Re: "perf list" built from git repository Segmentation fault with 4.4.0 kernels

[Arnaldo Carvalho de Melo <acme@kernel.org>]

Em Thu, Dec 10, 2015 at 05:54:03PM -0500, William Cohen escreveu:

> I have noticed that with the 4.4.0 kernels "perf list" has been
> segfaulting on both aarch64 and x86_64.  Below is the failure on
> fedora rawhide running on x86_64:

Ok, this was fixed, but just on the perf/core branch:

"perf list: Add support for PERF_COUNT_SW_BPF_OUT"
https://git.kernel.org/cgit/linux/kernel/git/acme/linux.git/commit/tools/perf/util/parse-events.c?h=perf/core&id=bae9cc41105b9edd74d68a9636be2ba240e74b9e

And then this one on top to avoid such things in the future:
"perf list: Robustify event printing routine"
https://git.kernel.org/cgit/linux/kernel/git/acme/linux.git/commit/tools/perf/util/parse-events.c?h=perf/core&id=e37df6c76cb19971f1228bfaff504d8a3ea6f748

I'll cherry-pick those and push to Ingo.

Thanks for the report!

- Arnaldo

> [wcohen@localhost ~]$ rpm -q perf
> perf-4.4.0-0.rc4.git0.1.fc24.x86_64
> [wcohen@localhost ~]$ gdb /bin/perf ./core.2110 
> GNU gdb (GDB) Fedora 7.10.50.20151113-33.fc24
> Copyright (C) 2015 Free Software Foundation, Inc.
> License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
> This is free software: you are free to change and redistribute it.
> There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
> and "show warranty" for details.
> This GDB was configured as "x86_64-redhat-linux-gnu".
> Type "show configuration" for configuration details.
> For bug reporting instructions, please see:
> <http://www.gnu.org/software/gdb/bugs/>.
> Find the GDB manual and other documentation resources online at:
> <http://www.gnu.org/software/gdb/documentation/>.
> For help, type "help".
> Type "apropos word" to search for commands related to "word"...
> Reading symbols from /bin/perf...Reading symbols from /usr/lib/debug/usr/bin/perf.debug...done.
> done.
> [New LWP 2110]
> [Thread debugging using libthread_db enabled]
> Using host libthread_db library "/lib64/libthread_db.so.1".
> Core was generated by `perf list'.
> Program terminated with signal SIGSEGV, Segmentation fault.
> #0  0x000055b1fc97fdce in print_symbol_events (type=<optimized out>, 
>     syms=0x55b1fcd81580 <event_symbols_sw+160>, max=<optimized out>, 
>     name_only=<optimized out>, event_glob=<optimized out>)
>     at util/parse-events.c:1895
> 1895			if (!name_only && strlen(syms->alias))
> Missing separate debuginfos, use: dnf debuginfo-install bzip2-libs-1.0.6-19.fc24.x86_64 nss-softokn-freebl-3.21.0-3.fc24.x86_64
> (gdb) where
> #0  0x000055b1fc97fdce in print_symbol_events (type=<optimized out>, 
>     syms=0x55b1fcd81580 <event_symbols_sw+160>, max=<optimized out>, 
>     name_only=<optimized out>, event_glob=<optimized out>)
>     at util/parse-events.c:1895
> #1  print_events (event_glob=event_glob@entry=0x0, name_only=false)
>     at util/parse-events.c:1943
> #2  0x000055b1fc90a63e in cmd_list (argc=0, argv=0x7ffde1523f80, 
>     prefix=<optimized out>) at builtin-list.c:43
> #3  0x000055b1fc95c413 in run_builtin (p=0x55b1fcd803d8 <commands+120>, 
>     argc=1, argv=0x7ffde1523f80) at perf.c:386
> #4  0x000055b1fc8fe308 in handle_internal_command (argv=0x7ffde1523f80, argc=1)
>     at perf.c:447
> #5  run_argv (argv=0x7ffde1523d10, argcp=0x7ffde1523d1c) at perf.c:491
> #6  main (argc=1, argv=0x7ffde1523f80) at perf.c:608
> (gdb) 
> 
> 
> I did a git bisect on the kernel source code to find out when "perf list" broke and the git bisect pointed to the following commit causing the problem:
> 
> # git bisect good
> a43eec304259a6c637f4014a6d4767159b6a3aa3 is the first bad commit
> commit a43eec304259a6c637f4014a6d4767159b6a3aa3
> Author: Alexei Starovoitov <ast@plumgrid.com>
> Date:   Tue Oct 20 20:02:34 2015 -0700
> 
>     bpf: introduce bpf_perf_event_output() helper
>     
>     This helper is used to send raw data from eBPF program into
>     special PERF_TYPE_SOFTWARE/PERF_COUNT_SW_BPF_OUTPUT perf_event.
>     User space needs to perf_event_open() it (either for one or all cpus) and
>     store FD into perf_event_array (similar to bpf_perf_event_read() helper)
>     before eBPF program can send data into it.
>     
>     Today the programs triggered by kprobe collect the data and either store
>     it into the maps or print it via bpf_trace_printk() where latter is the debug
>     facility and not suitable to stream the data. This new helper replaces
>     such bpf_trace_printk() usage and allows programs to have dedicated
>     channel into user space for post-processing of the raw data collected.
>     
>     Signed-off-by: Alexei Starovoitov <ast@kernel.org>
>     Signed-off-by: David S. Miller <davem@davemloft.net>
> 
> :040000 040000 322abd03170df11e43da73d3fa4f74f1826788e9 14f3ee14b7fb4b6c76daf18e026cf67acd3154fa M	include
> :040000 040000 f84baec2884fe97ec501ae8d45bbb8a7aad3a578 ea440aeb13f1dd25aa8f2e1dadfc32dbf4b3d2b3 M	kernel
> 
> 
> -Will
> --
> To unsubscribe from this list: send the line "unsubscribe linux-perf-users" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html