similar to: [AArch64] Address computation folding

Hi, Indeed, the complex add is more expensive on all Cortex cores I know of. However there is an important point here that the code sequence we generate requires two registers live instead of one. In high regpressure loops, were probably losing performance. James On Wed, 11 Nov 2015 at 21:09, Tim Northover via llvm-dev < llvm-dev at lists.llvm.org> wrote: > On 11 November 2015 at

When I pass an empty string for cpu and features to createTargetMachine, and then use LLVMGetTargetMachineCPU() and LLVMGetTargetMachineFeatureString() to get the strings back, they are still empty. Is there a way to have llvm compute the effective cpu/features string, and provide it so that I can inspect it? I'm trying to figure out how the cpu/features string that I am explicitly passing,

I am building LLVM on OS X 10.7.5 with cmake. Under this configuration some ASan and UBSan tests are failing: $ make check-ubsan … ******************** Testing Time: 2.36s ******************** Failing Tests (11): UndefinedBehaviorSanitizer :: Float/cast-overflow.cpp UndefinedBehaviorSanitizer :: Integer/add-overflow.cpp UndefinedBehaviorSanitizer :: Integer/div-zero.cpp

On Thu, Jul 23, 2015 at 9:53 AM, Sky Flyer <skylake007 at googlemail.com> wrote: > Hello all, > > I don't find anything helping me understand the llvm bang operator. In the > llvm TableGen language reference, it only says: > > 'TableGen also has “bang operators” which have a wide variety of meanings:' > > I would be very thankful if someone can explain it

Hello, For a very simple loop where all IV users are post-inc users, I observed redundant add instructions in AArch64. From LSR debug, I can see initial formula for icmp is the one that transformed to a post-inc form in OptimizeLoopTermCond() and later expanded in post-inc mode. Based on the observation that the icmp is already a post-inc user, I hacked LSR to prevent the icmp from being

Hello all, I don't find anything helping me understand the llvm bang operator. In the llvm TableGen language reference, it only says: 'TableGen also has “bang operators” which have a wide variety of meanings: *'* I would be very thankful if someone can explain it to me. e.g. what does "!strconcat" or "!if" mean? Cheers ES -------------- next part --------------

Hi folks, Moving the discussion to llvm.dev. None of the changes we talked earlier help. Find attached the C source code that you can use to reproduce the issue. clang --target=aarch64-linux-gnu -c -mcpu=cortex-a57 -Ofast -fno-math-errno test.c -S -o test.s -mllvm -debug-only=licm LICM hoisting to while.body.lr.ph: %21 = load double** %arrayidx8, align 8, !tbaa !5 LICM hoisting to

Some more information … On Jan 4, 2013, at 6:56 PM, Meador Inge wrote: > I am building LLVM on OS X 10.7.5 with cmake. Under this configuration some ASan and UBSan tests > are failing: > > $ make check-ubsan > > … > > ******************** > Testing Time: 2.36s > ******************** > Failing Tests (11): > UndefinedBehaviorSanitizer ::

I also encounter this issue and solved it locally by implementing this 2 functions. - The linux version of StartSymbolizerSubprocess uses only POSIX function and can be reused as is on OS X (maybe we should move it in a new sanitizer_symbolizer_posix.cc file) - I have a simple implementation of GetListOfModules (see the attached file) but it required 10.6 at least. That said, implementing this

Hi All, I recently ran into an issue that was posted several years back on LLVMdev [1]. A brief recap of the problem is that when cross-compiling LLVM itself the configure/make scripts get confused when creating the needed build host tools. For example, building and configuring like: CC_FOR_BUILD='i686-pc-linux-gnu-gcc' CXX_FOR_BUILD='i686-pc-linux-gnu-g++'

On 08/02/2012 11:11 AM, Chris Lattner wrote: >> A new self-contained `LibCallSimplifier` class will be created. An instance >> of the class will be instantiated when running the `InstCombiner` pass. It's >> folding functionality will be invoked from `InstCombiner::tryOptimizeCall` and >> the implementation will be table-driven like `SimplifyLibCalls`. All of the

Can you send me actual LLVM IR or a preprocessed source from using -E? I don't have a machine handy that has headers that target that arch. On Tue Jan 13 2015 at 4:33:29 PM Daniel Berlin <dberlin at dberlin.org> wrote: > Anything other than noalias or mustalias should be getting passed down the > stack, so either that is not happening or CFL aa is giving better answers > and

> On May 27, 2016, at 2:50 PM, via llvm-dev <llvm-dev at lists.llvm.org> wrote: > > Hello, > > For a very simple loop where all IV users are post-inc users, I observed redundant add instructions in AArch64. > > From LSR debug, I can see initial formula for icmp is the one that transformed to a post-inc form in OptimizeLoopTermCond() and later expanded in post-inc

On 11/12/2012 09:35 AM, Duncan Sands wrote: > Hi, > > On 12/11/12 15:11, Meador Inge wrote: >> On 11/11/2012 11:58 PM, Chris Lattner wrote: >> >>>> Is there a particular sub-system size that makes sense to mark as owned? I have been >>>> reworking the library call simplification infrastructure recently and will be happy >>>> to sign up as

Hi All, I found some cases where a spill of a live range in a block is reloaded only in one of its successors, and there is no reload in other paths through other successors. Since the spill is reloaded only in a certain path, it must be okay to sink such spill close to its reloads. In the AArch64 code below, there is a spill(x2) in the entry, but this value is reloaded only in %bb.1, not in

On Jan 6, 2013, at 10:54 AM, Kostya Serebryany wrote: > Alex (in CC) is working on asan for OS X. I think he can handle ubsan as well. > (but give us a few days, we are all out of office due to holidays). OK, thanks. If there are no objections, then I can go ahead and commit the attached test case fix I mentioned before. Adding ALWAYS_INLINE to 'free_common' in r158885 changed

On 13 January 2015 at 22:11, Daniel Berlin <dberlin at dberlin.org> wrote: > This is caused by CFLAA returning PartialAlias for a query that BasicAA > can prove is NoAlias. > One of them is wrong. Which one? I'm not sure from your description that this is a chaining issue. PartialAlias doesn't chain and isn't supposed to, it's a final answer just like NoAlias and

From: llvm-dev [mailto:llvm-dev-bounces at lists.llvm.org] On Behalf Of Jun Lim via llvm-dev Sent: Thursday, February 22, 2018 11:05 AM Hi All, I found some cases where a spill of a live range in a block is reloaded only in one of its successors, and there is no reload in other paths through other successors. Since the spill is reloaded only in a certain path, it must be okay to sink such

On Aug 3, 2012, at 12:56 PM, Meador Inge <meadori at codesourcery.com> wrote: >>> An option for enabling/disabling library call simplification in `InstCombiner` >>> will be available. For backwards compatibility perhaps it should remain >>> '-simplify-libcalls'. The `NumSimplified` and `NumAnnotated` statistics shall >>> be added to

On 2018-02-22 11:14, gberry at codeaurora.org wrote: > FROM: llvm-dev [mailto:llvm-dev-bounces at lists.llvm.org] ON BEHALF OF > Jun Lim via llvm-dev > SENT: Thursday, February 22, 2018 11:05 AM > > Hi All, > > I found some cases where a spill of a live range in a block is > reloaded only in one of its successors, and there is no reload in > other paths through other