similar to: LLVM X86 AVX Backend

Hello, After generating AVX code for large no of iterations i came to realize that it still uses only 2 registers zmm0 and zmm1 when the loop urnroll factor=1024, i wonder if this register allocation allows operations in parallel? Also i know all the elements within a single vector instruction are computed in parallel but does the elements of multiple instructions computed in parallel? like are

On 10/30/13 03:17 AM, Chandler Carruth wrote: > On Mon, Oct 28, 2013 at 6:07 PM, "C. Bergström" > <cbergstrom at pathscale.com <mailto:cbergstrom at pathscale.com>> wrote: > > On 10/29/13 07:27 AM, Chandler Carruth wrote: > > On Mon, Oct 28, 2013 at 5:06 PM, "C. Bergström" > <cbergstrom at pathscale.com

On Wed, Jul 8, 2015 at 7:08 AM, Ilia Mirkin <imirkin at alum.mit.edu> wrote: > On Tue, Jul 7, 2015 at 8:07 PM, C Bergström <cbergstrom at pathscale.com> wrote: >> On Wed, Jul 8, 2015 at 6:58 AM, Ben Skeggs <skeggsb at gmail.com> wrote: >>> On 8 July 2015 at 09:53, C Bergström <cbergstrom at pathscale.com> wrote: >>>> regarding >>>>

Hi Ahmed, >From what can be seen in the code snippet you provided, the reuse of XMM0 and XMM1 across loop-unroll instances does not inhibit instruction-level parallelism. Modern X86 processors use register renaming that can eliminate the dependencies in the instruction stream. In the example you provided, the processor should be able to identify the 2-vloads + vadd + vstore sequences as

On 10/29/13 07:27 AM, Chandler Carruth wrote: > On Mon, Oct 28, 2013 at 5:06 PM, "C. Bergström" > <cbergstrom at pathscale.com <mailto:cbergstrom at pathscale.com>> wrote: > > fuzzy://How much "heads up" > > > One full release cycle, so approximately 6 months before a release If it's 3-6 months from *today* before something hits clang

Hi, Zvi, I agree. In the context of targeting the KNL, however, I'm a bit concerned about the addressing, and specifically, the size of the resulting encoding: > vmovdqu32 zmm0, zmmword ptr [rax + c+401280] ;load b[401280] in > zmm0 > > vpaddd zmm1, zmm1, zmmword ptr [rax + b+401344] > ; zmm1<-zmm1+b[401344] The KNL can only

On Sun, Jun 7, 2015 at 2:34 AM, Eric Christopher <echristo at gmail.com> wrote: > > > On Sat, Jun 6, 2015 at 12:31 PM C Bergström <cbergstrom at pathscale.com> > wrote: >> >> On Sun, Jun 7, 2015 at 2:22 AM, Eric Christopher <echristo at gmail.com> >> wrote: >> > >> > >> > On Sat, Jun 6, 2015 at 5:02 AM C Bergström

On Sun, Jun 7, 2015 at 2:22 AM, Eric Christopher <echristo at gmail.com> wrote: > > > On Sat, Jun 6, 2015 at 5:02 AM C Bergström <cbergstrom at pathscale.com> wrote: >> >> On Sat, Jun 6, 2015 at 6:24 PM, Christos Margiolas >> <chrmargiolas at gmail.com> wrote: >> > Hello, >> > >> > Thank you a lot for the feedback. I believe

On Mon, Oct 28, 2013 at 6:07 PM, "C. Bergström" <cbergstrom at pathscale.com>wrote: > On 10/29/13 07:27 AM, Chandler Carruth wrote: > > On Mon, Oct 28, 2013 at 5:06 PM, "C. Bergström" <cbergstrom at pathscale.com<mailto: >> cbergstrom at pathscale.**com <cbergstrom at pathscale.com>>> wrote: >> >> fuzzy://How much

On Jun 10, 2014, at 5:36 PM, C. Bergström <cbergstrom at pathscale.com> wrote: > On 06/11/14 06:58 AM, Ted Kremenek wrote: >> ** NOTE: This is a compiler job announcement. ** >> >> The Apple Source Languages team is looking for exceptional engineers to work on the Swift programming language: > Maybe this has been asked already and I missed it - Will Swift be open

On Tue, Jul 07, 2015 at 08:13:28PM -0400, Ilia Mirkin wrote: > On Tue, Jul 7, 2015 at 8:11 PM, C Bergström <cbergstrom at pathscale.com> wrote: > > On Wed, Jul 8, 2015 at 7:08 AM, Ilia Mirkin <imirkin at alum.mit.edu> wrote: > >> On Tue, Jul 7, 2015 at 8:07 PM, C Bergström <cbergstrom at pathscale.com> wrote: > >>> On Wed, Jul 8, 2015 at 6:58 AM, Ben

So that people have a shorter link to go to that entry page directly. It's especially useful for non-LLVM folks who want to try out LLVM's CUDA support. Many researchers fall into this category btw because LLVM used to support very little CUDA. They don't like to search llvm.org for what they want. On Tue, Nov 10, 2015 at 2:59 PM, C Bergström <cbergstrom at pathscale.com> wrote:

On Wed, Jul 8, 2015 at 6:58 AM, Ben Skeggs <skeggsb at gmail.com> wrote: > On 8 July 2015 at 09:53, C Bergström <cbergstrom at pathscale.com> wrote: >> regarding >> -------- >> Fixed address allocations weren't going to be part of that, but I see >> that it makes sense for a variety of use cases. One question I have >> here is how this is intended

On 04/11/15 09:08, cbergstrom at pathscale.com wrote: > Is anyone actually and or actively working on this? > Github.com/pathscale/pscnv is totally bitrot but waaay more portable > base. Nouveau made hard Linux assumptions that will be difficult to > overcome afaik. As pointed out by Ilia, this is not true anymore. Nouveau can also partially run in the userspace, the hard

On 06/11/14 08:39 PM, Chandler Carruth wrote: > > On Wed, Jun 11, 2014 at 1:44 AM, Ted Kremenek <kremenek at apple.com > <mailto:kremenek at apple.com>> wrote: > > On Jun 10, 2014, at 5:36 PM, C. Bergström > <cbergstrom at pathscale.com <mailto:cbergstrom at pathscale.com>> wrote: > > > On 06/11/14 06:58 AM, Ted Kremenek wrote: >

Dear MS, Will you be cleaning up and encouraging people to get the work on github upstream? https://github.com/Microsoft/llvm/commits/MS Thanks

Hi - Not sure if anyone else saw this or cares about a decompiler (not personally tested) https://github.com/draperlaboratory/fracture I wonder if they have been in contact with anyone in the community in getting this upstreamed. Does it look interesting to anyone else? (thoughts/random comments/feedback)

On 5 May 2016 at 13:23, C Bergström <cbergstrom at pathscale.com> wrote: > Is the list PG, PG-13, R or at what level do "we" adults all consider > "ok". Even on broadcast tv (in the US) you'll hear some profanity. > (context) > https://www.fcc.gov/consumers/guides/obscene-indecent-and-profane-broadcasts Excellent context! > Some people have pointed

For those driving c++11 in clang/llvm - Would it generally be acceptable to have a "sunrise" period where the preliminary evaluation has been done (buildbots, compiler evaluate.. etc) and the 1st actual c++11 commit hits the repo. (30-60 days?) ------------- My concern/thoughts - When we swap out STDCXX for libc++ - We aren't able to self host clang. This could be entirely *our*

On Sun, Oct 16, 2016 at 11:39 PM, C Bergström <cbergstrom at pathscale.com> wrote: > Is there any particular reason you're not using the open source NVIDIA > Tegra drivers? > > This is what is being used for the TX1 (Sorry I don't know if it's the > same for the TK1) > http://nv-tegra.nvidia.com/gitweb/?p=linux-3.10.git;a=summary That's ... not particularly