Use compiler intrinsics for byte swapping #20139
Merged
Conversation
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
yuyichao
reviewed
Jan 20, 2017
| #ifdef __x86_64__ | ||
| # define LEGACY_REGS "=Q" | ||
| #if defined(__clang__) || (defined(__GNUC__) && (__GNUC__ > 4 || __GNUC_MINOR__ >= 8)) | ||
| #define bswap_16(x) __builtin_bswap16(x) |
There was a problem hiding this comment.
This seems to be available in https://gcc.gnu.org/onlinedocs/gcc-4.5.0/gcc/Other-Builtins.html which is older than our minimum gcc version requirement so remove the version check (or at least use it on 4.7)?
There was a problem hiding this comment.
Linux seemed to think 4.8 was the first one that had it. https://github.com/torvalds/linux/blob/master/include/linux/compiler-gcc.h#L261
There was a problem hiding this comment.
Ah, you are right. bswap16 is only in 4.8....
There was a problem hiding this comment.
Apparently gcc 4.7 is mostly-in-theory the minimum for llvm since it wasn't the default system compiler on many particularly commonly tested distros.
yuyichao
requested changes
Jan 20, 2017
| #define bswap_32(x) _byteswap_ulong(x) | ||
| #define bswap_64(x) _byteswap_uint64(x) | ||
| #elif defined(__INTEL_COMPILER) | ||
| #define bswap_16(X) _bswap16 |
There is really no advantage to messing with inline assembly here, but plenty of disadvantages. The compiler can't optimize this for whatever domain the data happens to be in (integer vs SSE+ registers), some of the code violates the C standard, and it generates warnings on newer gcc platforms (on i686 at least). Just get rid of it, use the compiler intrinsics, and fall back to generic C instructions for unknown compilers.
yuyichao
approved these changes
Jan 20, 2017
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
There is really no advantage to messing with inline assembly here,
but plenty of disadvantages. The compiler can't optimize this for
whatever domain the data happens to be in (integer vs SSE+ registers),
some of the code violates the C standard, and it generates warnings
on newer gcc platforms (on i686 at least). Just get rid of it, use
the compiler intrinsics, and fall back to generic C instructions for
unknown compilers.