Last active
January 5, 2022 21:44
-
-
Save pelletier/637c1cc16ec8fc08dd19009343976998 to your computer and use it in GitHub Desktop.
AVX2 right shift cross lanes fill with zeroes
This file contains 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
| // Go AVO code that generates a right shift cross lanes with zeroes fill when offset is known and less than 16 bytes. | |
| Comment("Shift right that works if remaining bytes >= 16") | |
| VPMOVQU(Mem{Base: d}, currentBlockY) | |
| zeroes := YMM() | |
| VPXOR(zeroes, zeroes, zeroes) | |
| src1 := YMM() | |
| VPERM2I128(Imm(3), currentBlockY, zeroes, src1) | |
| ymm3 := YMM() | |
| VPALIGNR(Imm(garbageSize), currentBlockY, src1, ymm3) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment