[llvm-dev] PGO information at LTO/thinLTO link step
Graham Yiu via llvm-dev
llvm-dev at lists.llvm.org
Tue Oct 3 14:44:02 PDT 2017
Awesome. Thanks David, Teresa.
Graham Yiu
LLVM Compiler Development
IBM Toronto Software Lab
Office: (905) 413-4077 C2-707/8200/Markham
Email: gyiu at ca.ibm.com
From: Xinliang David Li <xinliangli at gmail.com>
To: Teresa Johnson <tejohnson at google.com>
Cc: Graham Yiu <gyiu at ca.ibm.com>, llvm-dev
<llvm-dev at lists.llvm.org>, yamauchi at google.com
Date: 10/03/2017 05:34 PM
Subject: Re: [llvm-dev] PGO information at LTO/thinLTO link step
On Tue, Oct 3, 2017 at 2:15 PM, Teresa Johnson via llvm-dev <
llvm-dev at lists.llvm.org> wrote:
On Tue, Oct 3, 2017 at 2:10 PM, Graham Yiu <gyiu at ca.ibm.com> wrote:
Thanks Easwaran. This is what we've observed as well, where the old PM
inliner was only looking hot/cold callee information, which have
signficantly smaller boosts/penalties compared to callsite information.
Teresa, do you know if there is some documentation/video/presentation on
how PGO information is represented in LLVM and what information is
passed via the IR? I'm finding some difficulty in getting the big
picture via the code.
In a nutshell, there are two main types of profile data (BB/edge related).
1) branch probability
2) function entry count.
PGO instrumentation actually collects BB count. During profile annotation
pass, the branch probability (weights) profile data is computed and
annotated to the IR (the branch instructions). After profile annotation,
the block count information is dropped except for the function entry count
which is annotated to the function entry.
The BPI/BFI pass can recompute the block frequency data from the branch
probability info. The BB frequency is intra-function and does not have
global meaning. To recompute the profile count for a BB, function entry
profile count is used, which is multiplied by the ratio of the BB freq and
entry freq.
This scheme works really well except for CFG with irreducible loops, for
which BFI's frequency propagation pass can not reconstruct the
frequency/count properly. Hiroshi is working on a patch to fix the
problem.
For AutoFDO, there is also a sample count profile for callsites which is
needed because function entry may not have any samples.
The value profiler also has its meta data for value histograms of a given
value site. At module level, while program profile summary data is also
represented so that optimization passes can query for global hotness info.
David
The documentation I am aware of is in the Language Ref and a subpage
linked from here:
https://llvm.org/docs/LangRef.html#prof-metadata
If that doesn't help let me know and I can point you to someone who would
know (if I can't answer it myself).
Teresa
Graham Yiu
LLVM Compiler Development
IBM Toronto Software Lab
Office: (905) 413-4077 C2-707/8200/Markham
Email: gyiu at ca.ibm.com
Inactive hide details for Teresa Johnson ---10/03/2017 05:00:11 PM---On
Tue, Oct 3, 2017 at 1:55 PM, Easwaran Raman <eraman at gooTeresa Johnson
---10/03/2017 05:00:11 PM---On Tue, Oct 3, 2017 at 1:55 PM, Easwaran
Raman <eraman at google.com> wrote: >
From: Teresa Johnson <tejohnson at google.com>
To: Easwaran Raman <eraman at google.com>
Cc: Graham Yiu <gyiu at ca.ibm.com>, llvm-dev <llvm-dev at lists.llvm.org>
Date: 10/03/2017 05:00 PM
Subject: Re: [llvm-dev] PGO information at LTO/thinLTO link step
On Tue, Oct 3, 2017 at 1:55 PM, Easwaran Raman <eraman at google.com>
wrote:
On Tue, Oct 3, 2017 at 1:46 PM, Teresa Johnson via llvm-dev <
llvm-dev at lists.llvm.org> wrote:
On Tue, Oct 3, 2017 at 1:38 PM, Graham Yiu <gyiu at ca.ibm.com>
wrote:
Hi Teresa,
Actually, enabling the new pass manager manually seems to
have solved this issue, so this problem is only valid for
the old pass manager.
It should not be an issue in the old PM either - the callsite
hotness is passed via IR.
More precisely, the function entry counts are passed via IR. With
the old PM, we don't have callsite hotness information, but
callee's entry count is used to boost the threshold.
Thanks for the clarification. (But essentially there should be no
difference between the profile info in the IR in the compile step vs the
link aka ThinLTO backend steps - the inliner in both cases is working
off the same profile info in the IR)
As David mentioned, the new PM inliner does a better job of
updating call hotness after inlining, but it should be there (some
things might look hotter than then should, which seems to be the
opposite of the problem you are hitting). Can you send me a
reproducer with the old PM?
Teresa
Thanks,
Graham Yiu
LLVM Compiler Development
IBM Toronto Software Lab
Office: (905) 413-4077 C2-707/8200/Markham
Email: gyiu at ca.ibm.com
Inactive hide details for Teresa Johnson ---10/03/2017
04:18:17 PM---On Tue, Oct 3, 2017 at 12:54 PM, Graham Yiu
via llvm-dev <Teresa Johnson ---10/03/2017 04:18:17 PM---On
Tue, Oct 3, 2017 at 12:54 PM, Graham Yiu via llvm-dev <
llvm-dev at lists.llvm.org> wrote:
From: Teresa Johnson <tejohnson at google.com>
To: Graham Yiu <gyiu at ca.ibm.com>
Cc: llvm-dev <llvm-dev at lists.llvm.org>
Date: 10/03/2017 04:18 PM
Subject: Re: [llvm-dev] PGO information at LTO/thinLTO link
step
On Tue, Oct 3, 2017 at 12:54 PM, Graham Yiu via llvm-dev <
llvm-dev at lists.llvm.org> wrote:
Hello,
My team and I noticed that callsite hotness
information is not preserved from compile to
link step with LTO/thinLTO enabled. As a result,
the link step inlining pass remains conservative
when inlining callsites known to be hot (ie.
without the 'HotCallSiteThreshold' which is set
at 3000 by default). There are likely many
cross-module inlining opportunities lost this
way, and diminishes the benefit of using
LTO/thinLTO+PGO together.
The callsite hotness is passed via the IR, so it should be
there in the LTO/ThinLTO backends (during the link step).
Can you provide a reproducer where that isn't happening?
Teresa
In general, does LLVM pass profiling information
through the IR to the link step other than
branch probabilities and function entry counts?
If not, are there plans to do so in the future?
For inlining specifically, perhaps we can mark
callsites with hot/cold attributes during
compile step to ensure LTO inlining will give
appropriate threshold bonuses/penalties.
Any thoughts/insights/comments would be
appreciated.
Cheers,
Graham Yiu
LLVM Compiler Development
IBM Toronto Software Lab
Office: (905) 413-4077 C2-707/8200/Markham
Email: gyiu at ca.ibm.com
_______________________________________________
LLVM Developers mailing list
llvm-dev at lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev
--
|-----------+-------------+--------------------+------------|
|Teresa | Software | | |
|Johnson | |Engineer | |tejohnson at google.com|408-460-2413|
| | || | |
|-----------+-------------+--------------------+------------|
--
|-----------+-------------+--------------------+------------|
|Teresa | Software | | |
|Johnson | |Engineer | |tejohnson at google.com|408-460-2413|
| | || | |
|-----------+-------------+--------------------+------------|
_______________________________________________
LLVM Developers mailing list
llvm-dev at lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev
--
|-----------+-------------+--------------------+------------|
|Teresa | Software | | |
|Johnson | |Engineer | |tejohnson at google.com|408-460-2413|
| | || | |
|-----------+-------------+--------------------+------------|
--
-----------------------------------------------------------
Teresa Software
Johnson | Engineer | tejohnson at google.com 408-460-2413
|
_______________________________________________
LLVM Developers mailing list
llvm-dev at lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20171003/878ba447/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: graycol.gif
Type: image/gif
Size: 105 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20171003/878ba447/attachment.gif>
More information about the llvm-dev
mailing list