search for: veclib

...let's start only with the directly-legal calls". Apparently, what we have right now in the trunk is "not legal enough". I'll work on the patch to stop bleeding while we continue to discuss legalization topic. I suppose 1) LV only solution (let LV emit already legalized VECLIB calls) is certainly not scalable. It won't help if VECLIB calls are generated elsewhere. Also, keeping VF low enough to prevent the legalization problem is only a workaround, not a solution. 2) Assuming that we have to go to IR to IR pass route, there are 3 ways to think: a. Go wit...

...tly, what we have right now >> >> in the trunk is “not legal enough”. I’ll work on the patch to stop >> bleeding while we continue to discuss legalization topic. >> >> >> >> I suppose >> >> 1) LV only solution (let LV emit already legalized VECLIB calls) is >> certainly not scalable. It won’t help if VECLIB calls >> are generated elsewhere. Also, keeping VF low enough to prevent the >> legalization problem is only a workaround, >> not a solution. >> >> 2) Assuming that we have to go to IR to IR pass r...

...o solve, but I don't know why adding to include/llvm/CodeGen/RuntimeLibcalls.def is a problem in itself. Certainly, it's a mess that could be organized, especially so we're not repeating everything for each data type as we do right now. So yes, I think that would allow us to remove the VecLib mappings because we are always waiting until codegen to make the translation from generic IR to target-specific libcall. Or is there some reason that the vectorizer needs to be aware of those libcalls? On Mon, Jul 2, 2018 at 11:52 AM, Saito, Hideki <hideki.saito at intel.com> wrote: > &g...

Illustrative Example: clang -fveclib=SVML -O3 svml.c -mavx #include <math.h> void foo(double *a, int N){ int i; #pragma clang loop vectorize_width(8) for (i=0;i<N;i++){ a[i] = sin(i); } } Currently, this results in a call to <8 x double> __svml_sin8(<8 x double>) after the vectorizer. This is 8-element...

...2018 11:49 AM > *To:* Saito, Hideki <hideki.saito at intel.com> > *Cc:* Venkataramanan Kumar <venkataramanan.kumar.llvm at gmail.com>; > llvm-dev at lists.llvm.org; Masten, Matt <matt.masten at intel.com>; > dccitaliano at gmail.com > *Subject:* Re: [llvm-dev] [RFC][VECLIB] how should we legalize VECLIB > calls? > >   > > It may not be a full solution for the problems you're trying to solve, > but I don't know why adding to > include/llvm/CodeGen/RuntimeLibcalls.def is a problem in itself. > Certainly, it's a mess that could be org...

...12:12 PM > To: Hal Finkel <hfinkel at anl.gov>; Saito, Hideki <hideki.saito at intel.com>; > Sanjay Patel <spatel at rotateright.com>; mzolotukhin at apple.com > Cc: dccitaliano at gmail.com; Masten, Matt <matt.masten at intel.com> > Subject: RE: [llvm-dev] [RFC][VECLIB] how should we legalize VECLIB calls? > > Hi Hal, > > > __svml_sin8 (plus whatever shuffles are necessary). > > The vectorizer should do this. > > It should not generate calls to functions that don't exist. > > I'm not sure how vectorizer will do this, consi...

R-devel has the --with-lapack flag for configure. If you build R with --with-blas="-framework vecLib" --with-lapack="--framework vecLib" then the build goes through, using the native optimized BLAS and Lapack in /System/Library/Frameworks/vecLib.framework. Also, it works, in the sense that it does eigenvalue problems correctly. I don't have any timings yet, because I just upgr...

Hi all, I am submitting the following RFC [1] to re-implement -fveclib via OpenMP constructs. The RFC was discussed during a round table at the last LLVM developer meeting, and presented during the BoF [2]. The proposal is published on Phabricator, for the purpose of keeping track of the comments, and it now ready for a review from a wider audience after being polish...

OS X 10.2 and higher comes standard with highly optimized versions of BLAS and Lapack in /Systems/Library/Frameworks/vecLib.framework. It seems that even for double precision they do much better. See http://sthmac.magnet.fsu.edu/benchmarks/ I am not sure how these numbers would look on G3 Macs, but obviously for double precision there is not much reliance on Altivec. So I tried to configure R-devel with --with-blas=&...

...ccelerate 1.2.2 (Accelerate 1.2.2) /System/Library/Frameworks/Accelerate.framework/Versions/A/Accelerate 0x91e4e000 - 0x91f33fff com.apple.vImage 2.4 /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vImage.framework/Versions/A/vImage 0x91f3b000 - 0x91f5afff com.apple.Accelerate.vecLib 3.2.2 (vecLib 3.2.2) /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/A/vecLib 0x91fc6000 - 0x92034fff libvMisc.dylib /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/A/libvMisc.dylib 0x9203f000 - 0x920d4...

...ccelerate 1.2.2 (Accelerate 1.2.2) /System/Library/Frameworks/Accelerate.framework/Versions/A/Accelerate 0x91c9f000 - 0x91d84fff com.apple.vImage 2.4 /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vImage.framework/Versions/A/vImage 0x91d8c000 - 0x91dabfff com.apple.Accelerate.vecLib 3.2.2 (vecLib 3.2.2) /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/A/vecLib 0x91e17000 - 0x91e85fff libvMisc.dylib /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/A/libvMisc.dylib 0x91e90000 - 0x91f25...

...e 1.1.1 (Accelerate 1.1.1) /System/Library/Frameworks/Accelerate.framework/Versions/A/ Accelerate 0x91c46000 - 0x91d30fff com.apple.vImage 2.0 /System/Library/ Frameworks/Accelerate.framework/Versions/A/Frameworks/ vImage.framework/Versions/A/vImage 0x91d38000 - 0x91d57fff com.apple.Accelerate.vecLib 3.1.1 (vecLib 3.1.1) /System/Library/Frameworks/Accelerate.framework/Versions/A/ Frameworks/vecLib.framework/Versions/A/vecLib 0x91dc3000 - 0x91e28fff libvMisc.dylib /System/Library/Frameworks/ Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/ A/libvMisc.dylib 0x91e32000 -...

...s/SearchKit.framework/Versions/A/SearchKit 0x94b20000 - 0x94badfff com.apple.ink.framework 55.8 /System/Library/Frameworks/Carbon.framework/Versions/A/Frameworks/Ink.framework/Versions/A/Ink 0x954c0000 - 0x95ac6fff libBLAS.dylib /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/A/libBLAS.dylib 0x95b20000 - 0x95df0fff libLAPACK.dylib /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/A/libLAPACK.dylib 0x95e40000 - 0x95eadfff libvDSP.dylib /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/...

No, blas/veclib is tested, so aprt this extreme case you should report some other more commonly used cases in which something fails on OS X. This will help us to work it out. As said, I'll try some tests without using veclib and let you know. I've fowarded this mail to r-devel, which seems to be the ri...

...l work for user driven ones, but user driven have additional complexities that will only get in the way of the standard discussion. Comments inline... On Wed, 12 Dec 2018 at 03:47, Francesco Petrogalli via llvm-dev <llvm-dev at lists.llvm.org> wrote: > Summary > ======= > > New `veclib` directives in clang I know this is not new but, why "fveclib"? From the review, I take this is the same as GCC's "mveclibabi", and if it is, why come up with a new name for the same thing? If it's not, what justifies implementing a different way of handling the same c...

...nts/MacOS/ColorManagement_W 0x555b000 - 0x5595fff vfonts.so /Library/Frameworks/R.framework/Resources/modules/vfonts.so 0x5650000 - 0x5658fff com.apple.print.converter.plugin 3.0 /System/Library/Printers/CVs/Converter.plugin/Contents/MacOS/Converter 0x76c02000 - 0x76c12fff com.apple.Accelerate.vecLib 3.0.3 (vecLib 3.0.3) /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/A/vecLib 0x806c0000 - 0x806e9fff libxslt.1.dylib /usr/lib/libxslt.1.dylib 0x80830000 - 0x8090efff libxml2.2.dylib /usr/lib/libxml2.2.dylib 0x86e30000 - 0x86e52fff libCFilter.A.dyli...

...ccelerate 1.2.1 (Accelerate 1.2.1) /System/Library/Frameworks/Accelerate.framework/Versions/A/Accelerate 0x91cd7000 - 0x91db7fff com.apple.vImage 2.3 /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vImage.framework/Versions/A/vImage 0x91dbf000 - 0x91ddefff com.apple.Accelerate.vecLib 3.2.1 (vecLib 3.2.1) /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/A/vecLib 0x91e4a000 - 0x91eb8fff libvMisc.dylib /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/A/libvMisc.dylib 0x91ec3000 - 0x91f57...

...e 1.1.1 (Accelerate 1.1.1) /System/Library/Frameworks/Accelerate.framework/Versions/A/ Accelerate 0x91c46000 - 0x91d30fff com.apple.vImage 2.0 /System/Library/ Frameworks/Accelerate.framework/Versions/A/Frameworks/ vImage.framework/Versions/A/vImage 0x91d38000 - 0x91d57fff com.apple.Accelerate.vecLib 3.1.1 (vecLib 3.1.1) /System/Library/Frameworks/Accelerate.framework/Versions/A/ Frameworks/vecLib.framework/Versions/A/vecLib 0x91dc3000 - 0x91e28fff libvMisc.dylib /System/Library/Frameworks/ Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/ A/libvMisc.dylib 0x91e32000 -...

After I upgraded my installation of R for OS X (Tiger) to 1.12, it consistently crashes when calling the function acf() [stats lib]. Roberto Osorio

...lbox.framework/Versions/A/HIToolbox > 0x92e70000 - 0x9336ffff com.apple.AppKit 6.3.8 (743.39) /System/ > Library/Frameworks/AppKit.framework/Versions/C/AppKit > 0x9398f000 - 0x939fcfff libvDSP.dylib /System/Library/ > Frameworks/Accelerate.framework/Versions/A/Frameworks/ > vecLib.framework/Versions/A/libvDSP.dylib > 0x93a20000 - 0x93a34fff libcups.2.dylib /usr/lib/libcups.2.dylib > 0x93a50000 - 0x93a54fff libmathCommon.A.dylib /usr/lib/system/ > libmathCommon.A.dylib > 0x940f0000 - 0x94108fff com.apple.WebServices 1.1.1 (1.1.0) / > System/Library...