similar to: Showing hotness in LLVM optimization remarks using AutoFDO sampling profile data?

> On Jul 14, 2017, at 10:22 AM, Davide Italiano <davide at freebsd.org> wrote: > > On Fri, Jul 14, 2017 at 10:10 AM, Adam Nemet <anemet at apple.com <mailto:anemet at apple.com>> wrote: >> >> >> On Jul 14, 2017, at 8:21 AM, Davide Italiano via llvm-dev <llvm-dev at lists.llvm.org> wrote: >> >> On Mon, Jun 19, 2017 at 4:13 PM, Brian

Hello all, In https://www.youtube.com/watch?v=qq0q1hfzidg, Adam Nemet (cc'ed) describes optimization remarks and some future plans for the project. I had a few follow-up questions: 1. As an example of future work to be done, the talk mentions expanding the set of optimization passes that emit remarks. However, the Clang User Manual mentions that "optimization remarks do not really make

> On Jul 14, 2017, at 8:21 AM, Davide Italiano via llvm-dev <llvm-dev at lists.llvm.org> wrote: > > On Mon, Jun 19, 2017 at 4:13 PM, Brian Gesiak via llvm-dev > <llvm-dev at lists.llvm.org <mailto:llvm-dev at lists.llvm.org>> wrote: >> Hello all, >> >> In https://www.youtube.com/watch?v=qq0q1hfzidg, Adam Nemet (cc'ed) describes >>

> On Wed, Jun 28, 2017 at 8:13 AM, Hal Finkel <hfinkel at anl.gov> wrote: > > I don't object to adding some kind of filtering option, but in general it won't help. An important goal here is to provide analysis (and other) tools to users that present this information at a higher level. The users won't, and shouldn't, know exactly what kinds of messages the tools use.

Adam, thanks for all the suggestions! One nice aspect of the `-Rpass` family of options is that I can filter based on what I want. If I only want to see which inlines I missed, I could use `clang -Rpass-missed="inline"`, for example. On the other hand, optimization remark YAML always include remarks from all passes (as far as I can tell), which increases the amount of time it takes

This idea came up a few times recently [1][2] so I’d like start prototyping it. To summarize, we can emit optimization remarks using the -Rpass* options. These are currently emitted by optimizations like vectorization[3], unrolling, inlining and since last week loop distribution. For large programs however this can amount to a lot of diagnostics output to sift through. Filtering this by the

Hi Hal, > On May 10, 2016, at 5:39 PM, Hal Finkel <hfinkel at anl.gov> wrote: > > Hi Adam, > > I think would be a really useful feature to have. I don't think that the backend should be responsible for filtering, but should pass the relative hotness information to the frontend. Given that these diagnostics are not just going to be used for -Rpass and friends, but also

> On May 11, 2016, at 3:37 AM, Hal Finkel <hfinkel at anl.gov> wrote: > > ----- Original Message ----- >> From: "Adam Nemet" <anemet at apple.com> >> To: "Hal Finkel" <hfinkel at anl.gov> >> Cc: "llvm-dev (llvm-dev at lists.llvm.org)" <llvm-dev at lists.llvm.org> >> Sent: Wednesday, May 11, 2016 1:15:42 AM

> On May 11, 2016, at 10:45 AM, Adam Nemet via llvm-dev <llvm-dev at lists.llvm.org> wrote: > >> >> On May 11, 2016, at 3:37 AM, Hal Finkel <hfinkel at anl.gov <mailto:hfinkel at anl.gov>> wrote: >> >> ----- Original Message ----- >>> From: "Adam Nemet" <anemet at apple.com <mailto:anemet at apple.com>> >>>

With recent bug fixes and performance tunings, AutoFDO at llvm has reached a usable state. To evaluate performance, we used O3/-fprofile-use/-fprofile-sample-use respectively to optimize clang itself, and measure its speed. clang built with -fprofile-use is ~20% faster than clang built with O3 clang built with -fprofile-sample-use is ~10% faster than clang built with O3 AutoFDO can deliver 50%

Hi, all Recently I'm trying to use LLVM PGO and autoFDO. However I have some problems in the process. LLVM source code is updated on April 9th. Operating system is SUSE x86_64 1. Problems in instrumentation based PGO: clang -O2 -fprofile-instr-generate test.c -o a.out ./a.out (then default.profraw is generated) clang -O2 -fprofile-instr-use=default.profraw test.c -o a.out

I am looking for advice on a problem observed with -fprofile-sample-use for samples built with the AutoFDO tool I took the "hmmer" benchmark out of SPEC2006 It is initially compiled clnag++ -o hmmer -O3 -std=gnu89 -DSPEC_CPU -DNDEBUG -fno-strict-aliasing -w -g *.c This baseline binary runs in about 164.2 seconds as reported by "perf stat" We build a sample file from this

On Fri, Aug 12, 2016 at 12:15 PM, Xinliang David Li via llvm-dev < llvm-dev at lists.llvm.org> wrote: > +dehao. > > There are two potential problems: > > 1) the branch gets eliminated in the binary that is being profiled, so > there is no profile data > This seems like a fundamental problem for PGO. Maybe it is also responsible for this bug:

Thank you for the link and the suggestion to try master! I did so and discovered that it reproduces on master for me as well. The repro script I used (unchanged from before) and the output can be found here: https://gist.github.com/modocache/d9700166067f4a155820bc57d9bee1f3 (Note that the output looks nearly identical, but it's using clang-10 from the master branch of llvm-project.) I wonder

Hi Rong, This is a very interesting proposal. We've also observed profile quality degradation from CFG destructive pass like loop rotate, and I can see how this proposal would help improve quality of profile that drives later optimization passes in the pipeline. I have a few questions. * How does this affect today's AutoFDO? Specifically, can users upgrade compiler with FS-AutoFDO

Hello all! I find that using lldb to debug LLVM libraries can be super frustrating, because a lot of LLVM classes, like the constructor for StringRef, are marked LLVM_ATTRIBUTE_ALWAYS_INLINE. So when I attempt to have lldb evaluate an expression that implicitly instantiates a StringRef, I get 'error: Couldn't lookup symbols: __ZN4llvm9StringRefC1EPKc'. As an example, most recently

Hello all, I'm trying to better understand the state of Mach-O support in lld. The lld docs state that "the linker supports ELF (Unix), PE/COFF (Windows), Mach-O (macOS) and WebAssembly in descending order of completeness." [1] True to that statement, I found an email on this list from Jan 2018 stating that "MachO support in lld is not really ready for real world usage. It was

Hello all, It's been a month since my previous email on the topic, and since then I've done some initial work on porting the coroutines passes to the new pass manager. In total there are 6 patches -- that's a lot to review, so allow me to introduce the changes being made in each of them. # What's finished In these first 6 patches, I focused on lowering coroutine intrinsics

Vedant, Jeremy, Thanks a ton! I copied ASan's use of 'replaceDbgDeclare', think that worked! https://github.com/modocache/llvm-project/commit/afbc04e1dcba has some extremely quick and dirty changes I made (with no tests!), and a link to a Gist with the LLVM IR and DWARF produced, https://gist.github.com/modocache/6f29093ba2827946011b422ed3bd2903. There's only one kink: the spot

Hi Mikhail, As Brian noted, stuff like this works better in the new pass manager. Even in the old pass manager I thought it should work though. Did you initialize the pass, via `INITIALIZE_PASS_DEPENDENCY(PostDominatorTreeWrapperPass)`? Did you require it, via ` AU.addRequired<PostDominatorTreeWrapperPass>();`? Btw. May I ask what you are planning to do? Cheers, Johannes On 01/07,