Re: Implement lbr-as-callgraph v10

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

Em Wed, Nov 19, 2014 at 11:54:50AM +0100, Jiri Olsa escreveu:
> > > BUILD_BUG_ON then?
> > 
> > sounds gut
> 
> hum, acme/perf/core changed and so has the compile error ;-)
> we dont overload the <linux/bug.h>, so the kernel one got
> included, which is wrong.. attached patch fixes that

I had fixed this but not force pushed out, sorry.

Now I mistakenly tried running:

perf report --stdio --no-children --branch-history

on a file that has no BRANCH_STACK, i.e. a perf.data file on a wrong
directory since I'm comparing the output of --stdio, --tui and --gtk,
since it looks --gtk is wrong, still unsure about what the problem is in
that case, but stumbled on:

--branch-history works with this perf.data:

[acme@zoo ~]$ perf evlist --verbose
cycles: sample_freq=4000, size: 96, sample_type:
IP|TID|TIME|CALLCHAIN|PERIOD|BRANCH_STACK, disabled: 1, inherit: 1,
mmap: 1, mmap2: 1, comm: 1, comm_exec: 1, freq: 1, enable_on_exec: 1,
sample_id_all: 1, exclude_guest: 1, branch_sample_type: 8
[acme@zoo ~]$


But not with this one:

[root@zoo ~]# perf evlist --verbose
cycles: sample_freq=4000, size: 96, sample_type:
IP|TID|TIME|CALLCHAIN|CPU|PERIOD, disabled: 1, inherit: 1, mmap: 1,
mmap2: 1, comm: 1, comm_exec: 1, freq: 1, sample_id_all: 1,
exclude_guest: 1
[root@zoo ~]# 

Where I get tons of:

[root@zoo ~]# perf report --stdio --no-children --branch-history
# To display the perf.data header info, please use --header/--header-only options.
#
BFD: Dwarf Error: Offset (994443) greater than or equal to .debug_str size (156124).
BFD: Dwarf Error: Offset (1162200) greater than or equal to .debug_str size (156124).
BFD: Dwarf Error: Offset (1929446660) greater than or equal to .debug_str size (1940537).
BFD: Dwarf Error: Offset (465055236) greater than or equal to .debug_str size (156124).
BFD: Dwarf Error: Could not find abbrev number 736412.
BFD: Dwarf Error: Offset (470186753) greater than or equal to .debug_str size (1940537).
<SNIP>
BFD: Dwarf Error: Offset (3525378816) greater than or equal to .debug_str size (156124).
BFD: Dwarf Error: Offset (1483770) greater than or equal to .debug_str size (156124).
BFD: Dwarf Error: Could not find abbrev number 127.
^C^C^C^C^C^C
^C^C^C^Z

and finally a:


Segmentation fault (core dumped)
[root@zoo ~]#

Rebuilding with DEBUG=1 I get this backtrace:

Using host libthread_db library "/lib64/libthread_db.so.1".
Core was generated by `perf report --stdio --no-children --branch-history'.
Program terminated with signal SIGSEGV, Segmentation fault.
#0  0x000000000053ab10 in read_unsigned_leb128 ()
Missing separate debuginfos, use: debuginfo-install bzip2-libs-1.0.6-9.fc20.x86_64 elfutils-libelf-0.160-1.fc20.x86_64 elfutils-libs-0.160-1.fc20.x86_64 libunwind-1.1-3.fc20.x86_64 nss-softokn-freebl-3.17.2-1.fc20.x86_64 numactl-libs-2.0.9-2.fc20.x86_64 slang-2.2.4-11.fc20.x86_64 xz-libs-5.1.2-12alpha.fc20.x86_64
(gdb) bt
#0  0x000000000053ab10 in read_unsigned_leb128 ()
#1  0x00000000005485d7 in find_abstract_instance_name.isra ()
#2  0x0000000000548756 in find_abstract_instance_name.isra ()
#3  0x0000000000548756 in find_abstract_instance_name.isra ()
#4  0x0000000000548d31 in scan_unit_for_symbols ()
#5  0x00000000005496c9 in comp_unit_find_nearest_line ()
#6  0x000000000054a6ca in find_line.part ()
#7  0x00000000005606ca in _bfd_elf_find_nearest_line_discriminator ()
#8  0x00000000005607db in _bfd_elf_find_nearest_line ()
#9  0x00000000004dabf1 in find_address_in_section (abfd=0xfd522f0, section=0xe8be528, data=0xfa50c30) at util/srcline.c:100
#10 0x000000000053cbdc in bfd_map_over_sections ()
#11 0x00000000004dae84 in addr2line (dso_name=0xe89fe70 "/usr/lib/debug/usr/lib64/firefox/libxul.so.debug", addr=24667432, file=0x7fff3981f9c0, line=0x7fff3981f9bc, dso=0x22a0b50) at util/srcline.c:168
#12 0x00000000004db017 in get_srcline (dso=0x22a0b50, addr=24667432, sym=0xf6b93f0, show_sym=true) at util/srcline.c:276
#13 0x00000000004c5a31 in sort__srcline_cmp (left=0x39494a0, right=0x7fff3981fb80) at util/sort.c:303
#14 0x00000000004cbe96 in hist_entry__cmp (left=0x39494a0, right=0x7fff3981fb80) at util/hist.c:916
#15 0x00000000004caa97 in add_hist_entry (hists=0x229e688, entry=0x7fff3981fb80, al=0x7fff3981fe00, sample_self=true) at util/hist.c:397
#16 0x00000000004cae19 in __hists__add_entry (hists=0x229e688, al=0x7fff3981fe00, sym_parent=0x0, bi=0x0, mi=0x0, period=770586, weight=0, transaction=0, sample_self=true) at util/hist.c:477
#17 0x00000000004cb585 in iter_add_single_normal_entry (iter=0x7fff3981fe30, al=0x7fff3981fe00) at util/hist.c:668
#18 0x00000000004cbcfe in hist_entry_iter__add (iter=0x7fff3981fe30, al=0x7fff3981fe00, evsel=0x229e570, sample=0x7fff39820020, max_stack_depth=127, arg=0x7fff39820510) at util/hist.c:876
#19 0x000000000043206c in process_sample_event (tool=0x7fff39820510, event=0x7fcb0c4d8040, sample=0x7fff39820020, evsel=0x229e570, machine=0x229c9d0) at builtin-report.c:182
#20 0x00000000004afe4e in perf_session__deliver_sample (session=0x229c910, tool=0x7fff39820510, event=0x7fcb0c4d8040, sample=0x7fff39820020, evsel=0x229e570, machine=0x229c9d0) at util/session.c:805
#21 0x00000000004affeb in perf_session__deliver_event (session=0x229c910, event=0x7fcb0c4d8040, sample=0x7fff39820020, tool=0x7fff39820510, file_offset=1146944) at util/session.c:842
#22 0x00000000004b2ee4 in __ordered_events__flush (s=0x229c910, tool=0x7fff39820510) at util/ordered-events.c:186
#23 0x00000000004b31e6 in ordered_events__flush (s=0x229c910, tool=0x7fff39820510, how=OE_FLUSH__FINAL) at util/ordered-events.c:251
#24 0x00000000004b1236 in __perf_session__process_events (session=0x229c910, data_offset=216, data_size=1220024, file_size=1220240, tool=0x7fff39820510) at util/session.c:1318
#25 0x00000000004b132e in perf_session__process_events (session=0x229c910, tool=0x7fff39820510) at util/session.c:1337
#26 0x0000000000432d5b in __cmd_report (rep=0x7fff39820510) at builtin-report.c:483
#27 0x00000000004341ca in cmd_report (argc=0, argv=0x7fff39821950, prefix=0x0) at builtin-report.c:851
#28 0x000000000041c8bc in run_builtin (p=0x857ec8 <commands+168>, argc=4, argv=0x7fff39821950) at perf.c:331
#29 0x000000000041cb1b in handle_internal_command (argc=4, argv=0x7fff39821950) at perf.c:390
#30 0x000000000041cc67 in run_argv (argcp=0x7fff398217ac, argv=0x7fff398217a0) at perf.c:434
#31 0x000000000041cfbe in main (argc=4, argv=0x7fff39821950) at perf.c:549
(gdb)

Haven't checked yet, but I guess we need to check, when --branch-history is
passed, if the sample_type has BRANCH_STACK, etc.

Now back to checking why the --gtk output doesn't match, then I go back to this
other bug.

- Arnaldo