Cap descriptor width to 64 bytes in FindChild. - #3443
Merged
Merged
Conversation
copybara-service
Bot
force-pushed
the
test_991798243
branch
from
October 2, 2026 11:31
ede2762 to
124f09a
Compare
previously we were hard capping at 16, that becomes a problem lets say at SVE 2048 bytes, which when capped at 16 lanes and lets say 64 bite keys would give us a capped vector of 16 lanes each containing 64 bit keys thus vector size reaching 128 bytes. and we do Load(d, internal->keys + i) below, for this to be safe the ptr has to be a multiple of vector byte size of d, but internal nodes are only aligned to 64 bytes, thus 64 is not a multiple of 128 , so this Load was unsafe on wider vectors >=1024 PiperOrigin-RevId: 992242309
copybara-service
Bot
force-pushed
the
test_991798243
branch
from
October 2, 2026 11:32
124f09a to
741eaa5
Compare
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Cap descriptor width to 64 bytes in FindChild.
previously we were hard capping at 16, that becomes a problem lets say at SVE 2048 bytes, which when capped at 16 lanes and lets say 64 bite keys would give us a capped vector of 16 lanes each containing 64 bit keys thus vector size reaching 128 bytes.
and we do Load(d, internal->keys + i) below, for this to be safe the ptr has to be a multiple of vector byte size of d, but internal nodes are only aligned to 64 bytes, thus 64 is not a multiple of 128 , so this Load was unsafe on wider vectors >=1024