search for: reconverging

Dear all, I would like announce a round table planned for the upcoming LLVM Dev meeting next week that will cover various topics related to the use of LLVM in the compiler stacks for the GPUs. Here is the initial list of discussion topics: - Canonicalization vs. GPUs: Type mutation; - Control flow mutation (graphics shaders are more sensitive to this); - Divergence/reconvergence sensitivity;

Hello, I am new to this forum and have done my best to search for my answer before emailing, but have been unsuccessful. I have an application where I would like to utilize the speex AEC library for canceling hybrid echo. I thought it would be useful to save the coefficients of the filter for reuse, especially since the characteristics of the echo path will not vary. The specific application for

...mance impacts. So I reject this type of proposal as well. The literature from practicioners on SPMD/SIMT control flow (a lot of it targeting a hardware audience rather than a compiler audience) does not concern itself with this problem to my knowledge, but there is a commonly recurring theme of reconverging or rejoining threads at explicit instructions and/or post-dominators. This suggests a viable path towards a solution to me. The SPIR-V spec has a notion of explicitly structured control flow with merge basic blocks. It also defines "dynamic instances" of instructions that are disting...

Thanks, Marco! If there is enough interest in this topic we can also organize a separate round table for this discussion. Cheers, Anastasia ________________________________ From: Marco Antognini <Marco.Antognini at arm.com> Sent: 18 October 2019 14:42 To: Anastasia Stulova <Anastasia.Stulova at arm.com>; Simone Atzeni via llvm-dev <llvm-dev at lists.llvm.org>; clang developer

On 20.12.18 18:03, Connor Abbott wrote: > We already have the notion of "convergent" functions like > syncthreads(), to which we cannot add control-flow dependencies. > That is, it's legal to hoist syncthreads out of an "if", but it's > not legal to sink it into an "if".  It's not clear to me why we > can't have

Hey everyone, My company is beginning to look at using SSD drives in our CentOS based servers. Does C5 and C6 support TRIM and other "required" functions for the SSD to operate? Thanks, Andrew Reis Microsoft Certified Technology Specialist CompTIA Network+ Networking/Systems Analyst Webmaster DBMS Inc.

I don't see how this would fix the continue vs. nested loop problem I explained earlier. That is, how would this prevent turning: for (...) { ballot(); if (... /* non-uniform */) continue; } into for (...) { do { ballot(); } while (... /* non-uniform */); } and vice versa? Note that there's no duplication going on here, and the single-threaded flow of control is

...hing you'd want to do with a GPU. > > Irreducible control flow has to be handled somehow, and linearization > is the only transform I know of, that will handle everything. I'm not > sure what the execution model says about irreducible control flow. For what it's worth, the reconverging CFG approach can also handle arbitrary irreducible control flow. Whether (or how) it can do this while being compatible with whatever semantics we come up with for cross-lane operations remains to be seen. >> >  Another case is DCE, >> >  where a ballot() could be eliminated,...

...t; >> Irreducible control flow has to be handled somehow, and linearization > >> is the only transform I know of, that will handle everything. I'm not > >> sure what the execution model says about irreducible control flow. > > > > For what it's worth, the reconverging CFG approach can also handle > > arbitrary irreducible control flow. > > > > Whether (or how) it can do this while being compatible with whatever > > semantics we come up with for cross-lane operations remains to be seen. > > I am wondering about execution order and if...

On Fri, Jan 25, 2019 at 3:05 AM Jan Sjodin <jan_sjodin at yahoo.com> wrote: > > > for (...) { > > ballot(); > > if (... /* non-uniform */) continue; > > } > > > > into > > > > for (...) { > > do { > > ballot(); > > } while (... /* non-uniform */); > > } > > I'm not sure if I follow

On Mon, Jan 28, 2019 at 9:09 PM Jan Sjodin <jan_sjodin at yahoo.com> wrote: > > > for (int i = 0; i < 2; i++) { > > foo = ballot(true); // ballot 1 > > > > if (threadID /* ID of the thread within a wavefront/warp */ % 2 == 0) continue; > > > > bar = ballot(true); // ballot 2 > > } > > > > versus: > > > > int i =

On Wed, Jan 30, 2019 at 4:20 PM Jan Sjodin <jan_sjodin at yahoo.com> wrote: > > > > > > > for (int i = 0; i < 2; i++) { > > > > foo = ballot(true); // ballot 1 > > > > > > > > if (threadID /* ID of the thread within a wavefront/warp */ % 2 > == 0) continue; > > > > > > > > bar = ballot(true); //

1) Everytime a participant speaks there is a echo for a short duration (maybe a word or two) but as the participant continues to speak without a any break the echo is 95% cancelled (i.e there is a feeble echo still present if observed very carefully). 2) The moment the participant stops / pauses speaking and start talking again, scenario 1 is repeated as if the echo state has been re-initialized

On Wed, Jan 30, 2019 at 7:20 AM Jan Sjodin via llvm-dev < llvm-dev at lists.llvm.org> wrote: > > > > > > > for (int i = 0; i < 2; i++) { > > > > foo = ballot(true); // ballot 1 > > > > > > > > if (threadID /* ID of the thread within a wavefront/warp */ % 2 > == 0) continue; > > > > > > > > bar =

Strong agree with Mehdi, I am also not really sure what is the proposal at this point so it's hard to comment further. > There are a number of questions that I have. Do we need better machine descriptions so that various resources can be considered? Do we need the capability to reason about the machine state for the cross-lane operations to enable more optimizations? Are intrinsics the

PuppetDB, a component of the Puppet Data Library, is a centralized storage daemon for auto-generated data. This initial release of PuppetDB targets the storage of catalogs and facts: * It’s a drop-in, 100% compatible replacement for storeconfigs * It’s a drop-in, 100% compatible replacement for inventory service * It hooks into your Puppet infrastructure using Puppet’s pre-existing