lkml.org 
[lkml]   [2014]   [Nov]   [6]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
Date
SubjectRe: [PATCH 1/1] perf tools: perf diff for different binaries
Hi Kan,

On Thu, Nov 6, 2014 at 2:28 AM, Liang, Kan <kan.liang@intel.com> wrote:
>
>> Hi Kan,
>>
>> On Tue, 4 Nov 2014 17:07:43 +0000, Kan Liang wrote:
>> >> What about setting the
>> >> sort_sym.se_collapse in data_process() so that hists__match() can use
>> >> symbol names?
>> >
>> > Yes, we can set it if we only do function level diff. But I'd like to
>> > keep both. So I defined two sort keys.
>> > "symbol" means "symbol address executed at the time of sample "
>> > "symbol_name" means "name of function executed at the time of
>> sample"
>>
>> Hmm.. I don't think the symbol sort key provides the instruction level diff
>> that you want. If it finds a symbol it just use the start address of the
>> symbol, not the exact address of the sample. Am I missing something?
>>
>
> No, the meaning of symbol in perf diff is different as in perf report. It uses
> the exact address of the sample.

Hmm.. are you sure? I think it uses same code..


>
> Here is current perf diff result and objdump fragment of the first binary.
> You can see the exact address of the sample in function f3 was used, not the
> start address.
> 33.55% [unknown] [.] 0x0000000000400554
> 48.13% [unknown] [.] 0x000000000040056b

But I think it's because it failed to find a symbol (f3) for some
reason. The symbol sort key will use the ip address only if it didn't
find a symbol for the address. Please see sort__sym_cmp (and
_sort__sym_cmp too).

Thanks,
Namhyung


>
> [perf-test]#./perf diff -s dso,symbol --compute ratio ../../../v1_1_6perf.data
> ../../../v2_1_7perf.data
> # Event 'cycles'
> #
> # Baseline Ratio Shared Object Symbol
> # ........ .............. ................. .........................
> #
> [kernel.kallsyms] [k] __update_cpu_load
> [kernel.kallsyms] [k] mod_timer
> 0.01% [kernel.kallsyms] [k] native_write_msr_safe
> 0.01% [kernel.kallsyms] [k] notifier_call_chain
> 0.01% [kernel.kallsyms] [k] perf_event_task_tick
> [kernel.kallsyms] [k] rt_mutex_slowlock
> 0.01% [kernel.kallsyms] [k] run_posix_cpu_timers
> 0.01% [kernel.kallsyms] [k] run_timer_softirq
> 0.01% [kernel.kallsyms] [k] trigger_load_balance
> 0.01% [kernel.kallsyms] [k] update_vsyscall
> [kernel.kallsyms] [k] wait_for_common
> [kernel.kallsyms] [k] 0x000000000028482a
> 0.05% [unknown] [.] 0x0000000000400540
> 0.04% [unknown] [.] 0x0000000000400541
> 0.03% [unknown] [.] 0x000000000040054b
> 0.04% [unknown] [.] 0x0000000000400552
> 33.55% [unknown] [.] 0x0000000000400554
> 1.22% [unknown] [.] 0x000000000040055a
> 8.00% [unknown] [.] 0x000000000040055e
> 0.02% [unknown] [.] 0x0000000000400562
> 8.41% [unknown] [.] 0x0000000000400564
> 48.13% [unknown] [.] 0x000000000040056b
> 0.16% [unknown] [.] 0x0000000000400570
> 0.17% [unknown] [.] 0x0000000000400571
> [unknown] [.] 0x0000000000400580
> [unknown] [.] 0x0000000000400581
> 0.01% [unknown] [.] 0x0000000000400583
> 0.01% [unknown] [.] 0x0000000000400588
> [unknown] [.] 0x000000000040058b
> 0.01% 1240.990221 [unknown] [.] 0x000000000040058d
> [unknown] [.] 0x0000000000400590
> 0.06% [unknown] [.] 0x0000000000400591
> [unknown] [.] 0x0000000000400593
> 0.04% [unknown] [.] 0x0000000000400595
> 0.01% 1240.603148 [unknown] [.] 0x0000000000400597
> [unknown] [.] 0x000000000040059b
> [unknown] [.] 0x000000000040059d
> [unknown] [.] 0x00000000004005a1
> [unknown] [.] 0x00000000004005a5
> [unknown] [.] 0x00000000004005a7
> [unknown] [.] 0x00000000004005a8
> [unknown] [.] 0x00000000004005aa
> [unknown] [.] 0x00000000004005ba
> [unknown] [.] 0x00000000004005bf
> [unknown] [.] 0x00000000004005c4
> [unknown] [.] 0x00000000004005c9
> [unknown] [.] 0x00000000004005ce
> [unknown] [.] 0x00000000004005d2
> [unknown] [.] 0x00000000004005d6
> [unknown] [.] 0x00000000004005d8
> [unknown] [.] 0x00000000004005f5
>
>
> [perf-test]#objdump -d tchain_edit_1
>
> 0000000000400540 <f3>:
> 400540: 55 push %rbp
> 400541: 48 89 e5 mov %rsp,%rbp
> 400544: c7 45 f8 e8 03 00 00 movl $0x3e8,-0x8(%rbp)
> 40054b: c7 45 fc 00 00 00 00 movl $0x0,-0x4(%rbp)
> 400552: eb 14 jmp 400568 <f3+0x28>
> 400554: 8b 45 fc mov -0x4(%rbp),%eax
> 400557: 83 e0 01 and $0x1,%eax
> 40055a: 85 c0 test %eax,%eax
> 40055c: 74 06 je 400564 <f3+0x24>
> 40055e: 83 45 fc 01 addl $0x1,-0x4(%rbp)
> 400562: eb 04 jmp 400568 <f3+0x28>
> 400564: 83 45 fc 01 addl $0x1,-0x4(%rbp)
> 400568: 8b 45 fc mov -0x4(%rbp),%eax
> 40056b: 3b 45 f8 cmp -0x8(%rbp),%eax
> 40056e: 7c e4 jl 400554 <f3+0x14>
> 400570: 5d pop %rbp
> 400571: c3 retq
>
>
> Thanks,
> Kan
>
>> Thanks,
>> Namhyung



--
Thanks,
Namhyung


\
 
 \ /
  Last update: 2014-11-06 09:01    [W:0.100 / U:0.684 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site