sha256 x86_64 optimization v2/vectorized checksum (01.09.2015) rebased to 09.03.2016 [take1] #4400
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
Signed-off-by: Chunwei Chen <tuxoko@gmail.com>
Arch specific include is located at include/asm. include/asm is generated during configure time. It first copy every *.h from include/asm-generic, and then copy every *.h from include/$TARGET_ASM_DIR over them. Thus, when introducing new arch specific include file foo.h, you only need to add a generic version in include/asm-generic. Later on, each architecture can add its own version of foo.h in its respective directory. Signed-off-by: Chunwei Chen <tuxoko@gmail.com>
The asm optimization for sha256 we'll be using transforms multiple blocks in a single function call. Thus, we refactor the existing generic code in preparation for the upcoming change. Signed-off-by: Chunwei Chen <tuxoko@gmail.com>
The function pointer sha256_transform decides which arch specific optimization to use. It defaults to sha256_transform_generic(). To override it, you need to: 1. Override funtion prototype for arch_sha256_init in include/asm-xxx/sha256.h 2. Implement arch_sha256_init() to assign sha256_transform properly and put it in module/zfs/asm-xxxx/ Signed-off-by: Chunwei Chen <tuxoko@gmail.com>
The sha256-{ssse3,avx,avx2}-asm.S files are copied from linux-3.14.
Currently, we only support optimization in kernel space. User programs will
use generic code.
Signed-off-by: Chunwei Chen <tuxoko@gmail.com>
Run some simple tests on predefined strings when ZFS_DEBUG is on. Signed-off-by: Chunwei Chen <tuxoko@gmail.com>
kernelOfTruth
changed the title
|
@kernelOfTruth |
|
@tuxoko thank you very much ! |
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.
#2351 sha256 x86_64 optimization v2 rebased to current master
Not sure of the compilation and parallel build changes were done properly
(47a4a6fd),
but for now this has to suffice - pretty late & time to get some rest,
not compile-tested yet.
Buildbots Go ! ;)
thanks to @tomgarcia