Add SSSE3 acceleration for S32_{opaque,alpha}_D32_filter_DX; should yield
25% speedup on 64b architectures, something smaller but still valuable on
32b.
Adds new files, breaking client gyps.
http://codereview.appspot.com/5515044/
git-svn-id: http://skia.googlecode.com/svn/trunk@3193 2bbb7eff-a529-9590-31e7-b0007b416f81
diff --git a/gyp/opts.gyp b/gyp/opts.gyp
index 1f67e00..66348f6 100644
--- a/gyp/opts.gyp
+++ b/gyp/opts.gyp
@@ -31,18 +31,23 @@
'../src/opts',
],
'conditions': [
- [ 'skia_os in ["linux", "freebsd", "openbsd", "solaris"]', {
- 'cflags': [
- '-msse2',
- ],
- }],
[ 'skia_target_arch != "arm"', {
+ 'conditions': [
+ [ 'skia_os in ["linux", "freebsd", "openbsd", "solaris"]', {
+ 'cflags': [
+ '-msse2',
+ ],
+ }],
+ ],
'sources': [
'../src/opts/opts_check_SSE2.cpp',
'../src/opts/SkBitmapProcState_opts_SSE2.cpp',
'../src/opts/SkBlitRow_opts_SSE2.cpp',
'../src/opts/SkUtils_opts_SSE2.cpp',
],
+ 'dependencies': [
+ 'opts_ssse3',
+ ],
}],
[ 'skia_target_arch == "arm" and armv7 == 1', {
# The assembly uses the frame pointer register (r7 in Thumb/r11 in
@@ -71,6 +76,31 @@
}],
],
},
+ # For the same lame reasons as what is done for skia_opts, we have to
+ # create another target specifically for SSSE3 code as we would not want
+ # to compile the SSE2 code with -mssse3 which would potentially allow
+ # gcc to generate SSSE3 code.
+ {
+ 'target_name': 'opts_ssse3',
+ 'type': 'static_library',
+ 'include_dirs': [
+ '../include/config',
+ '../include/core',
+ '../src/core',
+ ],
+ 'conditions': [
+ [ 'skia_os in ["linux", "freebsd", "openbsd", "solaris"]', {
+ 'cflags': [
+ '-mssse3',
+ ],
+ }],
+ [ 'skia_target_arch != "arm"', {
+ 'sources': [
+ '../src/opts/SkBitmapProcState_opts_SSSE3.cpp',
+ ],
+ }],
+ ],
+ },
],
}