Use clmul for CRC32. #28511
Open
Use clmul for CRC32. #28511
+23
−12
Conversation
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
|
|
The crc32 instructions are non-standard and not supported by upstream clang. According to perftest, this only slows things down by a few cycles before: I00003 crc32_perftest.c:33] CRC32 computed in 11343 cycles. I00004 crc32_perftest.c:33] CRC32 computed in 11317 cycles. I00005 crc32_perftest.c:33] CRC32 computed in 11314 cycles. I00006 crc32_perftest.c:33] CRC32 computed in 11314 cycles. I00007 crc32_perftest.c:33] CRC32 computed in 11314 cycles. I00008 crc32_perftest.c:33] CRC32 computed in 11314 cycles. I00009 crc32_perftest.c:33] CRC32 computed in 11314 cycles. I00010 crc32_perftest.c:33] CRC32 computed in 11314 cycles. I00011 crc32_perftest.c:33] CRC32 computed in 11314 cycles. I00012 crc32_perftest.c:33] CRC32 computed in 11314 cycles. after: I00003 crc32_perftest.c:33] CRC32 computed in 11358 cycles. I00004 crc32_perftest.c:33] CRC32 computed in 11319 cycles. I00005 crc32_perftest.c:33] CRC32 computed in 11318 cycles. I00006 crc32_perftest.c:33] CRC32 computed in 11318 cycles. I00007 crc32_perftest.c:33] CRC32 computed in 11318 cycles. I00008 crc32_perftest.c:33] CRC32 computed in 11318 cycles. I00009 crc32_perftest.c:33] CRC32 computed in 11318 cycles. I00010 crc32_perftest.c:33] CRC32 computed in 11318 cycles. I00011 crc32_perftest.c:33] CRC32 computed in 11318 cycles. Signed-off-by: Kor Nielsen <[email protected]>
luismarques
approved these changes
Oct 18, 2025
| asm(".option push;" | ||
| ".option arch, +zbr0p93;" | ||
| "crc32.b %0, %1;" | ||
| ".option arch, +zbc;" |
There was a problem hiding this comment.
The -march arch string we use already includes zbc, so the .option arch, +zbc shouldn't be needed here.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 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.
The crc32 instructions are non-standard and not supported by upstream clang.
According to perftest, this only slows things down by a few cycles