Conversation
Anerudhan
commented
May 13, 2021
Anerudhan
commented
Collaborator
- [New feature] Support reduction operation in the frontend.
- [New feature] Add engine runtime compilation filter in the frontend as a behavior filter.
- [New feature] Adding fallback list for convBiasAct
- [New feature Beta] Adding Errata filter with an sample.
- [Samples] Add ConvBnstats and ConvColReduction tests
- [Bug Fix] Clamp upper_clip for float compute type to float max for pointwise descriptor when computeType is float.
- [Bug Fix] Compilation fix for newer gcc toolchain (gcc 9+).
- [Bug Fix] Add operation tag to the Plan generated by cudnnFind and cudnnGet
- [Maintenance] Added default fallback lists to newer versions of cudnn.
Anerudhan
commented
May 14, 2021
| version : 1 - Mandatory. Tells the format version of the json. | ||
| rules : [] - Mandatory. Array of rule object which identifies the engine config | ||
| rule_id : "" - Optional. Used to uniquely identify a rule. Has no purpose other than being easy to debug. | ||
| operation : "" - Mandatory. Stringified version of the operation graph. |
Collaborator
Author
There was a problem hiding this comment.
Add an indendation for the rule_id and other fields.
zkyue
added a commit
to zkyue/cudnn-frontend
that referenced
this pull request
Jul 18, 2026
Completes the remove/relocate edits from the code review that the inline suggestion blocks could not express on their own: - NVIDIA#2: remove the now-duplicate same_hdim t2r_dKV4 wait before dKV1 - NVIDIA#4: remove the now-duplicate same_hdim t2r_dKV01 wait before dKV3 - NVIDIA#6: switch same_hdim dKV2/dKV3 drain to split T2R + reduce_dKV_from_reg Also repairs line-drift from batch-applying the earlier suggestions (their anchors shifted after the first suggestion inserted lines): - NVIDIA#3: remove a leftover dangling `if cutlass.const_expr(not self.same_hdim_kv):` with no body (IndentationError) left because the original lines were not replaced - NVIDIA#5: remove the leftover old dKV0/dKV1 if/else block that sat before the new unified block, which would otherwise double the reduce/atomic_add and the t2r_dKV01 barrier arrival Final state is the intended split-T2R + signal drain refactor: every reduce site loads dKV into registers, fences, signals the MMA warp that the TMEM columns are free, then does the global atomic_add; same_hdim and not-same_hdim now follow the same protocol. dq is bitwise-identical to ef50ddc; head_dim 576 and 512 both verified race-free (compute-sanitizer racecheck 0).
zkyue
added a commit
to zkyue/cudnn-frontend
that referenced
this pull request
Jul 20, 2026
Completes the remove/relocate edits from the code review that the inline suggestion blocks could not express on their own: - NVIDIA#2: remove the now-duplicate same_hdim t2r_dKV4 wait before dKV1 - NVIDIA#4: remove the now-duplicate same_hdim t2r_dKV01 wait before dKV3 - NVIDIA#6: switch same_hdim dKV2/dKV3 drain to split T2R + reduce_dKV_from_reg Also repairs line-drift from batch-applying the earlier suggestions (their anchors shifted after the first suggestion inserted lines): - NVIDIA#3: remove a leftover dangling `if cutlass.const_expr(not self.same_hdim_kv):` with no body (IndentationError) left because the original lines were not replaced - NVIDIA#5: remove the leftover old dKV0/dKV1 if/else block that sat before the new unified block, which would otherwise double the reduce/atomic_add and the t2r_dKV01 barrier arrival Final state is the intended split-T2R + signal drain refactor: every reduce site loads dKV into registers, fences, signals the MMA warp that the TMEM columns are free, then does the global atomic_add; same_hdim and not-same_hdim now follow the same protocol. dq is bitwise-identical to ef50ddc; head_dim 576 and 512 both verified race-free (compute-sanitizer racecheck 0). Signed-off-by: zky <51477259+zkyue@users.noreply.github.com>
Anerudhan
pushed a commit
that referenced
this pull request
Jul 20, 2026
…m 576) (#396) * Fix latent TMEM WAR race in DSA backward SM100 dKV drain (head_dim 576 path) In the not-same_hdim_kv path (head_dim 576 / head_dim_v 512), the reduce warps read dKV2/dKV3 from TMEM inside store_dKV, after their t2r_dKV4_done arrive. The MMA warp's next-iteration dKV0/dKV1 gemms overwrite the same TMEM columns (tmem_dKV2_offset == tmem_dKV0_offset, tmem_dKV3_offset == tmem_dKV1_offset) with no happens-before edge to those reads: mma_reduce_dKV_pipeline has 2 stages, so the part1 producer_acquire of generation 3i+3 only orders against the consumer_release of generation 3i+1 (the dKV4 generation), and no named barrier separates part2's TMEM reads (generation 3i+2) from the issue of the next part1's overwrites. Correctness currently depends on the tensor core still draining queued work when the reduce warps issue their loads. The window reproduces deterministically under adversarial delay: a 1 us spin inserted in the reduce warps between the two part2 store_dKV calls (no other change) corrupts dkv columns [384:512) on every run at S=2048 H=64 D=576 topk=512 bf16 (rel L2 vs fp32 autograd 1.0e0 vs 5.6e-3 baseline); a 10 us spin before both calls corrupts [256:384) as well. Delays at already-synchronized points (after part1's register-staged T2Rs, or before the MMA warp's dKV0 issue) never corrupt. Fix, mirroring part1's existing register-staging pattern: T2R dKV2/dKV3 into registers, fence, arrive on a new t2r_dKV23_done named barrier (id 9), and only then run the global-memory atomic reduction; the MMA warp waits on that barrier before issuing the next iteration's dKV0 (skipped on the first iteration, balanced by a final arrive after the loop, like t2r_dKV4_done). Register peak is unchanged (part1 already holds two fragments). The same_hdim_kv (head_dim 512) path is untouched and compiles identically. Measured on B200 (S=8192 H=64 topk=1024 bf16, ABAB paired): head_dim 576: 5.398 -> 5.840 ms median (+8.0%), the cost of no longer overlapping the next iteration's dKV issue with the dKV2/dKV3 readback; head_dim 512: +0.03% (noise). dq stays bitwise-identical to develop on both head dims; dkv/d_sink match the fp32 autograd reference at the baseline rel L2 on both, and the adversarial-delay matrix that trips develop is clean on the fixed kernel. Signed-off-by: zky <51477259+zkyue@users.noreply.github.com> * Apply suggestions from code review Co-authored-by: Jie Fang <jief@nvidia.com> Signed-off-by: zky <51477259+zkyue@users.noreply.github.com> * Complete review remove/relocate edits and repair suggestion line-drift Completes the remove/relocate edits from the code review that the inline suggestion blocks could not express on their own: - #2: remove the now-duplicate same_hdim t2r_dKV4 wait before dKV1 - #4: remove the now-duplicate same_hdim t2r_dKV01 wait before dKV3 - #6: switch same_hdim dKV2/dKV3 drain to split T2R + reduce_dKV_from_reg Also repairs line-drift from batch-applying the earlier suggestions (their anchors shifted after the first suggestion inserted lines): - #3: remove a leftover dangling `if cutlass.const_expr(not self.same_hdim_kv):` with no body (IndentationError) left because the original lines were not replaced - #5: remove the leftover old dKV0/dKV1 if/else block that sat before the new unified block, which would otherwise double the reduce/atomic_add and the t2r_dKV01 barrier arrival Final state is the intended split-T2R + signal drain refactor: every reduce site loads dKV into registers, fences, signals the MMA warp that the TMEM columns are free, then does the global atomic_add; same_hdim and not-same_hdim now follow the same protocol. dq is bitwise-identical to ef50ddc; head_dim 576 and 512 both verified race-free (compute-sanitizer racecheck 0). Signed-off-by: zky <51477259+zkyue@users.noreply.github.com> --------- Signed-off-by: zky <51477259+zkyue@users.noreply.github.com> Co-authored-by: Jie Fang <jief@nvidia.com>
1 task
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
1 participant
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.