similar to: BSWAP matching in codegen

>> Are you sure there isn't any test coverage? As far as I can tell, the tests from https://reviews.llvm.org/rL133503 are still in the tree. I looked at those, but none of them include the full pattern that decomposes into bswap and rol. I debugged through the X86 bswap.ll test and verified none of those cases make it through MatchBSwapHWord (they get handled in MatchBSwapHWordLow

On 12/9/2016 11:03 AM, Jim Lewis via llvm-dev wrote: > > Thanks, that helps enormously! The issue is that the match is supposed > to support both cascade and tree OR patterns, but there appears to be > a problem with the tree matching. Both test1 and test6 in the ARM > tests exercise the cascade pattern, and I remember now our fix is > confined to the tree case. > > I

Hi all, As of r141677 I have a failing regression test, see below. This is for LLVM built with clang on a Intel Atom running FreeBSD8.2. Should I file a bug for this? Thanks, Ed. ******************** TEST 'LLVM :: CodeGen/X86/bswap.ll' FAILED ******************** Script: -- /usr/home/emeewis/build/llvm-debug-clang-configure/Debug+Asserts/bin/llc <

On 13.10.2011, at 15:57, Edward Meewis wrote: > Hi all, > > As of r141677 I have a failing regression test, see below. > > This is for LLVM built with clang on a Intel Atom running FreeBSD8.2. > > Should I file a bug for this? Fixed in r141863. - Ben

Hi, Can I ask what is the status of load widening. It seems there is no load widening on IR at all. // Paweł On Wed, Oct 5, 2016 at 1:49 PM Artur Pilipenko via llvm-dev < llvm-dev at lists.llvm.org> wrote: > Philip and I talked about this is person. Given the fact that load > widening in presence of atomics is irreversible transformation we agreed > that we don't want to do

Ok, thanks. Are there any plans to reintroduce it on the IR level? I'm not confident this is strictly necessary, but in some cases not having load widening ends up really bad. Like in the case where vectorizer tries to do something about it: https://godbolt.org/z/60RuEw https://bugs.llvm.org/show_bug.cgi?id=42708 At the current state I'm forced to use memset() to express uint64 load from

On 5/2/2013 4:20 PM, Rafael Espíndola wrote: > > I don't get any errors compiling this. Do you know if it was fixed? Do > you get an error only for some particular target? What is the 'clang > -cc1' command line that you seed in the error? I see this error in our compiler, but not with the trunk. I guess it's been fixed, but I don't know which commit did it.

If we do load combining at the IR level, one thing we'll need to give some thought to is atomicity.  Combining two atomic loads into a wider (legal) atomic load is not a reversible transformation given our current specification. I've been thinking about a concept I've been tentatively calling "element wise atomicity" which would make this a reversible transform by

Hi. There are few one-line bugs Andrey Karpov have found with static analisys. He wrote a big article in russian on http://habrahabr.ru/blogs/compilers/125626/ for advertising purposes of static analyzer for Visual Studio his company developed. Most of the problems are easy to fix, so I list them in here for trunk version. Also few problems in clang code were found, I don't list them in here.

bswapdi2 for i386 is correct Bits 31:0 of the source are loaded into edx. Bits 63:32 are loaded into eax. Those are each bswapped. The ABI for the return is edx contains bits [63:32] and eax contains [31:0]. This is opposite of how the register were loaded. ~Craig On Sun, Nov 25, 2018 at 10:36 AM Craig Topper <craig.topper at gmail.com> wrote: > bswapsi2 on the x86-64 isn't using

Hi @ll, targetting i386, LLVM/clang generates wrong code for the following functions: unsigned long __bswapsi2 (unsigned long ul) { return (((ul) & 0xff000000ul) >> 3 * 8) | (((ul) & 0x00ff0000ul) >> 8) | (((ul) & 0x0000ff00ul) << 8) | (((ul) & 0x000000fful) << 3 * 8); } unsigned long long __bswapdi2(unsigned long

Hello List, I am on the hook to instrument a piece of legacy LLVM IR code, and then we are planning to feed to the SeaHorn framework for some model checking tasks. After the instrumentation, I tried to use llc (version 3.9) to compile the IR code, and it works fine. However, when I try to use llc (version 3.8.1, the default llvm version of SeaHorn) to compile the IR code, it shows the following

The LegalizeDAG.cpp file has this code in SelectionDAGLegalize::PromoteNode: case ISD::BSWAP: { unsigned DiffBits = NVT.getSizeInBits() - OVT.getSizeInBits(); Tmp1 = DAG.getNode(ISD::ZERO_EXTEND, dl, NVT, Tmp1); Tmp1 = DAG.getNode(ISD::BSWAP, dl, NVT, Tmp1); Tmp1 = DAG.getNode(ISD::SRL, dl, NVT, Tmp1, DAG.getConstant(DiffBits, TLI.getShiftAmountTy()));

I am not entirely sure this is safe. Transforming this to an fsub could change the value stored on platforms that implement negates using arithmetic instead of with bitmath (such as ours) and either canonicalize NaNs or don’t support denormals. This is actually important because this kind of bitmath on floats is very commonly used as part of algorithms for complex math functions that need to get

Hi! Anyone else experiencing performance problems when using the new monorepo? My experience is that performance of gitk (and git log) sometimes is really bad when working in the monorepo. I've mainly seen it when using gitk on specific files/directories, but since gitk seems to be using "git log --no-color -z --pretty=raw --show-notes --parents --boundary HEAD -- <file>" it

I just compiled the two attached files in 32-bit mode and ran it. It printed efcdab8967452301. I verified via objdump that the my_bswap function contains the follow assembly which I believe matches the assembly you linked to on godbolt. _my_bswap: 1f70: 55 pushl %ebp 1f71: 89 e5 movl %esp, %ebp 1f73: 8b 55 08 movl 8(%ebp), %edx 1f76: 8b 45 0c movl 12(%ebp), %eax 1f79: 0f c8

Have you looked at these? https://llvm.org/docs/LangRef.html https://llvm.org/docs/ExtendingLLVM.html On Tue, Mar 26, 2019 at 9:06 AM div code via llvm-dev < llvm-dev at lists.llvm.org> wrote: > Of course, in this sense they are not platform-dependent. I just want to > write a semantics-correct version of such intrinsics and let my static > analyzer goes smoothly. > > On

On 10/03/2011 09:36 PM, Villmow, Micah wrote: > One of the projects I am working on with others is to make LLVM-IR endian agnostic. > > > > So, I am sending out this proposal for feedback to the LLVM community. I’ve attached > > pretty version of the proposal in PDF format and pasted a 80-column safe text version > > below. > > > > A second smaller

One of the arguments for doing this earlier is inline cost perception of the original pattern. Reading i32/i64 by bytes look much more expensive than it is and can prevent inlining of interesting function. Inhibiting other optimizations concern can be addressed by careful selection of the pattern we’d like to match. I limit the transformation to the case when all the individual have no uses other

On Feb 23, 2009, at 2:49 PM, Scott Michel wrote: > On Mon, Feb 23, 2009 at 2:19 PM, Nate Begeman <natebegeman at me.com> > wrote: > > On Feb 23, 2009, at 1:46 PM, Chris Lattner wrote: > > In my opinion, the proper direction for shuffles is: > > 1. Back out your patch. > 2. Move the functionality of "is splat" etc to method somewhere, e.g. > on