* Question about indirect functions and PGO
@ 2020-07-10 11:17 Erick Ochoa
2020-07-10 11:19 ` Erick Ochoa
0 siblings, 1 reply; 4+ messages in thread
From: Erick Ochoa @ 2020-07-10 11:17 UTC (permalink / raw)
To: GCC Development; +Cc: Christoph Müllner, Philipp Tomsich
Hello,
I'm working on an optimization and I encountered this interesting
behaviour. There are a couple of functions that are specialized when the
program is not compiled with PGO (-fprofile-generate and -fprofile-use)
However, when the program is compiled with PGO the compiler does not
specialize the function calls.
I printing the program just after materializing all clones.
I am running this version of GCC:
Author: GCC Administrator <gccadmin@gcc.gnu.org>
Date: Fri Jul 10 00:16:28 2020 +0000
Daily bump.
I can imagine that the profiling information was used to determine that
specializing these functions is a bad tradeoff between binary size and
speed. But I do not know this for sure. How can I find out why these
functions were not specialized? (I.e. is there a threshold that wasn't
met, and if so, where is it located and what's its value?)
Thanks!
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: Question about indirect functions and PGO
2020-07-10 11:17 Question about indirect functions and PGO Erick Ochoa
@ 2020-07-10 11:19 ` Erick Ochoa
2020-07-13 11:40 ` Erick Ochoa
0 siblings, 1 reply; 4+ messages in thread
From: Erick Ochoa @ 2020-07-10 11:19 UTC (permalink / raw)
To: GCC Development; +Cc: Christoph Müllner, Philipp Tomsich
Forgot to mention that these functions take a function pointer as a
parameter and as a result, the specialized functions are able to replace
the indirect function call with a direct function call.
On 10/07/2020 13:17, Erick Ochoa wrote:
> Hello,
>
> I'm working on an optimization and I encountered this interesting
> behaviour. There are a couple of functions that are specialized when the
> program is not compiled with PGO (-fprofile-generate and -fprofile-use)
>
> However, when the program is compiled with PGO the compiler does not
> specialize the function calls.
>
> I printing the program just after materializing all clones.
>
> I am running this version of GCC:
> Author: GCC Administrator <gccadmin@gcc.gnu.org>
> Date: Fri Jul 10 00:16:28 2020 +0000
>
> Daily bump.
>
> I can imagine that the profiling information was used to determine that
> specializing these functions is a bad tradeoff between binary size and
> speed. But I do not know this for sure. How can I find out why these
> functions were not specialized? (I.e. is there a threshold that wasn't
> met, and if so, where is it located and what's its value?)
>
> Thanks!
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: Question about indirect functions and PGO
2020-07-10 11:19 ` Erick Ochoa
@ 2020-07-13 11:40 ` Erick Ochoa
2020-07-14 0:35 ` Victor Rodriguez
0 siblings, 1 reply; 4+ messages in thread
From: Erick Ochoa @ 2020-07-13 11:40 UTC (permalink / raw)
To: GCC Development; +Cc: Christoph Müllner, Philipp Tomsich
Hi,
I just wanted to answer myself.
It seems that there are two thresholds that need to be met if a function
is to be specialized within a particular context:
1. --param=hot-bb-count-ws-permille=. This option controls the hotness
threshold of basic blocks and is needed for function specialization
during LTO. If a callsite is not in a hot basic block, it seems that the
callsite will not be specialized for a particular parameter.
2. --param=ipa-cp-eval-threshold=. This option controls a heuristic that
lets constant propagation happen if a function is a good candidate for
cloning. This parameter is used for both: function specialization within
a particular context and for all contexts.
On 10/07/2020 13:19, Erick Ochoa wrote:
> Forgot to mention that these functions take a function pointer as a
> parameter and as a result, the specialized functions are able to replace
> the indirect function call with a direct function call.
>
> On 10/07/2020 13:17, Erick Ochoa wrote:
>> Hello,
>>
>> I'm working on an optimization and I encountered this interesting
>> behaviour. There are a couple of functions that are specialized when
>> the program is not compiled with PGO (-fprofile-generate and
>> -fprofile-use)
>>
>> However, when the program is compiled with PGO the compiler does not
>> specialize the function calls.
>>
>> I printing the program just after materializing all clones.
>>
>> I am running this version of GCC:
>> Author: GCC Administrator <gccadmin@gcc.gnu.org>
>> Date: Fri Jul 10 00:16:28 2020 +0000
>>
>> Daily bump.
>>
>> I can imagine that the profiling information was used to determine
>> that specializing these functions is a bad tradeoff between binary
>> size and speed. But I do not know this for sure. How can I find out
>> why these functions were not specialized? (I.e. is there a threshold
>> that wasn't met, and if so, where is it located and what's its value?)
>>
>> Thanks!
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: Question about indirect functions and PGO
2020-07-13 11:40 ` Erick Ochoa
@ 2020-07-14 0:35 ` Victor Rodriguez
0 siblings, 0 replies; 4+ messages in thread
From: Victor Rodriguez @ 2020-07-14 0:35 UTC (permalink / raw)
To: Erick Ochoa; +Cc: GCC Development, Philipp Tomsich, Christoph Müllner
On Mon, Jul 13, 2020 at 6:41 AM Erick Ochoa
<erick.ochoa@theobroma-systems.com> wrote:
>
> Hi,
>
> I just wanted to answer myself.
> It seems that there are two thresholds that need to be met if a function
> is to be specialized within a particular context:
>
> 1. --param=hot-bb-count-ws-permille=. This option controls the hotness
> threshold of basic blocks and is needed for function specialization
> during LTO. If a callsite is not in a hot basic block, it seems that the
> callsite will not be specialized for a particular parameter.
> 2. --param=ipa-cp-eval-threshold=. This option controls a heuristic that
> lets constant propagation happen if a function is a good candidate for
> cloning. This parameter is used for both: function specialization within
> a particular context and for all contexts.
>
> On 10/07/2020 13:19, Erick Ochoa wrote:
> > Forgot to mention that these functions take a function pointer as a
> > parameter and as a result, the specialized functions are able to replace
> > the indirect function call with a direct function call.
> >
> > On 10/07/2020 13:17, Erick Ochoa wrote:
> >> Hello,
> >>
> >> I'm working on an optimization and I encountered this interesting
> >> behaviour. There are a couple of functions that are specialized when
> >> the program is not compiled with PGO (-fprofile-generate and
> >> -fprofile-use)
> >>
> >> However, when the program is compiled with PGO the compiler does not
> >> specialize the function calls.
> >>
> >> I printing the program just after materializing all clones.
> >>
> >> I am running this version of GCC:
> >> Author: GCC Administrator <gccadmin@gcc.gnu.org>
> >> Date: Fri Jul 10 00:16:28 2020 +0000
> >>
> >> Daily bump.
> >>
> >> I can imagine that the profiling information was used to determine
> >> that specializing these functions is a bad tradeoff between binary
> >> size and speed. But I do not know this for sure. How can I find out
> >> why these functions were not specialized? (I.e. is there a threshold
> >> that wasn't met, and if so, where is it located and what's its value?)
> >>
Have you tried the new option-fprofile-partial-training ?
-fprofile-partial-training can now be used to inform the compiler that
code paths not covered by the training run should not be optimized for
size.
I tested by myself this week and it literally do not optimize at all
the functions not touched by the training
Regards
Victor Rodriguez
> >> Thanks!
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2020-07-14 0:35 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-07-10 11:17 Question about indirect functions and PGO Erick Ochoa
2020-07-10 11:19 ` Erick Ochoa
2020-07-13 11:40 ` Erick Ochoa
2020-07-14 0:35 ` Victor Rodriguez
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).