Skip to content

Add ARM Neon and scalar implementations of SIMD functions #359

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
lh3 merged 4 commits into from
Aug 31, 2022
Merged

Add ARM Neon and scalar implementations of SIMD functions #359

Changes from 1 commit
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
ab01ab4
Make _mm_load_si128() explicit
jmarshall Jun 17, 2022
165e524
On ARM, rewrite SSE2 SIMD calls using Neon intrinsics
jmarshall Jun 17, 2022
ac612b6
On other platforms, emulate SSE2 SIMD calls using scalar code
jmarshall Jun 22, 2022
c77ace7
Use native SSE2 intrinsics on i386 as well as x86-64
jmarshall Jun 27, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Use native SSE2 intrinsics on i386 as well as x86-64 
Make the native SSE2 code conditional on __SSE2__, which is defined
by GCC/Clang/etc on x86-64 by default and on i386 with -msse2 etc.
  • Loading branch information
@jmarshall
jmarshall committed Jun 27, 2022
commit c77ace7059e0376d33c818f7d77b63ecf3739fa9
6 changes: 3 additions & 3 deletions ksw.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -26,7 +26,7 @@
#include <stdlib.h>
#include <stdint.h>
#include <assert.h>
#if defined __x86_64__
#if defined __SSE2__
#include <emmintrin.h>
#elif defined __ARM_NEON
#include "neon_sse.h"
Expand Down Expand Up @@ -126,7 +126,7 @@ kswr_t ksw_u8(kswq_t *q, int tlen, const uint8_t *target, int _o_del, int _e_del
__m128i zero, oe_del, e_del, oe_ins, e_ins, shift, *H0, *H1, *E, *Hmax;
kswr_t r;

#if defined __x86_64__
#if defined __SSE2__
#define __max_16(ret, xx) do { \
(xx) = _mm_max_epu8((xx), _mm_srli_si128((xx), 8)); \
(xx) = _mm_max_epu8((xx), _mm_srli_si128((xx), 4)); \
Expand Down Expand Up @@ -259,7 +259,7 @@ kswr_t ksw_i16(kswq_t *q, int tlen, const uint8_t *target, int _o_del, int _e_de
__m128i zero, oe_del, e_del, oe_ins, e_ins, *H0, *H1, *E, *Hmax;
kswr_t r;

#if defined __x86_64__
#if defined __SSE2__
#define __max_8(ret, xx) do { \
(xx) = _mm_max_epi16((xx), _mm_srli_si128((xx), 8)); \
(xx) = _mm_max_epi16((xx), _mm_srli_si128((xx), 4)); \
Expand Down