap-flush 1K -e cycles -- matrix.gcc
The option is independent from -z setting, doesn't vary with compression level and can serve two purposes.
The first purpose is to increase the compression ratio of a trace data. Larger data chunks are compressed more effectively so the implemented option allows specifying data chunk size to compress. Also at some cases executing more write syscalls with smaller data size can take longer than executing less write syscalls with bigger data size due to syscall overhead so extracting bigger data chunks specified by the option value could additionally decrease runtime overhead.
The second purpose is to avoid self monitoring live-lock issue in system wide (-a) profiling mode. Profiling in system wide mode with compression (-a -z) can additionally induce data into the kernel buffers along with the data from monitored processes. If performance data rate and volume from the monitored processes is high then trace streaming and compression activity in the tool is also high. High tool process activity can lead to subtle live-lock effect when compression of single new byte from some of mmaped kernel buffer leads to generation of the next single byte at some mmaped buffer. So perf tool process ends up in endless self monitoring.
Implemented sync parameter is the mean to force data move independently from the specified flush threshold value. Despite the provided flush value the tool needs capability to unconditionally drain memory buffers, at least in the end of the collection.
--- tools/perf/Documentation/perf-record.txt | 12 +++++ tools/perf/builtin-record.c | 65 +++++++++++++++++++++--- tools/perf/perf.h | 1 + tools/perf/util/evlist.c | 6 +-- tools/perf/util/evlist.h | 3 +- tools/perf/util/mmap.c | 4 +- tools/perf/util/mmap.h | 3 +- 7 files changed, 82 insertions(+), 12 deletions(-)
tools/perf/Documentation/perf-record.txt +12 -0
@@ -459,6 +459,18 @@ Set affinity mask of trace reading thread according to the policy defined by 'mo node - thread affinity mask is set to NUMA node cpu mask of the processed mmap buffer cpu - thread affinity mask is set to cpu of the processed mmap buffer +--mmap-flush=number:: +Specify minimal number of bytes that is extracted from mmap data pages and stored +into a trace. The number specification is possible using B/K/M/G suffixes. Maximal allowed +value is a quarter of the size of mmaped data pages. The default option value is 1 byte +what means that every time trace writing thread finds some new data in the mmaped buffer +the data is extracted, possibly compressed (-z) and written to a trace. Larger data chunks +are compressed more effectively in comparison to smaller chunks so extraction of larger +chunks from the mmap data pages is preferable from perspective of trace size reduction. +Also at some cases executing less trace write syscalls with bigger data size can take +shorter than executing more trace write syscalls with smaller data size thus lowering +runtime profiling overhead. + --all-kernel:: Configure all used events to run in kernel space.
tools/perf/builtin-record.c +59 -6
147 lines not on this page · raw message
tools/perf/perf.h +1 -0
@@ -85,6 +85,7 @@ struct record_opts {u64 clockid_res_ns;int nr_cblocks;int affinity; +int mmap_flush; };enum perf_affinity {
tools/perf/util/evlist.c +3 -3
@@ -1037,7 +1037,7 @@ int perf_evlist__parse_mmap_pages(const struct option *opt, const char *str,*/ int perf_evlist__mmap_ex (struct perf_evlist *evlist,unsigned int pages,unsigned int auxtrace_pages, -bool auxtrace_overwrite,int nr_cblocks,int affinity) +bool auxtrace_overwrite,int nr_cblocks,int affinity,int flush) {struct perf_evsel *evsel;const struct cpu_map *cpus = evlist->cpus;
@@ -1047,7 +1047,7 @@ int perf_evlist__mmap_ex(struct perf_evlist *evlist, unsigned int pages,* Its value is decided by evsel's write_backward. * So &mp should not be passed through const pointer. */ -struct mmap_params mp = { .nr_cblocks = nr_cblocks, .affinity = affinity }; +struct mmap_params mp = { .nr_cblocks = nr_cblocks, .affinity = affinity, .flush = flush };if (!evlist->mmap) evlist->mmap =perf_evlist__alloc_mmap (evlist, false);
@@ -1079,7 +1079,7 @@ int perf_evlist__mmap_ex(struct perf_evlist *evlist, unsigned int pages,int perf_evlist__mmap (struct perf_evlist *evlist,unsigned int pages) { -return perf_evlist__mmap_ex (evlist, pages,0 , false,0 ,PERF_AFFINITY_SYS ); +return perf_evlist__mmap_ex (evlist, pages,0 , false,0 ,PERF_AFFINITY_SYS ,1 ); }int perf_evlist__create_maps (struct perf_evlist *evlist,struct target *target)
tools/perf/util/evlist.h +2 -1
@@ -165,7 +165,8 @@ unsigned long perf_event_mlock_kb_in_pages(void);int perf_evlist__mmap_ex (struct perf_evlist *evlist,unsigned int pages,unsigned int auxtrace_pages, -bool auxtrace_overwrite,int nr_cblocks,int affinity); +bool auxtrace_overwrite,int nr_cblocks, +int affinity,int flush);int perf_evlist__mmap (struct perf_evlist *evlist,unsigned int pages);void perf_evlist__munmap (struct perf_evlist *evlist);
tools/perf/util/mmap.c +3 -1
@@ -440,6 +440,8 @@ int perf_mmap__mmap(struct perf_mmap *map, struct mmap_params *mp, int fd, int cperf_mmap__setup_affinity_mask (map, mp); + map->flush = mp->flush; +if (auxtrace_mmap__mmap (&map->auxtrace_mmap, &mp->auxtrace_mp, map->base, fd))return -1 ;
@@ -492,7 +494,7 @@ static int __perf_mmap__read_init(struct perf_mmap *md) md->start = md->overwrite ? head : old; md->end = md->overwrite ? old : head; -if (md->start == md->end) +if ((md->end - md->start) < md->flush)return -EAGAIN ; size = md->end - md->start;
tools/perf/util/mmap.h +2 -1
@@ -39,6 +39,7 @@ struct perf_mmap { } aio;#endif cpu_set_t affinity_mask; +u64 flush; };/*
@@ -70,7 +71,7 @@ enum bkw_mmap_state { };struct mmap_params { -int prot, mask, nr_cblocks, affinity; +int prot, mask, nr_cblocks, affinity, flush;struct auxtrace_mmap_params auxtrace_mp; };
Signature · 1 line
2.20.1
Re: [PATCH v10 02/12] perf record: implement --mmap-flush=<number> option
Em Mon, Mar 18, 2019 at 08:40:26PM +0300, Alexey Budankov escreveu:
Quoted 53 lines
Implemented --mmap-flush option that specifies minimal number of bytes that is extracted from mmaped kernel buffer to store into a trace. The default option value is 1 byte what means every time trace writing thread finds some new data in the mmaped buffer the data is extracted, possibly compressed and written to a trace.
$ tools/perf/perf record --mmap-flush 1024 -e cycles -- matrix.gcc $ tools/perf/perf record --aio --mmap-flush 1K -e cycles -- matrix.gccThe option is independent from -z setting, doesn't vary with compression level and can serve two purposes.
The first purpose is to increase the compression ratio of a trace data. Larger data chunks are compressed more effectively so the implemented option allows specifying data chunk size to compress. Also at some cases executing more write syscalls with smaller data size can take longer than executing less write syscalls with bigger data size due to syscall overhead so extracting bigger data chunks specified by the option value could additionally decrease runtime overhead.
The second purpose is to avoid self monitoring live-lock issue in system wide (-a) profiling mode. Profiling in system wide mode with compression (-a -z) can additionally induce data into the kernel buffers along with the data from monitored processes. If performance data rate and volume from the monitored processes is high then trace streaming and compression activity in the tool is also high. High tool process activity can lead to subtle live-lock effect when compression of single new byte from some of mmaped kernel buffer leads to generation of the next single byte at some mmaped buffer. So perf tool process ends up in endless self monitoring.
Implemented sync parameter is the mean to force data move independently from the specified flush threshold value. Despite the provided flush value the tool needs capability to unconditionally drain memory buffers, at least in the end of the collection.
Signed-off-by Alexey Budankov <[email protected]>--- tools/perf/Documentation/perf-record.txt | 12 +++++ tools/perf/builtin-record.c | 65 +++++++++++++++++++++--- tools/perf/perf.h | 1 + tools/perf/util/evlist.c | 6 +-- tools/perf/util/evlist.h | 3 +- tools/perf/util/mmap.c | 4 +- tools/perf/util/mmap.h | 3 +- 7 files changed, 82 insertions(+), 12 deletions(-)tools/perf/Documentation/perf-record.txt +0
-0
tools/perf/Documentation/perf-record.txt Set affinity mask of trace reading thread according to the policy defined by 'mo · line 465
Quoted 7 lines
tools/perf/Documentation/perf-record.txt +4
-0@@ -459,6 +459,18 @@ Set affinity mask of trace reading thread according to the policy defined by 'mo node - thread affinity mask is set to NUMA node cpu mask of the processed mmap buffer cpu - thread affinity mask is set to cpu of the processed mmap buffer +--mmap-flush=number:: +Specify minimal number of bytes that is extracted from mmap data pages and stored +into a trace. The number specification is possible using B/K/M/G suffixes. Maximal allowed +value is a quarter of the size of mmaped data pages. The default option value is 1 byte
I found this annoying, I tried first with the default value:
perf trace -m 2048 --call-graph dwarf -e write -- perf record --mmap-flush <SNIP> the first writes for the synthesized data:
107.561 ( 0.005 ms): perf/25821 write(fd: 3</root/perf.data>, buf: 0x7f1befc02000, count: 336) = 336
__libc_write (/usr/lib64/libpthread-2.28.so)
ion (/home/acme/bin/perf)
record__write (inlined)
record__pushfn (/home/acme/bin/perf)
perf_mmap__push (/home/acme/bin/perf)
record__mmap_read_evlist (inlined)
record__mmap_read_all (inlined)
__cmd_record (inlined)
cmd_record (/home/acme/bin/perf)
12919.953 ( 0.136 ms): perf/25821 write(fd: 3</root/perf.data>, buf: 0x7f1befc83150, count: 184984) = 184984
<SNIP same backtrace as in the 107.561 timestamp>
12920.094 ( 0.155 ms): perf/25821 write(fd: 3</root/perf.data>, buf: 0x7f1befc02150, count: 261816) = 261816
<SNIP same backtrace as in the 107.561 timestamp>
12920.253 ( 0.093 ms): perf/25821 write(fd: 3</root/perf.data>, buf: 0x7f1befb81120, count: 170832) = 170832
<SNIP same backtrace as in the 107.561 timestamp>Then with --mmap-flush 16M, and then the writes to perf.data were always more than 132096, which is the limit that it silently set, I think we should warn this record__mmap_flush_parse, something like:
"max flush is a quarter of the mmap size, if wanting to bump the mmap flush further, bump the mmap size as well using -m/--mmap-pages"
Found this using -v, which shows the mmap size twice, one line after the next one:
mmap flush: 132096
mmap size 528384B
mmap size 528384B
I reflowed a bit the man page and added committer notes testing it, end result is at the bottom of this message, I also had to rename 'sync' to 'synch' to get it to build with other glibcs:
CC /tmp/build/perf/builtin-kmem.o
cc1: warnings being treated as errorsbuiltin-record.c: In function 'record__mmap_read_evlist':
builtin-record.c:775: warning: declaration of 'sync' shadows a global declaration /usr/include/unistd.h:933: warning: shadowed declaration is here builtin-record.c: In function 'record__mmap_read_all':
builtin-record.c:856: warning: declaration of 'sync' shadows a global declaration /usr/include/unistd.h:933: warning: shadowed declaration is here
mv: cannot stat `/tmp/build/perf/.builtin-record.o.tmp': No such file or directory
commit 221771de64b6bd0422f451e2c808d75eb3721814
Author: Alexey Budankov <[email protected]> Date: Mon Mar 18 20:40:26 2019 +0300
perf record: Implement --mmap-flush=<number> option
Implement a --mmap-flush option that specifies minimal number of bytes that is extracted from mmaped kernel buffer to store into a trace. The default option value is 1 byte what means every time trace writing thread finds some new data in the mmaped buffer the data is extracted, possibly compressed and written to a trace.
$ tools/perf/perf record --mmap-flush 1024 -e cycles -- matrix.gcc $ tools/perf/perf record --aio --mmap-flush 1K -e cycles -- matrix.gcc
The option is independent from -z setting, doesn't vary with compression level and can serve two purposes.
The first purpose is to increase the compression ratio of a trace data. Larger data chunks are compressed more effectively so the implemented option allows specifying data chunk size to compress. Also at some cases executing more write syscalls with smaller data size can take longer than executing less write syscalls with bigger data size due to syscall overhead so extracting bigger data chunks specified by the option value could additionally decrease runtime overhead.
The second purpose is to avoid self monitoring live-lock issue in system wide (-a) profiling mode. Profiling in system wide mode with compression (-a -z) can additionally induce data into the kernel buffers along with the data from monitored processes. If performance data rate and volume from the monitored processes is high then trace streaming and compression activity in the tool is also high. High tool process activity can lead to subtle live-lock effect when compression of single new byte from some of mmaped kernel buffer leads to generation of the next single byte at some mmaped buffer. So perf tool process ends up in endless self monitoring.
Implemented synch parameter is the mean to force data move independently from the specified flush threshold value. Despite the provided flush value the tool needs capability to unconditionally drain memory buffers, at least in the end of the collection.
Committer testing:
Running with the default value, i.e. as soon as there is something to read go on consuming, we first write the synthesized events, small chunks of about 128 bytes:
# perf trace -m 2048 --call-graph dwarf -e write -- perf record
<SNIP>
101.142 ( 0.004 ms): perf/25821 write(fd: 3</root/perf.data>, buf: 0x210db60, count: 120) = 120
__libc_write (/usr/lib64/libpthread-2.28.so) ion (/home/acme/bin/perf)
record__write (inlined) process_synthesized_event (/home/acme/bin/perf) perf_tool__process_synth_event (inlined) perf_event__synthesize_mmap_events (/home/acme/bin/perf)
Then we move to reading the mmap buffers consuming the events put there by the kernel perf infrastructure:
107.561 ( 0.005 ms): perf/25821 write(fd: 3</root/perf.data>, buf: 0x7f1befc02000, count: 336) = 336
__libc_write (/usr/lib64/libpthread-2.28.so)
ion (/home/acme/bin/perf)
record__write (inlined)
record__pushfn (/home/acme/bin/perf)
perf_mmap__push (/home/acme/bin/perf) record__mmap_read_evlist (inlined) record__mmap_read_all (inlined)
__cmd_record (inlined)
cmd_record (/home/acme/bin/perf)
12919.953 ( 0.136 ms): perf/25821 write(fd: 3</root/perf.data>, buf: 0x7f1befc83150, count: 184984) = 184984
<SNIP same backtrace as in the 107.561 timestamp>
12920.094 ( 0.155 ms): perf/25821 write(fd: 3</root/perf.data>, buf: 0x7f1befc02150, count: 261816) = 261816
<SNIP same backtrace as in the 107.561 timestamp>
12920.253 ( 0.093 ms): perf/25821 write(fd: 3</root/perf.data>, buf: 0x7f1befb81120, count: 170832) = 170832
<SNIP same backtrace as in the 107.561 timestamp>
If we limit it to write only when more than 16MB are available for reading, it throttles that to a quarter of the --mmap-pages set for 'perf record', which by default get to 528384 bytes, found out using 'record -v':
mmap flush: 132096
mmap size 528384B
With that in place all the writes coming from record__mmap_read_evlist(), i.e. from the mmap buffers setup by the kernel perf infrastructure were at least 132096 bytes long.
Trying with a bigger mmap size:
perf trace -e write perf record -v -m 2048 --mmap-flush 16M
74982.928 ( 2.471 ms): perf/26500 write(fd: 3</root/perf.data>, buf: 0x7ff94a6cc000, count: 3580888) = 3580888 74985.406 ( 2.353 ms): perf/26500 write(fd: 3</root/perf.data>, buf: 0x7ff949ecb000, count: 3453256) = 3453256 74987.764 ( 2.629 ms): perf/26500 write(fd: 3</root/perf.data>, buf: 0x7ff9496ca000, count: 3859232) = 3859232 74990.399 ( 2.341 ms): perf/26500 write(fd: 3</root/perf.data>, buf: 0x7ff948ec9000, count: 3769032) = 3769032 74992.744 ( 2.064 ms): perf/26500 write(fd: 3</root/perf.data>, buf: 0x7ff9486c8000, count: 3310520) = 3310520 74994.814 ( 2.619 ms): perf/26500 write(fd: 3</root/perf.data>, buf: 0x7ff947ec7000, count: 4194688) = 4194688 74997.439 ( 2.787 ms): perf/26500 write(fd: 3</root/perf.data>, buf: 0x7ff9476c6000, count: 4029760) = 4029760
Was again limited to a quarter of the mmap size:
mmap flush: 2098176
mmap size 8392704B
A warning about that would be good to have but can be added later, something like:
"max flush is a quarter of the mmap size, if wanting to bump the mmap
flush further, bump the mmap size as well using -m/--mmap-pages"
tools/perf/Documentation/perf-record.txt +19 -0
@@ -459,6 +459,25 @@ Set affinity mask of trace reading thread according to the policy defined by 'mo node - thread affinity mask is set to NUMA node cpu mask of the processed mmap buffer cpu - thread affinity mask is set to cpu of the processed mmap buffer +--mmap-flush=number:: + +Specify minimal number of bytes that is extracted from mmap data pages and +processed for output. One can specify the number using B/K/M/G suffixes. + +The maximal allowed value is a quarter of the size of mmaped data pages. + +The default option value is 1 byte which means that every time that the output +writing thread finds some new data in the mmaped buffer the data is extracted, +possibly compressed (-z) and written to the output, perf.data or pipe. + +Larger data chunks are compressed more effectively in comparison to smaller +chunks so extraction of larger chunks from the mmap data pages is preferable +from the perspective of output size reduction. + +Also at some cases executing less output write syscalls with bigger data size +can take less time than executing more output write syscalls with smaller data +size thus lowering runtime profiling overhead. + --all-kernel:: Configure all used events to run in kernel space.
tools/perf/builtin-record.c +59 -6
147 lines not on this page · raw message
tools/perf/perf.h +1 -0
@@ -85,6 +85,7 @@ struct record_opts {u64 clockid_res_ns;int nr_cblocks;int affinity; +int mmap_flush; };enum perf_affinity {
tools/perf/util/evlist.c +3 -3
@@ -1038,7 +1038,7 @@ int perf_evlist__parse_mmap_pages(const struct option *opt, const char *str,*/ int perf_evlist__mmap_ex (struct perf_evlist *evlist,unsigned int pages,unsigned int auxtrace_pages, -bool auxtrace_overwrite,int nr_cblocks,int affinity) +bool auxtrace_overwrite,int nr_cblocks,int affinity,int flush) {struct perf_evsel *evsel;const struct cpu_map *cpus = evlist->cpus;
@@ -1048,7 +1048,7 @@ int perf_evlist__mmap_ex(struct perf_evlist *evlist, unsigned int pages,* Its value is decided by evsel's write_backward. * So &mp should not be passed through const pointer. */ -struct mmap_params mp = { .nr_cblocks = nr_cblocks, .affinity = affinity }; +struct mmap_params mp = { .nr_cblocks = nr_cblocks, .affinity = affinity, .flush = flush };if (!evlist->mmap) evlist->mmap =perf_evlist__alloc_mmap (evlist, false);
@@ -1080,7 +1080,7 @@ int perf_evlist__mmap_ex(struct perf_evlist *evlist, unsigned int pages,int perf_evlist__mmap (struct perf_evlist *evlist,unsigned int pages) { -return perf_evlist__mmap_ex (evlist, pages,0 , false,0 ,PERF_AFFINITY_SYS ); +return perf_evlist__mmap_ex (evlist, pages,0 , false,0 ,PERF_AFFINITY_SYS ,1 ); }int perf_evlist__create_maps (struct perf_evlist *evlist,struct target *target)
tools/perf/util/evlist.h +2 -1
@@ -177,7 +177,8 @@ unsigned long perf_event_mlock_kb_in_pages(void);int perf_evlist__mmap_ex (struct perf_evlist *evlist,unsigned int pages,unsigned int auxtrace_pages, -bool auxtrace_overwrite,int nr_cblocks,int affinity); +bool auxtrace_overwrite,int nr_cblocks, +int affinity,int flush);int perf_evlist__mmap (struct perf_evlist *evlist,unsigned int pages);void perf_evlist__munmap (struct perf_evlist *evlist);
tools/perf/util/mmap.c +3 -1
@@ -440,6 +440,8 @@ int perf_mmap__mmap(struct perf_mmap *map, struct mmap_params *mp, int fd, int cperf_mmap__setup_affinity_mask (map, mp); + map->flush = mp->flush; +if (auxtrace_mmap__mmap (&map->auxtrace_mmap, &mp->auxtrace_mp, map->base, fd))return -1 ;
@@ -492,7 +494,7 @@ static int __perf_mmap__read_init(struct perf_mmap *md) md->start = md->overwrite ? head : old; md->end = md->overwrite ? old : head; -if (md->start == md->end) +if ((md->end - md->start) < md->flush)return -EAGAIN ; size = md->end - md->start;
tools/perf/util/mmap.h +2 -1
@@ -39,6 +39,7 @@ struct perf_mmap { } aio;#endif cpu_set_t affinity_mask; +u64 flush; };/*
@@ -70,7 +71,7 @@ enum bkw_mmap_state { };struct mmap_params { -int prot, mask, nr_cblocks, affinity; +int prot, mask, nr_cblocks, affinity, flush;struct auxtrace_mmap_params auxtrace_mp; };
Re: [PATCH v10 02/12] perf record: implement --mmap-flush=<number> option
Quoted 84 lines · Arnaldo Carvalho de Melo
Em Mon, Mar 18, 2019 at 08:40:26PM +0300, Alexey Budankov escreveu:
Quoted 53 lines
Implemented --mmap-flush option that specifies minimal number of bytes that is extracted from mmaped kernel buffer to store into a trace. The default option value is 1 byte what means every time trace writing thread finds some new data in the mmaped buffer the data is extracted, possibly compressed and written to a trace.
$ tools/perf/perf record --mmap-flush 1024 -e cycles -- matrix.gcc $ tools/perf/perf record --aio --mmap-flush 1K -e cycles -- matrix.gccThe option is independent from -z setting, doesn't vary with compression level and can serve two purposes.
The first purpose is to increase the compression ratio of a trace data. Larger data chunks are compressed more effectively so the implemented option allows specifying data chunk size to compress. Also at some cases executing more write syscalls with smaller data size can take longer than executing less write syscalls with bigger data size due to syscall overhead so extracting bigger data chunks specified by the option value could additionally decrease runtime overhead.
The second purpose is to avoid self monitoring live-lock issue in system wide (-a) profiling mode. Profiling in system wide mode with compression (-a -z) can additionally induce data into the kernel buffers along with the data from monitored processes. If performance data rate and volume from the monitored processes is high then trace streaming and compression activity in the tool is also high. High tool process activity can lead to subtle live-lock effect when compression of single new byte from some of mmaped kernel buffer leads to generation of the next single byte at some mmaped buffer. So perf tool process ends up in endless self monitoring.
Implemented sync parameter is the mean to force data move independently from the specified flush threshold value. Despite the provided flush value the tool needs capability to unconditionally drain memory buffers, at least in the end of the collection.
Signed-off-by Alexey Budankov <[email protected]>--- tools/perf/Documentation/perf-record.txt | 12 +++++ tools/perf/builtin-record.c | 65 +++++++++++++++++++++--- tools/perf/perf.h | 1 + tools/perf/util/evlist.c | 6 +-- tools/perf/util/evlist.h | 3 +- tools/perf/util/mmap.c | 4 +- tools/perf/util/mmap.h | 3 +- 7 files changed, 82 insertions(+), 12 deletions(-)tools/perf/Documentation/perf-record.txt +0
-0tools/perf/Documentation/perf-record.txt Set affinity mask of trace reading thread according to the policy defined by 'mo · line 465
Quoted 7 lines
tools/perf/Documentation/perf-record.txt +4
-0@@ -459,6 +459,18 @@ Set affinity mask of trace reading thread according to the policy defined by 'mo node - thread affinity mask is set to NUMA node cpu mask of the processed mmap buffer cpu - thread affinity mask is set to cpu of the processed mmap buffer +--mmap-flush=number:: +Specify minimal number of bytes that is extracted from mmap data pages and stored +into a trace. The number specification is possible using B/K/M/G suffixes. Maximal allowed +value is a quarter of the size of mmaped data pages. The default option value is 1 byteI found this annoying, I tried first with the default value:
perf trace -m 2048 --call-graph dwarf -e write -- perf record --mmap-flush <SNIP> the first writes for the synthesized data:
107.561 ( 0.005 ms): perf/25821 write(fd: 3</root/perf.data>, buf: 0x7f1befc02000, count: 336) = 336 __libc_write (/usr/lib64/libpthread-2.28.so) ion (/home/acme/bin/perf) record__write (inlined) record__pushfn (/home/acme/bin/perf) perf_mmap__push (/home/acme/bin/perf) record__mmap_read_evlist (inlined) record__mmap_read_all (inlined) __cmd_record (inlined) cmd_record (/home/acme/bin/perf) 12919.953 ( 0.136 ms): perf/25821 write(fd: 3</root/perf.data>, buf: 0x7f1befc83150, count: 184984) = 184984 <SNIP same backtrace as in the 107.561 timestamp> 12920.094 ( 0.155 ms): perf/25821 write(fd: 3</root/perf.data>, buf: 0x7f1befc02150, count: 261816) = 261816 <SNIP same backtrace as in the 107.561 timestamp> 12920.253 ( 0.093 ms): perf/25821 write(fd: 3</root/perf.data>, buf: 0x7f1befb81120, count: 170832) = 170832 <SNIP same backtrace as in the 107.561 timestamp>
Quoted 6 lines
Then with --mmap-flush 16M, and then the writes to perf.data were always more than 132096, which is the limit that it silently set, I think we should warn this
record__mmap_flush_parse, something like:"max flush is a quarter of the mmap size, if wanting to bump the mmap flush further, bump the mmap size as well using -m/--mmap-pages"
Makes sense.
Quoted 12 lines
Found this using -v, which shows the mmap size twice, one line after the next one:
mmap flush: 132096
mmap size 528384B
mmap size 528384BI reflowed a bit the man page and added committer notes testing it, end result is at the bottom of this message, I also had to rename 'sync' to 'synch' to get it to build with other glibcs:
Quoted 9 lines
CC
/tmp/build/perf/builtin-kmem.o
cc1: warnings being treated as errorsbuiltin-record.c: In function 'record__mmap_read_evlist':
builtin-record.c:775: warning: declaration of 'sync' shadows a global declaration/usr/include/unistd.h:933: warning: shadowed declaration is herebuiltin-record.c: In function 'record__mmap_read_all':
builtin-record.c:856: warning: declaration of 'sync' shadows a global declaration/usr/include/unistd.h:933: warning: shadowed declaration is here
mv: cannot stat `/tmp/build/perf/.builtin-record.o.tmp': No such file or directory
Thanks for applied corrections.
~Alexey
Quoted 417 lines
commit
221771de64b6bd0422f451e2c808d75eb3721814Author: Alexey Budankov <[email protected]> Date: Mon Mar 18 20:40:26 2019 +0300perf record: Implement --mmap-flush=<number> option
Implement a --mmap-flush option that specifies minimal number of bytes that is extracted from mmaped kernel buffer to store into a trace. The default option value is 1 byte what means every time trace writing thread finds some new data in the mmaped buffer the data is extracted, possibly compressed and written to a trace.
$
tools/perf/perfrecord --mmap-flush 1024 -e cycles -- matrix.gcc $tools/perf/perfrecord --aio --mmap-flush 1K -e cycles -- matrix.gccThe option is independent from -z setting, doesn't vary with compression level and can serve two purposes.
The first purpose is to increase the compression ratio of a trace data. Larger data chunks are compressed more effectively so the implemented option allows specifying data chunk size to compress. Also at some cases executing more write syscalls with smaller data size can take longer than executing less write syscalls with bigger data size due to syscall overhead so extracting bigger data chunks specified by the option value could additionally decrease runtime overhead.
The second purpose is to avoid self monitoring live-lock issue in system wide (-a) profiling mode. Profiling in system wide mode with compression (-a -z) can additionally induce data into the kernel buffers along with the data from monitored processes. If performance data rate and volume from the monitored processes is high then trace streaming and compression activity in the tool is also high. High tool process activity can lead to subtle live-lock effect when compression of single new byte from some of mmaped kernel buffer leads to generation of the next single byte at some mmaped buffer. So perf tool process ends up in endless self monitoring.
Implemented synch parameter is the mean to force data move independently from the specified flush threshold value. Despite the provided flush value the tool needs capability to unconditionally drain memory buffers, at least in the end of the collection.
Committer testing:
Running with the default value, i.e. as soon as there is something to read go on consuming, we first write the synthesized events, small chunks of about 128 bytes:
# perf trace -m 2048 --call-graph dwarf -e write -- perf record
<SNIP>
101.142 ( 0.004 ms): perf/25821write(fd: 3</root/perf.data>, buf: 0x210db60, count: 120)= 120
__libc_write(/usr/lib64/libpthread-2.28.so) ion (/home/acme/bin/perf)
record__write(inlined)process_synthesized_event(/home/acme/bin/perf)perf_tool__process_synth_event(inlined)perf_event__synthesize_mmap_events(/home/acme/bin/perf)Then we move to reading the mmap buffers consuming the events put there by the kernel perf infrastructure:
107.561 ( 0.005 ms): perf/25821
write(fd: 3</root/perf.data>, buf: 0x7f1befc02000, count: 336)= 336
__libc_write(/usr/lib64/libpthread-2.28.so)
ion (/home/acme/bin/perf)
record__write(inlined)
record__pushfn(/home/acme/bin/perf)
perf_mmap__push(/home/acme/bin/perf)record__mmap_read_evlist(inlined)record__mmap_read_all(inlined)
__cmd_record(inlined)
cmd_record(/home/acme/bin/perf)
12919.953 ( 0.136 ms): perf/25821write(fd: 3</root/perf.data>, buf: 0x7f1befc83150, count: 184984)= 184984
<SNIP same backtrace as in the 107.561 timestamp>
12920.094 ( 0.155 ms): perf/25821write(fd: 3</root/perf.data>, buf: 0x7f1befc02150, count: 261816)= 261816
<SNIP same backtrace as in the 107.561 timestamp>
12920.253 ( 0.093 ms): perf/25821write(fd: 3</root/perf.data>, buf: 0x7f1befb81120, count: 170832)= 170832
<SNIP same backtrace as in the 107.561 timestamp>If we limit it to write only when more than 16MB are available for reading, it throttles that to a quarter of the --mmap-pages set for 'perf record', which by default get to 528384 bytes, found out using 'record -v':
mmap flush: 132096
mmap size 528384BWith that in place all the writes coming from
record__mmap_read_evlist(), i.e. from the mmap buffers setup by the kernel perf infrastructure were at least 132096 bytes long.Trying with a bigger mmap size:
perf trace -e write perf record -v -m 2048 --mmap-flush 16M
74982.928 ( 2.471 ms): perf/26500write(fd: 3</root/perf.data>, buf: 0x7ff94a6cc000, count: 3580888)= 3580888 74985.406 ( 2.353 ms): perf/26500write(fd: 3</root/perf.data>, buf: 0x7ff949ecb000, count: 3453256)= 3453256 74987.764 ( 2.629 ms): perf/26500write(fd: 3</root/perf.data>, buf: 0x7ff9496ca000, count: 3859232)= 3859232 74990.399 ( 2.341 ms): perf/26500write(fd: 3</root/perf.data>, buf: 0x7ff948ec9000, count: 3769032)= 3769032 74992.744 ( 2.064 ms): perf/26500write(fd: 3</root/perf.data>, buf: 0x7ff9486c8000, count: 3310520)= 3310520 74994.814 ( 2.619 ms): perf/26500write(fd: 3</root/perf.data>, buf: 0x7ff947ec7000, count: 4194688)= 4194688 74997.439 ( 2.787 ms): perf/26500write(fd: 3</root/perf.data>, buf: 0x7ff9476c6000, count: 4029760)= 4029760Was again limited to a quarter of the mmap size:
mmap flush: 2098176
mmap size 8392704BA warning about that would be good to have but can be added later, something like:
"max flush is a quarter of the mmap size, if wanting to bump the mmap
flush further, bump the mmap size as well using -m/--mmap-pages"Signed-off-by Alexey Budankov <[email protected]>Reviewed-by Jiri Olsa <[email protected]>Cc Alexander Shishkin <[email protected]>Cc Andi Kleen <[email protected]>Cc Namhyung Kim <[email protected]>Cc Peter Zijlstra <[email protected]>Signed-off-by Arnaldo Carvalho de Melo <[email protected]>tools/perf/Documentation/perf-record.txt +19
-0@@ -459,6 +459,25 @@ Set affinity mask of trace reading thread according to the policy defined by 'mo node - thread affinity mask is set to NUMA node cpu mask of the processed mmap buffer cpu - thread affinity mask is set to cpu of the processed mmap buffer +--mmap-flush=number:: + +Specify minimal number of bytes that is extracted from mmap data pages and +processed for output. One can specify the number using B/K/M/G suffixes. + +The maximal allowed value is a quarter of the size of mmaped data pages. + +The default option value is 1 byte which means that every time that the output +writing thread finds some new data in the mmaped buffer the data is extracted, +possibly compressed (-z) and written to the output, perf.data or pipe. + +Larger data chunks are compressed more effectively in comparison to smaller +chunks so extraction of larger chunks from the mmap data pages is preferable +from the perspective of output size reduction. + +Also at some cases executing less output write syscalls with bigger data size +can take less time than executing more output write syscalls with smaller data +size thus lowering runtime profiling overhead. + --all-kernel:: Configure all used events to run in kernel space.tools/perf/builtin-record.c +59
-6@@ -337,6 +337,41 @@ static int record__aio_enabled(struct record *rec)return rec->opts.nr_cblocks >0 ; } +#define MMAP_FLUSH_DEFAULT 1 +static int record__mmap_flush_parse (const struct option *opt, +const char *str, +int unset) +{ +int flush_max; +struct record_opts *opts = (struct record_opts *)opt->value; +static struct parse_tag tags[] = { + { .tag ='B' , .mult =1 }, + { .tag ='K' , .mult =1 <<10 }, + { .tag ='M' , .mult =1 <<20 }, + { .tag ='G' , .mult =1 <<30 }, + { .tag =0 }, + }; + +if (unset) +return 0 ; + +if (str) { + opts->mmap_flush =parse_tag_value (str, tags); +if (opts->mmap_flush == (int )-1 ) + opts->mmap_flush =strtol (str,NULL ,0 ); + } + +if (!opts->mmap_flush) + opts->mmap_flush =MMAP_FLUSH_DEFAULT ; + + flush_max =perf_evlist__mmap_size (opts->mmap_pages); + flush_max /=4 ; +if (opts->mmap_flush > flush_max) + opts->mmap_flush = flush_max; + +return 0 ; +} + static int process_synthesized_event(struct perf_tool *tool, union perf_event *event, struct perf_sample *sample __maybe_unused,@@ -546,7 +581,8 @@ static int record__mmap_evlist(struct record *rec,if (perf_evlist__mmap_ex (evlist, opts->mmap_pages, opts->auxtrace_mmap_pages, opts->auxtrace_snapshot_mode, - opts->nr_cblocks, opts->affinity) <0 ) { + opts->nr_cblocks, opts->affinity, + opts->mmap_flush) <0 ) {if (errno ==EPERM ) {pr_err ("Permission error mapping pages.\n" "Consider increasing "@@ -736,7 +772,7 @@ static void record__adjust_affinity(struct record *rec, struct perf_mmap *map) }static int record__mmap_read_evlist (struct record *rec,struct perf_evlist *evlist, -bool overwrite) +bool overwrite,bool synch) {u64 bytes_written = rec->bytes_written;int i;@@ -759,12 +795,19 @@ static int record__mmap_read_evlist(struct record *rec, struct perf_evlist *evli off =record__aio_get_pos (trace_fd);for (i =0 ; i < evlist->nr_mmaps; i++) { +u64 flush =0 ;struct perf_mmap *map = &maps[i];if (map->base) {record__adjust_affinity (rec, map); +if (synch) { + flush = map->flush; + map->flush =1 ; + }if (!record__aio_enabled (rec)) {if (perf_mmap__push (map, rec, record__pushfn) !=0 ) { +if (synch) + map->flush = flush; rc =-1 ; goto out; }@@ -777,10 +820,14 @@ static int record__mmap_read_evlist(struct record *rec, struct perf_evlist *evli idx =record__aio_sync (map, false);if (perf_mmap__aio_push (map, rec, idx, record__aio_pushfn, &off) !=0 ) {record__aio_set_pos (trace_fd, off); +if (synch) + map->flush = flush; rc =-1 ; goto out; } } +if (synch) + map->flush = flush; } if (map->auxtrace_mmap.base && !rec->opts.auxtrace_snapshot_mode &&@@ -806,15 +853,15 @@ static int record__mmap_read_evlist(struct record *rec, struct perf_evlist *evlireturn rc; } -static int record__mmap_read_all (struct record *rec) +static int record__mmap_read_all (struct record *rec,bool synch) {int err; - err =record__mmap_read_evlist (rec, rec->evlist, false); + err =record__mmap_read_evlist (rec, rec->evlist, false, synch);if (err)return err; -return record__mmap_read_evlist (rec, rec->evlist, true); +return record__mmap_read_evlist (rec, rec->evlist, true, synch); }static void record__init_features (struct record *rec)@@ -1340,7 +1387,7 @@ static int __cmd_record(struct record *rec, int argc, const char **argv)if (trigger_is_hit (&switch_output_trigger) || done || draining)perf_evlist__toggle_bkw_mmap (rec->evlist,BKW_MMAP_DATA_PENDING ); -if (record__mmap_read_all (rec) <0 ) { +if (record__mmap_read_all (rec, false) <0 ) {trigger_error (&auxtrace_snapshot_trigger);trigger_error (&switch_output_trigger); err =-1 ;@@ -1441,6 +1488,7 @@ static int __cmd_record(struct record *rec, int argc, const char **argv)record__synthesize_workload (rec, true); out_child: +record__mmap_read_all (rec, true);record__aio_mmap_read_sync (rec);if (forks) {@@ -1846,6 +1894,7 @@ static struct record record = { .uses_mmap = true, .default_per_cpu = true, }, + .mmap_flush = MMAP_FLUSH_DEFAULT, }, .tool = { .sample = process_sample_event,@@ -1912,6 +1961,9 @@ static struct option __record_options[] = {OPT_CALLBACK ('m' ,"mmap-pages" , &record.opts,"pages[,pages]" ,"number of mmap data pages and AUX area tracing mmap pages" , record__parse_mmap_pages), +OPT_CALLBACK (0 ,"mmap-flush" , &record.opts,"number" , +"Minimal number of bytes that is extracted from mmap data pages (default: 1)" , + record__mmap_flush_parse),OPT_BOOLEAN (0 ,"group" , &record.opts.group,"put the counters into a counter group" ), OPT_CALLBACK_NOOPT('g', NULL, &callchain_param,@@ -2224,6 +2276,7 @@ int cmd_record(int argc, const char **argv)pr_info ("nr_cblocks: %d\n" , rec->opts.nr_cblocks);pr_debug ("affinity: %s\n" , affinity_tags[rec->opts.affinity]); +pr_debug ("mmap flush: %d\n" , rec->opts.mmap_flush); err =__cmd_record (&record, argc, argv); out:tools/perf/perf.h +1
-0@@ -85,6 +85,7 @@ struct record_opts {u64 clockid_res_ns;int nr_cblocks;int affinity; +int mmap_flush; };enum perf_affinity {tools/perf/util/evlist.c +3
-3@@ -1038,7 +1038,7 @@ int perf_evlist__parse_mmap_pages(const struct option *opt, const char *str,*/ int perf_evlist__mmap_ex (struct perf_evlist *evlist,unsigned int pages,unsigned int auxtrace_pages, -bool auxtrace_overwrite,int nr_cblocks,int affinity) +bool auxtrace_overwrite,int nr_cblocks,int affinity,int flush) {struct perf_evsel *evsel;const struct cpu_map *cpus = evlist->cpus;@@ -1048,7 +1048,7 @@ int perf_evlist__mmap_ex(struct perf_evlist *evlist, unsigned int pages,* Its value is decided by evsel's write_backward. * So &mp should not be passed through const pointer. */ -struct mmap_params mp = { .nr_cblocks = nr_cblocks, .affinity = affinity }; +struct mmap_params mp = { .nr_cblocks = nr_cblocks, .affinity = affinity, .flush = flush };if (!evlist->mmap) evlist->mmap =perf_evlist__alloc_mmap (evlist, false);@@ -1080,7 +1080,7 @@ int perf_evlist__mmap_ex(struct perf_evlist *evlist, unsigned int pages,int perf_evlist__mmap (struct perf_evlist *evlist,unsigned int pages) { -return perf_evlist__mmap_ex (evlist, pages,0 , false,0 ,PERF_AFFINITY_SYS ); +return perf_evlist__mmap_ex (evlist, pages,0 , false,0 ,PERF_AFFINITY_SYS ,1 ); }int perf_evlist__create_maps (struct perf_evlist *evlist,struct target *target)tools/perf/util/evlist.h +2
-1@@ -177,7 +177,8 @@ unsigned long perf_event_mlock_kb_in_pages(void);int perf_evlist__mmap_ex (struct perf_evlist *evlist,unsigned int pages,unsigned int auxtrace_pages, -bool auxtrace_overwrite,int nr_cblocks,int affinity); +bool auxtrace_overwrite,int nr_cblocks, +int affinity,int flush);int perf_evlist__mmap (struct perf_evlist *evlist,unsigned int pages);void perf_evlist__munmap (struct perf_evlist *evlist);tools/perf/util/mmap.c +3
-1@@ -440,6 +440,8 @@ int perf_mmap__mmap(struct perf_mmap *map, struct mmap_params *mp, int fd, int cperf_mmap__setup_affinity_mask (map, mp); + map->flush = mp->flush; +if (auxtrace_mmap__mmap (&map->auxtrace_mmap, &mp->auxtrace_mp, map->base, fd))return -1 ;@@ -492,7 +494,7 @@ static int __perf_mmap__read_init(struct perf_mmap *md) md->start = md->overwrite ? head : old; md->end = md->overwrite ? old : head; -if (md->start == md->end) +if ((md->end - md->start) < md->flush)return -EAGAIN ; size = md->end - md->start;tools/perf/util/mmap.h +2
-1@@ -39,6 +39,7 @@ struct perf_mmap { } aio;#endif cpu_set_t affinity_mask; +u64 flush; };/* @@ -70,7 +71,7 @@ enum bkw_mmap_state { };struct mmap_params { -int prot, mask, nr_cblocks, affinity; +int prot, mask, nr_cblocks, affinity, flush;struct auxtrace_mmap_params auxtrace_mp; };
tip-bot for Alexey Budankov
merge bot[tip:perf/core] perf record: Implement --mmap-flush=<number> option
In reply to Alexey Budankovperf record: Implement --mmap-flush=<number> option
Commit-ID: 470530bbb8fbbf2a09bd1d7150bb92501c5c54e6 Gitweb: https://git.kernel.org/tip/470530bbb8fbbf2a09bd1d7150bb92501c5c54e6 Author: Alexey Budankov <[email protected]> AuthorDate: Mon, 18 Mar 2019 20:40:26 +0300 Committer: Arnaldo Carvalho de Melo <[email protected]> CommitDate: Mon, 1 Apr 2019 15:18:10 -0300
perf record: Implement --mmap-flush=<number> option
Implement a --mmap-flush option that specifies minimal number of bytes that is extracted from mmaped kernel buffer to store into a trace. The default option value is 1 byte what means every time trace writing thread finds some new data in the mmaped buffer the data is extracted, possibly compressed and written to a trace.
$ tools/perf/perf record --mmap-flush 1024 -e cycles -- matrix.gcc $ tools/perf/perf record --aio --mmap-flush 1K -e cycles -- matrix.gcc
The option is independent from -z setting, doesn't vary with compression level and can serve two purposes.
The first purpose is to increase the compression ratio of a trace data. Larger data chunks are compressed more effectively so the implemented option allows specifying data chunk size to compress. Also at some cases executing more write syscalls with smaller data size can take longer than executing less write syscalls with bigger data size due to syscall overhead so extracting bigger data chunks specified by the option value could additionally decrease runtime overhead.
The second purpose is to avoid self monitoring live-lock issue in system wide (-a) profiling mode. Profiling in system wide mode with compression (-a -z) can additionally induce data into the kernel buffers along with the data from monitored processes. If performance data rate and volume from the monitored processes is high then trace streaming and compression activity in the tool is also high. High tool process activity can lead to subtle live-lock effect when compression of single new byte from some of mmaped kernel buffer leads to generation of the next single byte at some mmaped buffer. So perf tool process ends up in endless self monitoring.
Implemented synch parameter is the mean to force data move independently from the specified flush threshold value. Despite the provided flush value the tool needs capability to unconditionally drain memory buffers, at least in the end of the collection.
Committer testing:
Running with the default value, i.e. as soon as there is something to read go on consuming, we first write the synthesized events, small chunks of about 128 bytes:
# perf trace -m 2048 --call-graph dwarf -e write -- perf record
<SNIP>
101.142 ( 0.004 ms): perf/25821 write(fd: 3</root/perf.data>, buf: 0x210db60, count: 120) = 120
__libc_write (/usr/lib64/libpthread-2.28.so)
ion (/home/acme/bin/perf)
record__write (inlined)
process_synthesized_event (/home/acme/bin/perf)
perf_tool__process_synth_event (inlined)
perf_event__synthesize_mmap_events (/home/acme/bin/perf)Then we move to reading the mmap buffers consuming the events put there by the kernel perf infrastructure:
107.561 ( 0.005 ms): perf/25821 write(fd: 3</root/perf.data>, buf: 0x7f1befc02000, count: 336) = 336
__libc_write (/usr/lib64/libpthread-2.28.so)
ion (/home/acme/bin/perf)
record__write (inlined)
record__pushfn (/home/acme/bin/perf)
perf_mmap__push (/home/acme/bin/perf)
record__mmap_read_evlist (inlined)
record__mmap_read_all (inlined)
__cmd_record (inlined)
cmd_record (/home/acme/bin/perf)
12919.953 ( 0.136 ms): perf/25821 write(fd: 3</root/perf.data>, buf: 0x7f1befc83150, count: 184984) = 184984
<SNIP same backtrace as in the 107.561 timestamp>
12920.094 ( 0.155 ms): perf/25821 write(fd: 3</root/perf.data>, buf: 0x7f1befc02150, count: 261816) = 261816
<SNIP same backtrace as in the 107.561 timestamp>
12920.253 ( 0.093 ms): perf/25821 write(fd: 3</root/perf.data>, buf: 0x7f1befb81120, count: 170832) = 170832<SNIP same backtrace as in the 107.561 timestamp>
If we limit it to write only when more than 16MB are available for reading, it throttles that to a quarter of the --mmap-pages set for 'perf record', which by default get to 528384 bytes, found out using 'record -v':
mmap flush: 132096
mmap size 528384B
With that in place all the writes coming from record__mmap_read_evlist(), i.e. from the mmap buffers setup by the kernel perf infrastructure were at least 132096 bytes long.
Trying with a bigger mmap size:
perf trace -e write perf record -v -m 2048 --mmap-flush 16M 74982.928 ( 2.471 ms): perf/26500 write(fd: 3</root/perf.data>, buf: 0x7ff94a6cc000, count: 3580888) = 3580888 74985.406 ( 2.353 ms): perf/26500 write(fd: 3</root/perf.data>, buf: 0x7ff949ecb000, count: 3453256) = 3453256 74987.764 ( 2.629 ms): perf/26500 write(fd: 3</root/perf.data>, buf: 0x7ff9496ca000, count: 3859232) = 3859232 74990.399 ( 2.341 ms): perf/26500 write(fd: 3</root/perf.data>, buf: 0x7ff948ec9000, count: 3769032) = 3769032 74992.744 ( 2.064 ms): perf/26500 write(fd: 3</root/perf.data>, buf: 0x7ff9486c8000, count: 3310520) = 3310520 74994.814 ( 2.619 ms): perf/26500 write(fd: 3</root/perf.data>, buf: 0x7ff947ec7000, count: 4194688) = 4194688 74997.439 ( 2.787 ms): perf/26500 write(fd: 3</root/perf.data>, buf: 0x7ff9476c6000, count: 4029760) = 4029760
Was again limited to a quarter of the mmap size:
mmap flush: 2098176
mmap size 8392704B
A warning about that would be good to have but can be added later, something like:
"max flush is a quarter of the mmap size, if wanting to bump the mmap
flush further, bump the mmap size as well using -m/--mmap-pages"
Also rename the 'sync' parameters to 'synch' to keep tools/perf building with older glibcs:
cc1: warnings being treated as errors
builtin-record.c: In function 'record__mmap_read_evlist':
builtin-record.c:775: warning: declaration of 'sync' shadows a global declaration /usr/include/unistd.h:933: warning: shadowed declaration is here builtin-record.c: In function 'record__mmap_read_all':
builtin-record.c:856: warning: declaration of 'sync' shadows a global declaration /usr/include/unistd.h:933: warning: shadowed declaration is here
--- tools/perf/Documentation/perf-record.txt | 19 ++++++++++ tools/perf/builtin-record.c | 65 +++++++++++++++++++++++++++++--- tools/perf/perf.h | 1 + tools/perf/util/evlist.c | 6 +-- tools/perf/util/evlist.h | 3 +- tools/perf/util/mmap.c | 4 +- tools/perf/util/mmap.h | 3 +- 7 files changed, 89 insertions(+), 12 deletions(-)
tools/perf/Documentation/perf-record.txt +19 -0
@@ -459,6 +459,25 @@ Set affinity mask of trace reading thread according to the policy defined by 'mo node - thread affinity mask is set to NUMA node cpu mask of the processed mmap buffer cpu - thread affinity mask is set to cpu of the processed mmap buffer +--mmap-flush=number:: + +Specify minimal number of bytes that is extracted from mmap data pages and +processed for output. One can specify the number using B/K/M/G suffixes. + +The maximal allowed value is a quarter of the size of mmaped data pages. + +The default option value is 1 byte which means that every time that the output +writing thread finds some new data in the mmaped buffer the data is extracted, +possibly compressed (-z) and written to the output, perf.data or pipe. + +Larger data chunks are compressed more effectively in comparison to smaller +chunks so extraction of larger chunks from the mmap data pages is preferable +from the perspective of output size reduction. + +Also at some cases executing less output write syscalls with bigger data size +can take less time than executing more output write syscalls with smaller data +size thus lowering runtime profiling overhead. + --all-kernel:: Configure all used events to run in kernel space.
tools/perf/builtin-record.c +59 -6
147 lines not on this page · raw message
tools/perf/perf.h +1 -0
@@ -85,6 +85,7 @@ struct record_opts {u64 clockid_res_ns;int nr_cblocks;int affinity; +int mmap_flush; };enum perf_affinity {
tools/perf/util/evlist.c +3 -3
@@ -1009,7 +1009,7 @@ int perf_evlist__parse_mmap_pages(const struct option *opt, const char *str,*/ int perf_evlist__mmap_ex (struct perf_evlist *evlist,unsigned int pages,unsigned int auxtrace_pages, -bool auxtrace_overwrite,int nr_cblocks,int affinity) +bool auxtrace_overwrite,int nr_cblocks,int affinity,int flush) {struct perf_evsel *evsel;const struct cpu_map *cpus = evlist->cpus;
@@ -1019,7 +1019,7 @@ int perf_evlist__mmap_ex(struct perf_evlist *evlist, unsigned int pages,* Its value is decided by evsel's write_backward. * So &mp should not be passed through const pointer. */ -struct mmap_params mp = { .nr_cblocks = nr_cblocks, .affinity = affinity }; +struct mmap_params mp = { .nr_cblocks = nr_cblocks, .affinity = affinity, .flush = flush };if (!evlist->mmap) evlist->mmap =perf_evlist__alloc_mmap (evlist, false);
@@ -1051,7 +1051,7 @@ int perf_evlist__mmap_ex(struct perf_evlist *evlist, unsigned int pages,int perf_evlist__mmap (struct perf_evlist *evlist,unsigned int pages) { -return perf_evlist__mmap_ex (evlist, pages,0 , false,0 ,PERF_AFFINITY_SYS ); +return perf_evlist__mmap_ex (evlist, pages,0 , false,0 ,PERF_AFFINITY_SYS ,1 ); }int perf_evlist__create_maps (struct perf_evlist *evlist,struct target *target)
tools/perf/util/evlist.h +2 -1
@@ -177,7 +177,8 @@ unsigned long perf_event_mlock_kb_in_pages(void);int perf_evlist__mmap_ex (struct perf_evlist *evlist,unsigned int pages,unsigned int auxtrace_pages, -bool auxtrace_overwrite,int nr_cblocks,int affinity); +bool auxtrace_overwrite,int nr_cblocks, +int affinity,int flush);int perf_evlist__mmap (struct perf_evlist *evlist,unsigned int pages);void perf_evlist__munmap (struct perf_evlist *evlist);
tools/perf/util/mmap.c +3 -1
@@ -440,6 +440,8 @@ int perf_mmap__mmap(struct perf_mmap *map, struct mmap_params *mp, int fd, int cperf_mmap__setup_affinity_mask (map, mp); + map->flush = mp->flush; +if (auxtrace_mmap__mmap (&map->auxtrace_mmap, &mp->auxtrace_mp, map->base, fd))return -1 ;
@@ -492,7 +494,7 @@ static int __perf_mmap__read_init(struct perf_mmap *md) md->start = md->overwrite ? head : old; md->end = md->overwrite ? old : head; -if (md->start == md->end) +if ((md->end - md->start) < md->flush)return -EAGAIN ; size = md->end - md->start;
tools/perf/util/mmap.h +2 -1
@@ -39,6 +39,7 @@ struct perf_mmap { } aio;#endif cpu_set_t affinity_mask; +u64 flush; };/*
@@ -70,7 +71,7 @@ enum bkw_mmap_state { };struct mmap_params { -int prot, mask, nr_cblocks, affinity; +int prot, mask, nr_cblocks, affinity, flush;struct auxtrace_mmap_params auxtrace_mp; };
PATCH v10 3/12 perf session: define bytes_transferred and bytes_compressed metrics
In reply to Alexey BudankovDefine bytes_transferred and bytes_compressed metrics to calculate ratio in the end of the data collection:
compression ratio = bytes_transferred / bytes_compressed
bytes_transferred accumulates the amount of bytes that was extracted from the mmaped kernel buffers for compression. bytes_compressed accumulates the amount of bytes that was received after applying compression.
--- tools/perf/builtin-record.c | 14 +++++++++++++- tools/perf/util/env.h | 1 + tools/perf/util/session.h | 2 ++ 3 files changed, 16 insertions(+), 1 deletion(-)
tools/perf/builtin-record.c +13 -1
@@ -1166,6 +1166,7 @@ static int __cmd_record(struct record *rec, int argc, const char **argv)struct perf_session *session;bool disabled = false, draining = false;int fd; +float ratio =0 ;atexit (record__sig_exit);signal (SIGCHLD , sig_handler);
@@ -1463,6 +1464,11 @@ static int __cmd_record(struct record *rec, int argc, const char **argv)record__mmap_read_all (rec, true);record__aio_mmap_read_sync (rec); +if (rec->session->bytes_transferred && rec->session->bytes_compressed) { + ratio = (float )rec->session->bytes_transferred/(float )rec->session->bytes_compressed; + session->header.env.comp_ratio = ratio +0.5 ; + } +if (forks) {int exit_status;
@@ -1509,9 +1515,15 @@ static int __cmd_record(struct record *rec, int argc, const char **argv)else samples[0 ] ='\0' ; -fprintf (stderr,"[ perf record: Captured and wrote %.3f MB %s%s%s ]\n" , +fprintf (stderr,"[ perf record: Captured and wrote %.3f MB %s%s%s" ,perf_data__size (data) /1024.0 /1024.0 , data->path, postfix, samples); +if (ratio) { +fprintf (stderr,", compressed (original %.3f MB, ratio is %.3f)" , + rec->session->bytes_transferred /1024.0 /1024.0 , + ratio); + } +fprintf (stderr," ]\n" ); } out_delete_session:
tools/perf/util/env.h +1 -0
@@ -64,6 +64,7 @@ struct perf_env {struct memory_node *memory_nodes;unsigned long long memory_bsize;u64 clockid_res_ns; +u32 comp_ratio; };extern struct perf_env perf_env;
tools/perf/util/session.h +2 -0
@@ -35,6 +35,8 @@ struct perf_session {struct ordered_events ordered_events;struct perf_data *data;struct perf_tool *tool; +u64 bytes_transferred; +u64 bytes_compressed; };struct perf_tool ;
Signature · 1 line
2.20.1
tip-bot for Alexey Budankov
merge bot[tip:perf/core] perf session: Define 'bytes_transferred' and 'bytes_compressed' metrics
perf session: Define 'bytes_transferred' and 'bytes_compressed' metrics
Commit-ID: d3c8c08e75c4cbb6a940323092b40fcfd1de5380 Gitweb: https://git.kernel.org/tip/d3c8c08e75c4cbb6a940323092b40fcfd1de5380 Author: Alexey Budankov <[email protected]> AuthorDate: Mon, 18 Mar 2019 20:41:02 +0300 Committer: Arnaldo Carvalho de Melo <[email protected]> CommitDate: Wed, 15 May 2019 16:36:49 -0300
perf session: Define 'bytes_transferred' and 'bytes_compressed' metrics
Define 'bytes_transferred' and 'bytes_compressed' metrics to calculate ratio in the end of the data collection:
compression ratio = bytes_transferred / bytes_compressed
The 'bytes_transferred' metric accumulates the amount of bytes that was extracted from the mmaped kernel buffers for compression, while 'bytes_compressed' accumulates the amount of bytes that was received after applying compression.
--- tools/perf/builtin-record.c | 14 +++++++++++++- tools/perf/util/env.h | 1 + tools/perf/util/session.h | 2 ++ 3 files changed, 16 insertions(+), 1 deletion(-)
tools/perf/builtin-record.c +13 -1
@@ -1186,6 +1186,7 @@ static int __cmd_record(struct record *rec, int argc, const char **argv)bool disabled = false, draining = false;struct perf_evlist *sb_evlist =NULL ;int fd; +float ratio =0 ;atexit (record__sig_exit);signal (SIGCHLD , sig_handler);
@@ -1491,6 +1492,11 @@ out_child:record__mmap_read_all (rec, true);record__aio_mmap_read_sync (rec); +if (rec->session->bytes_transferred && rec->session->bytes_compressed) { + ratio = (float )rec->session->bytes_transferred/(float )rec->session->bytes_compressed; + session->header.env.comp_ratio = ratio +0.5 ; + } +if (forks) {int exit_status;
@@ -1537,9 +1543,15 @@ out_child:else samples[0 ] ='\0' ; -fprintf (stderr,"[ perf record: Captured and wrote %.3f MB %s%s%s ]\n" , +fprintf (stderr,"[ perf record: Captured and wrote %.3f MB %s%s%s" ,perf_data__size (data) /1024.0 /1024.0 , data->path, postfix, samples); +if (ratio) { +fprintf (stderr,", compressed (original %.3f MB, ratio is %.3f)" , + rec->session->bytes_transferred /1024.0 /1024.0 , + ratio); + } +fprintf (stderr," ]\n" ); } out_delete_session:
tools/perf/util/env.h +1 -0
@@ -62,6 +62,7 @@ struct perf_env {struct cpu_topology_map *cpu;struct cpu_cache_level *caches;int caches_cnt; +u32 comp_ratio;struct numa_node *numa_nodes;struct memory_node *memory_nodes;unsigned long long memory_bsize;
tools/perf/util/session.h +2 -0
@@ -35,6 +35,8 @@ struct perf_session {struct ordered_events ordered_events;struct perf_data *data;struct perf_tool *tool; +u64 bytes_transferred; +u64 bytes_compressed; };struct perf_tool ;
PATCH v10 4/12 perf record: implement COMPRESSED event record and its attributes
In reply to Alexey BudankovImplemented PERF_RECORD_COMPRESSED event, related data types, header feature and functions to write, read and print feature attributes from the trace header section.
comp_mmap_len preserves the size of mmaped kernel buffer that was used during collection. comp_mmap_len size is used on loading stage as the size of decomp buffer for decompression of COMPRESSED events content.
--- .../Documentation/perf.data-file-format.txt | 24 ++++++++ tools/perf/builtin-record.c | 8 +++ tools/perf/perf.h | 1 + tools/perf/util/env.h | 10 ++++ tools/perf/util/event.c | 1 + tools/perf/util/event.h | 7 +++ tools/perf/util/header.c | 55 ++++++++++++++++++- tools/perf/util/header.h | 1 + 8 files changed, 106 insertions(+), 1 deletion(-)
tools/perf/Documentation/perf.data-file-format.txt +24 -0
@@ -272,6 +272,19 @@ struct { Two uint64_t for the time of first sample and the time of last sample. + HEADER_COMPRESSED = 24, + +struct { + u32 version; + u32 type; + u32 level; + u32 ratio; + u32 mmap_len; +}; + +Indicates that trace contains records of PERF_RECORD_COMPRESSED type +that have perf_events records in compressed form. + other bits are reserved and should ignored for now HEADER_FEAT_BITS = 256,
@@ -437,6 +450,17 @@ struct auxtrace_error_event { Describes a header feature. These are records used in pipe-mode that contain information that otherwise would be in perf.data file's header. + PERF_RECORD_COMPRESSED = 81, + +struct compressed_event { + struct perf_event_header header; + char data[]; +}; + +The header is followed by compressed data frame that can be decompressed +into array of perf trace records. The size of the entire compressed event +record including the header is limited by the max value of header.size. + Event types Define the event attributes with their IDs.
tools/perf/builtin-record.c +8 -0
@@ -369,6 +369,11 @@ static int record__mmap_flush_parse(const struct option *opt,return 0 ; } +static int record__comp_enabled (struct record *rec) +{ +return rec->opts.comp_level >0 ; +} + static int process_synthesized_event(struct perf_tool *tool, union perf_event *event, struct perf_sample *sample __maybe_unused,
@@ -885,6 +890,8 @@ static void record__init_features(struct record *rec)perf_header__clear_feat (&session->header,HEADER_CLOCKID );perf_header__clear_feat (&session->header,HEADER_DIR_FORMAT ); +if (!record__comp_enabled (rec)) +perf_header__clear_feat (&session->header,HEADER_COMPRESSED );perf_header__clear_feat (&session->header,HEADER_STAT ); }
@@ -1225,6 +1232,7 @@ static int __cmd_record(struct record *rec, int argc, const char **argv) err =-1 ; goto out_child; } + session->header.env.comp_mmap_len = session->evlist->mmap_len; err =bpf__apply_obj_config ();if (err) {
tools/perf/perf.h +1 -0
@@ -86,6 +86,7 @@ struct record_opts {int nr_cblocks;int affinity;int mmap_flush; +unsigned int comp_level; };enum perf_affinity {
tools/perf/util/env.h +10 -0
@@ -65,6 +65,16 @@ struct perf_env {unsigned long long memory_bsize;u64 clockid_res_ns;u32 comp_ratio; +u32 comp_ver; +u32 comp_type; +u32 comp_level; +u32 comp_mmap_len; +}; + +enum perf_compress_type { +PERF_COMP_NONE =0 , +PERF_COMP_ZSTD , +PERF_COMP_MAX };extern struct perf_env perf_env;
tools/perf/util/event.c +1 -0
@@ -68,6 +68,7 @@ static const char *perf_event__names[] = { [PERF_RECORD_EVENT_UPDATE] = "EVENT_UPDATE", [PERF_RECORD_TIME_CONV] = "TIME_CONV", [PERF_RECORD_HEADER_FEATURE] = "FEATURE", + [PERF_RECORD_COMPRESSED] = "COMPRESSED", };static const char *perf_ns__names[] = {
tools/perf/util/event.h +7 -0
@@ -254,6 +254,7 @@ enum perf_user_event_type { /* above any possible kernel type */PERF_RECORD_EVENT_UPDATE =78 ,PERF_RECORD_TIME_CONV =79 ,PERF_RECORD_HEADER_FEATURE =80 , +PERF_RECORD_COMPRESSED =81 , PERF_RECORD_HEADER_MAX };
@@ -626,6 +627,11 @@ struct feature_event {char data[]; }; +struct compressed_event { +struct perf_event_header header; +char data[]; +}; +union perf_event {struct perf_event_header header;struct mmap_event mmap;
@@ -659,6 +665,7 @@ union perf_event {struct feature_event feat;struct ksymbol_event ksymbol_event;struct bpf_event bpf_event; +struct compressed_event pack; };void perf_event__print_totals (void );
tools/perf/util/header.c +54 -1
79 lines not on this page · raw message
tools/perf/util/header.h +1 -0
@@ -40,6 +40,7 @@ enum {HEADER_MEM_TOPOLOGY ,HEADER_CLOCKID ,HEADER_DIR_FORMAT , +HEADER_COMPRESSED ,HEADER_LAST_FEATURE , HEADER_FEAT_BITS = 256, };
Signature · 1 line
2.20.1
tip-bot for Alexey Budankov
merge bot[tip:perf/core] perf record: Implement COMPRESSED event record and its attributes
perf record: Implement COMPRESSED event record and its attributes
Commit-ID: 42e1fd80a5b8bf9188ddb502b788433ece189aae Gitweb: https://git.kernel.org/tip/42e1fd80a5b8bf9188ddb502b788433ece189aae Author: Alexey Budankov <[email protected]> AuthorDate: Mon, 18 Mar 2019 20:41:33 +0300 Committer: Arnaldo Carvalho de Melo <[email protected]> CommitDate: Wed, 15 May 2019 16:36:49 -0300
perf record: Implement COMPRESSED event record and its attributes
Implemented PERF_RECORD_COMPRESSED event, related data types, header feature and functions to write, read and print feature attributes from the trace header section.
comp_mmap_len preserves the size of mmaped kernel buffer that was used during collection. comp_mmap_len size is used on loading stage as the size of decomp buffer for decompression of COMPRESSED events content.
Committer notes:
Fixed up conflict with BPF_PROG_INFO and BTF_BTF header features.
--- tools/perf/Documentation/perf.data-file-format.txt | 24 ++++++++++ tools/perf/builtin-record.c | 8 ++++ tools/perf/perf.h | 1 + tools/perf/util/env.h | 10 ++++ tools/perf/util/event.c | 1 + tools/perf/util/event.h | 7 +++ tools/perf/util/header.c | 53 ++++++++++++++++++++++ tools/perf/util/header.h | 1 + 8 files changed, 105 insertions(+)
tools/perf/Documentation/perf.data-file-format.txt +24 -0
@@ -272,6 +272,19 @@ struct { Two uint64_t for the time of first sample and the time of last sample. + HEADER_COMPRESSED = 27, + +struct { + u32 version; + u32 type; + u32 level; + u32 ratio; + u32 mmap_len; +}; + +Indicates that trace contains records of PERF_RECORD_COMPRESSED type +that have perf_events records in compressed form. + other bits are reserved and should ignored for now HEADER_FEAT_BITS = 256,
@@ -437,6 +450,17 @@ struct auxtrace_error_event { Describes a header feature. These are records used in pipe-mode that contain information that otherwise would be in perf.data file's header. + PERF_RECORD_COMPRESSED = 81, + +struct compressed_event { + struct perf_event_header header; + char data[]; +}; + +The header is followed by compressed data frame that can be decompressed +into array of perf trace records. The size of the entire compressed event +record including the header is limited by the max value of header.size. + Event types Define the event attributes with their IDs.
tools/perf/builtin-record.c +8 -0
@@ -372,6 +372,11 @@ static int record__mmap_flush_parse(const struct option *opt,return 0 ; } +static int record__comp_enabled (struct record *rec) +{ +return rec->opts.comp_level >0 ; +} + static int process_synthesized_event(struct perf_tool *tool, union perf_event *event, struct perf_sample *sample __maybe_unused,
@@ -888,6 +893,8 @@ static void record__init_features(struct record *rec)perf_header__clear_feat (&session->header,HEADER_CLOCKID );perf_header__clear_feat (&session->header,HEADER_DIR_FORMAT ); +if (!record__comp_enabled (rec)) +perf_header__clear_feat (&session->header,HEADER_COMPRESSED );perf_header__clear_feat (&session->header,HEADER_STAT ); }
@@ -1245,6 +1252,7 @@ static int __cmd_record(struct record *rec, int argc, const char **argv) err =-1 ; goto out_child; } + session->header.env.comp_mmap_len = session->evlist->mmap_len; err =bpf__apply_obj_config ();if (err) {
tools/perf/perf.h +1 -0
@@ -86,6 +86,7 @@ struct record_opts {int nr_cblocks;int affinity;int mmap_flush; +unsigned int comp_level; };enum perf_affinity {
tools/perf/util/env.h +10 -0
@@ -63,6 +63,10 @@ struct perf_env {struct cpu_cache_level *caches;int caches_cnt;u32 comp_ratio; +u32 comp_ver; +u32 comp_type; +u32 comp_level; +u32 comp_mmap_len;struct numa_node *numa_nodes;struct memory_node *memory_nodes;unsigned long long memory_bsize;
@@ -81,6 +85,12 @@ struct perf_env { } bpf_progs; }; +enum perf_compress_type { +PERF_COMP_NONE =0 , +PERF_COMP_ZSTD , +PERF_COMP_MAX +}; +struct bpf_prog_info_node ;struct btf_node ;
tools/perf/util/event.c +1 -0
@@ -68,6 +68,7 @@ static const char *perf_event__names[] = { [PERF_RECORD_EVENT_UPDATE] = "EVENT_UPDATE", [PERF_RECORD_TIME_CONV] = "TIME_CONV", [PERF_RECORD_HEADER_FEATURE] = "FEATURE", + [PERF_RECORD_COMPRESSED] = "COMPRESSED", };static const char *perf_ns__names[] = {
tools/perf/util/event.h +7 -0
@@ -255,6 +255,7 @@ enum perf_user_event_type { /* above any possible kernel type */PERF_RECORD_EVENT_UPDATE =78 ,PERF_RECORD_TIME_CONV =79 ,PERF_RECORD_HEADER_FEATURE =80 , +PERF_RECORD_COMPRESSED =81 , PERF_RECORD_HEADER_MAX };
@@ -627,6 +628,11 @@ struct feature_event {char data[]; }; +struct compressed_event { +struct perf_event_header header; +char data[]; +}; +union perf_event {struct perf_event_header header;struct mmap_event mmap;
@@ -660,6 +666,7 @@ union perf_event {struct feature_event feat;struct ksymbol_event ksymbol_event;struct bpf_event bpf_event; +struct compressed_event pack; };void perf_event__print_totals (void );
tools/perf/util/header.c +53 -0
77 lines not on this page · raw message
tools/perf/util/header.h +1 -0
@@ -42,6 +42,7 @@ enum {HEADER_DIR_FORMAT ,HEADER_BPF_PROG_INFO ,HEADER_BPF_BTF , +HEADER_COMPRESSED ,HEADER_LAST_FEATURE , HEADER_FEAT_BITS = 256, };
PATCH v10 5/12 perf mmap: implement dedicated memory buffer for data compression
In reply to Alexey BudankovImplemented mmap data buffer that is used as the memory to operate on when compressing data in case of serial trace streaming.
--- tools/perf/builtin-record.c | 8 +++++++- tools/perf/util/evlist.c | 8 +++++--- tools/perf/util/evlist.h | 2 +- tools/perf/util/mmap.c | 30 ++++++++++++++++++++++++++++-- tools/perf/util/mmap.h | 4 +++- 5 files changed, 44 insertions(+), 8 deletions(-)
tools/perf/builtin-record.c +7 -1
@@ -369,6 +369,8 @@ static int record__mmap_flush_parse(const struct option *opt,return 0 ; } +static unsigned int comp_level_max =22 ; +static int record__comp_enabled (struct record *rec) {return rec->opts.comp_level >0 ;
@@ -584,7 +586,7 @@ static int record__mmap_evlist(struct record *rec, opts->auxtrace_mmap_pages, opts->auxtrace_snapshot_mode, opts->nr_cblocks, opts->affinity, - opts->mmap_flush) < 0) { + opts->mmap_flush, opts->comp_level) < 0) {if (errno ==EPERM ) {pr_err ("Permission error mapping pages.\n" "Consider increasing "
@@ -2258,6 +2260,10 @@ int cmd_record(int argc, const char **argv)pr_debug ("affinity: %s\n" , affinity_tags[rec->opts.affinity]);pr_debug ("mmap flush: %d\n" , rec->opts.mmap_flush); +if (rec->opts.comp_level > comp_level_max) + rec->opts.comp_level = comp_level_max; +pr_debug ("comp level: %d\n" , rec->opts.comp_level); + err =__cmd_record (&record, argc, argv); out:perf_evlist__delete (rec->evlist);
tools/perf/util/evlist.c +5 -3
@@ -1037,7 +1037,8 @@ int perf_evlist__parse_mmap_pages(const struct option *opt, const char *str,*/ int perf_evlist__mmap_ex (struct perf_evlist *evlist,unsigned int pages,unsigned int auxtrace_pages, -bool auxtrace_overwrite,int nr_cblocks,int affinity,int flush) +bool auxtrace_overwrite,int nr_cblocks,int affinity,int flush, +int comp_level) {struct perf_evsel *evsel;const struct cpu_map *cpus = evlist->cpus;
@@ -1047,7 +1048,8 @@ int perf_evlist__mmap_ex(struct perf_evlist *evlist, unsigned int pages,* Its value is decided by evsel's write_backward. * So &mp should not be passed through const pointer. */ -struct mmap_params mp = { .nr_cblocks = nr_cblocks, .affinity = affinity, .flush = flush }; +struct mmap_params mp = { .nr_cblocks = nr_cblocks, .affinity = affinity, .flush = flush, + .comp_level = comp_level };if (!evlist->mmap) evlist->mmap =perf_evlist__alloc_mmap (evlist, false);
@@ -1079,7 +1081,7 @@ int perf_evlist__mmap_ex(struct perf_evlist *evlist, unsigned int pages,int perf_evlist__mmap (struct perf_evlist *evlist,unsigned int pages) { -return perf_evlist__mmap_ex (evlist, pages,0 , false,0 ,PERF_AFFINITY_SYS ,1 ); +return perf_evlist__mmap_ex (evlist, pages,0 , false,0 ,PERF_AFFINITY_SYS ,1 ,0 ); }int perf_evlist__create_maps (struct perf_evlist *evlist,struct target *target)
tools/perf/util/evlist.h +1 -1
@@ -166,7 +166,7 @@ unsigned long perf_event_mlock_kb_in_pages(void);int perf_evlist__mmap_ex (struct perf_evlist *evlist,unsigned int pages,unsigned int auxtrace_pages,bool auxtrace_overwrite,int nr_cblocks, -int affinity,int flush); +int affinity,int flush,int comp_level);int perf_evlist__mmap (struct perf_evlist *evlist,unsigned int pages);void perf_evlist__munmap (struct perf_evlist *evlist);
tools/perf/util/mmap.c +28 -2
@@ -157,6 +157,10 @@ void __weak auxtrace_mmap_params__set_idx(struct auxtrace_mmap_params *mp __mayb }#ifdef HAVE_AIO_SUPPORT +static int perf_mmap__aio_enabled (struct perf_mmap *map) +{ +return map->aio.nr_cblocks >0 ; +}#ifdef HAVE_LIBNUMA_SUPPORT static int perf_mmap__aio_alloc (struct perf_mmap *map,int idx)
@@ -198,7 +202,7 @@ static int perf_mmap__aio_bind(struct perf_mmap *map, int idx, int cpu, int affi return 0; } -#else +#else /* !HAVE_LIBNUMA_SUPPORT */ static int perf_mmap__aio_alloc (struct perf_mmap *map,int idx) { map->aio.data[idx] =malloc (perf_mmap__mmap_len (map));
@@ -359,7 +363,12 @@ int perf_mmap__aio_push(struct perf_mmap *md, void *to, int idx, return rc; } -#else +#else /* !HAVE_AIO_SUPPORT */ +static int perf_mmap__aio_enabled (struct perf_mmap *map__maybe_unused ) +{ +return 0 ; +} +static int perf_mmap__aio_mmap (struct perf_mmap *map__maybe_unused ,struct mmap_params *mp__maybe_unused ) {
@@ -374,6 +383,10 @@ static void perf_mmap__aio_munmap(struct perf_mmap *map __maybe_unused)void perf_mmap__munmap (struct perf_mmap *map) {perf_mmap__aio_munmap (map); +if (map->data !=NULL ) { +munmap (map->data,perf_mmap__mmap_len (map)); + map->data =NULL ; + }if (map->base !=NULL ) {munmap (map->base,perf_mmap__mmap_len (map)); map->base =NULL ;
@@ -442,6 +455,19 @@ int perf_mmap__mmap(struct perf_mmap *map, struct mmap_params *mp, int fd, int c map->flush = mp->flush; + map->comp_level = mp->comp_level; + +if (map->comp_level && !perf_mmap__aio_enabled (map)) { + map->data =mmap (NULL ,perf_mmap__mmap_len (map),PROT_READ |PROT_WRITE , +MAP_PRIVATE |MAP_ANONYMOUS ,0 ,0 ); +if (map->data ==MAP_FAILED ) { +pr_debug2 ("failed to mmap data buffer, error %d\n" , + errno); + map->data =NULL ; +return -1 ; + } + } +if (auxtrace_mmap__mmap (&map->auxtrace_mmap, &mp->auxtrace_mp, map->base, fd))return -1 ;
tools/perf/util/mmap.h +3 -1
@@ -40,6 +40,8 @@ struct perf_mmap {#endif cpu_set_t affinity_mask;u64 flush; +void *data; +int comp_level; };/*
@@ -71,7 +73,7 @@ enum bkw_mmap_state { };struct mmap_params { -int prot, mask, nr_cblocks, affinity, flush; +int prot, mask, nr_cblocks, affinity, flush, comp_level;struct auxtrace_mmap_params auxtrace_mp; };
Signature · 1 line
2.20.1
tip-bot for Alexey Budankov
merge bot[tip:perf/core] perf mmap: Implement dedicated memory buffer for data compression
perf mmap: Implement dedicated memory buffer for data compression
Commit-ID: 51255a8af7c41c876c2d715a35ab03c13302a607 Gitweb: https://git.kernel.org/tip/51255a8af7c41c876c2d715a35ab03c13302a607 Author: Alexey Budankov <[email protected]> AuthorDate: Mon, 18 Mar 2019 20:42:19 +0300 Committer: Arnaldo Carvalho de Melo <[email protected]> CommitDate: Wed, 15 May 2019 16:36:49 -0300
perf mmap: Implement dedicated memory buffer for data compression
Implemented mmap data buffer that is used as the memory to operate on when compressing data in case of serial trace streaming.
--- tools/perf/builtin-record.c | 8 +++++++- tools/perf/util/evlist.c | 8 +++++--- tools/perf/util/evlist.h | 2 +- tools/perf/util/mmap.c | 30 ++++++++++++++++++++++++++++-- tools/perf/util/mmap.h | 4 +++- 5 files changed, 44 insertions(+), 8 deletions(-)
tools/perf/builtin-record.c +7 -1
@@ -372,6 +372,8 @@ static int record__mmap_flush_parse(const struct option *opt,return 0 ; } +static unsigned int comp_level_max =22 ; +static int record__comp_enabled (struct record *rec) {return rec->opts.comp_level >0 ;
@@ -587,7 +589,7 @@ static int record__mmap_evlist(struct record *rec, opts->auxtrace_mmap_pages, opts->auxtrace_snapshot_mode, opts->nr_cblocks, opts->affinity, - opts->mmap_flush) < 0) { + opts->mmap_flush, opts->comp_level) < 0) {if (errno ==EPERM ) {pr_err ("Permission error mapping pages.\n" "Consider increasing "
@@ -2298,6 +2300,10 @@ int cmd_record(int argc, const char **argv)pr_debug ("affinity: %s\n" , affinity_tags[rec->opts.affinity]);pr_debug ("mmap flush: %d\n" , rec->opts.mmap_flush); +if (rec->opts.comp_level > comp_level_max) + rec->opts.comp_level = comp_level_max; +pr_debug ("comp level: %d\n" , rec->opts.comp_level); + err =__cmd_record (&record, argc, argv); out:perf_evlist__delete (rec->evlist);
tools/perf/util/evlist.c +5 -3
@@ -1009,7 +1009,8 @@ int perf_evlist__parse_mmap_pages(const struct option *opt, const char *str,*/ int perf_evlist__mmap_ex (struct perf_evlist *evlist,unsigned int pages,unsigned int auxtrace_pages, -bool auxtrace_overwrite,int nr_cblocks,int affinity,int flush) +bool auxtrace_overwrite,int nr_cblocks,int affinity,int flush, +int comp_level) {struct perf_evsel *evsel;const struct cpu_map *cpus = evlist->cpus;
@@ -1019,7 +1020,8 @@ int perf_evlist__mmap_ex(struct perf_evlist *evlist, unsigned int pages,* Its value is decided by evsel's write_backward. * So &mp should not be passed through const pointer. */ -struct mmap_params mp = { .nr_cblocks = nr_cblocks, .affinity = affinity, .flush = flush }; +struct mmap_params mp = { .nr_cblocks = nr_cblocks, .affinity = affinity, .flush = flush, + .comp_level = comp_level };if (!evlist->mmap) evlist->mmap =perf_evlist__alloc_mmap (evlist, false);
@@ -1051,7 +1053,7 @@ int perf_evlist__mmap_ex(struct perf_evlist *evlist, unsigned int pages,int perf_evlist__mmap (struct perf_evlist *evlist,unsigned int pages) { -return perf_evlist__mmap_ex (evlist, pages,0 , false,0 ,PERF_AFFINITY_SYS ,1 ); +return perf_evlist__mmap_ex (evlist, pages,0 , false,0 ,PERF_AFFINITY_SYS ,1 ,0 ); }int perf_evlist__create_maps (struct perf_evlist *evlist,struct target *target)
tools/perf/util/evlist.h +1 -1
@@ -178,7 +178,7 @@ unsigned long perf_event_mlock_kb_in_pages(void);int perf_evlist__mmap_ex (struct perf_evlist *evlist,unsigned int pages,unsigned int auxtrace_pages,bool auxtrace_overwrite,int nr_cblocks, -int affinity,int flush); +int affinity,int flush,int comp_level);int perf_evlist__mmap (struct perf_evlist *evlist,unsigned int pages);void perf_evlist__munmap (struct perf_evlist *evlist);
tools/perf/util/mmap.c +28 -2
@@ -157,6 +157,10 @@ void __weak auxtrace_mmap_params__set_idx(struct auxtrace_mmap_params *mp __mayb }#ifdef HAVE_AIO_SUPPORT +static int perf_mmap__aio_enabled (struct perf_mmap *map) +{ +return map->aio.nr_cblocks >0 ; +}#ifdef HAVE_LIBNUMA_SUPPORT static int perf_mmap__aio_alloc (struct perf_mmap *map,int idx)
@@ -198,7 +202,7 @@ static int perf_mmap__aio_bind(struct perf_mmap *map, int idx, int cpu, int affi return 0; } -#else +#else /* !HAVE_LIBNUMA_SUPPORT */ static int perf_mmap__aio_alloc (struct perf_mmap *map,int idx) { map->aio.data[idx] =malloc (perf_mmap__mmap_len (map));
@@ -359,7 +363,12 @@ int perf_mmap__aio_push(struct perf_mmap *md, void *to, int idx, return rc; } -#else +#else /* !HAVE_AIO_SUPPORT */ +static int perf_mmap__aio_enabled (struct perf_mmap *map__maybe_unused ) +{ +return 0 ; +} +static int perf_mmap__aio_mmap (struct perf_mmap *map__maybe_unused ,struct mmap_params *mp__maybe_unused ) {
@@ -374,6 +383,10 @@ static void perf_mmap__aio_munmap(struct perf_mmap *map __maybe_unused)void perf_mmap__munmap (struct perf_mmap *map) {perf_mmap__aio_munmap (map); +if (map->data !=NULL ) { +munmap (map->data,perf_mmap__mmap_len (map)); + map->data =NULL ; + }if (map->base !=NULL ) {munmap (map->base,perf_mmap__mmap_len (map)); map->base =NULL ;
@@ -442,6 +455,19 @@ int perf_mmap__mmap(struct perf_mmap *map, struct mmap_params *mp, int fd, int c map->flush = mp->flush; + map->comp_level = mp->comp_level; + +if (map->comp_level && !perf_mmap__aio_enabled (map)) { + map->data =mmap (NULL ,perf_mmap__mmap_len (map),PROT_READ |PROT_WRITE , +MAP_PRIVATE |MAP_ANONYMOUS ,0 ,0 ); +if (map->data ==MAP_FAILED ) { +pr_debug2 ("failed to mmap data buffer, error %d\n" , + errno); + map->data =NULL ; +return -1 ; + } + } +if (auxtrace_mmap__mmap (&map->auxtrace_mmap, &mp->auxtrace_mp, map->base, fd))return -1 ;
tools/perf/util/mmap.h +3 -1
@@ -40,6 +40,8 @@ struct perf_mmap {#endif cpu_set_t affinity_mask;u64 flush; +void *data; +int comp_level; };/*
@@ -71,7 +73,7 @@ enum bkw_mmap_state { };struct mmap_params { -int prot, mask, nr_cblocks, affinity, flush; +int prot, mask, nr_cblocks, affinity, flush, comp_level;struct auxtrace_mmap_params auxtrace_mp; };
PATCH v10 6/12 perf util: introduce Zstd streaming based compression API
In reply to Alexey BudankovImplemented functions are based on Zstd streaming compression API. The functions are used in runtime to compress data that come from mmaped kernel buffer. zstd_init(), zstd_fini() are used for initialization and finalization to allocate and deallocate internal zstd objects. zstd_compress_stream_to_records() is used to convert parts of mmaped kernel buffer into an array of PERF_RECORD_COMPRESSED records.
--- tools/perf/util/Build | 2 ++ tools/perf/util/compress.h | 43 +++++++++++++++++++++++ tools/perf/util/zstd.c | 71 ++++++++++++++++++++++++++++++++++++++ 3 files changed, 116 insertions(+) create mode 100644 tools/perf/util/zstd.c
tools/perf/util/Build +2 -0
@@ -145,6 +145,8 @@ perf-y += scripting-engines/ perf-$(CONFIG_ZLIB) += zlib.o perf-$(CONFIG_LZMA) += lzma.o +perf-$(CONFIG_ZSTD) += zstd.o + perf-y += demangle-java.o perf-y += demangle-rust.o
tools/perf/util/compress.h +43 -0
53 lines not on this page · raw message
tools/perf/util/zstd.c new file +71 -0
71 lines not on this page · raw message
Signature · 1 line
2.20.1
tip-bot for Alexey Budankov
merge bot[tip:perf/core] perf tools: Introduce Zstd streaming based compression API
perf tools: Introduce Zstd streaming based compression API
Commit-ID: f24c1d7523e6db26ec2115a308750c875927741b Gitweb: https://git.kernel.org/tip/f24c1d7523e6db26ec2115a308750c875927741b Author: Alexey Budankov <[email protected]> AuthorDate: Mon, 18 Mar 2019 20:42:55 +0300 Committer: Arnaldo Carvalho de Melo <[email protected]> CommitDate: Wed, 15 May 2019 16:36:49 -0300
perf tools: Introduce Zstd streaming based compression API
Implemented functions are based on Zstd streaming compression API.
The functions are used in runtime to compress data that come from mmaped kernel buffer. zstd_init(), zstd_fini() are used for initialization and finalization to allocate and deallocate internal zstd objects. zstd_compress_stream_to_records() is used to convert parts of mmaped kernel buffer into an array of PERF_RECORD_COMPRESSED records.
--- tools/perf/util/Build | 2 ++ tools/perf/util/compress.h | 42 ++++++++++++++++++++++++++++ tools/perf/util/zstd.c | 70 ++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 114 insertions(+)
tools/perf/util/Build +2 -0
@@ -145,6 +145,8 @@ perf-y += scripting-engines/ perf-$(CONFIG_ZLIB) += zlib.o perf-$(CONFIG_LZMA) += lzma.o +perf-$(CONFIG_ZSTD) += zstd.o + perf-y += demangle-java.o perf-y += demangle-rust.o
tools/perf/util/compress.h +42 -0
52 lines not on this page · raw message
tools/perf/util/zstd.c new file +70 -0
70 lines not on this page · raw message
PATCH v10 7/12 perf record: implement compression for serial trace streaming
In reply to Alexey BudankovCompression is implemented using the functions from zstd.c. As the memory to operate on the compression uses mmap->data buffer. If Zstd streaming compression API fails for some reason the data to be compressed are just copied into the memory buffers using plain memcpy().
Compressed trace frame consists of an array of PERF_RECORD_COMPRESSED records. Each element of the array is not longer that PERF_SAMPLE_MAX_SIZE and consists of perf_event_header followed by the compressed chunk that is decompressed on the loading stage.
--- tools/perf/builtin-record.c | 53 +++++++++++++++++++++++++++++++++++-- tools/perf/util/session.h | 2 ++ 2 files changed, 53 insertions(+), 2 deletions(-)
tools/perf/builtin-record.c +51 -2
@@ -130,6 +130,9 @@ static int record__write(struct record *rec, struct perf_mmap *map __maybe_unusereturn 0 ; } +static size_t zstd_compress (struct perf_session *session,void *dst,size_t dst_size, +void *src,size_t src_size); +#ifdef HAVE_AIO_SUPPORT static int record__aio_write (struct aiocb *cblock,int trace_fd,void *buf,size_t size,off_t off)
@@ -389,6 +392,12 @@ static int record__pushfn(struct perf_mmap *map, void *to, void *bf, size_t size {struct record *rec = to; +if (record__comp_enabled (rec)) { + size =zstd_compress (rec->session, map->data, +perf_mmap__mmap_len (map), bf, size); + bf = map->data; + } + rec->samples++;return record__write (rec, map, bf, size); }
@@ -775,6 +784,38 @@ static void record__adjust_affinity(struct record *rec, struct perf_mmap *map) } } +static size_t process_comp_header (void *record,size_t increment) +{ +struct compressed_event *event = record; +size_t size =sizeof (struct compressed_event ); + +if (increment) { + event->header.size += increment; +return increment; + } + + event->header.type =PERF_RECORD_COMPRESSED ; + event->header.size = size; + +return size; +} + +static size_t zstd_compress (struct perf_session *session,void *dst,size_t dst_size, +void *src,size_t src_size) +{ +size_t compressed; +size_t max_record_size =PERF_SAMPLE_MAX_SIZE -sizeof (struct compressed_event ) -1 ; + + compressed =zstd_compress_stream_to_records (&(session->zstd_data), + dst, dst_size, src, src_size, max_record_size, + process_comp_header); + + session->bytes_transferred += src_size; + session->bytes_compressed += compressed; + +return compressed; +} +static int record__mmap_read_evlist (struct record *rec,struct perf_evlist *evlist,bool overwrite,bool sync) {
@@ -1205,6 +1246,14 @@ static int __cmd_record(struct record *rec, int argc, const char **argv) fd =perf_data__fd (data); rec->session = session; +if (zstd_init (&(session->zstd_data), rec->opts.comp_level) <0 ) { +pr_err ("Compression initialization failed.\n" ); +return -1 ; + } + + session->header.env.comp_type =PERF_COMP_ZSTD ; + session->header.env.comp_level = rec->opts.comp_level; +record__init_features (rec);if (rec->opts.use_clockid && rec->opts.clockid_res_ns)
@@ -1537,6 +1586,7 @@ static int __cmd_record(struct record *rec, int argc, const char **argv) } out_delete_session: +zstd_fini (&(session->zstd_data));perf_session__delete (session);return status; }
@@ -2254,8 +2304,7 @@ int cmd_record(int argc, const char **argv)if (rec->opts.nr_cblocks > nr_cblocks_max) rec->opts.nr_cblocks = nr_cblocks_max; -if (verbose >0 ) -pr_info ("nr_cblocks: %d\n" , rec->opts.nr_cblocks); +pr_debug ("nr_cblocks: %d\n" , rec->opts.nr_cblocks);pr_debug ("affinity: %s\n" , affinity_tags[rec->opts.affinity]);pr_debug ("mmap flush: %d\n" , rec->opts.mmap_flush);
tools/perf/util/session.h +2 -0
@@ -8,6 +8,7 @@#include "machine.h" #include "data.h" #include "ordered-events.h" +#include "util/compress.h" #include <linux/kernel.h> #include <linux/rbtree.h> #include <linux/perf_event.h>
@@ -37,6 +38,7 @@ struct perf_session {struct perf_tool *tool;u64 bytes_transferred;u64 bytes_compressed; +struct zstd_data zstd_data; };struct perf_tool ;
Signature · 1 line
2.20.1
tip-bot for Alexey Budankov
merge bot[tip:perf/core] perf record: Implement compression for serial trace streaming
perf record: Implement compression for serial trace streaming
Commit-ID: 5d7f41164930ecc1797702b7f9728ac702609ef3 Gitweb: https://git.kernel.org/tip/5d7f41164930ecc1797702b7f9728ac702609ef3 Author: Alexey Budankov <[email protected]> AuthorDate: Mon, 18 Mar 2019 20:43:35 +0300 Committer: Arnaldo Carvalho de Melo <[email protected]> CommitDate: Wed, 15 May 2019 16:36:49 -0300
perf record: Implement compression for serial trace streaming
Compression is implemented using the functions from zstd.c. As the memory to operate on the compression uses mmap->data buffer.
If Zstd streaming compression API fails for some reason the data to be compressed are just copied into the memory buffers using plain memcpy().
Compressed trace frame consists of an array of PERF_RECORD_COMPRESSED records. Each element of the array is not longer that PERF_SAMPLE_MAX_SIZE and consists of perf_event_header followed by the compressed chunk that is decompressed on the loading stage.
Comitter notes:
Undo some unnecessary line breaks, remove some unnecessary () around zstd_data to then just get its address, and fix conflicts with BPF_PROG_INFO/BPF_BTF patchkits.
--- tools/perf/builtin-record.c | 51 +++++++++++++++++++++++++++++++++++++++++++-- tools/perf/util/session.h | 2 ++ 2 files changed, 51 insertions(+), 2 deletions(-)
tools/perf/builtin-record.c +49 -2
@@ -133,6 +133,9 @@ static int record__write(struct record *rec, struct perf_mmap *map __maybe_unusereturn 0 ; } +static size_t zstd_compress (struct perf_session *session,void *dst,size_t dst_size, +void *src,size_t src_size); +#ifdef HAVE_AIO_SUPPORT static int record__aio_write (struct aiocb *cblock,int trace_fd,void *buf,size_t size,off_t off)
@@ -392,6 +395,11 @@ static int record__pushfn(struct perf_mmap *map, void *to, void *bf, size_t size {struct record *rec = to; +if (record__comp_enabled (rec)) { + size =zstd_compress (rec->session, map->data,perf_mmap__mmap_len (map), bf, size); + bf = map->data; + } + rec->samples++;return record__write (rec, map, bf, size); }
@@ -778,6 +786,37 @@ static void record__adjust_affinity(struct record *rec, struct perf_mmap *map) } } +static size_t process_comp_header (void *record,size_t increment) +{ +struct compressed_event *event = record; +size_t size =sizeof (*event); + +if (increment) { + event->header.size += increment; +return increment; + } + + event->header.type =PERF_RECORD_COMPRESSED ; + event->header.size = size; + +return size; +} + +static size_t zstd_compress (struct perf_session *session,void *dst,size_t dst_size, +void *src,size_t src_size) +{ +size_t compressed; +size_t max_record_size =PERF_SAMPLE_MAX_SIZE -sizeof (struct compressed_event ) -1 ; + + compressed =zstd_compress_stream_to_records (&session->zstd_data, dst, dst_size, src, src_size, + max_record_size, process_comp_header); + + session->bytes_transferred += src_size; + session->bytes_compressed += compressed; + +return compressed; +} +static int record__mmap_read_evlist (struct record *rec,struct perf_evlist *evlist,bool overwrite,bool synch) {
@@ -1225,6 +1264,14 @@ static int __cmd_record(struct record *rec, int argc, const char **argv) fd =perf_data__fd (data); rec->session = session; +if (zstd_init (&session->zstd_data, rec->opts.comp_level) <0 ) { +pr_err ("Compression initialization failed.\n" ); +return -1 ; + } + + session->header.env.comp_type =PERF_COMP_ZSTD ; + session->header.env.comp_level = rec->opts.comp_level; +record__init_features (rec);if (rec->opts.use_clockid && rec->opts.clockid_res_ns)
@@ -1565,6 +1612,7 @@ out_child: } out_delete_session: +zstd_fini (&session->zstd_data);perf_session__delete (session);if (!opts->no_bpf_event)
@@ -2294,8 +2342,7 @@ int cmd_record(int argc, const char **argv)if (rec->opts.nr_cblocks > nr_cblocks_max) rec->opts.nr_cblocks = nr_cblocks_max; -if (verbose >0 ) -pr_info ("nr_cblocks: %d\n" , rec->opts.nr_cblocks); +pr_debug ("nr_cblocks: %d\n" , rec->opts.nr_cblocks);pr_debug ("affinity: %s\n" , affinity_tags[rec->opts.affinity]);pr_debug ("mmap flush: %d\n" , rec->opts.mmap_flush);
tools/perf/util/session.h +2 -0
@@ -8,6 +8,7 @@#include "machine.h" #include "data.h" #include "ordered-events.h" +#include "util/compress.h" #include <linux/kernel.h> #include <linux/rbtree.h> #include <linux/perf_event.h>
@@ -37,6 +38,7 @@ struct perf_session {struct perf_tool *tool;u64 bytes_transferred;u64 bytes_compressed; +struct zstd_data zstd_data; };struct perf_tool ;
PATCH v10 8/12 perf record: implement compression for AIO trace streaming
In reply to Alexey BudankovCompression is implemented using the functions from zstd.c. As the memory to operate on the compression uses mmap->aio.data[] buffers. If Zstd streaming compression API fails for some reason the data to be compressed are just copied into the memory buffers using plain memcpy().
Compressed trace frame consists of an array of PERF_RECORD_COMPRESSED records. Each element of the array is not longer that PERF_SAMPLE_MAX_SIZE and consists of perf_event_header followed by the compressed chunk that is decompressed on the loading stage.
perf_mmap__aio_push() is replaced by perf_mmap__push() which is now used in the both serial and AIO streaming cases. perf_mmap__push() is extended with positive return values to signify absence of data ready for processing.
--- tools/perf/builtin-record.c | 114 ++++++++++++++++++++++++++++-------- tools/perf/util/mmap.c | 76 +----------------------- tools/perf/util/mmap.h | 12 ---- 3 files changed, 89 insertions(+), 113 deletions(-)
tools/perf/builtin-record.c +88 -26
@@ -130,6 +130,8 @@ static int record__write(struct record *rec, struct perf_mmap *map __maybe_unusereturn 0 ; } +static int record__aio_enabled (struct record *rec); +static int record__comp_enabled (struct record *rec);static size_t zstd_compress (struct perf_session *session,void *dst,size_t dst_size,void *src,size_t src_size);
@@ -183,9 +185,9 @@ static int record__aio_complete(struct perf_mmap *md, struct aiocb *cblock)if (rem_size ==0 ) { cblock->aio_fildes =-1 ;/* -* md->refcount is incremented in perf_mmap__push() for -* every enqueued aio write request so decrement it because -* the request is now complete. +* md->refcount is incremented in record__aio_pushfn() for +* every aio write request started in record__aio_push() so +* decrement it because the request is now complete. */ perf_mmap__put (md); rc =1 ;
@@ -240,18 +242,89 @@ static int record__aio_sync(struct perf_mmap *md, bool sync_all) }while (1 ); } -static int record__aio_pushfn (void *to,struct aiocb *cblock,void *bf,size_t size,off_t off) +struct record_aio { +struct record *rec; +void *data; +size_t size; +}; + +static int record__aio_pushfn (struct perf_mmap *map,void *to,void *buf,size_t size) { -struct record *rec = to; -int ret, trace_fd = rec->session->data->file.fd; +struct record_aio *aio = to; - rec->samples++; +/* +* map->base data pointed by buf is copied into free map->aio.data[] buffer +* to release space in the kernel buffer as fast as possible, calling +* perf_mmap__consume() from perf_mmap__push() function. +* +* That lets the kernel to proceed with storing more profiling data into +* the kernel buffer earlier than other per-cpu kernel buffers are handled. +* +* Coping can be done in two steps in case the chunk of profiling data +* crosses the upper bound of the kernel buffer. In this case we first move +* part of data from map->start till the upper bound and then the reminder +* from the beginning of the kernel buffer till the end of the data chunk. +*/ - ret =record__aio_write (cblock, trace_fd, bf, size, off); +if (record__comp_enabled (aio->rec)) { + size =zstd_compress (aio->rec->session, aio->data + aio->size, +perf_mmap__mmap_len (map) - aio->size, + buf, size); + }else { +memcpy (aio->data + aio->size, buf, size); + } + +if (!aio->size) { +/* +* Increment map->refcount to guard map->aio.data[] buffer +* from premature deallocation because map object can be +* released earlier than aio write request started on +* map->aio.data[] buffer is complete. +* +* perf_mmap__put() is done at record__aio_complete() +* after started aio request completion or at record__aio_push() +* if the request failed to start. +*/ +perf_mmap__get (map); + } + + aio->size += size; + +return size; +} + +static int record__aio_push (struct record *rec,struct perf_mmap *map,off_t *off) +{ +int ret, idx; +int trace_fd = rec->session->data->file.fd; +struct record_aio aio = { .rec = rec, .size =0 }; + +/* +* Call record__aio_sync() to wait till map->aio.data[] buffer +* becomes available after previous aio write operation. +*/ + + idx =record__aio_sync (map, false); + aio.data = map->aio.data[idx]; + ret =perf_mmap__push (map, &aio, record__aio_pushfn); +if (ret !=0 )/* ret > 0 - no data, ret < 0 - error */ +return ret; + + rec->samples++; + ret =record__aio_write (&(map->aio.cblocks[idx]), trace_fd, aio.data, aio.size, *off);if (!ret) { - rec->bytes_written += size; + *off += aio.size; + rec->bytes_written += aio.size;if (switch_output_size (rec))trigger_hit (&switch_output_trigger); + }else { +/* +* Decrement map->refcount incremented in record__aio_pushfn() +* back if record__aio_write() operation failed to start, otherwise +* map->refcount is decremented in record__aio_complete() after +* aio write operation finishes successfully. +*/ +perf_mmap__put (map); }return ret;
@@ -273,7 +346,7 @@ static void record__aio_mmap_read_sync(struct record *rec)struct perf_evlist *evlist = rec->evlist;struct perf_mmap *maps = evlist->mmap; -if (!rec->opts.nr_cblocks) +if (!record__aio_enabled (rec))return ;for (i =0 ; i < evlist->nr_mmaps; i++) {
@@ -307,13 +380,8 @@ static int record__aio_parse(const struct option *opt,#else /* HAVE_AIO_SUPPORT */ static int nr_cblocks_max =0 ; -static int record__aio_sync (struct perf_mmap *md__maybe_unused ,bool sync_all__maybe_unused ) -{ -return -1 ; -} - -static int record__aio_pushfn (void *to__maybe_unused ,struct aiocb *cblock__maybe_unused , -void *bf__maybe_unused ,size_t size__maybe_unused ,off_t off__maybe_unused ) +static int record__aio_push (struct record *rec__maybe_unused ,struct perf_mmap *map__maybe_unused , +off_t *off__maybe_unused ) {return -1 ; }
@@ -824,7 +892,7 @@ static int record__mmap_read_evlist(struct record *rec, struct perf_evlist *evliint rc =0 ;struct perf_mmap *maps;int trace_fd = rec->data.file.fd; -off_t off; +off_t off =0 ;if (!evlist)return 0 ;
@@ -850,20 +918,14 @@ static int record__mmap_read_evlist(struct record *rec, struct perf_evlist *evli map->flush =1 ; }if (!record__aio_enabled (rec)) { -if (perf_mmap__push (map, rec, record__pushfn) !=0 ) { +if (perf_mmap__push (map, rec, record__pushfn) <0 ) {if (sync) map->flush = flush; rc =-1 ; goto out; } }else { -int idx; -/* -* Call record__aio_sync() to wait till map->data buffer -* becomes available after previous aio write request. -*/ - idx =record__aio_sync (map, false); -if (perf_mmap__aio_push (map, rec, idx, record__aio_pushfn, &off) !=0 ) { +if (record__aio_push (rec, map, &off) <0 ) {record__aio_set_pos (trace_fd, off);if (sync) map->flush = flush;
tools/perf/util/mmap.c +1 -75
88 lines not on this page · raw message
tools/perf/util/mmap.h +0 -12
@@ -101,18 +101,6 @@ union perf_event *perf_mmap__read_event(struct perf_mmap *map);int perf_mmap__push (struct perf_mmap *md,void *to,int push (struct perf_mmap *map,void *to,void *buf,size_t size)); -#ifdef HAVE_AIO_SUPPORT -int perf_mmap__aio_push (struct perf_mmap *md,void *to,int idx, -int push (void *to,struct aiocb *cblock,void *buf,size_t size,off_t off), -off_t *off); -#else -static inline int perf_mmap__aio_push (struct perf_mmap *md__maybe_unused ,void *to__maybe_unused ,int idx__maybe_unused , -int push (void *to,struct aiocb *cblock,void *buf,size_t size,off_t off)__maybe_unused , -off_t *off__maybe_unused ) -{ -return 0 ; -} -#endif size_t perf_mmap__mmap_len (struct perf_mmap *map);
Signature · 1 line
2.20.1
tip-bot for Alexey Budankov
merge bot[tip:perf/core] perf record: Implement compression for AIO trace streaming
perf record: Implement compression for AIO trace streaming
Commit-ID: ef781128e47e73f0e5b2ad385cfa685a0719456a Gitweb: https://git.kernel.org/tip/ef781128e47e73f0e5b2ad385cfa685a0719456a Author: Alexey Budankov <[email protected]> AuthorDate: Mon, 18 Mar 2019 20:44:12 +0300 Committer: Arnaldo Carvalho de Melo <[email protected]> CommitDate: Wed, 15 May 2019 16:36:49 -0300
perf record: Implement compression for AIO trace streaming
Compression is implemented using the functions from zstd.c. As the memory to operate on the compression uses mmap->aio.data[] buffers. If Zstd streaming compression API fails for some reason the data to be compressed are just copied into the memory buffers using plain memcpy().
Compressed trace frame consists of an array of PERF_RECORD_COMPRESSED records. Each element of the array is not longer that PERF_SAMPLE_MAX_SIZE and consists of perf_event_header followed by the compressed chunk that is decompressed on the loading stage.
perf_mmap__aio_push() is replaced by perf_mmap__push() which is now used in the both serial and AIO streaming cases. perf_mmap__push() is extended with positive return values to signify absence of data ready for processing.
--- tools/perf/builtin-record.c | 114 ++++++++++++++++++++++++++++++++++---------- tools/perf/util/mmap.c | 76 +---------------------------- tools/perf/util/mmap.h | 12 ----- 3 files changed, 89 insertions(+), 113 deletions(-)
tools/perf/builtin-record.c +88 -26
@@ -133,6 +133,8 @@ static int record__write(struct record *rec, struct perf_mmap *map __maybe_unusereturn 0 ; } +static int record__aio_enabled (struct record *rec); +static int record__comp_enabled (struct record *rec);static size_t zstd_compress (struct perf_session *session,void *dst,size_t dst_size,void *src,size_t src_size);
@@ -186,9 +188,9 @@ static int record__aio_complete(struct perf_mmap *md, struct aiocb *cblock)if (rem_size ==0 ) { cblock->aio_fildes =-1 ;/* -* md->refcount is incremented in perf_mmap__push() for -* every enqueued aio write request so decrement it because -* the request is now complete. +* md->refcount is incremented in record__aio_pushfn() for +* every aio write request started in record__aio_push() so +* decrement it because the request is now complete. */ perf_mmap__put (md); rc =1 ;
@@ -243,18 +245,89 @@ static int record__aio_sync(struct perf_mmap *md, bool sync_all) }while (1 ); } -static int record__aio_pushfn (void *to,struct aiocb *cblock,void *bf,size_t size,off_t off) +struct record_aio { +struct record *rec; +void *data; +size_t size; +}; + +static int record__aio_pushfn (struct perf_mmap *map,void *to,void *buf,size_t size) { -struct record *rec = to; -int ret, trace_fd = rec->session->data->file.fd; +struct record_aio *aio = to; - rec->samples++; +/* +* map->base data pointed by buf is copied into free map->aio.data[] buffer +* to release space in the kernel buffer as fast as possible, calling +* perf_mmap__consume() from perf_mmap__push() function. +* +* That lets the kernel to proceed with storing more profiling data into +* the kernel buffer earlier than other per-cpu kernel buffers are handled. +* +* Coping can be done in two steps in case the chunk of profiling data +* crosses the upper bound of the kernel buffer. In this case we first move +* part of data from map->start till the upper bound and then the reminder +* from the beginning of the kernel buffer till the end of the data chunk. +*/ - ret =record__aio_write (cblock, trace_fd, bf, size, off); +if (record__comp_enabled (aio->rec)) { + size =zstd_compress (aio->rec->session, aio->data + aio->size, +perf_mmap__mmap_len (map) - aio->size, + buf, size); + }else { +memcpy (aio->data + aio->size, buf, size); + } + +if (!aio->size) { +/* +* Increment map->refcount to guard map->aio.data[] buffer +* from premature deallocation because map object can be +* released earlier than aio write request started on +* map->aio.data[] buffer is complete. +* +* perf_mmap__put() is done at record__aio_complete() +* after started aio request completion or at record__aio_push() +* if the request failed to start. +*/ +perf_mmap__get (map); + } + + aio->size += size; + +return size; +} + +static int record__aio_push (struct record *rec,struct perf_mmap *map,off_t *off) +{ +int ret, idx; +int trace_fd = rec->session->data->file.fd; +struct record_aio aio = { .rec = rec, .size =0 }; + +/* +* Call record__aio_sync() to wait till map->aio.data[] buffer +* becomes available after previous aio write operation. +*/ + + idx =record__aio_sync (map, false); + aio.data = map->aio.data[idx]; + ret =perf_mmap__push (map, &aio, record__aio_pushfn); +if (ret !=0 )/* ret > 0 - no data, ret < 0 - error */ +return ret; + + rec->samples++; + ret =record__aio_write (&(map->aio.cblocks[idx]), trace_fd, aio.data, aio.size, *off);if (!ret) { - rec->bytes_written += size; + *off += aio.size; + rec->bytes_written += aio.size;if (switch_output_size (rec))trigger_hit (&switch_output_trigger); + }else { +/* +* Decrement map->refcount incremented in record__aio_pushfn() +* back if record__aio_write() operation failed to start, otherwise +* map->refcount is decremented in record__aio_complete() after +* aio write operation finishes successfully. +*/ +perf_mmap__put (map); }return ret;
@@ -276,7 +349,7 @@ static void record__aio_mmap_read_sync(struct record *rec)struct perf_evlist *evlist = rec->evlist;struct perf_mmap *maps = evlist->mmap; -if (!rec->opts.nr_cblocks) +if (!record__aio_enabled (rec))return ;for (i =0 ; i < evlist->nr_mmaps; i++) {
@@ -310,13 +383,8 @@ static int record__aio_parse(const struct option *opt,#else /* HAVE_AIO_SUPPORT */ static int nr_cblocks_max =0 ; -static int record__aio_sync (struct perf_mmap *md__maybe_unused ,bool sync_all__maybe_unused ) -{ -return -1 ; -} - -static int record__aio_pushfn (void *to__maybe_unused ,struct aiocb *cblock__maybe_unused , -void *bf__maybe_unused ,size_t size__maybe_unused ,off_t off__maybe_unused ) +static int record__aio_push (struct record *rec__maybe_unused ,struct perf_mmap *map__maybe_unused , +off_t *off__maybe_unused ) {return -1 ; }
@@ -825,7 +893,7 @@ static int record__mmap_read_evlist(struct record *rec, struct perf_evlist *evliint rc =0 ;struct perf_mmap *maps;int trace_fd = rec->data.file.fd; -off_t off; +off_t off =0 ;if (!evlist)return 0 ;
@@ -851,20 +919,14 @@ static int record__mmap_read_evlist(struct record *rec, struct perf_evlist *evli map->flush =1 ; }if (!record__aio_enabled (rec)) { -if (perf_mmap__push (map, rec, record__pushfn) !=0 ) { +if (perf_mmap__push (map, rec, record__pushfn) <0 ) {if (synch) map->flush = flush; rc =-1 ; goto out; } }else { -int idx; -/* -* Call record__aio_sync() to wait till map->data buffer -* becomes available after previous aio write request. -*/ - idx =record__aio_sync (map, false); -if (perf_mmap__aio_push (map, rec, idx, record__aio_pushfn, &off) !=0 ) { +if (record__aio_push (rec, map, &off) <0 ) {record__aio_set_pos (trace_fd, off);if (synch) map->flush = flush;
tools/perf/util/mmap.c +1 -75
88 lines not on this page · raw message
tools/perf/util/mmap.h +0 -12
@@ -101,18 +101,6 @@ union perf_event *perf_mmap__read_event(struct perf_mmap *map);int perf_mmap__push (struct perf_mmap *md,void *to,int push (struct perf_mmap *map,void *to,void *buf,size_t size)); -#ifdef HAVE_AIO_SUPPORT -int perf_mmap__aio_push (struct perf_mmap *md,void *to,int idx, -int push (void *to,struct aiocb *cblock,void *buf,size_t size,off_t off), -off_t *off); -#else -static inline int perf_mmap__aio_push (struct perf_mmap *md__maybe_unused ,void *to__maybe_unused ,int idx__maybe_unused , -int push (void *to,struct aiocb *cblock,void *buf,size_t size,off_t off)__maybe_unused , -off_t *off__maybe_unused ) -{ -return 0 ; -} -#endif size_t perf_mmap__mmap_len (struct perf_mmap *map);
PATCH v10 9/12 perf record: implement -z,--compression_level[=<n>] option
In reply to Alexey BudankovImplemented -z,--compression_level[=<n>] option that enables compression of mmaped kernel data buffers content in runtime during perf record mode collection. Default option value is 1 (fastest compression).
Compression overhead has been measured for serial and AIO streaming when profiling matrix multiplication workload:
-------------------------------------------------------------
| SERIAL | AIO-1 |
----------------------------------------------------------------| |-z | OVH(x) | ratio(x) size(MiB) | OVH(x) | ratio(x) size(MiB) | |---------------------------------------------------------------| | 0 | 1,00 | 1,000 179,424 | 1,00 | 1,000 187,527 | | 1 | 1,04 | 8,427 181,148 | 1,01 | 8,474 188,562 | | 2 | 1,07 | 8,055 186,953 | 1,03 | 7,912 191,773 | | 3 | 1,04 | 8,283 181,908 | 1,03 | 8,220 191,078 | | 5 | 1,09 | 8,101 187,705 | 1,05 | 7,780 190,065 | | 8 | 1,05 | 9,217 179,191 | 1,12 | 6,111 193,024 |
-----------------------------------------------------------------
OVH = (Execution time with -z N) / (Execution time with -z 0)
ratio - compression ratio
size - number of bytes that was compressed
size ~= trace size x ratio
--- tools/perf/Documentation/perf-record.txt | 5 +++++ tools/perf/builtin-record.c | 25 ++++++++++++++++++++++++ 2 files changed, 30 insertions(+)
tools/perf/Documentation/perf-record.txt +5 -0
@@ -471,6 +471,11 @@ Also at some cases executing less trace write syscalls with bigger data size can shorter than executing more trace write syscalls with smaller data size thus lowering runtime profiling overhead. +-z:: +--compression-level[=n]:: +Produce compressed trace using specified level n (default: 1 - fastest compression, +22 - smallest trace) + --all-kernel:: Configure all used events to run in kernel space.
tools/perf/builtin-record.c +25 -0
@@ -440,6 +440,26 @@ static int record__mmap_flush_parse(const struct option *opt,return 0 ; } +#ifdef HAVE_ZSTD_SUPPORT +static unsigned int comp_level_default =1 ; +static int record__parse_comp_level (const struct option *opt, +const char *str, +int unset) +{ +struct record_opts *opts = (struct record_opts *)opt->value; + +if (unset) { + opts->comp_level =0 ; + }else { +if (str) + opts->comp_level =strtol (str,NULL ,0 ); +if (!opts->comp_level) + opts->comp_level = comp_level_default; + } + +return 0 ; +} +#endif static unsigned int comp_level_max =22 ;static int record__comp_enabled (struct record *rec)
@@ -2169,6 +2189,11 @@ static struct option __record_options[] = { OPT_CALLBACK(0, "affinity", &record.opts, "node|cpu", "Set affinity mask of trace reading thread to NUMA node cpu mask or cpu of processed mmap buffer", record__parse_affinity), +#ifdef HAVE_ZSTD_SUPPORT +OPT_CALLBACK_OPTARG ('z' ,"compression-level" , &record.opts, &comp_level_default, +"n" ,"Produce compressed trace using specified level (default: 1 - fastest compression, 22 - smallest trace)" , + record__parse_comp_level), +#endifOPT_END () };
Signature · 1 line
2.20.1
Re: [PATCH v10 09/12] perf record: implement -z,--compression_level[=<n>] option
Em Mon, Mar 18, 2019 at 08:44:42PM +0300, Alexey Budankov escreveu:
Quoted 11 lines
Implemented -z,--compression_level[=<n>] option that enables compression of mmaped kernel data buffers content in runtime during perf record mode collection. Default option value is 1 (fastest compression).
Compression overhead has been measured for serial and AIO streaming when profiling matrix multiplication workload:
-------------------------------------------------------------
| SERIAL | AIO-1 |
----------------------------------------------------------------|
Please don't have lines starting with --- in the cset comment log message, breaks scripts, fixing it up now.
- Arnaldo
Quoted 84 lines
|-z | OVH(x) | ratio(x) size(MiB) | OVH(x) | ratio(x) size(MiB) | |---------------------------------------------------------------| | 0 | 1,00 | 1,000 179,424 | 1,00 | 1,000 187,527 | | 1 | 1,04 | 8,427 181,148 | 1,01 | 8,474 188,562 | | 2 | 1,07 | 8,055 186,953 | 1,03 | 7,912 191,773 | | 3 | 1,04 | 8,283 181,908 | 1,03 | 8,220 191,078 | | 5 | 1,09 | 8,101 187,705 | 1,05 | 7,780 190,065 | | 8 | 1,05 | 9,217 179,191 | 1,12 | 6,111 193,024 |-----------------------------------------------------------------
OVH = (Execution time with -z N) / (Execution time with -z 0)
ratio - compression ratio
size - number of bytes that was compressedsize ~= trace size x ratio
Signed-off-by Alexey Budankov <[email protected]>--- tools/perf/Documentation/perf-record.txt | 5 +++++ tools/perf/builtin-record.c | 25 ++++++++++++++++++++++++ 2 files changed, 30 insertions(+)tools/perf/Documentation/perf-record.txt +5
-0@@ -471,6 +471,11 @@ Also at some cases executing less trace write syscalls with bigger data size can shorter than executing more trace write syscalls with smaller data size thus lowering runtime profiling overhead. +-z:: +--compression-level[=n]:: +Produce compressed trace using specified level n (default: 1 - fastest compression, +22 - smallest trace) + --all-kernel:: Configure all used events to run in kernel space.tools/perf/builtin-record.c +25
-0@@ -440,6 +440,26 @@ static int record__mmap_flush_parse(const struct option *opt,return 0 ; } +#ifdef HAVE_ZSTD_SUPPORT +static unsigned int comp_level_default =1 ; +static int record__parse_comp_level (const struct option *opt, +const char *str, +int unset) +{ +struct record_opts *opts = (struct record_opts *)opt->value; + +if (unset) { + opts->comp_level =0 ; + }else { +if (str) + opts->comp_level =strtol (str,NULL ,0 ); +if (!opts->comp_level) + opts->comp_level = comp_level_default; + } + +return 0 ; +} +#endif static unsigned int comp_level_max =22 ;static int record__comp_enabled (struct record *rec)@@ -2169,6 +2189,11 @@ static struct option __record_options[] = { OPT_CALLBACK(0, "affinity", &record.opts, "node|cpu", "Set affinity mask of trace reading thread to NUMA node cpu mask or cpu of processed mmap buffer", record__parse_affinity), +#ifdef HAVE_ZSTD_SUPPORT +OPT_CALLBACK_OPTARG ('z' ,"compression-level" , &record.opts, &comp_level_default, +"n" ,"Produce compressed trace using specified level (default: 1 - fastest compression, 22 - smallest trace)" , + record__parse_comp_level), +#endifOPT_END () };Signature · 1 line
2.20.1
Signature · 1 line
- Arnaldo
Re: [PATCH v10 09/12] perf record: implement -z,--compression_level[=<n>] option
Quoted 9 lines · Arnaldo Carvalho de Melo
Em Mon, Mar 18, 2019 at 08:44:42PM +0300, Alexey Budankov escreveu:
Quoted 8 lines
Implemented -z,--compression_level[=<n>] option that enables compression of mmaped kernel data buffers content in runtime during perf record mode collection. Default option value is 1 (fastest compression).
Compression overhead has been measured for serial and AIO streaming when profiling matrix multiplication workload:
Quoted 6 lines
Quoted 3 lines
-------------------------------------------------------------
| SERIAL | AIO-1 |
----------------------------------------------------------------|Please don't have lines starting with --- in the cset comment log message, breaks scripts, fixing it up now.
Oops, will do my best about that. Thanks.
~Alexey
Quoted 88 lines
- Arnaldo
Quoted 84 lines
|-z | OVH(x) | ratio(x) size(MiB) | OVH(x) | ratio(x) size(MiB) | |---------------------------------------------------------------| | 0 | 1,00 | 1,000 179,424 | 1,00 | 1,000 187,527 | | 1 | 1,04 | 8,427 181,148 | 1,01 | 8,474 188,562 | | 2 | 1,07 | 8,055 186,953 | 1,03 | 7,912 191,773 | | 3 | 1,04 | 8,283 181,908 | 1,03 | 8,220 191,078 | | 5 | 1,09 | 8,101 187,705 | 1,05 | 7,780 190,065 | | 8 | 1,05 | 9,217 179,191 | 1,12 | 6,111 193,024 |-----------------------------------------------------------------
OVH = (Execution time with -z N) / (Execution time with -z 0)
ratio - compression ratio
size - number of bytes that was compressedsize ~= trace size x ratio
Signed-off-by Alexey Budankov <[email protected]>--- tools/perf/Documentation/perf-record.txt | 5 +++++ tools/perf/builtin-record.c | 25 ++++++++++++++++++++++++ 2 files changed, 30 insertions(+)tools/perf/Documentation/perf-record.txt +5
-0@@ -471,6 +471,11 @@ Also at some cases executing less trace write syscalls with bigger data size can shorter than executing more trace write syscalls with smaller data size thus lowering runtime profiling overhead. +-z:: +--compression-level[=n]:: +Produce compressed trace using specified level n (default: 1 - fastest compression, +22 - smallest trace) + --all-kernel:: Configure all used events to run in kernel space.tools/perf/builtin-record.c +25
-0@@ -440,6 +440,26 @@ static int record__mmap_flush_parse(const struct option *opt,return 0 ; } +#ifdef HAVE_ZSTD_SUPPORT +static unsigned int comp_level_default =1 ; +static int record__parse_comp_level (const struct option *opt, +const char *str, +int unset) +{ +struct record_opts *opts = (struct record_opts *)opt->value; + +if (unset) { + opts->comp_level =0 ; + }else { +if (str) + opts->comp_level =strtol (str,NULL ,0 ); +if (!opts->comp_level) + opts->comp_level = comp_level_default; + } + +return 0 ; +} +#endif static unsigned int comp_level_max =22 ;static int record__comp_enabled (struct record *rec)@@ -2169,6 +2189,11 @@ static struct option __record_options[] = { OPT_CALLBACK(0, "affinity", &record.opts, "node|cpu", "Set affinity mask of trace reading thread to NUMA node cpu mask or cpu of processed mmap buffer", record__parse_affinity), +#ifdef HAVE_ZSTD_SUPPORT +OPT_CALLBACK_OPTARG ('z' ,"compression-level" , &record.opts, &comp_level_default, +"n" ,"Produce compressed trace using specified level (default: 1 - fastest compression, 22 - smallest trace)" , + record__parse_comp_level), +#endifOPT_END () };Signature · 1 line
2.20.1
Re: [PATCH v10 09/12] perf record: implement -z,--compression_level[=<n>] option
In reply to Alexey BudankovEm Mon, Mar 18, 2019 at 08:44:42PM +0300, Alexey Budankov escreveu:
Quoted 19 lines
Implemented -z,--compression_level[=<n>] option that enables compression of mmaped kernel data buffers content in runtime during perf record mode collection. Default option value is 1 (fastest compression).
Compression overhead has been measured for serial and AIO streaming when profiling matrix multiplication workload:
-------------------------------------------------------------
| SERIAL | AIO-1 |----------------------------------------------------------------| |-z | OVH(x) | ratio(x) size(MiB) | OVH(x) | ratio(x) size(MiB) | |---------------------------------------------------------------| | 0 | 1,00 | 1,000 179,424 | 1,00 | 1,000 187,527 | | 1 | 1,04 | 8,427 181,148 | 1,01 | 8,474 188,562 | | 2 | 1,07 | 8,055 186,953 | 1,03 | 7,912 191,773 | | 3 | 1,04 | 8,283 181,908 | 1,03 | 8,220 191,078 | | 5 | 1,09 | 8,101 187,705 | 1,05 | 7,780 190,065 | | 8 | 1,05 | 9,217 179,191 | 1,12 | 6,111 193,024 |
Quoted 8 lines
-----------------------------------------------------------------
OVH = (Execution time with -z N) / (Execution time with -z 0)
ratio - compression ratio
size - number of bytes that was compressedsize ~= trace size x ratio
[root@quaco ~]# perf record -z2
^C[ perf record: Woken up 1 times to write data ]0x1746e0 [0x76]: failed to process type: 81 [Invalid argument]
[ perf record: Captured and wrote 1.568 MB perf.data, compressed (original 0.452 MB, ratio is 3.995) ]
[root@quaco ~]#
I've pushed what I have to the tmp.perf/core branch, please try to see if I made any mistake in fixing up conflicts with BPF_PROG_INFO and BPF_BTF header features. I'll continue tomorrow with 10-12/12.
- Arnaldo
Quoted 67 lines
Signed-off-by Alexey Budankov <[email protected]>--- tools/perf/Documentation/perf-record.txt | 5 +++++ tools/perf/builtin-record.c | 25 ++++++++++++++++++++++++ 2 files changed, 30 insertions(+)tools/perf/Documentation/perf-record.txt +5
-0@@ -471,6 +471,11 @@ Also at some cases executing less trace write syscalls with bigger data size can shorter than executing more trace write syscalls with smaller data size thus lowering runtime profiling overhead. +-z:: +--compression-level[=n]:: +Produce compressed trace using specified level n (default: 1 - fastest compression, +22 - smallest trace) + --all-kernel:: Configure all used events to run in kernel space.tools/perf/builtin-record.c +25
-0@@ -440,6 +440,26 @@ static int record__mmap_flush_parse(const struct option *opt,return 0 ; } +#ifdef HAVE_ZSTD_SUPPORT +static unsigned int comp_level_default =1 ; +static int record__parse_comp_level (const struct option *opt, +const char *str, +int unset) +{ +struct record_opts *opts = (struct record_opts *)opt->value; + +if (unset) { + opts->comp_level =0 ; + }else { +if (str) + opts->comp_level =strtol (str,NULL ,0 ); +if (!opts->comp_level) + opts->comp_level = comp_level_default; + } + +return 0 ; +} +#endif static unsigned int comp_level_max =22 ;static int record__comp_enabled (struct record *rec)@@ -2169,6 +2189,11 @@ static struct option __record_options[] = { OPT_CALLBACK(0, "affinity", &record.opts, "node|cpu", "Set affinity mask of trace reading thread to NUMA node cpu mask or cpu of processed mmap buffer", record__parse_affinity), +#ifdef HAVE_ZSTD_SUPPORT +OPT_CALLBACK_OPTARG ('z' ,"compression-level" , &record.opts, &comp_level_default, +"n" ,"Produce compressed trace using specified level (default: 1 - fastest compression, 22 - smallest trace)" , + record__parse_comp_level), +#endifOPT_END () };Signature · 1 line
2.20.1
Signature · 1 line
- Arnaldo
Re: [PATCH v10 09/12] perf record: implement -z,--compression_level[=<n>] option
Em Tue, May 14, 2019 at 05:20:41PM -0300, Arnaldo Carvalho de Melo escreveu:
Quoted 5 lines
Em Mon, Mar 18, 2019 at 08:44:42PM +0300, Alexey Budankov escreveu:
Quoted 4 lines
Implemented -z,--compression_level[=<n>] option that enables compression of mmaped kernel data buffers content in runtime during perf record mode collection. Default option value is 1 (fastest compression).
<SNIP>
Quoted 6 lines
[root@quaco ~]# perf record -z2
^C[ perf record: Woken up 1 times to write data ]0x1746e0[0x76]: failed to process type: 81 [Invalid argument]
[ perf record: Captured and wrote 1.568 MB perf.data, compressed (original 0.452 MB, ratio is 3.995) ][root@quaco ~]#
So, its the buildid processing at the end, so we can't do build-id processing when using PERF_RECORD_COMPRESSED, otherwise we'd have to uncompress at the end to find the PERF_RECORD_FORK/PERF_RECORD_MMAP, etc.
[root@quaco ~]# perf record -z2 --no-buildid sleep 1
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.020 MB perf.data, compressed (original 0.001 MB, ratio is 2.153) ] [root@quaco ~]# perf report -D | grep PERF_RECORD_COMP0x4f40 [0x195]: failed to process type: 81 [Invalid argument]
Error:
failed to process sample
0 0x4f40 [0x195]: PERF_RECORD_COMPRESSED
[root@quaco ~]#
I'll play with it tomorrow.
- Arnaldo
Quoted 3 lines
I've pushed what I have to the tmp.perf/core branch, please try to see if I made any mistake in fixing up conflicts with
BPF_PROG_INFOandBPF_BTFheader features. I'll continue tomorrow with 10-12/12.
Re: [PATCH v10 09/12] perf record: implement -z,--compression_level[=<n>] option
Quoted 31 lines · Arnaldo Carvalho de Melo
Em Tue, May 14, 2019 at 05:20:41PM -0300, Arnaldo Carvalho de Melo escreveu:
Quoted 5 lines
Em Mon, Mar 18, 2019 at 08:44:42PM +0300, Alexey Budankov escreveu:
Quoted 4 lines
Implemented -z,--compression_level[=<n>] option that enables compression of mmaped kernel data buffers content in runtime during perf record mode collection. Default option value is 1 (fastest compression).
<SNIP>
Quoted 6 lines
[root@quaco ~]# perf record -z2
^C[ perf record: Woken up 1 times to write data ]0x1746e0[0x76]: failed to process type: 81 [Invalid argument]
[ perf record: Captured and wrote 1.568 MB perf.data, compressed (original 0.452 MB, ratio is 3.995) ][root@quaco ~]#
So, its the buildid processing at the end, so we can't do build-id processing when using
PERF_RECORD_COMPRESSED, otherwise we'd have to uncompress at the end to find the PERF_RECORD_FORK/PERF_RECORD_MMAP, etc.[root@quaco ~]# perf record -z2 --no-buildid sleep 1
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.020 MB perf.data, compressed (original 0.001 MB, ratio is 2.153) ] [root@quaco ~]# perf report -D | grepPERF_RECORD_COMP0x4f40[0x195]: failed to process type: 81 [Invalid argument]
Error:
failed to process sample
00x4f40[0x195]:PERF_RECORD_COMPRESSED
[root@quaco ~]#
Quoted 1 line
I'll play with it tomorrow.
Applied the whole patch set on top of the current perf/core and the whole thing functions as expected.
~Alexey
Quoted 7 lines
- Arnaldo
Quoted 3 lines
I've pushed what I have to the tmp.perf/core branch, please try to see if I made any mistake in fixing up conflicts with
BPF_PROG_INFOandBPF_BTFheader features. I'll continue tomorrow with 10-12/12.
tip-bot for Alexey Budankov
merge bot[tip:perf/core] perf record: Implement -z,--compression_level[=<n>] option
In reply to Alexey Budankovperf record: Implement -z,--compression_level[=<n>] option
Commit-ID: 504c1ad11691d1a16e92285bb961728a80c06014 Gitweb: https://git.kernel.org/tip/504c1ad11691d1a16e92285bb961728a80c06014 Author: Alexey Budankov <[email protected]> AuthorDate: Mon, 18 Mar 2019 20:44:42 +0300 Committer: Arnaldo Carvalho de Melo <[email protected]> CommitDate: Wed, 15 May 2019 16:36:49 -0300
perf record: Implement -z,--compression_level[=<n>] option
Implemented -z,--compression_level[=<n>] option that enables compression of mmaped kernel data buffers content in runtime during perf record mode collection. Default option value is 1 (fastest compression).
Compression overhead has been measured for serial and AIO streaming when profiling matrix multiplication workload:
-------------------------------------------------------------
| SERIAL | AIO-1 |
----------------------------------------------------------------| |-z | OVH(x) | ratio(x) size(MiB) | OVH(x) | ratio(x) size(MiB) | |---------------------------------------------------------------| | 0 | 1,00 | 1,000 179,424 | 1,00 | 1,000 187,527 | | 1 | 1,04 | 8,427 181,148 | 1,01 | 8,474 188,562 | | 2 | 1,07 | 8,055 186,953 | 1,03 | 7,912 191,773 | | 3 | 1,04 | 8,283 181,908 | 1,03 | 8,220 191,078 | | 5 | 1,09 | 8,101 187,705 | 1,05 | 7,780 190,065 | | 8 | 1,05 | 9,217 179,191 | 1,12 | 6,111 193,024 |
-----------------------------------------------------------------
OVH = (Execution time with -z N) / (Execution time with -z 0)
ratio - compression ratio
size - number of bytes that was compressed
size ~= trace size x ratio
Committer notes:
Testing it I noticed that it failed to disable build id processing when compression is enabled, and as we'd have to uncompress everything to look for the PERF_RECORD_{MMAP,SAMPLE,etc} to figure out which build ids to read from DSOs, we better disable build id processing when compression is enabled, logging with pr_debug() when doing so:
Original patch:
# perf record -z2 ^C[ perf record: Woken up 1 times to write data ] 0x1746e0 [0x76]: failed to process type: 81 [Invalid argument] [ perf record: Captured and wrote 1.568 MB perf.data, compressed (original 0.452 MB, ratio is 3.995) ] #
After auto-disabling build id processing when compression is enabled:
$ perf record -z2 sleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.001 MB perf.data, compressed (original 0.001 MB, ratio is 2.292) ] $ perf record -v -z2 sleep 1 Compression enabled, disabling build id collection at the end of the session. <SNIP extra -v pr_debug() messages> [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.001 MB perf.data, compressed (original 0.001 MB, ratio is 2.305) ] $
Also, with parts of the patch originally after this one moved to just before this one we get:
$ perf record -z2 sleep 1
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.001 MB perf.data, compressed (original 0.001 MB, ratio is 2.371) ]
$ perf report -D | grep COMPRESS
0 0x1b8 [0x155]: PERF_RECORD_COMPRESSED: unhandled!
0 0x30d [0x80]: PERF_RECORD_COMPRESSED: unhandled!
COMPRESSED events: 2
COMPRESSED events: 0
$I.e. when faced with PERF_RECORD_COMPRESSED that we still have no code to process, we just show it as not being handled, skip them and continue, while before we had:
$ perf report -D | grep COMPRESS 0x1b8 [0x169]: failed to process type: 81 [Invalid argument] Error: failed to process sample 0 0x1b8 [0x169]: PERF_RECORD_COMPRESSED $
--- tools/perf/Documentation/perf-record.txt | 5 +++++ tools/perf/builtin-record.c | 30 ++++++++++++++++++++++++++++++ 2 files changed, 35 insertions(+)
tools/perf/Documentation/perf-record.txt +5 -0
@@ -478,6 +478,11 @@ Also at some cases executing less output write syscalls with bigger data size can take less time than executing more output write syscalls with smaller data size thus lowering runtime profiling overhead. +-z:: +--compression-level[=n]:: +Produce compressed trace using specified level n (default: 1 - fastest compression, +22 - smallest trace) + --all-kernel:: Configure all used events to run in kernel space.
tools/perf/builtin-record.c +30 -0
@@ -443,6 +443,25 @@ static int record__mmap_flush_parse(const struct option *opt,return 0 ; } +#ifdef HAVE_ZSTD_SUPPORT +static unsigned int comp_level_default =1 ; + +static int record__parse_comp_level (const struct option *opt,const char *str,int unset) +{ +struct record_opts *opts = opt->value; + +if (unset) { + opts->comp_level =0 ; + }else { +if (str) + opts->comp_level =strtol (str,NULL ,0 ); +if (!opts->comp_level) + opts->comp_level = comp_level_default; + } + +return 0 ; +} +#endif static unsigned int comp_level_max =22 ;static int record__comp_enabled (struct record *rec)
@@ -2200,6 +2219,11 @@ static struct option __record_options[] = { OPT_CALLBACK(0, "affinity", &record.opts, "node|cpu", "Set affinity mask of trace reading thread to NUMA node cpu mask or cpu of processed mmap buffer", record__parse_affinity), +#ifdef HAVE_ZSTD_SUPPORT +OPT_CALLBACK_OPTARG ('z' ,"compression-level" , &record.opts, &comp_level_default, +"n" ,"Compressed records using specified level (default: 1 - fastest compression, 22 - greatest compression)" , + record__parse_comp_level), +#endifOPT_END () };
@@ -2259,6 +2283,12 @@ int cmd_record(int argc, const char **argv) "cgroup monitoring only available in system-wide mode"); } + +if (rec->opts.comp_level !=0 ) { +pr_debug ("Compression enabled, disabling build id collection at the end of the session.\n" ); + rec->no_buildid = true; + } +if (rec->opts.record_switch_events && !perf_can_record_switch_events ()) {ui__error ("kernel does not support recording context switch events\n" );
PATCH v10 10/12 perf report: implement record trace decompression
In reply to Alexey Budankovzstd_init(, comp_level = 0) initializes decompression part of API only that now consists of zstd_decompress_stream() function.
Trace frames containing PERF_RECORD_COMPRESSED records are decompressed using zstd_decompress_stream() function into a linked list of mmaped memory regions of mmap_comp_len size (struct decomp).
After decompression of one COMPRESSED record its content is iterated and fetched for usual processing. The mmaped memory regions with decompressed events are kept in the linked list till the tool process termination.
When dumping raw trace (e.g., perf report -D --header) file offsets of events from compressed records are printed as zero.
--- tools/perf/builtin-report.c | 5 +- tools/perf/util/compress.h | 11 +++ tools/perf/util/session.c | 129 +++++++++++++++++++++++++++++++++++- tools/perf/util/session.h | 10 +++ tools/perf/util/tool.h | 2 + tools/perf/util/zstd.c | 40 +++++++++++ 6 files changed, 195 insertions(+), 2 deletions(-)
tools/perf/builtin-report.c +4 -1
@@ -1260,6 +1260,9 @@ int cmd_report(int argc, const char **argv)if (session ==NULL )return -1 ; +if (zstd_init (&(session->zstd_data),0 ) <0 ) +pr_warning ("Decompression initialization failed. Reported data may be incomplete.\n" ); +if (report.queue_size) {ordered_events__set_alloc_size (&session->ordered_events, report.queue_size);
@@ -1450,7 +1453,7 @@ int cmd_report(int argc, const char **argv) error:if (report.ptime_range)zfree (&report.ptime_range); - +zstd_fini (&(session->zstd_data));perf_session__delete (session);return ret; }
tools/perf/util/compress.h +11 -0
@@ -20,6 +20,7 @@ bool lzma_is_compressed(const char *input);struct zstd_data {#ifdef HAVE_ZSTD_SUPPORT ZSTD_CStream *cstream; +ZSTD_DStream *dstream;#endif };
@@ -32,6 +33,9 @@ size_t zstd_compress_stream_to_records(struct zstd_data *data, void *dst, size_t dst_size, void *src, size_t src_size, size_t max_record_size, size_t process_header(void *record, size_t increment)); +size_t zstd_decompress_stream (struct zstd_data *data, +void *src,size_t src_size,void *dst,size_t dst_size); +#else /* !HAVE_ZSTD_SUPPORT */ static inline int zstd_init (struct zstd_data *data__maybe_unused ,int level__maybe_unused )
@@ -53,6 +57,13 @@ static inline size_t zstd_compress_stream_to_records(struct zstd_data *data __mareturn 0 ; } +static inline size_t zstd_decompress_stream (struct zstd_data *data__maybe_unused ,void *src__maybe_unused , +size_t src_size__maybe_unused ,void *dst__maybe_unused , +size_t dst_size__maybe_unused ) +{ +return 0 ; +} +#endif #endif /* PERF_COMPRESS_H */
tools/perf/util/session.c +128 -1
189 lines not on this page · raw message
tools/perf/util/session.h +10 -0
@@ -39,6 +39,16 @@ struct perf_session {u64 bytes_transferred;u64 bytes_compressed;struct zstd_data zstd_data; +struct decomp *decomp; +struct decomp *decomp_last; +}; + +struct decomp { +struct decomp *next; +u64 file_pos; +u64 head; +size_t size; +char data[]; };struct perf_tool ;
tools/perf/util/tool.h +2 -0
@@ -28,6 +28,7 @@ typedef int (*event_attr_op)(struct perf_tool *tool,typedef int (*event_op2 )(struct perf_session *session,union perf_event *event);typedef s64 (*event_op3 )(struct perf_session *session,union perf_event *event); +typedef int (*event_op4 )(struct perf_session *session,union perf_event *event,u64 data);typedef int (*event_oe )(struct perf_tool *tool,union perf_event *event,struct ordered_events *oe);
@@ -72,6 +73,7 @@ struct perf_tool { stat, stat_round, feature; +event_op4 compressed;event_op3 auxtrace;bool ordered_events;bool ordering_requires_timestamps;
tools/perf/util/zstd.c +40 -0
@@ -9,6 +9,21 @@ int zstd_init(struct zstd_data *data, int level) {size_t ret; + data->dstream =ZSTD_createDStream (); +if (data->dstream ==NULL ) { +pr_err ("Couldn't create decompression stream.\n" ); +return -1 ; + } + + ret =ZSTD_initDStream (data->dstream); +if (ZSTD_isError (ret)) { +pr_err ("Failed to initialize decompression stream: %s\n" ,ZSTD_getErrorName (ret)); +return -1 ; + } + +if (!level) +return 0 ; + data->cstream =ZSTD_createCStream ();if (data->cstream ==NULL ) {pr_err ("Couldn't create compression stream.\n" );
@@ -26,6 +41,11 @@ int zstd_init(struct zstd_data *data, int level)int zstd_fini (struct zstd_data *data) { +if (data->dstream) { +ZSTD_freeDStream (data->dstream); + data->dstream =NULL ; + } +if (data->cstream) {ZSTD_freeCStream (data->cstream); data->cstream =NULL ;
@@ -69,3 +89,23 @@ size_t zstd_compress_stream_to_records(struct zstd_data *data,return compressed; } +size_t zstd_decompress_stream (struct zstd_data *data, +void *src,size_t src_size,void *dst,size_t dst_size) +{ +size_t ret; +ZSTD_inBuffer input = { src, src_size,0 }; +ZSTD_outBuffer output = { dst, dst_size,0 }; + +while (input.pos < input.size) { + ret =ZSTD_decompressStream (data->dstream, &output, &input); +if (ZSTD_isError (ret)) { +pr_err ("failed to decompress (B): %ld -> %ld : %s\n" , + src_size, output.size,ZSTD_getErrorName (ret)); +break ; + } + output.dst = dst + output.pos; + output.size = dst_size - output.pos; + } + +return output.pos; +}
Signature · 1 line
2.20.1
tip-bot for Alexey Budankov
merge bot[tip:perf/core] perf report: Add stub processing of compressed events for -D
perf report: Add stub processing of compressed events for -D
Commit-ID: 61a7773ca88f32ef7e185fdf9fc0d44e8ec18a66 Gitweb: https://git.kernel.org/tip/61a7773ca88f32ef7e185fdf9fc0d44e8ec18a66 Author: Alexey Budankov <[email protected]> AuthorDate: Mon, 18 Mar 2019 20:45:11 +0300 Committer: Arnaldo Carvalho de Melo <[email protected]> CommitDate: Wed, 15 May 2019 16:36:49 -0300
perf report: Add stub processing of compressed events for -D
Committer note:
Split from a larger patch, this only dumps PERF_RECORD_COMPRESSED as unhandled, so that when we introduce the record part in the next patch, we don't see unhandled events when using 'perf record -D'.
Changed it so that we dump the event if the handler is just a stub, i.e. for the case where we don't have ZSTD linked but we're processing a perf.data file generated by a tool with that linked.
Also when failing to decompress we can't just dump the uncompressed event and return 0, we have to propagate the error.
--- tools/perf/util/session.c | 19 ++++++++++++++++++- tools/perf/util/tool.h | 2 ++ 2 files changed, 20 insertions(+), 1 deletion(-)
tools/perf/util/session.c +18 -1
@@ -358,6 +358,14 @@ static int process_stat_round_stub(struct perf_session *perf_session __maybe_unureturn 0 ; } +static int perf_session__process_compressed_event_stub (struct perf_session *session__maybe_unused , +union perf_event *event__maybe_unused , +u64 file_offset__maybe_unused ) +{ +dump_printf (": unhandled!\n" ); +return 0 ; +} +void perf_tool__fill_defaults (struct perf_tool *tool) {if (tool->sample ==NULL )
@@ -430,6 +438,8 @@ void perf_tool__fill_defaults(struct perf_tool *tool) tool->time_conv = process_event_op2_stub;if (tool->feature ==NULL ) tool->feature = process_event_op2_stub; +if (tool->compressed ==NULL ) + tool->compressed = perf_session__process_compressed_event_stub; }static void swap_sample_id_all (union perf_event *event,void *data)
@@ -1373,7 +1383,9 @@ static s64 perf_session__process_user_event(struct perf_session *session,int fd =perf_data__fd (session->data);int err; -dump_event (session->evlist, event, file_offset, &sample); +if (event->header.type !=PERF_RECORD_COMPRESSED || + tool->compressed == perf_session__process_compressed_event_stub) +dump_event (session->evlist, event, file_offset, &sample);/* These events are processed right away */ switch (event->header.type) {
@@ -1426,6 +1438,11 @@ static s64 perf_session__process_user_event(struct perf_session *session,return tool->time_conv (session, event);case PERF_RECORD_HEADER_FEATURE :return tool->feature (session, event); +case PERF_RECORD_COMPRESSED : + err = tool->compressed (session, event, file_offset); +if (err) +dump_event (session->evlist, event, file_offset, &sample); +return err;default :return -EINVAL ; }
tools/perf/util/tool.h +2 -0
@@ -28,6 +28,7 @@ typedef int (*event_attr_op)(struct perf_tool *tool,typedef int (*event_op2 )(struct perf_session *session,union perf_event *event);typedef s64 (*event_op3 )(struct perf_session *session,union perf_event *event); +typedef int (*event_op4 )(struct perf_session *session,union perf_event *event,u64 data);typedef int (*event_oe )(struct perf_tool *tool,union perf_event *event,struct ordered_events *oe);
@@ -72,6 +73,7 @@ struct perf_tool { stat, stat_round, feature; +event_op4 compressed;event_op3 auxtrace;bool ordered_events;bool ordering_requires_timestamps;
tip-bot for Alexey Budankov
merge bot[tip:perf/core] perf report: Implement perf.data record decompression
In reply to Alexey Budankovperf report: Implement perf.data record decompression
Commit-ID: cb62c6f1f59232457414ecbbf2337a1cb67b4ce2 Gitweb: https://git.kernel.org/tip/cb62c6f1f59232457414ecbbf2337a1cb67b4ce2 Author: Alexey Budankov <[email protected]> AuthorDate: Mon, 18 Mar 2019 20:45:11 +0300 Committer: Arnaldo Carvalho de Melo <[email protected]> CommitDate: Wed, 15 May 2019 16:36:49 -0300
perf report: Implement perf.data record decompression
zstd_init(, comp_level = 0) initializes decompression part of API only hat now consists of zstd_decompress_stream() function.
The perf.data PERF_RECORD_COMPRESSED records are decompressed using zstd_decompress_stream() function into a linked list of mmaped memory regions of mmap_comp_len size (struct decomp).
After decompression of one COMPRESSED record its content is iterated and fetched for usual processing. The mmaped memory regions with decompressed events are kept in the linked list till the tool process termination.
When dumping raw records (e.g., perf report -D --header) file offsets of events from compressed records are printed as zero.
Committer notes:
Since now we have support for processing PERF_RECORD_COMPRESSED, we see none, in raw form, like we saw in the previous patch commiter notes, they were decompressed into the usual PERF_RECORD_{FORK,MMAP,COMM,etc} records, we only see the stats for those PERF_RECORD_COMPRESSED events, and since I used the file generated in the commiter notes for the previous patch, there they are, 2 compressed records:
$ perf report --header-only | grep cmdline
# cmdline : /home/acme/bin/perf record -z2 sleep 1
$ perf report -D | grep COMPRESS
COMPRESSED events: 2
COMPRESSED events: 0
$ perf report --stdio
# To display the perf.data header info, please use --header/--header-only options.
#
#
# Total Lost Samples: 0
#
# Samples: 15 of event 'cycles:u'
# Event count (approx.): 962227
#
# Overhead Command Shared Object Symbol
# ........ ....... ................ ...........................
#
46.99% sleep libc-2.28.so [.] _dl_addr
29.24% sleep [unknown] [k] 0xffffffffaea00a67
16.45% sleep libc-2.28.so [.] __GI__IO_un_link.part.1
5.92% sleep ld-2.28.so [.] _dl_setup_hash
1.40% sleep libc-2.28.so [.] __nanosleep
0.00% sleep [unknown] [k] 0xffffffffaea00163 #
# (Tip: To see callchains in a more compact form: perf report -g folded) #
$
--- tools/perf/builtin-report.c | 5 +- tools/perf/util/compress.h | 11 +++++ tools/perf/util/session.c | 116 +++++++++++++++++++++++++++++++++++++++++++- tools/perf/util/session.h | 10 ++++ tools/perf/util/zstd.c | 41 ++++++++++++++++ 5 files changed, 181 insertions(+), 2 deletions(-)
tools/perf/builtin-report.c +4 -1
@@ -1258,6 +1258,9 @@ repeat:if (session ==NULL )return -1 ; +if (zstd_init (&(session->zstd_data),0 ) <0 ) +pr_warning ("Decompression initialization failed. Reported data may be incomplete.\n" ); +if (report.queue_size) {ordered_events__set_alloc_size (&session->ordered_events, report.queue_size);
@@ -1448,7 +1451,7 @@ repeat: error:if (report.ptime_range)zfree (&report.ptime_range); - +zstd_fini (&(session->zstd_data));perf_session__delete (session);return ret; }
tools/perf/util/compress.h +11 -0
@@ -20,6 +20,7 @@ bool lzma_is_compressed(const char *input);struct zstd_data {#ifdef HAVE_ZSTD_SUPPORT ZSTD_CStream *cstream; +ZSTD_DStream *dstream;#endif };
@@ -31,6 +32,9 @@ int zstd_fini(struct zstd_data *data);size_t zstd_compress_stream_to_records (struct zstd_data *data,void *dst,size_t dst_size,void *src,size_t src_size,size_t max_record_size,size_t process_header (void *record,size_t increment)); + +size_t zstd_decompress_stream (struct zstd_data *data,void *src,size_t src_size, +void *dst,size_t dst_size);#else /* !HAVE_ZSTD_SUPPORT */ static inline int zstd_init (struct zstd_data *data__maybe_unused ,int level__maybe_unused )
@@ -52,6 +56,13 @@ size_t zstd_compress_stream_to_records(struct zstd_data *data __maybe_unused, {return 0 ; } + +static inline size_t zstd_decompress_stream (struct zstd_data *data__maybe_unused ,void *src__maybe_unused , +size_t src_size__maybe_unused ,void *dst__maybe_unused , +size_t dst_size__maybe_unused ) +{ +return 0 ; +}#endif #endif /* PERF_COMPRESS_H */
tools/perf/util/session.c +115 -1
164 lines not on this page · raw message
tools/perf/util/session.h +10 -0
@@ -39,6 +39,16 @@ struct perf_session {u64 bytes_transferred;u64 bytes_compressed;struct zstd_data zstd_data; +struct decomp *decomp; +struct decomp *decomp_last; +}; + +struct decomp { +struct decomp *next; +u64 file_pos; +u64 head; +size_t size; +char data[]; };struct perf_tool ;
tools/perf/util/zstd.c +41 -0
56 lines not on this page · raw message
PATCH v10 11/12 perf inject: enable COMPRESSED records decompression
In reply to Alexey BudankovInitialized decompression part of Zstd based API so COMPRESSED records would be decompressed into the resulting output data file.
--- tools/perf/builtin-inject.c | 4 ++++ 1 file changed, 4 insertions(+)
tools/perf/builtin-inject.c +4 -0
@@ -837,6 +837,9 @@ int cmd_inject(int argc, const char **argv)if (inject.session ==NULL )return -1 ; +if (zstd_init (&(inject.session->zstd_data),0 ) <0 ) +pr_warning ("Decompression initialization failed.\n" ); +if (inject.build_ids) {/* * to make sure the mmap records are ordered correctly
@@ -867,6 +870,7 @@ int cmd_inject(int argc, const char **argv) ret =__cmd_inject (&inject); out_delete: +zstd_fini (&(inject.session->zstd_data));perf_session__delete (inject.session);return ret; }
Signature · 1 line
2.20.1
tip-bot for Alexey Budankov
merge bot[tip:perf/core] perf inject: Enable COMPRESSED record decompression
perf inject: Enable COMPRESSED record decompression
Commit-ID: 371a3378d83a755add84b2dca730a3a641002f3a Gitweb: https://git.kernel.org/tip/371a3378d83a755add84b2dca730a3a641002f3a Author: Alexey Budankov <[email protected]> AuthorDate: Mon, 18 Mar 2019 20:45:44 +0300 Committer: Arnaldo Carvalho de Melo <[email protected]> CommitDate: Wed, 15 May 2019 16:36:49 -0300
perf inject: Enable COMPRESSED record decompression
Initialized decompression part of Zstd based API so COMPRESSED records would be decompressed into the resulting output data file.
--- tools/perf/builtin-inject.c | 4 ++++ 1 file changed, 4 insertions(+)
tools/perf/builtin-inject.c +4 -0
@@ -837,6 +837,9 @@ int cmd_inject(int argc, const char **argv)if (inject.session ==NULL )return -1 ; +if (zstd_init (&(inject.session->zstd_data),0 ) <0 ) +pr_warning ("Decompression initialization failed.\n" ); +if (inject.build_ids) {/* * to make sure the mmap records are ordered correctly
@@ -867,6 +870,7 @@ int cmd_inject(int argc, const char **argv) ret =__cmd_inject (&inject); out_delete: +zstd_fini (&(inject.session->zstd_data));perf_session__delete (inject.session);return ret; }
PATCH v10 12/12 perf tests: implement Zstd comp/decomp integration test
In reply to Alexey BudankovImplemented basic integration test for Zstd based trace compression/decompression in record and report modes.
--- .../tests/shell/record+zstd_comp_decomp.sh | 35 +++++++++++++++++++ 1 file changed, 35 insertions(+) create mode 100755 tools/perf/tests/shell/record+zstd_comp_decomp.sh
tools/perf/tests/shell/record+zstd_comp_decomp.sh new file +35 -0
@@ -0,0 +1,35 @@ +#!/bin/sh +# record trace Zstd compression/decompression + +trace_file=$(mktemp /tmp/perf.data.XXX) +perf_tool=perf +output=/dev/null + +skip_if_no_z_record() { + $perf_tool record -h 2>&1 | grep '\-z, \-\-compression\-level' +} + +collect_z_record() { + echo "Collecting compressed record trace file:" + $perf_tool record -o $trace_file -g -z -F 25000 -- \ + dd count=1000 if=/dev/random of=/dev/null > $output 2>&1 +} + +check_compressed_stats() { + echo "Checking compressed events stats:" + $perf_tool report -i $trace_file --header --stats | \ + grep -E "(# compressed : Zstd,)|(COMPRESSED events:)" > $output 2>&1 +} + +check_compressed_output() { + $perf_tool inject -i $trace_file -o $trace_file.decomp && + $perf_tool report -i $trace_file --stdio | head -n -3 > $trace_file.comp.output && + $perf_tool report -i $trace_file.decomp --stdio | head -n -3 > $trace_file.decomp.output && + diff $trace_file.comp.output $trace_file.decomp.output > $output 2>&1 +} + +skip_if_no_z_record || exit 2 +collect_z_record && check_compressed_stats && check_compressed_output +err=$? +rm -f $trace_file* +exit $err
Signature · 1 line
2.20.1
tip-bot for Alexey Budankov
merge bot[tip:perf/core] perf tests: Implement Zstd comp/decomp integration test
perf tests: Implement Zstd comp/decomp integration test
Commit-ID: bdc35cbc35c0b33428922503c7c85259510911a6 Gitweb: https://git.kernel.org/tip/bdc35cbc35c0b33428922503c7c85259510911a6 Author: Alexey Budankov <[email protected]> AuthorDate: Mon, 18 Mar 2019 20:46:17 +0300 Committer: Arnaldo Carvalho de Melo <[email protected]> CommitDate: Wed, 15 May 2019 16:36:49 -0300
perf tests: Implement Zstd comp/decomp integration test
Introduce a basic integration test for Zstd based record compression/decompression using 'perf record' and 'perf report'.
Committer notes:
Reduce a bit the freq (from 25 kHz to 5 kHz) and the number of /dev/null records read (from 1000 to 500), reducing the time it takes to something more in line with the time existing 'perf test' entries take to run.
With that in place:
$ time perf test zstd 68: Zstd perf.data compression/decompression : Ok
real 0m10.376s
user 0m0.105s
sys 0m0.440s
$ grep "model name" /proc/cpuinfo | head -1 model name : Intel(R) Core(TM) i7-8650U CPU @ 1.90GHz $
--- tools/perf/tests/shell/record+zstd_comp_decomp.sh | 35 +++++++++++++++++++++++ 1 file changed, 35 insertions(+)
tools/perf/tests/shell/record+zstd_comp_decomp.sh new file +35 -0
@@ -0,0 +1,35 @@ +#!/bin/sh +# Zstd perf.data compression/decompression + +trace_file=$(mktemp /tmp/perf.data.XXX) +perf_tool=perf +output=/dev/null + +skip_if_no_z_record() { + $perf_tool record -h 2>&1 | grep '\-z, \-\-compression\-level' +} + +collect_z_record() { + echo "Collecting compressed record file:" + $perf_tool record -o $trace_file -g -z -F 5000 -- \ + dd count=500 if=/dev/random of=/dev/null > $output 2>&1 +} + +check_compressed_stats() { + echo "Checking compressed events stats:" + $perf_tool report -i $trace_file --header --stats | \ + grep -E "(# compressed : Zstd,)|(COMPRESSED events:)" > $output 2>&1 +} + +check_compressed_output() { + $perf_tool inject -i $trace_file -o $trace_file.decomp && + $perf_tool report -i $trace_file --stdio | head -n -3 > $trace_file.comp.output && + $perf_tool report -i $trace_file.decomp --stdio | head -n -3 > $trace_file.decomp.output && + diff $trace_file.comp.output $trace_file.decomp.output > $output 2>&1 +} + +skip_if_no_z_record || exit 2 +collect_z_record && check_compressed_stats && check_compressed_output +err=$? +rm -f $trace_file* +exit $err
Hi,
This is a gentle reminder regarding the patch set below.
Thanks,
Alexey
Quoted 621 lines · Alexey Budankov
The patch set implements runtime trace compression (-z option) in record mode and trace auto decompression in report and inject modes. Streaming Zstd API [1] is used for compression and decompression of data that come from kernel mmaped data buffers.
Usage of implemented -z,--compression_level=n option provides ~3-5x avg. trace file size reduction on variety of tested workloads what saves storage space on larger server systems where trace file size can easily reach several tens or even hundreds of GiBs, especially when profiling with dwarf-based stacks and tracing of context switches. Default option value is 1 (fastest compression).
Implemented --mmap-flush option can be used to specify minimal size of data chunk that is extracted from mmaped kernel buffer to store into a trace. The option is independent from -z setting and doesn't vary with compression level. The default option value is 1 byte what means every time trace writing thread finds some new data in the mmaped buffer the data is extracted, possibly compressed and written to a trace. The option serves two purposes the first one is to increase the compression ratio of trace data and the second one is to avoid live-lock self tool process monitoring in system wide (-a) profiling mode. Profiling in system wide mode with compression (-a -z) can additionally induce data into the kernel buffers along with the data from monitored processes. If performance data rate and volume from the monitored processes is high then trace streaming and compression activity in the tool is also high. It can lead to subtle live-lock effect of endless activity when compression of single new byte from some of mmaped kernel buffer induces the next single byte at some mmaped buffer. So perf tool thread never stops on polling event file descriptors. Varying data chunk size to be extracted from mmap buffers allows avoiding live-locking self monitoring in system wide mode and makes mmap buffers polling loop manageable. Possible usage examples:
$ tools/perf/perf record -z -e cycles -- matrix.gcc $ tools/perf/perf record --aio -z -e cycles -- matrix.gcc $ tools/perf/perf record -z --mmap-flush 1024 -e cycles -- matrix.gcc $ tools/perf/perf record --aio -z --mmap-flush 1K -e cycles -- matrix.gccRuntime compression overhead has been measured for serial and AIO trace writing modes when profiling matrix multiplication workload:
-------------------------------------------------------------
| SERIAL | AIO-1 |----|-----------------------------|-----------------------------| |-z | OVH(x) | ratio(x) size(MiB) | OVH(x) | ratio(x) size(MiB) | |---|--------|--------------------|--------|--------------------| | 0 | 1,00 | 1,000 179,424 | 1,00 | 1,000 187,527 | | 1 | 1,04 | 8,427 181,148 | 1,01 | 8,474 188,562 | | 2 | 1,07 | 8,055 186,953 | 1,03 | 7,912 191,773 | | 3 | 1,04 | 8,283 181,908 | 1,03 | 8,220 191,078 | | 5 | 1,09 | 8,101 187,705 | 1,05 | 7,780 190,065 | | 8 | 1,05 | 9,217 179,191 | 1,12 | 6,111 193,024 |-----------------------------------------------------------------
OVH = (Execution time with -z N) / (Execution time with -z 0)
ratio - compression ratio
size - number of bytes that was compressedsize ~= trace file x ratio
See complete description of measurement conditions with details below.
Introduced compression functionality can be disabled or configured from the command line using
NO_LIBZSTDandLIBZSTD_DIRdefines:$ make -C tools/perf NO_LIBZSTD=1 clean all $ make -C tools/perf LIBZSTD_DIR=/path/to/zstd/sources/ clean allIf your system has some version of the zstd package preinstalled then the build system finds and uses it during the build. Auto detection feature status is reported just before compilation starts, as usual. If you still prefer to compile with some other version of zstd you have capability to refer the compilation to that version using
LIBZSTD_DIRdefine.See 'perf test' results below for enabled and disabled (NO_LIBZSTD=1) feature configurations.
---
Alexey Budankov (12):
feature: implement libzstd check,LIBZSTD_DIRandNO_LIBZSTDdefines perf record: implement --mmap-flush=<number> option
perf session: definebytes_transferredandbytes_compressedmetrics perf record: implement COMPRESSED event record and its attributes perf mmap: implement dedicated memory buffer for data compression perf util: introduce Zstd streaming based compression API perf record: implement compression for serial trace streaming perf record: implement compression for AIO trace streaming perf record: implement -z,--compression_level[=<n>] option perf report: implement record trace decompression
perf inject: enable COMPRESSED records decompression
perf tests: implement Zstd comp/decomp integration testtools/build/Makefile.feature | 6 +- tools/build/feature/Makefile | 6 +- tools/build/feature/test-all.c | 5 + tools/build/feature/test-libzstd.c | 12 + tools/perf/Documentation/perf-record.txt | 17 ++ .../Documentation/perf.data-file-format.txt | 24 ++ tools/perf/Makefile.config | 20 ++ tools/perf/Makefile.perf | 3 + tools/perf/builtin-inject.c | 4 + tools/perf/builtin-record.c | 285 +++++++++++++++--- tools/perf/builtin-report.c | 5 +- tools/perf/builtin-version.c | 2 + tools/perf/perf.h | 2 + .../tests/shell/record+zstd_comp_decomp.sh | 35 +++ tools/perf/util/Build | 2 + tools/perf/util/compress.h | 54 ++++ tools/perf/util/env.h | 11 + tools/perf/util/event.c | 1 + tools/perf/util/event.h | 7 + tools/perf/util/evlist.c | 8 +- tools/perf/util/evlist.h | 3 +- tools/perf/util/header.c | 55 +++- tools/perf/util/header.h | 1 + tools/perf/util/mmap.c | 106 ++----- tools/perf/util/mmap.h | 17 +- tools/perf/util/session.c | 129 +++++++- tools/perf/util/session.h | 14 + tools/perf/util/tool.h | 2 + tools/perf/util/zstd.c | 111 +++++++ 29 files changed, 813 insertions(+), 134 deletions(-) create mode 100644 tools/build/feature/test-libzstd.c create mode 100755 tools/perf/tests/shell/record+zstd_comp_decomp.sh create mode 100644 tools/perf/util/zstd.c---
Changes in v10:
- separated decomp list deallocation intoperf_session__release_decomp_events
- extended the test with suggested decompression validationChanges in v9:
- fixed issue with improper max COMPRESSED record size calculation
- moved up calculation of ratio variable in 03/12
- made minor corrections in changelogs
- corrected several checkpatch.pl warnings and errorsChanges in v8:
- avoid using -f for --mmap-flush option
- move stubs tocompress.hand avoid unconditional compiling ofzstd.c
- fixed silent interruption for perf record collection
- implemented -z 1 as defaultChanges in v7:
- rebased to Arnaldo's perf/core tip
- implemented B/K/M/G suffixes for -f option
- reworkedrecord__mmap_read_evlist()to replaceperf_mmap__aio_push()byperf_mmap__push()in AIO case
- extended "[ perf record: Captured ... ]" message with compression statistics
- extended changelog for v5 06/10
- usedPERF_SAMPLE_MAX_SIZEfor compressed record size calculations
- renamedrecord__zstd_compresstozstd_compressandrecord__process_comp_headertoprocess_comp_header
- separatednr_cblocks_maxapplyingChanges in v6:
- extended docs with description ofPERF_RECORD_COMPRESSEDrecord andHEADER_COMPRESSEDfeature layoutsChanges in v5:
- implemented perf version --build-options extension for aio and zstd - see TESTING below
- adjusted commit message and perf-record.txt content for -f option
- fixed build errors in case of NO_AIO=1 and NO_LIBZSTD=1Changes in v4:
- implemented integration tests
- adjustedzstd_stub functions
- rebased on tip of Arnaldo's perf/coreChanges in v3:
- moved -f,--mmap-flush option implementation into a separate patch
- moved definition and printing ofbytes_transferredandbytes_compressedinto a separate patch
- moved COMPRESSED feature into a separate patch
- added versioning and stored COMPRESSED feature attributes as u32
- implemented dedicated memory buffer for compression in case of serial streaming
- moved low level Zstd based compression functions into util/{compress.h,zstd.c}
- made compress function to be a param of__push(),__aio_push()functions
- enabled perf inject to decompress COMPRESSED records
- measured compression overhead for serial and AIO streaming using
basic matrix multiplication workload on 8 core skylakeChanges in v2:
- moved compression/decompression code to session layer
- enabled allocation aio data buffers for compression
- enabled trace compression for serial trace streaming---
[1] https://github.com/facebook/zstd---
OVERHEAD MEASUREMENTS:uname -a
Linux localhost 4.20.7-200.fc29.x86_64 #1 SMP Wed Feb 6 19:16:42 UTC 2019x86_64x86_64x86_64GNU/Linuxcat
/proc/cpuinfo
processor : 7vendor_id: GenuineIntel
cpu family : 6
model : 94
model name : Intel(R) Core(TM) i7-6700K CPU @ 4.00GHz
stepping : 3
microcode :0xc6
cpu MHz : 4021.884
cache size : 8192 KB
physical id : 0
siblings : 8
core id : 3
cpu cores : 4
apicid : 7
initial apicid : 7
fpu : yesfpu_exception: yes
cpuid level : 22
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lmconstant_tscartarch_perfmonpebs btsrep_goodnopl xtopologynonstop_tsccpuid aperfmperftsc_known_freqpni pclmulqdq dtes64 monitords_cplvmx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcidsse4_1sse4_2x2apic movbe popcnttsc_deadline_timeraes xsave avx f16c rdrandlahf_lmabm 3dnowprefetchcpuid_faultinvpcid_singlepti ssbd ibrs ibpb stibptpr_shadowvnmi flexpriority ept vpidept_adfsgsbasetsc_adjustbmi1 hle avx2 smep bmi2 erms invpcid rtm mpx rdseed adx smap clflushoptintel_ptxsaveopt xsavec xgetbv1 xsaves dtherm ida arat pln pts hwphwp_notifyhwp_act_windowhwp_eppflush_l1d
bugs :cpu_meltdownspectre_v1spectre_v2spec_store_bypassl1tf
bogomips : 8016.00
clflush size : 64cache_alignment: 64
address sizes : 39 bits physical, 48 bits virtual
power management:-----------------------------------------------------------------
#!/bin/bash -xvecho 0 >
/proc/sys/kernel/perf_event_paranoid
+ echo 0
cat/proc/sys/kernel/perf_event_paranoid
+ cat/proc/sys/kernel/perf_event_paranoid
0echo performance | tee
/sys/devices/system/cpu/cpu*/cpufreq/scaling_governor
+ echo performance
+ tee/sys/devices/system/cpu/cpu0/cpufreq/scaling_governor/sys/devices/system/cpu/cpu1/cpufreq/scaling_governor/sys/devices/system/cpu/cpu2/cpufreq/scaling_governor/sys/devices/system/cpu/cpu3/cpufreq/scaling_governor/sys/devices/system/cpu/cpu4/cpufreq/scaling_governor/sys/devices/system/cpu/cpu5/cpufreq/scaling_governor/sys/devices/system/cpu/cpu6/cpufreq/scaling_governor/sys/devices/system/cpu/cpu7/cpufreq/scaling_governorperformancefor i in 0 1 2 3 5 8
do/usr/bin/time tools/perf/perf record -z $i -F 25000 -N -B -T -R -e cycles -- ../../matrix/linux/matrix.gccdone
+ for i in 0 1 2 3 5 8
+/usr/bin/timetools/perf/perfrecord -z 0 -F 25000 -N -B -T -R -e cycles -- ../../matrix/linux/matrix.gcc Addr of buf1 =0x7fe36de5c010
Offs of buf1 =0x7fe36de5c180
Addr of buf2 =0x7fe36be5b010
Offs of buf2 =0x7fe36be5b1c0
Addr of buf3 =0x7fe369e5a010
Offs of buf3 =0x7fe369e5a100
Addr of buf4 =0x7fe367e59010
Offs of buf4 =0x7fe367e59140
Threads #: 8 Pthreads
Matrix size: 2048
Using multiply kernel: multiply1
Execution time = 16.949 seconds
[ perf record: Woken up 309 times to write data ]
[ perf record: Captured and wrote 179.424 MB perf.data ]
133.67user 0.35system 0:17.08elapsed 784%CPU (0avgtext+0avgdata 100580maxresident)k 0inputs+367480outputs (0major+34737minor)pagefaults 0swaps
+ for i in 0 1 2 3 5 8
+/usr/bin/timetools/perf/perfrecord -z 1 -F 25000 -N -B -T -R -e cycles -- ../../matrix/linux/matrix.gcc Addr of buf1 =0x7fcaec334010
Offs of buf1 =0x7fcaec334180
Addr of buf2 =0x7fcaea333010
Offs of buf2 =0x7fcaea3331c0
Addr of buf3 =0x7fcae8332010
Offs of buf3 =0x7fcae8332100
Addr of buf4 =0x7fcae6331010
Offs of buf4 =0x7fcae6331140
Threads #: 8 Pthreads
Matrix size: 2048
Using multiply kernel: multiply1
Execution time = 17.608 seconds
[ perf record: Woken up 595 times to write data ]
[ perf record: Compressed 181.148 MB to 21.497 MB, ratio is 8.427 ]
[ perf record: Captured and wrote 21.527 MB perf.data ]
135.69user 0.24system 0:17.73elapsed 766%CPU (0avgtext+0avgdata 100500maxresident)k 0inputs+44112outputs (0major+35033minor)pagefaults 0swaps
+ for i in 0 1 2 3 5 8
+/usr/bin/timetools/perf/perfrecord -z 2 -F 25000 -N -B -T -R -e cycles -- ../../matrix/linux/matrix.gcc Addr of buf1 =0x7f1336f8d010
Offs of buf1 =0x7f1336f8d180
Addr of buf2 =0x7f1334f8c010
Offs of buf2 =0x7f1334f8c1c0
Addr of buf3 =0x7f1332f8b010
Offs of buf3 =0x7f1332f8b100
Addr of buf4 =0x7f1330f8a010
Offs of buf4 =0x7f1330f8a140
Threads #: 8 Pthreads
Matrix size: 2048
Using multiply kernel: multiply1
Execution time = 18.175 seconds
[ perf record: Woken up 521 times to write data ]
[ perf record: Compressed 186.953 MB to 23.210 MB, ratio is 8.055 ]
[ perf record: Captured and wrote 23.239 MB perf.data ]
140.21user 0.25system 0:18.32elapsed 766%CPU (0avgtext+0avgdata 100560maxresident)k 0inputs+47608outputs (0major+35263minor)pagefaults 0swaps
+ for i in 0 1 2 3 5 8
+/usr/bin/timetools/perf/perfrecord -z 3 -F 25000 -N -B -T -R -e cycles -- ../../matrix/linux/matrix.gcc Addr of buf1 =0x7f97060e3010
Offs of buf1 =0x7f97060e3180
Addr of buf2 =0x7f97040e2010
Offs of buf2 =0x7f97040e21c0
Addr of buf3 =0x7f97020e1010
Offs of buf3 =0x7f97020e1100
Addr of buf4 =0x7f97000e0010
Offs of buf4 =0x7f97000e0140
Threads #: 8 Pthreads
Matrix size: 2048
Using multiply kernel: multiply1
Execution time = 17.688 seconds
[ perf record: Woken up 485 times to write data ]
[ perf record: Compressed 181.908 MB to 21.962 MB, ratio is 8.283 ]
[ perf record: Captured and wrote 21.991 MB perf.data ]
136.87user 0.23system 0:17.81elapsed 769%CPU (0avgtext+0avgdata 100616maxresident)k 0inputs+45064outputs (0major+35773minor)pagefaults 0swaps
+ for i in 0 1 2 3 5 8
+/usr/bin/timetools/perf/perfrecord -z 5 -F 25000 -N -B -T -R -e cycles -- ../../matrix/linux/matrix.gcc Addr of buf1 =0x7f477b444010
Offs of buf1 =0x7f477b444180
Addr of buf2 =0x7f4779443010
Offs of buf2 =0x7f47794431c0
Addr of buf3 =0x7f4777442010
Offs of buf3 =0x7f4777442100
Addr of buf4 =0x7f4775441010
Offs of buf4 =0x7f4775441140
Threads #: 8 Pthreads
Matrix size: 2048
Using multiply kernel: multiply1
Execution time = 18.406 seconds
[ perf record: Woken up 416 times to write data ]
[ perf record: Compressed 187.705 MB to 23.170 MB, ratio is 8.101 ]
[ perf record: Captured and wrote 23.200 MB perf.data ]
142.72user 0.26system 0:18.53elapsed 771%CPU (0avgtext+0avgdata 100520maxresident)k 0inputs+47528outputs (0major+36928minor)pagefaults 0swaps
+ for i in 0 1 2 3 5 8
+/usr/bin/timetools/perf/perfrecord -z 8 -F 25000 -N -B -T -R -e cycles -- ../../matrix/linux/matrix.gcc Addr of buf1 =0x7fb5bf032010
Offs of buf1 =0x7fb5bf032180
Addr of buf2 =0x7fb5bd031010
Offs of buf2 =0x7fb5bd0311c0
Addr of buf3 =0x7fb5bb030010
Offs of buf3 =0x7fb5bb030100
Addr of buf4 =0x7fb5b902f010
Offs of buf4 =0x7fb5b902f140
Threads #: 8 Pthreads
Matrix size: 2048
Using multiply kernel: multiply1
Execution time = 17.751 seconds
[ perf record: Woken up 391 times to write data ]
[ perf record: Compressed 179.191 MB to 19.441 MB, ratio is 9.217 ]
[ perf record: Captured and wrote 19.502 MB perf.data ]
138.90user 0.29system 0:17.88elapsed 778%CPU (0avgtext+0avgdata 100612maxresident)k 0inputs+39968outputs (0major+37436minor)pagefaults 0swapsfor i in 0 1 2 3 5 8
do/usr/bin/time tools/perf/perf record --aio=1 -z $i -F 25000 -N -B -T -R -e cycles -- ../../matrix/linux/matrix.gccdone
+ for i in 0 1 2 3 5 8
+/usr/bin/timetools/perf/perfrecord --aio=1 -z 0 -F 25000 -N -B -T -R -e cycles -- ../../matrix/linux/matrix.gcc Addr of buf1 =0x7feee4519010
Offs of buf1 =0x7feee4519180
Addr of buf2 =0x7feee2518010
Offs of buf2 =0x7feee25181c0
Addr of buf3 =0x7feee0517010
Offs of buf3 =0x7feee0517100
Addr of buf4 =0x7feede516010
Offs of buf4 =0x7feede516140
Threads #: 8 Pthreads
Matrix size: 2048
Using multiply kernel: multiply1
Execution time = 17.912 seconds
[ perf record: Woken up 390 times to write data ]
[ perf record: Captured and wrote 187.527 MB perf.data ]
139.70user 0.39system 0:18.04elapsed 776%CPU (0avgtext+0avgdata 100624maxresident)k 0inputs+384072outputs (0major+35257minor)pagefaults 0swaps
+ for i in 0 1 2 3 5 8
+/usr/bin/timetools/perf/perfrecord --aio=1 -z 1 -F 25000 -N -B -T -R -e cycles -- ../../matrix/linux/matrix.gcc Addr of buf1 =0x7f72b93ac010
Offs of buf1 =0x7f72b93ac180
Addr of buf2 =0x7f72b73ab010
Offs of buf2 =0x7f72b73ab1c0
Addr of buf3 =0x7f72b53aa010
Offs of buf3 =0x7f72b53aa100
Addr of buf4 =0x7f72b33a9010
Offs of buf4 =0x7f72b33a9140
Threads #: 8 Pthreads
Matrix size: 2048
Using multiply kernel: multiply1
Execution time = 18.198 seconds
[ perf record: Woken up 416 times to write data ]
[ perf record: Compressed 188.562 MB to 22.252 MB, ratio is 8.474 ]
[ perf record: Captured and wrote 22.284 MB perf.data ]
141.12user 0.32system 0:18.32elapsed 771%CPU (0avgtext+0avgdata 100576maxresident)k 0inputs+45664outputs (0major+35040minor)pagefaults 0swaps
+ for i in 0 1 2 3 5 8
+/usr/bin/timetools/perf/perfrecord --aio=1 -z 2 -F 25000 -N -B -T -R -e cycles -- ../../matrix/linux/matrix.gcc Addr of buf1 =0x7ffb9caf3010
Offs of buf1 =0x7ffb9caf3180
Addr of buf2 =0x7ffb9aaf2010
Offs of buf2 =0x7ffb9aaf21c0
Addr of buf3 =0x7ffb98af1010
Offs of buf3 =0x7ffb98af1100
Addr of buf4 =0x7ffb96af0010
Offs of buf4 =0x7ffb96af0140
Threads #: 8 Pthreads
Matrix size: 2048
Using multiply kernel: multiply1
Execution time = 18.360 seconds
[ perf record: Woken up 442 times to write data ]
[ perf record: Compressed 191.773 MB to 24.238 MB, ratio is 7.912 ]
[ perf record: Captured and wrote 24.290 MB perf.data ]
143.76user 0.49system 0:18.50elapsed 779%CPU (0avgtext+0avgdata 100596maxresident)k 0inputs+49760outputs (0major+35276minor)pagefaults 0swaps
+ for i in 0 1 2 3 5 8
+/usr/bin/timetools/perf/perfrecord --aio=1 -z 3 -F 25000 -N -B -T -R -e cycles -- ../../matrix/linux/matrix.gcc Addr of buf1 =0x7f13f2df2010
Offs of buf1 =0x7f13f2df2180
Addr of buf2 =0x7f13f0df1010
Offs of buf2 =0x7f13f0df11c0
Addr of buf3 =0x7f13eedf0010
Offs of buf3 =0x7f13eedf0100
Addr of buf4 =0x7f13ecdef010
Offs of buf4 =0x7f13ecdef140
Threads #: 8 Pthreads
Matrix size: 2048
Using multiply kernel: multiply1
Execution time = 18.383 seconds
[ perf record: Woken up 499 times to write data ]
[ perf record: Compressed 191.078 MB to 23.246 MB, ratio is 8.220 ]
[ perf record: Captured and wrote 23.282 MB perf.data ]
143.72user 0.34system 0:18.51elapsed 778%CPU (0avgtext+0avgdata 100616maxresident)k 0inputs+47704outputs (0major+35783minor)pagefaults 0swaps
+ for i in 0 1 2 3 5 8
+/usr/bin/timetools/perf/perfrecord --aio=1 -z 5 -F 25000 -N -B -T -R -e cycles -- ../../matrix/linux/matrix.gcc Addr of buf1 =0x7fca0d091010
Offs of buf1 =0x7fca0d091180
Addr of buf2 =0x7fca0b090010
Offs of buf2 =0x7fca0b0901c0
Addr of buf3 =0x7fca0908f010
Offs of buf3 =0x7fca0908f100
Addr of buf4 =0x7fca0708e010
Offs of buf4 =0x7fca0708e140
Threads #: 8 Pthreads
Matrix size: 2048
Using multiply kernel: multiply1
Execution time = 18.758 seconds
[ perf record: Woken up 535 times to write data ]
[ perf record: Compressed 190.065 MB to 24.430 MB, ratio is 7.780 ]
[ perf record: Captured and wrote 24.519 MB perf.data ]
144.62user 0.66system 0:18.88elapsed 769%CPU (0avgtext+0avgdata 100528maxresident)k 0inputs+50232outputs (0major+36942minor)pagefaults 0swaps
+ for i in 0 1 2 3 5 8
+/usr/bin/timetools/perf/perfrecord --aio=1 -z 8 -F 25000 -N -B -T -R -e cycles -- ../../matrix/linux/matrix.gcc Addr of buf1 =0x7f7e1f449010
Offs of buf1 =0x7f7e1f449180
Addr of buf2 =0x7f7e1d448010
Offs of buf2 =0x7f7e1d4481c0
Addr of buf3 =0x7f7e1b447010
Offs of buf3 =0x7f7e1b447100
Addr of buf4 =0x7f7e19446010
Offs of buf4 =0x7f7e19446140
Threads #: 8 Pthreads
Matrix size: 2048
Using multiply kernel: multiply1
Execution time = 20.103 seconds
[ perf record: Woken up 260 times to write data ]
[ perf record: Compressed 193.024 MB to 31.588 MB, ratio is 6.111 ]
[ perf record: Captured and wrote 32.139 MB perf.data ]
151.73user 4.21system 0:20.23elapsed 770%CPU (0avgtext+0avgdata 100616maxresident)k 0inputs+65848outputs (0major+37431minor)pagefaults 0swaps---
TESTING:$ tools/perf/perf version --build-options perf version 4.13.rc5.gd8d056b dwarf: [ on ] # HAVE_DWARF_SUPPORT dwarf_getlocations: [ on ] # HAVE_DWARF_GETLOCATIONS_SUPPORT glibc: [ on ] # HAVE_GLIBC_SUPPORT gtk2: [ on ] # HAVE_GTK2_SUPPORT syscall_table: [ on ] # HAVE_SYSCALL_TABLE_SUPPORT libbfd: [ on ] # HAVE_LIBBFD_SUPPORT libelf: [ on ] # HAVE_LIBELF_SUPPORT libnuma: [ on ] # HAVE_LIBNUMA_SUPPORT numa_num_possible_cpus: [ on ] # HAVE_LIBNUMA_SUPPORT libperl: [ on ] # HAVE_LIBPERL_SUPPORT libpython: [ on ] # HAVE_LIBPYTHON_SUPPORT libslang: [ on ] # HAVE_SLANG_SUPPORT libcrypto: [ on ] # HAVE_LIBCRYPTO_SUPPORT libunwind: [ on ] # HAVE_LIBUNWIND_SUPPORT libdw-dwarf-unwind: [ on ] # HAVE_DWARF_SUPPORT zlib: [ on ] # HAVE_ZLIB_SUPPORT lzma: [ on ] # HAVE_LZMA_SUPPORT get_cpuid: [ on ] # HAVE_AUXTRACE_SUPPORT bpf: [ on ] # HAVE_LIBBPF_SUPPORT aio: [ OFF ] # HAVE_AIO_SUPPORT zstd: [ OFF ] # HAVE_ZSTD_SUPPORT$ tools/perf/perf version --build-options perf version 4.13.rc5.gd8d056b dwarf: [ on ] # HAVE_DWARF_SUPPORT dwarf_getlocations: [ on ] # HAVE_DWARF_GETLOCATIONS_SUPPORT glibc: [ on ] # HAVE_GLIBC_SUPPORT gtk2: [ on ] # HAVE_GTK2_SUPPORT syscall_table: [ on ] # HAVE_SYSCALL_TABLE_SUPPORT libbfd: [ on ] # HAVE_LIBBFD_SUPPORT libelf: [ on ] # HAVE_LIBELF_SUPPORT libnuma: [ on ] # HAVE_LIBNUMA_SUPPORT numa_num_possible_cpus: [ on ] # HAVE_LIBNUMA_SUPPORT libperl: [ on ] # HAVE_LIBPERL_SUPPORT libpython: [ on ] # HAVE_LIBPYTHON_SUPPORT libslang: [ on ] # HAVE_SLANG_SUPPORT libcrypto: [ on ] # HAVE_LIBCRYPTO_SUPPORT libunwind: [ on ] # HAVE_LIBUNWIND_SUPPORT libdw-dwarf-unwind: [ on ] # HAVE_DWARF_SUPPORT zlib: [ on ] # HAVE_ZLIB_SUPPORT lzma: [ on ] # HAVE_LZMA_SUPPORT get_cpuid: [ on ] # HAVE_AUXTRACE_SUPPORT bpf: [ on ] # HAVE_LIBBPF_SUPPORT aio: [ on ] # HAVE_AIO_SUPPORT zstd: [ on ] # HAVE_ZSTD_SUPPORT$ make -C tools/perf clean all ... $ pushd tools/perf/ && ./perf test && popd ~/abudanko/kernel/acme/tools/perf ~/abudanko/kernel/acme 1: vmlinux symtab matches kallsyms : Skip 2: Detect openat syscall event : Ok 3: Detect openat syscall event on all cpus : Ok 4: Read samples using the mmap interface : Ok 5: Test data source output : Ok 6: Parse event definition strings : Ok 7: Simple expression parser : Ok 8: PERF_RECORD_* events & perf_sample fields : Ok 9: Parse perf pmu format : Ok 10: DSO data read : Ok 11: DSO data cache : Ok 12: DSO data reopen : Ok 13: Roundtrip evsel->name : Ok 14: Parse sched tracepoints fields : Ok 15: syscalls:sys_enter_openat event fields : Ok 16: Setup struct perf_event_attr : Ok 17: Match and link multiple hists : Ok 18: 'import perf' in python : Ok 19: Breakpoint overflow signal handler : Ok 20: Breakpoint overflow sampling : Ok 21: Breakpoint accounting : Ok 22: Watchpoint : 22.1: Read Only Watchpoint : Skip 22.2: Write Only Watchpoint : Ok 22.3: Read / Write Watchpoint : Ok 22.4: Modify Watchpoint : Ok 23: Number of exit events of a simple workload : Ok 24: Software clock events period values : Ok 25: Object code reading : Ok 26: Sample parsing : Ok 27: Use a dummy software event to keep tracking : Ok 28: Parse with no sample_id_all bit set : Ok 29: Filter hist entries : Ok 30: Lookup mmap thread : Ok 31: Share thread mg : Ok 32: Sort output of hist entries : Ok 33: Cumulate child hist entries : Ok 34: Track with sched_switch : Ok 35: Filter fds with revents mask in a fdarray : Ok 36: Add fd to a fdarray, making it autogrow : Ok 37: kmod_path__parse : Ok 38: Thread map : Ok 39: LLVM search and compile : 39.1: Basic BPF llvm compile : Skip 39.2: kbuild searching : Skip 39.3: Compile source for BPF prologue generation : Skip 39.4: Compile source for BPF relocation : Skip 40: Session topology : Ok 41: BPF filter : 41.1: Basic BPF filtering : Skip 41.2: BPF pinning : Skip 41.3: BPF prologue generation : Skip 41.4: BPF relocation checker : Skip 42: Synthesize thread map : Ok 43: Remove thread map : Ok 44: Synthesize cpu map : Ok 45: Synthesize stat config : Ok 46: Synthesize stat : Ok 47: Synthesize stat round : Ok 48: Synthesize attr update : Ok 49: Event times : Ok 50: Read backward ring buffer : Ok 51: Print cpu map : Ok 52: Probe SDT events : Ok 53: is_printable_array : Ok 54: Print bitmap : Ok 55: perf hooks : Ok 56: builtin clang support : Skip (not compiled in) 57: unit_number__scnprintf : Ok 58: mem2node : Ok 59: x86 rdpmc : Ok 60: Convert perf time to TSC : Ok 61: DWARF unwind : Ok 62: x86 instruction decoder - new instructions : Ok 63: x86 bp modify : Ok 64: Check open filename arg using perf trace + vfs_getname: Skip 65: Add vfs_getname probe to get syscall args filenames : Skip 66: probe libc's inet_pton & backtrace it with ping : Ok 67: Use vfs_getname probe to get syscall args filenames : Skip 68: record trace Zstd compression/decompression : Ok ~/abudanko/kernel/acme$ make -C tools/perf NO_LIBZSTD=1 clean all ... $ pushd tools/perf/ && ./perf test && popd ~/abudanko/kernel/acme/tools/perf ~/abudanko/kernel/acme ... 68: record trace Zstd compression/decompression : Skip ~/abudanko/kernel/acme