search for: chisophugi

This just failed again: http://lab.llvm.org:8011/builders/clang-ppc64le-linux-multistage/builds/1579 Bill, could you take a look at this? This is like the 3rd time I've been incorrectly pinged by this buildbot due to this issue. -- Sean Silva On Mon, Jun 13, 2016 at 8:07 PM, Sean Silva <chisophugis at gmail.com> wrote: > Thanks for taking a look. The flaky ASan test failure in the OP was stage1 > so it sounds like there may be multiple problems :( > > > -- Sean Silva > > On Sun, Jun 12, 2016 at 7:32 AM, Benjamin Kramer <benny.kra at gmail.com> > wrote: >...

.../lab.llvm.org:8011/builders/clang-ppc64le-linux-multistage/builds/1579 > > Bill, could you take a look at this? This is like the 3rd time I've been > incorrectly pinged by this buildbot due to this issue. > > -- Sean Silva > > On Mon, Jun 13, 2016 at 8:07 PM, Sean Silva <chisophugis at gmail.com > <mailto:chisophugis at gmail.com>> wrote: > > Thanks for taking a look. The flaky ASan test failure in the OP was > stage1 so it sounds like there may be multiple problems :( > > > -- Sean Silva > > On Sun, Jun 12, 2016 at 7:32 AM,...

Rafael, did you ever get a chance to look at this? -- Sean Silva On Thu, Feb 23, 2017 at 1:14 AM, Sean Silva <chisophugis at gmail.com> wrote: > > > On Wed, Feb 22, 2017 at 6:14 AM, Rafael Avila de Espindola < > rafael.espindola at gmail.com> wrote: > >> Sean Silva <chisophugis at gmail.com> writes: >> >> When __libc_start_main is hidden, it doesn't end up in the d...

By that logic we shouldn't have an 80 column limit in LLVM :) On Wed, Apr 8, 2015 at 3:35 PM Sean Silva <chisophugis at gmail.com> wrote: > On Wed, Apr 8, 2015 at 1:32 PM, Zachary Turner <zturner at google.com> wrote: > >> When the line is like 400 characters, it's pretty obnoxious. >> > > > I guess I'd need to see the particular situation, but in my experience > m...

...I had access to on this test and didn't find anything. The merging uses threads so I can't rule out nondeterminism. It's strange that it only happens on ppc64le and only on stage 2, so an actual miscompile wouldn't surprise me either. On Sun, Jun 12, 2016 at 4:26 PM, Sean Silva <chisophugis at gmail.com> wrote: > Also, `Clang Tools :: include-fixer/merge.test` seems to be a bit flaky: > > http://lab.llvm.org:8011/builders/clang-ppc64le-linux-multistage/builds/1304 > http://lab.llvm.org:8011/builders/clang-ppc64le-linux-multistage/builds/1315 > > Ben, it looks lik...

...built with > PGO already) and rebuild without xray nop insertion and function layout. > > David > > On Mon, Jan 16, 2017 at 3:40 PM, Sean Silva via llvm-dev < > llvm-dev at lists.llvm.org> wrote: > >> >> >> On Mon, Jan 16, 2017 at 3:34 PM, Sean Silva <chisophugis at gmail.com> >> wrote: >> >>> >>> >>> On Mon, Jan 16, 2017 at 3:32 PM, Sean Silva <chisophugis at gmail.com> >>> wrote: >>> >>>> >>>> >>>> On Mon, Jan 16, 2017 at 2:31 PM, Davide Italiano <dav...

.../uuencode/uuencode 5 MultiSource/Benchmarks/Trimaran/enc-rc4/rc4 11 SingleSource/Benchmarks/BenchmarkGame/n-body 2 SingleSource/Benchmarks/Shootout/ackermann Let me know if there are any specific benchmarks you would like me to test. -- Sean Silva On Wed, May 20, 2015 at 3:31 PM, Sean Silva <chisophugis at gmail.com> wrote: > I found an interesting datapoint: > > In the last 10,000 revisions of LLVM+Clang, only 10 revisions actually > caused the binary of MultiSource/Benchmarks/BitBench/five11 to change. So > if just store a hash of the binary in the database, we should be able...

Lets try this on the whole test suite? > On May 26, 2015, at 7:05 PM, Sean Silva <chisophugis at gmail.com> wrote: > > Update: in that same block of 10,000 LLVM/Clang revisions, this the number of distinct SHA1 hashes for the binaries of the following benchmarks: > > 7 MultiSource/Applications/aha/aha > 2 MultiSource/Benchmarks/BitBench/drop3/drop3 > 10 MultiSource/B...

...test and didn't > find anything. The merging uses threads so I can't rule out > nondeterminism. It's strange that it only happens on ppc64le and only > on stage 2, so an actual miscompile wouldn't surprise me either. > > On Sun, Jun 12, 2016 at 4:26 PM, Sean Silva <chisophugis at gmail.com> wrote: > > Also, `Clang Tools :: include-fixer/merge.test` seems to be a bit flaky: > > > > > http://lab.llvm.org:8011/builders/clang-ppc64le-linux-multistage/builds/1304 > > > http://lab.llvm.org:8011/builders/clang-ppc64le-linux-multistage/builds/13...

> On Sep 16, 2017, at 4:49 PM, Sean Silva <chisophugis at gmail.com> wrote: > > Actually maybe something like: > > if (auto &E = ORE.emitMissed(DEBUG_TYPE)) { > E.emit(...) << ...; > } Well, the point of this interface was exactly to avoid writing a conditional. If you’re willing to use a conditional you can alread...

...but the problem is that inlining decisions are different between profile-gen and profile-use compilations. This makes it hard to match call trace data from prof-gen run with post-inline callgraph of the profile-use build. > > David > > On Mon, Jan 16, 2017 at 7:41 PM, Sean Silva <chisophugis at gmail.com <mailto:chisophugis at gmail.com>> wrote: > Would it make sense for xray instrumentation be part of -fprofile-generate? PGO will affect inlining decisions etc for the optimized binary, but the collected traces during the instrumented build would still have quite a bit of u...

When the line is like 400 characters, it's pretty obnoxious. On Wed, Apr 8, 2015 at 1:32 PM Sean Silva <chisophugis at gmail.com> wrote: > On Wed, Apr 8, 2015 at 1:19 PM, Zachary Turner <zturner at google.com> wrote: > >> Just found another issue with CMake versions prior to 3.0. I need to >> write a long string, and prior to 3.0 there is no line continuation for >> quoted arg...

On Nov 14, 2012, at 10:52 AM, Sean Silva <chisophugis at gmail.com> wrote: > It references release 2.7... is this document worth keeping? > Nuke it from orbit. :) > CC'ing Jim Grosbach since IIRC he recently pulled out some stuff > related to llvm-gcc. > > -- Sean Silva

On Tue, Jan 7, 2014 at 8:48 PM, Sean Silva <chisophugis at gmail.com> wrote: > sean:~/pg/llvm/llvm % git log --oneline --since='1 month ago' | wc -l > 706 > sean:~/pg/llvm/llvm % git log --oneline --since='1 month ago' ./test | wc -l > 317 Wouldn't this also catch commits to code generation that added tests as well?...

I have the unified analysis manager implemented now and passing all tests at: https://github.com/chisophugis/llvm/commits/analysis-manager One caveat: It would be a layering violation for the analysis manager to know about Loop and SCC since those live in libAnalysis. So currently I have some stuff commented out for imitating the proxy downward invalidation for them. (and generally the proxy downward in...

...Thu, Jul 14, 2016 at 4:49 PM, Tanya Lattner <tanyalattner at llvm.org> wrote: > What were you thinking? I increased it to 5MB.. but it sounds like you > have 26MB attachments? > Is 50MB doable? -- Sean Silva > > -Tanya > > On Jul 13, 2016, at 2:46 PM, Sean Silva <chisophugis at gmail.com> wrote: > > Tanya, could you increase the max attachment size limit on LLVM's bugzilla? > > -- Sean Silva > > On Wed, Jul 13, 2016 at 7:19 AM, James Y Knight <jyknight at google.com> > wrote: > >> This sounds like a bad idea. Let's just i...

Sean, hopefully you’re OK with that reasoning. I went ahead and committed this in r313691. > On Sep 16, 2017, at 10:43 PM, Adam Nemet <anemet at apple.com> wrote: > > >> On Sep 16, 2017, at 4:49 PM, Sean Silva <chisophugis at gmail.com <mailto:chisophugis at gmail.com>> wrote: >> >> Actually maybe something like: >> >> if (auto &E = ORE.emitMissed(DEBUG_TYPE)) { >> E.emit(...) << ...; >> } > > Well, the point of this interface was exactly to avoid wri...

...> > > > MartinO > > > > PS: Movidius is now part of Intel, so I will be gradually switching to my > Intel email address. > > > > *From:* Ryan Taylor [mailto:ryta1203 at gmail.com] > *Sent:* Sunday, June 25, 2017 2:59 AM > *To:* Sean Silva <chisophugis at gmail.com> > *Cc:* Martin J. O'Riordan <martin.oriordan at movidius.com>; llvm-dev < > llvm-dev at lists.llvm.org> > *Subject:* Re: [llvm-dev] Definitive list of optimisations at each > optimisation level > > > > I agree, it's much clearer, it just...

Both arguments require a string. Sorry, I forgot to mention that. On Mon, Apr 4, 2016 at 12:02 PM, Sean Silva <chisophugis at gmail.com> wrote: > Could you do something like `-mode={if,od}` and make the option required? > > -- Sean Silva > > On Mon, Apr 4, 2016 at 11:54 AM, Daniel Dilts via llvm-dev < > llvm-dev at lists.llvm.org> wrote: > >> With the CommandLine library, is there...

...cooperation from the adaptors, because (like the example "IRUnit stacks" I showed above) the analysis manager would need a way to know whether to map back to a particular SCC or to a particular Module (or potentially both). -- Sean Silva On Tue, Jul 26, 2016 at 1:32 AM, Sean Silva <chisophugis at gmail.com> wrote: > I'm not quite sure what post to respond to for a status update, but I > guess this one will do (and you can check my log for more info of course). > > My current working branch for the analysis manager stuff is at > https://github.com/chisophugis/llvm/...