search for: thielemann

...is not enabled in your llc. This feature was enabled > just recently (late of April). I forgot to mention that I am using recent LLVM-3.1 and in principle my llc knows about avx as I have shown in the second example. But avx does not seem to be used by default. On Thu, 24 May 2012, Henning Thielemann wrote: > $ llc -o - -mattr avx avx.ll > .file "avx.ll" > .text > .globl _fun1 > .align 16, 0x90 > .type _fun1, at function > _fun1: # @_fun1 > .cfi_startproc > # BB#0:...

I wonder why AVX is not used automatically if available at the host machine. In contrast to that, SSE41 instructions (like pmulld) are automatically used if the host machine supports SSE41. E.g. $ cat avx.ll define void @_fun1(<8 x float>*, <8 x float>*) { _L1: %x = load <8 x float>* %0 %y = load <8 x float>* %1 %z = fadd <8 x float> %x, %y store

Henning, I believe the code that is supposed to do this is in: lib/Target/X86/X86Subtarget.cpp in X86Subtarget::AutoDetectSubtargetFeatures() Is there a bug in that function? -Hal On Thu, 24 May 2012 23:56:48 +0200 (CEST) Henning Thielemann <llvm at henning-thielemann.de> wrote: > > On Thu, 24 May 2012, Pan, Wei wrote: > > > Very likely AVX is not enabled in your llc. This feature was > > enabled just recently (late of April). > > I forgot to mention that I am using recent LLVM-3.1 and in principle...

On Thu, 24 May 2012, Hal Finkel wrote: > Henning, > > I believe the code that is supposed to do this is in: > lib/Target/X86/X86Subtarget.cpp in > X86Subtarget::AutoDetectSubtargetFeatures() > Is there a bug in that function? I read there: // FIXME: AVX codegen support is not ready. //if ((ECX >> 28) & 1) { X86SSELevel = AVX; ToggleFeature(X86::FeatureAVX); }

Henning, Are you looking at trunk? I believe that in trunk this has been uncommented. -Hal On Fri, 25 May 2012 00:47:42 +0200 (CEST) Henning Thielemann <llvm at henning-thielemann.de> wrote: > > On Thu, 24 May 2012, Hal Finkel wrote: > > > Henning, > > > > I believe the code that is supposed to do this is in: > > lib/Target/X86/X86Subtarget.cpp in > > X86Subtarget::AutoDetectSubtargetFeatures() >...

I want to generate Haskell bindings to LLVM intrinsics. In a first attempt I wrote a little parser that reads IntrinsicsX86.td and outputs a Haskell module. E.g. the definition def int_x86_avx_max_ps_256 : GCCBuiltin<"__builtin_ia32_maxps256">, Intrinsic<[llvm_v8f32_ty], [llvm_v8f32_ty, llvm_v8f32_ty], [IntrNoMem]>; is turned into

I have the module: target triple = "x86_64-pc-linux-gnu" define void @round(<8 x float>*) { _L1: %1 = load <8 x float>, <8 x float>* %0 %2 = fadd <8 x float> %1, zeroinitializer store <8 x float> %2, <8 x float>* %0 ret void } Optimizing it with -O3 does not remove the add with zero. Also options like -enable-no-infs-fp-math

Hi, I got my manual dsync finally working and can replicate mails by initiate it from comandline (doveadm sync -u mail at domain.dom -d), but it does not run by itself (per dovecot replication). When does dovecot invoke the replication normally? Thanks in advance Daniel

Hi, I bought repo access for 0,00$ one minute ago. If its free now then I wanna say thank you very much, otherwise somebody should fix it :) Take a look here: http://shop.dovecot.fi/home/8-dovecot-ee-repository-access.html Screen: http://www.directupload.net/file/d/3622/oxyo89zm_png.htm Regards

Hi all, I have some old code using the JIT via the LLVM-3.0-C API. I want to upgrade to newer versions of LLVM. As a simple example I wrote a C program that creates the following function and calls it: ; ModuleID = 'round-avx.bc' target triple = "x86_64-pc-linux-gnu" define void @round(<8 x float>*) { _L1: %1 = load <8 x float>* %0 %2 = call <8 x

Hi, i tried to setup replication since some days but I'am stuck now. My conf: # 2.2.10: /etc/dovecot/dovecot.conf # OS: Linux 2.6.32-431.3.1.el6.x86_64 x86_64 CentOS release 6.5 (Final) dsync_remote_cmd = ssh -l%{login} %{host} /usr/bin/doveadm dsync-server -u%u -l%{lock_timeout} -n%{namespace} log_path = /var/log/dovecot mail_gid = 5000 mail_home = /var/vmail/%d/%n mail_location =

I want to raise attention for my ticket http://llvm.org/bugs/show_bug.cgi?id=9405 Compiling the Haskell LLVM bindings is always a lot of trying around and patching, its configure procedure did not get better over the years. However, it would be straightforward if llvm paths could be queried via the pkg-config system. This is directly supported by Haskell's packaging system Cabal and I

Hi all, it's Advent time and I took the opportunity to program a song for you. The sound, i.e. speech synthesis and reverb, is rendered by the LLVM-JIT and the synthesizer is controlled via ALSA-MIDI by the Haskell live sequencer. http://www.youtube.com/watch?v=0EQCgi5qa3E "Alta trinita beata" You find additional information in the video description. Best, Henning