[mips] Fix floating point abs operation
Float32/64 abs operation should only clear the sign bit, but abs.s and abs.d instructions of mips64r2 would convert nan to canonical nan. Change-Id: Ibbd05cdb3a73acfe0e532030d1815d262c3ac433 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2439768 Commit-Queue: Zhao Jiazhong <zhaojiazhong-hf@loongson.cn> Reviewed-by: Jakob Gruber <jgruber@chromium.org> Cr-Commit-Position: refs/heads/master@{#70214}
Showing
Please
register
or
sign in
to comment