similar to: RFC: a practical mechanism for applying Machine Learning for optimization policies in LLVM

+Yundi Qian <yundi at google.com> +Eugene Brevdo <ebrevdo at google.com> , our team members from the ML side. To avoid formatting issues, here is a link to the RFC <https://docs.google.com/document/d/1BoSGQlmgAh-yUZMn4sCDoWuY6KWed2tV58P4_472mDE/edit?usp=sharing>, open to comments. Thanks! On Wed, Apr 8, 2020 at 2:34 PM Mircea Trofin <mtrofin at google.com> wrote: >

It turns out it's me, sorry. Let me see how I can sort this out. In the meantime, here is the csv: SPEC2006 data: binary,base -Oz size,ML -Oz size,ML size shrink by,,perf: base -Oz scores,perf: ML -Oz scores,ML improvement by 400.perlbench,2054200,2086776,-1.59%,,2.9,2.9,0.00% 401.bzip2,1129976,1095544,3.05%,,6.4,6.2,-3.13% 403.gcc,4078488,4130840,-1.28%,,11.6,11.7,0.86%

Sorry, I wasn't aware of that. I can make the google doc view-only, keeping the current comments. I'll wait a bit (few hrs) to see if there's any pushback to that. On Thu, Apr 9, 2020 at 9:57 AM Xinliang David Li <xinliangli at gmail.com> wrote: > One suggestion : should we consolidate the discussion into the main > thread? I know some folks are not willing to comment in

2017-07-24 16:14 GMT-07:00 Quentin Colombet via llvm-dev < llvm-dev at lists.llvm.org>: > Hi River, > > On Jul 24, 2017, at 2:36 PM, River Riddle <riddleriver at gmail.com> wrote: > > Hi Quentin, > I appreciate the feedback. When I reference the cost of Target Hooks it's > mainly for maintainability and cost on a target author. We want to keep the >

2017-07-26 9:31 GMT-07:00 Quentin Colombet <qcolombet at apple.com>: > > On Jul 25, 2017, at 10:36 PM, Mehdi AMINI <joker.eph at gmail.com> wrote: > > > > 2017-07-24 16:14 GMT-07:00 Quentin Colombet via llvm-dev < > llvm-dev at lists.llvm.org>: > >> Hi River, >> >> On Jul 24, 2017, at 2:36 PM, River Riddle <riddleriver at

Hi Quentin, I appreciate the feedback. When I reference the cost of Target Hooks it's mainly for maintainability and cost on a target author. We want to keep the intrusion into target information minimized. The heuristics used for the outliner are the same used by any other IR level pass seeking target information, i.e TTI for the most part. I can see where you are coming from with

Hi Evgeny, I know of the current machine outliner in LLVM. If you look in the "More detailed performance data" in the end section it includes performance comparisons to the machine outliner. As for the algorithmic approach they are kind of similar. Machine Outliner: - Builds a suffix tree based on identical equivalence between machine instrs. - Uses target specific cost model for

> On Jul 25, 2017, at 9:24 AM, Jessica Paquette <jpaquette at apple.com> wrote: > >> The two passes are pretty different in their approaches to congruency finding, so I don't think it helps to group them as though they were interchangeable "outliner technology". The two passes might be totally orthogonal. > > I think that based off how River described his

Hi Quentin, I understand your points and I believe that some meaning is being lost via email. For performance it's true that that cost isn't necessarily modeled, there is currently only support for using profile data to avoid mitigate that. I was working under the assumption, possibly incorrectly, that at Oz we favor small code over anything else including runtime performance. This is

Hi River, Thanks for the detailed explanation. If people are okay for you to move forward, like I said to Andrey, I won’t oppose. I feel sad we have to split our effort on outlining technology, but I certainly don’t pretend to know what is best! The bottom line is if people are happy with that going in, the conversation on the details can continue in parallel. > On Jul 24, 2017, at 4:56 PM,

On Mon, Jul 24, 2017 at 4:14 PM, Quentin Colombet via llvm-dev < llvm-dev at lists.llvm.org> wrote: > Hi River, > > On Jul 24, 2017, at 2:36 PM, River Riddle <riddleriver at gmail.com> wrote: > > Hi Quentin, > I appreciate the feedback. When I reference the cost of Target Hooks it's > mainly for maintainability and cost on a target author. We want to keep the

I’m River and I’m a compiler engineer at PlayStation. Recently, I’ve been working on an interprocedural outlining (code folding) pass for code size improvement at the IR level. We hit a couple of use cases that the current code size solutions didn’t handle well enough. Outlining is one of the avenues that seemed potentially beneficial. -- Algorithmic Approach -- The general implementation can be

Hi All, Our team at Facebook is building a new context-sensitive Sample PGO as an alternative to the existing AutoFDO. We’d like to share our motivation, propose a new design, and reveal preliminary results on benchmarks. We will refer to the proposed design as CSSPGO in this RFC. The new CSSPGO leverages simultaneous LBR and stack sampling to construct a full context-sensitive profile. It

See my answers inline. From: Xinliang David Li <davidxl at google.com> Date: Friday, August 7, 2020 at 7:57 PM To: Wenlei He <wenlei at fb.com> Cc: "llvm-dev at lists.llvm.org" <llvm-dev at lists.llvm.org>, Wei Mi <wmi at google.com>, Hongtao Yu <hoy at fb.com> Subject: Re: [RFC] Context-sensitive Sample PGO with Pseudo-Instrumentation On Fri, Aug 7,

Thanks for the thoughtful questions, David. See my answers inline. Thanks, Wenlei From: Xinliang David Li <davidxl at google.com> Date: Friday, August 7, 2020 at 1:24 PM To: Wenlei He <wenlei at fb.com> Cc: "llvm-dev at lists.llvm.org" <llvm-dev at lists.llvm.org>, Wei Mi <wmi at google.com>, Hongtao Yu <hoy at fb.com> Subject: Re: [RFC] Context-sensitive

Hal, Thank you for finding interest in our work! Please find my answers inlined below. > > ________________________________________ > From: Hal Finkel [hfinkel at anl.gov] > Sent: Monday, December 16, 2013 4:26 PM > To: Radovan Obradovic > Cc: llvmdev at cs.uiuc.edu; chandlerc; Andrew Trick > Subject: Re: [LLVMdev] [RFC] Iterrative compilation framework for Clang/LLVM >

Hi Rahman, Thanks for sharing the BB-info section proposal which is a shiny idea. I think the BB-info and pseudo probes deal with a similar problem in different spaces, i.e., mapping hardware samples to corresponding basic blocks. In the context of pseudo probes, we much focus on mapping samples back to source-level blocks which is the input to the optimizer. Therefore we are building a

Hi Jessica, The comparison to the inliner is an interesting one but we think it's important to note the difference in the use of heuristics. The inliner is juggling many different tasks at the same time, execution speed, code size, etc. which can cause the parameters to be very sensitive depending on the benchmark/platform/etc. The outliners heuristics are focused solely on the potential code

On Fri, Aug 7, 2020 at 11:28 PM Hongtao Yu <hoy at fb.com> wrote: > A few add-ons. > > > > *From: *Wenlei He <wenlei at fb.com> > *Date: *Friday, August 7, 2020 at 10:34 PM > *To: *Xinliang David Li <davidxl at google.com> > *Cc: *"llvm-dev at lists.llvm.org" <llvm-dev at lists.llvm.org>, Wei Mi < > wmi at

This is a first step towards to full iterative compilation framework for Clang/LLVM. The attached patch is work in progress and the idea of sending it to the list earlier rather than later is to open up a discussion on iterative compilation in Clang/LLVM. All comments are welcome, especially those related to integration of the framework into Clang/LLVM. Current compilers have pipeline structure