nested dispatching of segment_csr on cpu/gpu #3881
Closed
+49
−40
Conversation
This file contains 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
|
This pull request was exported from Phabricator. Differential Revision: D71663741 |
✅ Deploy Preview for pytorch-fbgemm-docs ready!
To edit notification comments on pull requests, go to your Netlify site configuration. |
jeetkanjani7
force-pushed
the
export-D71663741
branch
from
d7b1f02 to
4a2d487
Compare
jeetkanjani7
pushed a commit
to jeetkanjani7/FBGEMM-1
that referenced
this pull request
Mar 26, 2025
Summary: X-link: facebookresearch/FBGEMM#972 Updating the segment_sum_csr function (both CPU and GPU) in predictor to accept arbitrary input types (int32_t or int64_t) for the offset. In cases where the offsets overflow 31 bits, casting to int32 will result in negative numbers, causing unintended behavior from the function. Reviewed By: zhaozhul Differential Revision: D71663741
|
This pull request was exported from Phabricator. Differential Revision: D71663741 |
jeetkanjani7
force-pushed
the
export-D71663741
branch
from
4a2d487 to
758abae
Compare
jeetkanjani7
pushed a commit
to jeetkanjani7/FBGEMM-1
that referenced
this pull request
Apr 2, 2025
Summary: X-link: facebookresearch/FBGEMM#972 Updating the segment_sum_csr function (both CPU and GPU) in predictor to accept arbitrary input types (int32_t or int64_t) for the offset. In cases where the offsets overflow 31 bits, casting to int32 will result in negative numbers, causing unintended behavior from the function. Reviewed By: zhaozhul Differential Revision: D71663741
jeetkanjani7
pushed a commit
to jeetkanjani7/FBGEMM-1
that referenced
this pull request
Apr 2, 2025
Summary: X-link: facebookresearch/FBGEMM#972 Updating the segment_sum_csr function (both CPU and GPU) in predictor to accept arbitrary input types (int32_t or int64_t) for the offset. In cases where the offsets overflow 31 bits, casting to int32 will result in negative numbers, causing unintended behavior from the function. Reviewed By: zhaozhul Differential Revision: D71663741
jeetkanjani7
force-pushed
the
export-D71663741
branch
from
758abae to
f4e2063
Compare
|
This pull request was exported from Phabricator. Differential Revision: D71663741 |
jeetkanjani7
pushed a commit
to jeetkanjani7/FBGEMM-1
that referenced
this pull request
Apr 2, 2025
Summary: Pull Request resolved: pytorch#3881 X-link: facebookresearch/FBGEMM#972 Updating the segment_sum_csr function (both CPU and GPU) in predictor to accept arbitrary input types (int32_t or int64_t) for the offset. In cases where the offsets overflow 31 bits, casting to int32 will result in negative numbers, causing unintended behavior from the function. Reviewed By: zhaozhul Differential Revision: D71663741
jeetkanjani7
force-pushed
the
export-D71663741
branch
from
f4e2063 to
48d232f
Compare
|
This pull request was exported from Phabricator. Differential Revision: D71663741 |
jeetkanjani7
pushed a commit
to jeetkanjani7/FBGEMM-1
that referenced
this pull request
Apr 2, 2025
Summary: Pull Request resolved: pytorch#3881 X-link: facebookresearch/FBGEMM#972 Updating the segment_sum_csr function (both CPU and GPU) in predictor to accept arbitrary input types (int32_t or int64_t) for the offset. In cases where the offsets overflow 31 bits, casting to int32 will result in negative numbers, causing unintended behavior from the function. Reviewed By: zhaozhul Differential Revision: D71663741
jeetkanjani7
force-pushed
the
export-D71663741
branch
from
48d232f to
64db1f8
Compare
jeetkanjani7
pushed a commit
to jeetkanjani7/FBGEMM-1
that referenced
this pull request
Apr 2, 2025
Summary: X-link: facebookresearch/FBGEMM#972 Updating the segment_sum_csr function (both CPU and GPU) in predictor to accept arbitrary input types (int32_t or int64_t) for the offset. In cases where the offsets overflow 31 bits, casting to int32 will result in negative numbers, causing unintended behavior from the function. Reviewed By: zhaozhul Differential Revision: D71663741
jeetkanjani7
force-pushed
the
export-D71663741
branch
from
64db1f8 to
eecb684
Compare
jeetkanjani7
pushed a commit
to jeetkanjani7/FBGEMM-1
that referenced
this pull request
Apr 2, 2025
Summary: X-link: facebookresearch/FBGEMM#972 Updating the segment_sum_csr function (both CPU and GPU) in predictor to accept arbitrary input types (int32_t or int64_t) for the offset. In cases where the offsets overflow 31 bits, casting to int32 will result in negative numbers, causing unintended behavior from the function. Reviewed By: zhaozhul Differential Revision: D71663741
jeetkanjani7
force-pushed
the
export-D71663741
branch
from
eecb684 to
23bf831
Compare
|
This pull request was exported from Phabricator. Differential Revision: D71663741 |
jeetkanjani7
force-pushed
the
export-D71663741
branch
from
23bf831 to
2bbb5fe
Compare
jeetkanjani7
pushed a commit
to jeetkanjani7/FBGEMM-1
that referenced
this pull request
Apr 2, 2025
Summary: Pull Request resolved: pytorch#3881 X-link: facebookresearch/FBGEMM#972 Updating the segment_sum_csr function (both CPU and GPU) in predictor to accept arbitrary input types (int32_t or int64_t) for the offset. In cases where the offsets overflow 31 bits, casting to int32 will result in negative numbers, causing unintended behavior from the function. Reviewed By: zhaozhul Differential Revision: D71663741
|
This pull request was exported from Phabricator. Differential Revision: D71663741 |
jeetkanjani7
pushed a commit
to jeetkanjani7/FBGEMM-1
that referenced
this pull request
Apr 2, 2025
Summary: Pull Request resolved: pytorch#3881 X-link: facebookresearch/FBGEMM#972 Updating the segment_sum_csr function (both CPU and GPU) in predictor to accept arbitrary input types (int32_t or int64_t) for the offset. In cases where the offsets overflow 31 bits, casting to int32 will result in negative numbers, causing unintended behavior from the function. Reviewed By: zhaozhul Differential Revision: D71663741
jeetkanjani7
force-pushed
the
export-D71663741
branch
from
2bbb5fe to
9347a8b
Compare
jeetkanjani7
pushed a commit
to jeetkanjani7/FBGEMM-1
that referenced
this pull request
Apr 2, 2025
Summary: X-link: facebookresearch/FBGEMM#972 Updating the segment_sum_csr function (both CPU and GPU) in predictor to accept arbitrary input types (int32_t or int64_t) for the offset. In cases where the offsets overflow 31 bits, casting to int32 will result in negative numbers, causing unintended behavior from the function. Reviewed By: zhaozhul Differential Revision: D71663741
jeetkanjani7
force-pushed
the
export-D71663741
branch
from
9347a8b to
4ca3d74
Compare
jeetkanjani7
pushed a commit
to jeetkanjani7/FBGEMM-1
that referenced
this pull request
Apr 2, 2025
Summary: X-link: facebookresearch/FBGEMM#972 Updating the segment_sum_csr function (both CPU and GPU) in predictor to accept arbitrary input types (int32_t or int64_t) for the offset. In cases where the offsets overflow 31 bits, casting to int32 will result in negative numbers, causing unintended behavior from the function. Reviewed By: zhaozhul Differential Revision: D71663741
jeetkanjani7
force-pushed
the
export-D71663741
branch
from
4ca3d74 to
6ae3cbf
Compare
|
This pull request was exported from Phabricator. Differential Revision: D71663741 |
jeetkanjani7
pushed a commit
to jeetkanjani7/FBGEMM-1
that referenced
this pull request
Apr 2, 2025
Summary: Pull Request resolved: pytorch#3881 X-link: facebookresearch/FBGEMM#972 Updating the segment_sum_csr function (both CPU and GPU) in predictor to accept arbitrary input types (int32_t or int64_t) for the offset. In cases where the offsets overflow 31 bits, casting to int32 will result in negative numbers, causing unintended behavior from the function. Reviewed By: zhaozhul Differential Revision: D71663741
jeetkanjani7
force-pushed
the
export-D71663741
branch
from
6ae3cbf to
1e005e2
Compare
|
This pull request was exported from Phabricator. Differential Revision: D71663741 |
jeetkanjani7
pushed a commit
to jeetkanjani7/FBGEMM-1
that referenced
this pull request
Apr 2, 2025
Summary: Pull Request resolved: pytorch#3881 X-link: facebookresearch/FBGEMM#972 Updating the segment_sum_csr function (both CPU and GPU) in predictor to accept arbitrary input types (int32_t or int64_t) for the offset. In cases where the offsets overflow 31 bits, casting to int32 will result in negative numbers, causing unintended behavior from the function. Reviewed By: zhaozhul Differential Revision: D71663741
jeetkanjani7
force-pushed
the
export-D71663741
branch
2 times, most recently
from
349831c to
40ffdbf
Compare
jeetkanjani7
pushed a commit
to jeetkanjani7/FBGEMM-1
that referenced
this pull request
Apr 4, 2025
Summary: X-link: facebookresearch/FBGEMM#972 Updating the segment_sum_csr function (both CPU and GPU) in predictor to accept arbitrary input types (int32_t or int64_t) for the offset. In cases where the offsets overflow 31 bits, casting to int32 will result in negative numbers, causing unintended behavior from the function. Reviewed By: zhaozhul, YazhiGao Differential Revision: D71663741
jeetkanjani7
pushed a commit
to jeetkanjani7/FBGEMM-1
that referenced
this pull request
Apr 4, 2025
Summary: X-link: facebookresearch/FBGEMM#972 Updating the segment_sum_csr function (both CPU and GPU) in predictor to accept arbitrary input types (int32_t or int64_t) for the offset. In cases where the offsets overflow 31 bits, casting to int32 will result in negative numbers, causing unintended behavior from the function. Reviewed By: zhaozhul, YazhiGao Differential Revision: D71663741
jeetkanjani7
force-pushed
the
export-D71663741
branch
from
40ffdbf to
7bb0a0c
Compare
|
This pull request was exported from Phabricator. Differential Revision: D71663741 |
jeetkanjani7
pushed a commit
to jeetkanjani7/FBGEMM-1
that referenced
this pull request
Apr 4, 2025
Summary: Pull Request resolved: pytorch#3881 X-link: facebookresearch/FBGEMM#972 Updating the segment_sum_csr function (both CPU and GPU) in predictor to accept arbitrary input types (int32_t or int64_t) for the offset. In cases where the offsets overflow 31 bits, casting to int32 will result in negative numbers, causing unintended behavior from the function. Reviewed By: zhaozhul, YazhiGao Differential Revision: D71663741
jeetkanjani7
force-pushed
the
export-D71663741
branch
from
7bb0a0c to
1beea6b
Compare
Summary: Pull Request resolved: pytorch#3881 X-link: facebookresearch/FBGEMM#972 Updating the segment_sum_csr function (both CPU and GPU) in predictor to accept arbitrary input types (int32_t or int64_t) for the offset. In cases where the offsets overflow 31 bits, casting to int32 will result in negative numbers, causing unintended behavior from the function. Reviewed By: zhaozhul, YazhiGao Differential Revision: D71663741
|
This pull request was exported from Phabricator. Differential Revision: D71663741 |
jeetkanjani7
force-pushed
the
export-D71663741
branch
from
1beea6b to
9045b44
Compare
|
This pull request has been merged in 25f2aaa. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Summary:
X-link: https://github.com/facebookresearch/FBGEMM/pull/972
Updating the segment_sum_csr function (both CPU and GPU) to accept arbitrary input types (int32_t or int64_t) for the offset. In cases where the offsets overflow 31 bits, casting to int32 will result in negative numbers, causing unintended behavior from the function.