Create_gcov always dump 1600 bytes

[nitishroy84]

I am trying to covert captured perf data to gcc format.But always 1600 bytes are dumped.

Machine where I captured perf data is having LBR support. perf record -o /tmp//perf.data -b -e cpu/event=0xc4,umask=0x20,name=br_inst_retired_near_taken,period=400009/pp -p 11045 -F 10 -- sleep 900

Then I used following command to create gcov data create_gcov --binary=<> --profile=perf.data --gcov=perf-prod.afdo --profiler=perf -log_dir=. -gcov_version=1

Binary is unstripped and have debug symbols

In log following Log file created at: 2019/05/10 12:04:54 Running on machine: a107-2-2-3 Log line format: [IWEF]mmdd hh:mm:ss.uuuuuu threadid file:line] msg W0510 12:04:54.359030 5412 module_grouper.cc:60] Cannot read compilation info from binary. Please use -frecord-compilation-info-in-elf when building the binary I0510 12:04:55.822785 5412 perf_reader.cc:798] Number of events stored: 1164694 I0510 12:05:03.224366 5412 perf_parser.cc:256] Parser processed: 1066560 MMAP/MMAP2 events, 405 COMM events, 0 FORK events, 0 EXIT events, 90695 SAMPLE events, 73020 of these were mapped

There is no other information.I tried many combinations but it is not working.What could be the issue?

[nitishroy84]

I have used -ggdb for symbols at the time of binary compilation.Will it effect anyways?

[fwsGonzo]

I had the same issue. I solved this by using ocperf from pmu-tools, as it supports perf counters directly, instead of using perf directly. After that I got a crash in gcc-8 when building with auto-profile, but I solved it by upgrading to gcc-9.

[jwbee]

How does that solve the problem? I get the same behavior whether I use perf or ocperf.py.

[dyan2274]

How to solve this problem? I already used in gcc9.3 and ocperf.py.

[ganli2015]

Add option -use_lbr=0 Maybe your machine does not support lbr.