'branches' perf event mapping differs on ARC and ARM

'branches' perf event mapping differs on ARC and ARM

[Eugeniy Paltsev]

Hi,

While playing with perf tool on ARMv7 and ARCv2 processors and profiling the
same application I got interesting results. Even if we got pretty
similar total
execution time and instructions number the number of branches on ARC is about
three times more then on ARM.

I dug into architecture
specific perf sources and found that we map different
HW counters into generic 'branches' event on ARC and ARM.
- We use "ijmp" event on ARC which
counts all jump and branch instructions (regardless
of real execution flow - even if no real jump happens)
- We use "pc_write_retired" event on ARM
which counts only taken branches (Instruction
architecturally executed, condition check pass - software change of the PC)

So I was wondering do you know
which approach is correct?



I guess counting all jump and branch instructions is correct because we use
'branches' event value to calculate relative value of 'branch-misses' using

following formula:
----------------------------8----------------------------
branch-misses-ration = 'branch-misses' / 'branches' * 100.0
----------------
------------8----------------------------
And using only taken branches here is incorrect IMHO. So I guess we should
map 'br_immed_retired' instead of
"pc_write_retired" into generic 'branches'
event on ARM.
-- 
 Eugeniy Paltsev

'branches' perf event mapping differs on ARC and ARM

[Robin Murphy]

On 27/11/2018 14:36, Eugeniy Paltsev wrote:
> Hi,
> 
> While playing with perf tool on ARMv7 and ARCv2 processors and profiling the
> same application I got interesting results. Even if we got pretty
> similar total
> execution time and instructions number the number of branches on ARC is about
> three times more then on ARM.
> 
> I dug into architecture
> specific perf sources and found that we map different
> HW counters into generic 'branches' event on ARC and ARM.
> - We use "ijmp" event on ARC which
> counts all jump and branch instructions (regardless
> of real execution flow - even if no real jump happens)
> - We use "pc_write_retired" event on ARM
> which counts only taken branches (Instruction
> architecturally executed, condition check pass - software change of the PC)
> 
> So I was wondering do you know
> which approach is correct?
> 
> 
> 
> I guess counting all jump and branch instructions is correct because we use
> 'branches' event value to calculate relative value of 'branch-misses' using
> 
> following formula:
> ----------------------------8----------------------------
> branch-misses-ration = 'branch-misses' / 'branches' * 100.0
> ----------------
> ------------8----------------------------
> And using only taken branches here is incorrect IMHO. So I guess we should
> map 'br_immed_retired' instead of
> "pc_write_retired" into generic 'branches'
> event on ARM.

But register branches (including procedure returns) are also branches, 
so only counting immediate branches would be just as, if not more, 
misleading.

Robin.

'branches' perf event mapping differs on ARC and ARM

[Vineet Gupta]

On 11/27/18 6:36 AM, Eugeniy Paltsev wrote:
> Hi,
>
> While playing with perf tool on ARMv7 and ARCv2 processors and profiling the
> same application I got interesting results. Even if we got pretty
> similar total
> execution time and instructions number the number of branches on ARC is about
> three times more then on ARM.
>
> I dug into architecture
> specific perf sources and found that we map different
> HW counters into generic 'branches' event on ARC and ARM.
> - We use "ijmp" event on ARC which
> counts all jump and branch instructions (regardless
> of real execution flow - even if no real jump happens)

That doesn't seem correct IMO. A NOT taken conditional branch doesn't change
control flow, so semantically doesn't qualify as a branch.
On ARC, the generic branches event should be mapped to "actually taken branches"
condition i.e.  ijmptak


> - We use "pc_write_retired" event on ARM
> which counts only taken branches (Instruction
> architecturally executed, condition check pass - software change of the PC)

That seems correct.

> I guess counting all jump and branch instructions is correct because we use
> 'branches' event value to calculate relative value of 'branch-misses' using
>
> following formula:
> ----------------------------8----------------------------
> branch-misses-ration = 'branch-misses' / 'branches' * 100.0
> ----------------
> ------------8----------------------------
> And using only taken branches here is incorrect IMHO.

Why ? branch-misses is a CPU specific micro-arch state where the a changed control
flow was NOT predicted. If an implementation mispredicts NOT taken branches, those
should actually get counted and be fed to hardware folks to improve the micro-arch.


>  So I guess we should
> map 'br_immed_retired' instead of
> "pc_write_retired" into generic 'branches'
> event on ARM.