fix bit_count on arm - #25076
Conversation
Qodo reviews are paused for this user.Troubleshooting steps vary by plan Learn more → On a Teams plan? Using GitHub Enterprise Server, GitLab Self-Managed, or Bitbucket Data Center? |
XuPeng-SH
left a comment
There was a problem hiding this comment.
Re-checked the current head. The fix is narrowly scoped and correct for the actual cross-arch bug: negative FLOAT/DOUBLE values are now rounded first and then converted through int64 before the uint64 bit-count path, which avoids the ARM-specific zeroing behavior while preserving the existing MinInt64 clamp. The added regression cases hit the exact problematic boundary values, and I do not see a remaining correctness issue to block on.
|
Queued — the merge queue status continues in this comment ↓. |
Merge Queue Status
This pull request spent 1 hour 14 minutes 20 seconds in the queue, including 1 hour 13 minutes 54 seconds running CI. Required conditions to merge
|
What type of PR is this?
Which issue(s) this PR fixes:
issue #25046
What this PR does / why we need it:
原因:
修改方案: 先转in64, 再转uint64. 已经在arm64上验证了.