[ROCm] Improve reduction sum performance#160466
Closed
jerrymannil wants to merge 1 commit intopytorch:mainfrom
jerrymannil:patch-1
Closed
[ROCm] Improve reduction sum performance#160466jerrymannil wants to merge 1 commit intopytorch:mainfrom jerrymannil:patch-1
jerrymannil wants to merge 1 commit intopytorch:mainfrom
jerrymannil:patch-1
Conversation
* Use input vectorization for reduction_on_fastest_striding_dimension when dim >= 0
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/160466
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit bff4a8a with merge base 4d419a7 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
jerrymannil
added a commit
to ROCm/pytorch
that referenced
this pull request
Aug 12, 2025
* Use input vectorization for reduction_on_fastest_striding_dimension when dim0 >= 0
**Reproducer:**
```
import time
import torch
shapes = [
(5079670, 128)
]
dims = [
(1)
]
for i, shape in enumerate(shapes):
x = torch.randn(shape, device='cuda', dtype=torch.float)
for _ in range(10):
w = torch.sum(x, dims[i])
torch.cuda.synchronize()
print(w.size())
start_time = time.time()
for _ in range(50):
_ = torch.sum(x, dims[i])
torch.cuda.synchronize()
end_time = time.time()
mean_time = (end_time - start_time)/50
print(f"Avg time for shape {shape}: {mean_time * 1e6:.2f} us")
```
**Before (MI300X):**
Avg time for shape (5079670, 128): 1629.99 us
**After (MI300X)**
Avg time for shape (5079670, 128): 1008.59 us
cherry-pick of pytorch#160466
jeffdaily
approved these changes
Aug 12, 2025
pruthvistony
pushed a commit
to ROCm/pytorch
that referenced
this pull request
Aug 13, 2025
* Use input vectorization for reduction_on_fastest_striding_dimension
when dim0 >= 0
**Reproducer:**
```
import time
import torch
shapes = [
(5079670, 128)
]
dims = [
(1)
]
for i, shape in enumerate(shapes):
x = torch.randn(shape, device='cuda', dtype=torch.float)
for _ in range(10):
w = torch.sum(x, dims[i])
torch.cuda.synchronize()
print(w.size())
start_time = time.time()
for _ in range(50):
_ = torch.sum(x, dims[i])
torch.cuda.synchronize()
end_time = time.time()
mean_time = (end_time - start_time)/50
print(f"Avg time for shape {shape}: {mean_time * 1e6:.2f} us")
```
**Before (MI300X):**
Avg time for shape (5079670, 128): 1629.99 us
**After (MI300X)**
Avg time for shape (5079670, 128): 1008.59 us
cherry-pick of pytorch#160466
Fixes SWDEV-546136
Collaborator
|
@pytorchbot merge |
Collaborator
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
jerrymannil
added a commit
to ROCm/pytorch
that referenced
this pull request
Aug 13, 2025
* Use input vectorization for reduction_on_fastest_striding_dimension when dim0 >= 128
**Reproducer:**
```
import time
import torch
shapes = [
(5079670, 128)
]
dims = [
(1)
]
for i, shape in enumerate(shapes):
x = torch.randn(shape, device='cuda', dtype=torch.float)
for _ in range(10):
w = torch.sum(x, dims[i])
torch.cuda.synchronize()
print(w.size())
start_time = time.time()
for _ in range(50):
_ = torch.sum(x, dims[i])
torch.cuda.synchronize()
end_time = time.time()
mean_time = (end_time - start_time)/50
print(f"Avg time for shape {shape}: {mean_time * 1e6:.2f} us")
```
**Before (MI300X):**
Avg time for shape (5079670, 128): 1629.99 us
**After (MI300X)**
Avg time for shape (5079670, 128): 1008.59 us
cherry-pick of pytorch#160466
Fixes SWDEV-546136
dhonnappa-amd
pushed a commit
to ROCm/pytorch
that referenced
this pull request
Aug 13, 2025
* Use input vectorization for reduction_on_fastest_striding_dimension
when dim0 >= 0
**Reproducer:**
```
import time
import torch
shapes = [
(5079670, 128)
]
dims = [
(1)
]
for i, shape in enumerate(shapes):
x = torch.randn(shape, device='cuda', dtype=torch.float)
for _ in range(10):
w = torch.sum(x, dims[i])
torch.cuda.synchronize()
print(w.size())
start_time = time.time()
for _ in range(50):
_ = torch.sum(x, dims[i])
torch.cuda.synchronize()
end_time = time.time()
mean_time = (end_time - start_time)/50
print(f"Avg time for shape {shape}: {mean_time * 1e6:.2f} us")
```
**Before (MI300X):**
Avg time for shape (5079670, 128): 1629.99 us
**After (MI300X)**
Avg time for shape (5079670, 128): 1008.59 us
cherry-pick of pytorch#160466
Fixes SWDEV-546136
chuanhaozhuge
pushed a commit
that referenced
this pull request
Aug 14, 2025
* Use input vectorization for reduction_on_fastest_striding_dimension when dim0 >= 128
**Reproducer:**
```
import time
import torch
shapes = [
(5079670, 128)
]
dims = [
(1)
]
for i, shape in enumerate(shapes):
x = torch.randn(shape, device='cuda', dtype=torch.float)
for _ in range(10):
w = torch.sum(x, dims[i])
torch.cuda.synchronize()
print(w.size())
start_time = time.time()
for _ in range(50):
_ = torch.sum(x, dims[i])
torch.cuda.synchronize()
end_time = time.time()
mean_time = (end_time - start_time)/50
print(f"Avg time for shape {shape}: {mean_time * 1e6:.2f} us")
```
**Before (MI300X):**
Avg time for shape (5079670, 128): 1629.99 us
**After (MI300X)**
Avg time for shape (5079670, 128): 1008.59 us
Pull Request resolved: #160466
Approved by: https://github.com/petrex, https://github.com/jeffdaily
pruthvistony
pushed a commit
to ROCm/pytorch
that referenced
this pull request
Aug 15, 2025
* Use input vectorization for reduction_on_fastest_striding_dimension when dim0 >= 128
**Reproducer:**
```
import time
import torch
shapes = [
(5079670, 128)
]
dims = [
(1)
]
for i, shape in enumerate(shapes):
x = torch.randn(shape, device='cuda', dtype=torch.float)
for _ in range(10):
w = torch.sum(x, dims[i])
torch.cuda.synchronize()
print(w.size())
start_time = time.time()
for _ in range(50):
_ = torch.sum(x, dims[i])
torch.cuda.synchronize()
end_time = time.time()
mean_time = (end_time - start_time)/50
print(f"Avg time for shape {shape}: {mean_time * 1e6:.2f} us")
```
**Before (MI300X):**
Avg time for shape (5079670, 128): 1629.99 us
**After (MI300X)**
Avg time for shape (5079670, 128): 1008.59 us
cherry-pick of pytorch#160466
Fixes SWDEV-546136
chuanhaozhuge
pushed a commit
that referenced
this pull request
Aug 18, 2025
* Use input vectorization for reduction_on_fastest_striding_dimension when dim0 >= 128
**Reproducer:**
```
import time
import torch
shapes = [
(5079670, 128)
]
dims = [
(1)
]
for i, shape in enumerate(shapes):
x = torch.randn(shape, device='cuda', dtype=torch.float)
for _ in range(10):
w = torch.sum(x, dims[i])
torch.cuda.synchronize()
print(w.size())
start_time = time.time()
for _ in range(50):
_ = torch.sum(x, dims[i])
torch.cuda.synchronize()
end_time = time.time()
mean_time = (end_time - start_time)/50
print(f"Avg time for shape {shape}: {mean_time * 1e6:.2f} us")
```
**Before (MI300X):**
Avg time for shape (5079670, 128): 1629.99 us
**After (MI300X)**
Avg time for shape (5079670, 128): 1008.59 us
Pull Request resolved: #160466
Approved by: https://github.com/petrex, https://github.com/jeffdaily
can-gaa-hou
pushed a commit
to can-gaa-hou/pytorch
that referenced
this pull request
Aug 22, 2025
* Use input vectorization for reduction_on_fastest_striding_dimension when dim0 >= 128
**Reproducer:**
```
import time
import torch
shapes = [
(5079670, 128)
]
dims = [
(1)
]
for i, shape in enumerate(shapes):
x = torch.randn(shape, device='cuda', dtype=torch.float)
for _ in range(10):
w = torch.sum(x, dims[i])
torch.cuda.synchronize()
print(w.size())
start_time = time.time()
for _ in range(50):
_ = torch.sum(x, dims[i])
torch.cuda.synchronize()
end_time = time.time()
mean_time = (end_time - start_time)/50
print(f"Avg time for shape {shape}: {mean_time * 1e6:.2f} us")
```
**Before (MI300X):**
Avg time for shape (5079670, 128): 1629.99 us
**After (MI300X)**
Avg time for shape (5079670, 128): 1008.59 us
Pull Request resolved: pytorch#160466
Approved by: https://github.com/petrex, https://github.com/jeffdaily
jerrymannil
added a commit
to ROCm/pytorch
that referenced
this pull request
Sep 5, 2025
* Use input vectorization for reduction_on_fastest_striding_dimension when dim0 >= 128
**Reproducer:**
```
import time
import torch
shapes = [
(5079670, 128)
]
dims = [
(1)
]
for i, shape in enumerate(shapes):
x = torch.randn(shape, device='cuda', dtype=torch.float)
for _ in range(10):
w = torch.sum(x, dims[i])
torch.cuda.synchronize()
print(w.size())
start_time = time.time()
for _ in range(50):
_ = torch.sum(x, dims[i])
torch.cuda.synchronize()
end_time = time.time()
mean_time = (end_time - start_time)/50
print(f"Avg time for shape {shape}: {mean_time * 1e6:.2f} us")
```
**Before (MI300X):**
Avg time for shape (5079670, 128): 1629.99 us
**After (MI300X)**
Avg time for shape (5079670, 128): 1008.59 us
cherry-pick of pytorch#160466
Fixes SWDEV-546136
markc-614
pushed a commit
to markc-614/pytorch
that referenced
this pull request
Sep 17, 2025
* Use input vectorization for reduction_on_fastest_striding_dimension when dim0 >= 128
**Reproducer:**
```
import time
import torch
shapes = [
(5079670, 128)
]
dims = [
(1)
]
for i, shape in enumerate(shapes):
x = torch.randn(shape, device='cuda', dtype=torch.float)
for _ in range(10):
w = torch.sum(x, dims[i])
torch.cuda.synchronize()
print(w.size())
start_time = time.time()
for _ in range(50):
_ = torch.sum(x, dims[i])
torch.cuda.synchronize()
end_time = time.time()
mean_time = (end_time - start_time)/50
print(f"Avg time for shape {shape}: {mean_time * 1e6:.2f} us")
```
**Before (MI300X):**
Avg time for shape (5079670, 128): 1629.99 us
**After (MI300X)**
Avg time for shape (5079670, 128): 1008.59 us
Pull Request resolved: pytorch#160466
Approved by: https://github.com/petrex, https://github.com/jeffdaily
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.
Reproducer:
Before (MI300X):
Avg time for shape (5079670, 128): 1629.99 us
After (MI300X)
Avg time for shape (5079670, 128): 1008.59 us
cc @jeffdaily @sunway513 @jithunnair-amd @pruthvistony @ROCmSupport @dllehr-amd @jataylo @hongxiayang @naromero77amd