Re: [PATCH v2 6/7] perf diff: Print the basic block cycles diff

[Jin, Yao]

On 6/5/2019 7:44 PM, Jiri Olsa wrote:
> On Mon, Jun 03, 2019 at 10:36:16PM +0800, Jin Yao wrote:
> > perf record -b ./div
> > perf record -b ./div
> >
> > Following is the default perf diff output
> >
> >   # perf diff
> >
> >   # Event 'cycles'
> >   #
> >   # Baseline  Delta Abs  Shared Object     Symbol
> >   # ........  .........  ................  ....................................
> >   #
> >       49.03%     +0.30%  div               [.] main
> >       16.29%     -0.20%  libc-2.23.so      [.] __random
> >       18.82%     -0.07%  libc-2.23.so      [.] __random_r
> >        8.11%     -0.04%  div               [.] compute_flag
> >        2.25%     +0.01%  div               [.] rand@plt
> >        0.00%     +0.01%  [kernel.vmlinux]  [k] task_tick_fair
> >        5.46%     +0.01%  libc-2.23.so      [.] rand
> >        0.01%     -0.01%  [kernel.vmlinux]  [k] native_irq_return_iret
> >        0.00%     -0.00%  [kernel.vmlinux]  [k] interrupt_entry
> >
> > This patch creates a new computation selection 'cycles'.
> >
> >   # perf diff -c cycles
> >
> >   # Event 'cycles'
> >   #
> >   # Baseline         Block cycles diff [start:end]  Shared Object     Symbol
> >   # ........  ....................................  ................  ....................................
> >   #
> >       49.03%        -9 [         4ef:         520]  div               [.] main
> >       49.03%         0 [         4e8:         4ea]  div               [.] main
> >       49.03%         0 [         4ef:         500]  div               [.] main
> >       49.03%         0 [         4ef:         51c]  div               [.] main
> >       49.03%         0 [         4ef:         535]  div               [.] main
> >       18.82%         0 [       3ac40:       3ac4d]  libc-2.23.so      [.] __random_r
> >       18.82%         0 [       3ac40:       3ac5c]  libc-2.23.so      [.] __random_r
> >       18.82%         0 [       3ac40:       3ac76]  libc-2.23.so      [.] __random_r
> >       18.82%         0 [       3ac40:       3ac88]  libc-2.23.so      [.] __random_r
> >       18.82%         0 [       3ac90:       3ac9c]  libc-2.23.so      [.] __random_r
> >       16.29%        -8 [       3aac0:       3aac0]  libc-2.23.so      [.] __random
> >       16.29%         0 [       3aac0:       3aad2]  libc-2.23.so      [.] __random
> >       16.29%         0 [       3aae0:       3aae7]  libc-2.23.so      [.] __random
> >       16.29%         0 [       3ab03:       3ab0f]  libc-2.23.so      [.] __random
> >       16.29%         0 [       3ab14:       3ab1b]  libc-2.23.so      [.] __random
> >       16.29%         0 [       3ab28:       3ab2e]  libc-2.23.so      [.] __random
> >       16.29%         0 [       3ab4a:       3ab53]  libc-2.23.so      [.] __random
> >        8.11%         0 [         640:         644]  div               [.] compute_flag
> >        8.11%         0 [         649:         659]  div               [.] compute_flag
> >        5.46%         0 [       3af60:       3af60]  libc-2.23.so      [.] rand
> >        5.46%         0 [       3af60:       3af64]  libc-2.23.so      [.] rand
> >        2.25%         0 [         490:         490]  div               [.] rand@plt
> >        0.01%        26 [      c00a27:      c00a27]  [kernel.vmlinux]  [k] native_irq_return_iret
> >        0.00%      -157 [      2bf9f2:      2bfa63]  [kernel.vmlinux]  [k] update_blocked_averages
> >        0.00%       -56 [      2bf980:      2bf9d3]  [kernel.vmlinux]  [k] update_blocked_averages
> >        0.00%        48 [      2bf934:      2bf942]  [kernel.vmlinux]  [k] update_blocked_averages
> >        0.00%         3 [      2bfb38:      2bfb67]  [kernel.vmlinux]  [k] update_blocked_averages
> >        0.00%         0 [      2bf968:      2bf97b]  [kernel.vmlinux]  [k] update_blocked_averages
>
> so what I'd expect would be Baseline column with cycles and another
> column showing the differrence (in cycles) for given symbol
>
> > "[start:end]" indicates the basic block range. The output is sorted
> > by "Baseline" and the basic blocks in the same function are sorted
> > by cycles diff.
>
> hum, why is there multiple basic blocks [start:end] for a symbol?
>
> thanks,
> jirka

The basic block is the code between 2 branches (for one branch, for 
example, jmp, call, ret, interrupt, ...). So it's expected that one 
function (symbol is function) may contain multiple basic blocks.

The idea is, sorting by baseline to display the hottest functions and 
the second column shows the cycles diff of blocks in this function 
(comparing between different perf data files). This would allow to 
identify performance changes in specific code accurately and effectively.

Thanks
Jin Yao