search for: newgvn

...up to date with 'origin/master'. > >Changes not staged for commit: > (use "git add <file>..." to update what will be committed) > (use "git restore <file>..." to discard changes in working directory) > modified: llvm/test/Transforms/NewGVN/todo-pr42422-phi-of-ops.ll > >no changes added to commit (use "git add" and/or "git commit -a") > >D:\upstream\llvm-project>git diff >diff --git a/llvm/test/Transforms/NewGVN/todo-pr42422-phi-of-ops.ll b/llvm/test/Transforms/NewGVN/todo-pr42422-phi-of-ops.ll &g...

I think I understand the concepts, but I sure would appreciate a specific example, and I appreciate the offer. To make your life harder, could you start with what I should do given that I have not created a branch for the first patch? I just have the six files staged. I have GitHub Desktop installed, if that makes any of the steps easier. Thanks again, and no rush! At 8/10/2020 10:07 AM,

Hi everybody. NewGVN was recently committed and a few minute ago I added a flag to enable the new experimental pass. For the brave soul, passing `-mllvm -enable-newgvn` should do the trick. We'll be happy to receive bug reports to analyze/fix, bonus point if they contain a synthetic/reduced testcase. Open a bug li...

...ysis, instead it eliminates as it goes, so it's hard to reason about it. * GVN doesn't perform any phi predication, i.e. doesn't know about phi nodes, so later passes have to do some extra work to clean up * There are bugs, e.g. [2] which would require a rewrite of PRE to be fixed. ### NewGVN The new algorithm implements the ideas described in [3] with some engineering optimizations by Dan (for example the set of touched instructions is represented using a Bitvector instead of a set because it's not uncommon for large functions where a predicate change that thousands of instruction...

Hi, I’m curious if the test-suite running cleanly with this flag? (That would be a first step before wider testing). Thanks, — Mehdi > On Dec 26, 2016, at 10:45 AM, Davide Italiano via llvm-dev <llvm-dev at lists.llvm.org> wrote: > > Hi everybody. > NewGVN was recently committed and a few minute ago I added a flag to > enable the new experimental pass. > > For the brave soul, passing `-mllvm -enable-newgvn` should do the > trick. We'll be happy to receive bug reports to analyze/fix, bonus > point if they contain a synthetic/reduce...

...e guidance of your insights/prototype? I think I can spend more time on implementation. Thanks, Taewook ________________________________ From: Daniel Berlin <dberlin at dberlin.org> Sent: Tuesday, April 4, 2017 9:41:30 PM To: Taewook Oh Cc: llvm-dev at lists.llvm.org Subject: Re: [llvm-dev] [NewGVN] Plan for GVNPRE? Of course As a heads up, it's likely to be a while (ie 3-6 months or maybe longer) if you leave it to me. I try to stay out of the critical path, because my real day job is managing folks who work on llvm (and other things), not working on llvm, and that has a lot of interrup...

...t; >> >> Is there an umbrella PR or something that could help organise what >> someone would have to do/take on in order for it to be enabled? Outstanding >> design issues or critical blocking bugs etc. >> >> Amara >> >> > It seems like [META][GVN] NewGVN Integration [1] tries to keep track of > bugs/tasks for NewGVN. > > Cheers, > Florian > > > [1] https://bugs.llvm.org/show_bug.cgi?id=30995 > -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachme...

...2018, at 04:53, Daniel Berlin via llvm-dev <llvm-dev at lists.llvm.org> wrote: > > > > On Fri, Jan 5, 2018 at 8:39 PM, Andrew Kelley via llvm-dev <llvm-dev at lists.llvm.org <mailto:llvm-dev at lists.llvm.org>> wrote: > Greetings, > > I just found a bug in NewGVN: https://bugs.llvm.org/show_bug.cgi?id=35839 <https://bugs.llvm.org/show_bug.cgi?id=35839> > > I suspect this is a dupe of one of the three bugs we know about. > > > > I noticed that it's off by default in clang. > > Yup. > > I'm going to turn it...

Hello, In some of our internal benchmarks, I observe that LLVM performs worse than GCC because LLVM fails to perform PRE when GCC can. I hope this problem goes away when NewGVN equipped with PRE, and wonder if anyone has an idea about the status of PRE on top of NewGVN. Thanks! Best, Taewook -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20170404/f7fa37aa/attachment.html>

...est-suite running cleanly with this flag? > (That would be a first step before wider testing). > > Thanks, > > — > Mehdi > >> On Dec 26, 2016, at 10:45 AM, Davide Italiano via llvm-dev <llvm-dev at lists.llvm.org> wrote: >> >> Hi everybody. >> NewGVN was recently committed and a few minute ago I added a flag to >> enable the new experimental pass. >> >> For the brave soul, passing `-mllvm -enable-newgvn` should do the >> trick. We'll be happy to receive bug reports to analyze/fix, bonus >> point if they contai...

...9:26, Amara Emerson via llvm-dev wrote: > > Is there an umbrella PR or something that could help organise what > someone would have to do/take on in order for it to be enabled? > Outstanding design issues or critical blocking bugs etc. > > Amara > It seems like [META][GVN] NewGVN Integration [1] tries to keep track of bugs/tasks for NewGVN. Cheers, Florian [1] https://bugs.llvm.org/show_bug.cgi?id=30995

...on what to do about it it amounts to something like this (but again, it happens a lot): live = gep thing, 0 live2 = gep thing, 1 branch i1 provablytrue,, mergeblock, otherbb otherbb: dead = something else br mergeblock merge block a = phi(live, dead) b = live2 result = icmp sge a, b both GVN and NewGVN prove provablytrue to be true, and phi to be equivalent to live. GVN transforms this piece at time, and so by the time simplifycmpinst sees the icmp, it is result = icmp sge <live2, live> It proves result true. NewGVN is an analysis, and so it doesn't transform the ir, and simplifycmp...

...s ready. Please let me know if you need any help. Thanks, Taewook From: Daniel Berlin <dberlin at dberlin.org> Date: Tuesday, April 4, 2017 at 6:13 PM To: Taewook Oh <twoh at fb.com> Cc: "llvm-dev at lists.llvm.org" <llvm-dev at lists.llvm.org> Subject: Re: [llvm-dev] [NewGVN] Plan for GVNPRE? Hi Taewook, I have prototypes, and they work well enough that i was convinced it's not a big deal to implement. If you are interested in working on it, let me know. Otherwise, i'll get to it. I'm just knocking down the rest of the generated code perf differences i can...

...t would be a first step before wider testing). > > > > Thanks, > > > > — > > Mehdi > > > >> On Dec 26, 2016, at 10:45 AM, Davide Italiano via llvm-dev < > llvm-dev at lists.llvm.org> wrote: > >> > >> Hi everybody. > >> NewGVN was recently committed and a few minute ago I added a flag to > >> enable the new experimental pass. > >> > >> For the brave soul, passing `-mllvm -enable-newgvn` should do the > >> trick. We'll be happy to receive bug reports to analyze/fix, bonus > >&...

...rd to reason about it. > > * GVN doesn't perform any phi predication, i.e. doesn't know about phi > > nodes, so later passes have to do some extra work to clean up > > * There are bugs, e.g. [2] which would require a rewrite of PRE to be > fixed. > > > > ### NewGVN > > > > The new algorithm implements the ideas described in [3] with some > > engineering optimizations by Dan (for example the set of touched > > instructions is represented using a Bitvector instead of a set because > > it's not uncommon for large functions where...

Dump the IR before newgvn and run through opt -newgvn. It should crash. I'm not familiar with your frontend but you might want to use -mllvm -opt-bisect-limit. On Sep 17, 2017 1:06 PM, "Andrew Kelley" <superjoe30 at gmail.com> wrote: Valgrind is strictly better than address sanitizer, is that right? It...

...look right (just hiding the assertion failure) >> >> On Sep 17, 2017 10:45 AM, "Andrew Kelley via llvm-dev" < >> llvm-dev at lists.llvm.org> wrote: >> >>> What do you think about this patch? >>> >>> --- a/llvm/lib/Transforms/Scalar/NewGVN.cpp >>> +++ b/llvm/lib/Transforms/Scalar/NewGVN.cpp >>> @@ -732,7 +732,7 @@ private: >>> MemoryPhi *getMemoryAccess(const BasicBlock *) const; >>> template <class T, class Range> T *getMinDFSOfRange(const Range &) >>> const; >>>...

...). >> > >> > Thanks, >> > >> > — >> > Mehdi >> > >> >> On Dec 26, 2016, at 10:45 AM, Davide Italiano via llvm-dev < >> llvm-dev at lists.llvm.org> wrote: >> >> >> >> Hi everybody. >> >> NewGVN was recently committed and a few minute ago I added a flag to >> >> enable the new experimental pass. >> >> >> >> For the brave soul, passing `-mllvm -enable-newgvn` should do the >> >> trick. We'll be happy to receive bug reports to analyze/fix,...

...en a long time coming. >> >> Once this is in, please do not turn it on by default immediately. Let's >> call for volunteers to find some of the most egregious miscompiles, fix >> them, and then turn this on by default. >> >There are no immediate plans to enable NewGVN by default (at least, >not in the near future). In fact, the mail that I originally wrote >doesn't at all mention the switch, neither any follow-ups from me or >Daniel, so, I'm not entirely sure where you got that idea from. If you >take a look more closely (at the mail, or a th...

Greetings, I just found a bug in NewGVN: https://bugs.llvm.org/show_bug.cgi?id=35839 I noticed that it's off by default in clang. I'm going to turn it off by default in zig too, since it has a bug. What's the current status? Are we trying to move towards using it by default in llvm 6.0.0? 7.0.0? Regards, Andrew Kelley htt...