tp: scope gpu_counter custom-groups cache to per-sequence state#5742
Merged
Conversation
gpu_custom_groups_inserted_ in GpuEventParser was a global FlatHashMap<uint64_t, bool> keyed solely by counter_descriptor_iid, used to dedupe GpuCounterBlock insertions for the interned descriptor path. Interned ids are scoped to a packet sequence's incremental-state interval, not global, so two producers that legitimately share an iid on different packet sequences collide: whichever arrives first flips the cache, and the second producer's InsertCustomCounterGroups call is silently short-circuited. If the first producer's descriptor has no counter_groups, the second producer's blocks never reach gpu_counter_group_table and the UI renders those counters flat instead of grouped. Move the cache into a new GpuCounterSequenceState : CustomState on PacketSequenceStateGeneration, alongside the interned-data table the descriptor was looked up from. Cache key stays plain counter_descriptor_iid; correctness comes from CustomState being intrinsically per-sequence per-incremental-state-interval, matching where the iids are valid. Same pattern as AndroidKernelWakelockState.
🎨 Perfetto UI Builds
|
LalitMaganti
approved these changes
May 6, 2026
LalitMaganti
added a commit
that referenced
this pull request
May 6, 2026
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
2 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
gpu_custom_groups_inserted_ in GpuEventParser was a global FlatHashMap<uint64_t, bool> keyed solely by counter_descriptor_iid, used to dedupe GpuCounterBlock insertions for the interned descriptor path. Interned ids are scoped to a packet sequence's incremental-state interval, not global, so two producers that legitimately share an iid on different packet sequences collide: whichever arrives first flips the cache, and the second producer's InsertCustomCounterGroups call is silently short-circuited. If the first producer's descriptor has no counter_groups, the second producer's blocks never reach gpu_counter_group_table and the UI renders those counters flat instead of grouped.
Move the cache into a new GpuCounterSequenceState : CustomState on PacketSequenceStateGeneration, alongside the interned-data table the descriptor was looked up from. Cache key stays plain counter_descriptor_iid; correctness comes from CustomState being intrinsically per-sequence per-incremental-state-interval, matching where the iids are valid. Same pattern as AndroidKernelWakelockState.