[inductor] split reduction even if all reads are broadcasted#167894
Open
shunting314 wants to merge 4 commits intogh/shunting314/263/basefrom
Open
[inductor] split reduction even if all reads are broadcasted#167894shunting314 wants to merge 4 commits intogh/shunting314/263/basefrom
shunting314 wants to merge 4 commits intogh/shunting314/263/basefrom
Conversation
[ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/167894
Note: Links to docs will display an error until the docs builds have been completed. ❌ 13 New Failures, 4 Cancelled Jobs, 4 Unrelated FailuresAs of commit 0b28245 with merge base 4b9418a ( NEW FAILURES - The following jobs have failed:
CANCELLED JOBS - The following jobs were cancelled. Please retry:
FLAKY - The following jobs failed but were likely due to flakiness present on trunk:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
With split reduction we can speedup the following (extreme) kernel by 48x
```
# 56ms -> 1.163ms
import torch
from triton.testing import do_bench
def f(x):
return x.sum(dim=(0, 1))
x = torch.randn(100000000, 1, 2, device="cuda").expand(-1, 2, -1)
opt_f = torch.compile(f)
ref = f(x)
act = opt_f(x)
torch.testing.assert_close(ref, act, atol=1e-3, rtol=1e-3)
ms = do_bench(lambda: opt_f(x))
print(f"ms={ms:.3f}")
```
Not confident if this change will break things. Let's wait for CI
cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx ipiszy chenyang78 kadeng muchulee8 amjames chauhang aakhundov coconutruben
[ghstack-poisoned]
jansel
approved these changes
Nov 18, 2025
Contributor
Author
|
@pytorchbot merge -i |
Collaborator
Merge startedYour change will be merged while ignoring the following 4 checks: inductor / inductor-cpu-test / test (cpu_inductor_torchbench, 2, 2, linux.2xlarge.amx), inductor / inductor-cpu-test / test (dynamic_cpu_inductor_torchbench, 2, 2, linux.2xlarge.amx), inductor / unit-test / inductor-pallas-test / test (inductor-pallas, 1, 1, linux.g5.12xlarge.nvidia.gpu), inductor / inductor-test / test (inductor_torchbench, 2, 2, linux.g5.4xlarge.nvidia.gpu) Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
Collaborator
Merge failedReason: 2 jobs have failed, first few of them are: trunk / macos-py3-arm64 / test (mps, 1, 1, macos-m2-15), trunk / macos-py3-arm64 / test (mps, 1, 1, macos-m1-14) Details for Dev Infra teamRaised by workflow job |
eellison
approved these changes
Nov 18, 2025
v0i0
approved these changes
Dec 2, 2025
With split reduction we can speedup the following (extreme) kernel by 48x
```
# 56ms -> 1.163ms
import torch
from triton.testing import do_bench
def f(x):
return x.sum(dim=(0, 1))
x = torch.randn(100000000, 1, 2, device="cuda").expand(-1, 2, -1)
opt_f = torch.compile(f)
ref = f(x)
act = opt_f(x)
torch.testing.assert_close(ref, act, atol=1e-3, rtol=1e-3)
ms = do_bench(lambda: opt_f(x))
print(f"ms={ms:.3f}")
```
Not confident if this change will break things. Let's wait for CI
cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx ipiszy chenyang78 kadeng muchulee8 amjames chauhang aakhundov coconutruben
[ghstack-poisoned]
With split reduction we can speedup the following (extreme) kernel by 48x
```
# 56ms -> 1.163ms
import torch
from triton.testing import do_bench
def f(x):
return x.sum(dim=(0, 1))
x = torch.randn(100000000, 1, 2, device="cuda").expand(-1, 2, -1)
opt_f = torch.compile(f)
ref = f(x)
act = opt_f(x)
torch.testing.assert_close(ref, act, atol=1e-3, rtol=1e-3)
ms = do_bench(lambda: opt_f(x))
print(f"ms={ms:.3f}")
```
Not confident if this change will break things. Let's wait for CI
cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx ipiszy chenyang78 kadeng muchulee8 amjames chauhang aakhundov coconutruben
[ghstack-poisoned]
Contributor
|
Looks like this PR hasn't been updated in a while so we're going to go ahead and mark this as |
Contributor
Author
|
Didn't land because of accuracy failure for 'sam'. May find some time to further debug. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Stack from ghstack (oldest at bottom):
With split reduction we can speedup the following (extreme) kernel by 48x
Not confident if this change will break things. Let's wait for CI
cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @ipiszy @kadeng @muchulee8 @amjames @chauhang @aakhundov @coconutruben @jataylo @chenyang78