search for: sgemm

...tails: -------------- Sun Blade 1000, Solaris 8, Sun ONE Compiler Suite (rebranded Forte 7), R versions 1.6.0 and 1.6.2 - suspect others are affected, too Problem description: ------------------- Using the recommended env var setting from p.15 of the Admin Guide, the configure script fails the sgemm sunperf test, even though the sunperf libraries are available Not linking with libsunperf causes drastic performance hits on some simple benchmarks I've tried (e.g. matrix multiplication can take up to 100 times as long for a sufficiently large matrix (n=5000)) Solution: -------- Configure s...

...ock of operations is bounded at either end by an OpenCL barrier (an AMDIL fence instruction). The loads and multiply/adds are ordinarily interleaved... that is, the IR going in to code generation looks like: %39 = load float addrspace(3)* getelementptr inbounds ([16 x [17 x float]] addrspace(3)* @sgemm.b, i32 0, i32 0, i32 0), align 4 %40 = call float @llvm.fmuladd.f32(float %37, float %39, float %c0.037) nounwind %41 = load float addrspace(3)* getelementptr inbounds ([16 x [17 x float]] addrspace(3)* @sgemm.b, i32 0, i32 0, i32 1), align 4 %42 = call float @llvm.fmuladd.f32(float %37, floa...

...ng llvm-ir in the fat binary may have the same performance issues mentioned below. The fat binary discussed in the proposal has provision for storing the isa/llvm-ir. My point is instead of llvm-ir it shd be something like spir. Ok - so lets see some data. #1 Benchmarks showing at least SPIR dgemm/sgemm performance #2 Some logical explanation why all the extra work for SPIR when LLVM IR is native Basically besides an opinion or because it's "shiny" some solid technical reason. I hate to repeat myself, but again.. why on earth would a solution which is closed source be preferred...

I believe that actually means configure was able to use the BLAS you specified (MKL in this case). You can scan through config.log to make sure. Andy > From: > > Fedora Core 3 > > I installed the Intel MLk, > > and tried : --with-blas="-lmkl -lguide -lpthread" > > and got : External libraries: readline, BLAS(generic) > > thus I'm

On 08/11/14 01:03 PM, Das, Dibyendu wrote: > I didn’t see SPIR discussed anywhere. This isn't OpenCL and depending on OpenCL for OpenMP may not really make sense. While I have my own opinions - If you feel strongly that it will help enable higher performance somewhere please list those reasons. ---------- More specifically LLVM has a native AMD dGPU backend that is tightly coupled to the

...ne required checking for __setfpucw... no checking for working calloc... no checking for working finite... yes checking for working log... yes checking for working log1p... yes checking for working strptime... yes checking whether ftell works correctly on files opened for append... yes checking for sgemm... no checking for ATL_xerbla in -latlas... no checking for sgemm in -lblas... yes checking for dgemm in -ldgemm... no checking for sgemm in -lblas... (cached) yes checking for sgemm in -lessl... yes checking wchar.h usability... yes checking wchar.h presence... yes checking for wchar.h... yes chec...

On Tue, Mar 19, 2019 at 4:11 AM Graham Hunter <Graham.Hunter at arm.com> wrote: > Hi Eric and Chandler, > > I appreciate your concerns; I don't think the impact will be that great, > but then it's > rather easy for me to keep SVE in mind when working on other parts of the > codebase > given how long I've spent working on it. > > Are there any