jrbyrnes wrote:
> We should spent more energy making the scheduler sensible by default, instead
> of creating all of this complexity.
I would also prefer a more sensible default scheduler, but the driving usecase
for this is global scheduling. The scheduler is doing inefficient things since
@@ -437,16 +437,18 @@ void test_sched_group_barrier()
}
// CHECK-LABEL: @test_sched_group_barrier_rule
-// CHECK: call void @llvm.amdgcn.sched.group.barrier.rule(i32 0, i32 1, i32 2,
i32 0)
-// CHECK: call void @llvm.amdgcn.sched.group.barrier.rule(i32 1, i32 2, i32 4,
i32
https://github.com/jrbyrnes updated
https://github.com/llvm/llvm-project/pull/85304
>From 04dc59ff7757dea18e2202d1cbff1d675885fdae Mon Sep 17 00:00:00 2001
From: Jeffrey Byrnes
Date: Tue, 12 Mar 2024 10:22:24 -0700
Subject: [PATCH 1/4] [AMDGPU] Extend __builtin_amdgcn_sched_group_barrier to
jrbyrnes wrote:
Updated the PR as discussed offline.
Support the variadic builtin arg via combining into mask for intrinsic. This
sort of implies a limit of 64 rules, but we can workaround by add a new
intrinsic with two masks (to support rules 65-128), and so on.
For now, rules in this PR
https://github.com/jrbyrnes ready_for_review
https://github.com/llvm/llvm-project/pull/85304
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/jrbyrnes updated
https://github.com/llvm/llvm-project/pull/85304
>From 04dc59ff7757dea18e2202d1cbff1d675885fdae Mon Sep 17 00:00:00 2001
From: Jeffrey Byrnes
Date: Tue, 12 Mar 2024 10:22:24 -0700
Subject: [PATCH 1/2] [AMDGPU] Extend __builtin_amdgcn_sched_group_barrier to
https://github.com/jrbyrnes closed
https://github.com/llvm/llvm-project/pull/78775
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
jrbyrnes wrote:
Supersedes https://github.com/llvm/llvm-project/pull/78775
https://github.com/llvm/llvm-project/pull/85304
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/jrbyrnes edited
https://github.com/llvm/llvm-project/pull/85304
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/jrbyrnes created
https://github.com/llvm/llvm-project/pull/85304
I am still working with the user to define the actual rules, so it is still a
WIP. However, the current version contains the main machinery of the feature.
This helps bridge the gap between
https://github.com/jrbyrnes created
https://github.com/llvm/llvm-project/pull/78775
As stated, this simply adds and codegens the builtin/intrinsic. A subsequent
patch will interface it with IGroupLP.
The idea is to give the users more expression by allowing them to create
schedgroups which
https://github.com/jrbyrnes commented:
Just have a few questions about implementation details -- at a higher level,
seems like we are trading one heuristic for another w.r.t flagging regions as
ExcessRP -- so I'm curious about the relative performance.
https://github.com/jrbyrnes commented:
Just have a few questions about implementation details -- at a higher level,
seems like we are trading one heuristic for another w.r.t flagging regions as
ExcessRP -- so I'm curious about the relative performance.
@@ -894,10 +894,22 @@ void GCNSchedStage::setupNewBlock() {
void GCNSchedStage::finalizeGCNRegion() {
DAG.Regions[RegionIdx] = std::pair(DAG.RegionBegin, DAG.RegionEnd);
- DAG.RescheduleRegions[RegionIdx] = false;
jrbyrnes wrote:
Why was this removed?
@@ -959,16 +970,6 @@ void GCNSchedStage::checkScheduling() {
<< DAG.MinOccupancy << ".\n");
}
- unsigned MaxVGPRs = ST.getMaxNumVGPRs(MF);
- unsigned MaxSGPRs = ST.getMaxNumSGPRs(MF);
- if (PressureAfter.getVGPRNum(false) > MaxVGPRs ||
-
@@ -1117,16 +1118,23 @@ bool
OccInitialScheduleStage::shouldRevertScheduling(unsigned WavesAfter) {
bool UnclusteredHighRPStage::shouldRevertScheduling(unsigned WavesAfter) {
// If RP is not reduced in the unclustered reschedule stage, revert to the
// old schedule.
- if
@@ -702,7 +702,7 @@ bool UnclusteredHighRPStage::initGCNSchedStage() {
if (!GCNSchedStage::initGCNSchedStage())
return false;
- if (DAG.RegionsWithHighRP.none() && DAG.RegionsWithExcessRP.none())
+ if (DAG.RegionsWithExcessRP.none())
jrbyrnes wrote:
https://github.com/jrbyrnes commented:
Just have a few questions about implementation details -- at a higher level,
seems like we are trading one heuristic for another w.r.t flagging regions as
ExcessRP -- so I'm curious about the relative performance.
Author: Jeffrey Byrnes
Date: 2023-06-29T08:18:28-07:00
New Revision: be8a65b598b3b80f73e862a01c7eaafe84d853a0
URL:
https://github.com/llvm/llvm-project/commit/be8a65b598b3b80f73e862a01c7eaafe84d853a0
DIFF:
19 matches
Mail list logo
