[0.9.1][bugfix] fix deepseek memory bug#1551
Merged
ganyi1996ppo merged 1 commit intovllm-project:v0.9.1-devfrom Jul 2, 2025
Merged
[0.9.1][bugfix] fix deepseek memory bug#1551ganyi1996ppo merged 1 commit intovllm-project:v0.9.1-devfrom
ganyi1996ppo merged 1 commit intovllm-project:v0.9.1-devfrom
Conversation
Signed-off-by: zzzzwwjj <1183291235@qq.com>
ApsarasX
reviewed
Jul 1, 2025
Collaborator
There was a problem hiding this comment.
Can a compressed mask be used? If the sequence is too long, it might cause memory waste here.
Refer #1100
Collaborator
Author
There was a problem hiding this comment.
ringmla's mask size is equal to chunksize, it won't cause memory waste.
wangxiyuan
changed the title
ganyi1996ppo
approved these changes
Jul 2, 2025
Collaborator
|
Looks good, please cherry-pick this change back to main |
jianzs
reviewed
Jul 29, 2025
Collaborator
There was a problem hiding this comment.
Should we consider which scheduler is be using?
self.use_ring_mla = ascend_config.chunked_prefill_for_mla or \
not ascend_config.ascend_scheduler_config.enabled
jianzs
pushed a commit
to jianzs/vllm-ascend
that referenced
this pull request
Jul 29, 2025
fix OOM error when `chunked_prefill_for_mla` is enable and long input scene. Signed-off-by: zzzzwwjj <1183291235@qq.com>
jianzs
pushed a commit
to jianzs/vllm-ascend
that referenced
this pull request
Jul 29, 2025
fix OOM error when `chunked_prefill_for_mla` is enable and long input scene. Signed-off-by: zzzzwwjj <1183291235@qq.com> Signed-off-by: Jade Zheng <zheng.shoujian@outlook.com>
jianzs
pushed a commit
to jianzs/vllm-ascend
that referenced
this pull request
Aug 6, 2025
fix OOM error when `chunked_prefill_for_mla` is enable and long input scene. Signed-off-by: zzzzwwjj <1183291235@qq.com> Signed-off-by: Jade Zheng <zheng.shoujian@outlook.com>
jianzs
pushed a commit
to jianzs/vllm-ascend
that referenced
this pull request
Aug 7, 2025
fix OOM error when `chunked_prefill_for_mla` is enable and long input scene. Signed-off-by: zzzzwwjj <1183291235@qq.com> Signed-off-by: Jade Zheng <zheng.shoujian@outlook.com>
jianzs
pushed a commit
to jianzs/vllm-ascend
that referenced
this pull request
Aug 8, 2025
fix OOM error when `chunked_prefill_for_mla` is enable and long input scene. Signed-off-by: zzzzwwjj <1183291235@qq.com> Signed-off-by: Jade Zheng <zheng.shoujian@outlook.com>
jianzs
pushed a commit
to jianzs/vllm-ascend
that referenced
this pull request
Aug 11, 2025
fix OOM error when `chunked_prefill_for_mla` is enable and long input scene. Signed-off-by: zzzzwwjj <1183291235@qq.com> Signed-off-by: Jade Zheng <zheng.shoujian@outlook.com>
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.
What this PR does / why we need it?
fix OOM error when
chunked_prefill_for_mlais enable and long input scene.Does this PR introduce any user-facing change?
How was this patch tested?