You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@tvm.apache.org by "Hzfengsy (via GitHub)" <gi...@apache.org> on 2023/07/28 08:09:41 UTC
[GitHub] [tvm] Hzfengsy opened a new pull request, #15429: [Unity][DLight] Update GEMV rules
Hzfengsy opened a new pull request, #15429:
URL: https://github.com/apache/tvm/pull/15429
This PR updates the GEMV rules:
1. improve for large workloads, speedup Vicuna 13B from 43.0 tok/s to 60.8 tok/s
2. Fix the issue of unexpected unroll for allreduce parts.
cc @spectrometerHBH @MasterJH5574
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] junrushao merged pull request #15429: [Unity][DLight] Update GEMV rules
Posted by "junrushao (via GitHub)" <gi...@apache.org>.
junrushao merged PR #15429:
URL: https://github.com/apache/tvm/pull/15429
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] masahi commented on pull request #15429: [Unity][DLight] Update GEMV rules
Posted by "masahi (via GitHub)" <gi...@apache.org>.
masahi commented on PR #15429:
URL: https://github.com/apache/tvm/pull/15429#issuecomment-1656572692
On RTX 4080 this PR improved over my last week's result as follows:
7B
```
| mode | seqlen | genlen | p50: tok/s |
| FT | 128 | 128 | 121.7 |
| dlight (prev) | 128 | 128 | 141.9 |
| dlight (new) | 128 | 128 | 146.7 |
```
13B
```
| mode | seqlen | genlen | p50: tok/s |
| FT | 128 | 128 | 74.6 |
| dlight (prev) | 128 | 128 | 67.8 |
| dlight (new) | 128 | 128 | 80.6 |
```
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] tvm-bot commented on pull request #15429: [Unity][DLight] Update GEMV rules
Posted by "tvm-bot (via GitHub)" <gi...@apache.org>.
tvm-bot commented on PR #15429:
URL: https://github.com/apache/tvm/pull/15429#issuecomment-1655241892
<!---bot-comment-->
Thanks for contributing to TVM! Please refer to the contributing guidelines https://tvm.apache.org/docs/contribute/ for useful information and tips. Please request code reviews from [Reviewers](https://github.com/apache/incubator-tvm/blob/master/CONTRIBUTORS.md#reviewers) by @-ing them in a comment.
<!--bot-comment-ccs-start-->
* cc @quic-sanirudh <sub>See [#10317](https://github.com/apache/tvm/issues/10317) for details</sub><!--bot-comment-ccs-end-->
<sub>Generated by [tvm-bot](https://github.com/apache/tvm/blob/main/ci/README.md#github-actions)</sub>
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] junrushao commented on pull request #15429: [Unity][DLight] Update GEMV rules
Posted by "junrushao (via GitHub)" <gi...@apache.org>.
junrushao commented on PR #15429:
URL: https://github.com/apache/tvm/pull/15429#issuecomment-1656507395
CC @masahi
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org