User Details
- User Since
- Dec 31 2018, 12:07 PM (230 w, 2 d)
Tue, May 30
LGTM, thanks!
Thu, May 25
Use member function.
Wed, May 24
LGTM, thanks!
Mon, May 22
Mon, May 8
Apr 27 2023
Apr 3 2023
Apr 2 2023
I like the general approach. It seems like things could get unwieldy with larger SchedGroups. You would need to have lots of checks vs Collection.size() which could be somewhat hard to work with.
Mar 31 2023
Mar 19 2023
Mar 17 2023
Mar 13 2023
Added AMDGPU group to reviewers.
Mar 8 2023
Update tests.
Actually, this is breaking tests with non-HSA. Is TargetID relevant for pal/graphics/ect @foad, or should the default there be XNACK- in the absence of any explicit subtarget features being added?
Mar 7 2023
Add readelf run-line to test.
Mar 6 2023
Feb 14 2023
LGTM, can you renumber regs in the test please.
Feb 4 2023
Didn't this land?
Jan 25 2023
Jan 22 2023
LGTM.
Jan 20 2023
Can you add some debug printouts so we know when this is being triggered in logs?
Jan 18 2023
LGTM
LGTM, thanks!
Jan 15 2023
I think this makes sense, but it also makes the concept of an occupancy target a misnomer.
Jan 11 2023
Is it possible to just check for cumode? I don't really have a problem with this since on Navi we can't really take advantage of barriers being too conservative without addrspace info on fences, but it really only has an impact when compiling for cumode. Also, I thought the workaround was pretty simple.
Dec 12 2022
Dec 9 2022
Dec 2 2022
Oct 17 2022
Interesting. Why exactly does this improve compile time so much? I thought reverting scheduling wasn't exactly expensive and the RP tracking was the problem.
Sep 7 2022
LGTM
LGTM
Sep 6 2022
Aug 19 2022
Address comments.
Don't loop over all instructions again in pre-RA scheduler.
Address comment.
Update pipeline, remove edges from iglp_opt.
Aug 17 2022
LGTM
Aug 9 2022
Aug 4 2022
Aug 2 2022
Address comments.
Aug 1 2022
Jul 31 2022
Jul 30 2022
Jul 29 2022
Thanks! I like the idea behind the greedy solver. Not sure about SchedGroupSU. Maybe just a map between SUs and lists of schedgroups? I think trying to track sched_group_barriers by their order and assigning that an index is a bit confusing.
Jul 28 2022
Resubmit https://reviews.llvm.org/D120544.
Jul 27 2022
Why not make it a scheduling boundary?
Jul 26 2022
Jul 22 2022
Fix not running unclustred pass on non-excess RP regions.
Jul 21 2022
Minor fixes.
Jul 19 2022
Fix some bugs. Add better pipeline fitting. Address comments.
Jul 14 2022
I'm not objecting to the change, just pointing out that you may miss out on some optimization since this is lowered to inline ASM, and that you may want to lower it to intrinsics in the future.
Jul 13 2022
Can you add a test with the bug?
Jul 11 2022
Is this really necessary if we have https://reviews.llvm.org/D120544. A barrier has no requirement to wait for LDS and VMEM it only does so currently because of a bug. Using inline asm like this seems like it will eventually cause problems although I'm not too familiar with MLIR.
Jul 7 2022
Can we not rename initCandidate since it is mimicking GenericScheduler::initCandidate?