summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorJason Ekstrand <jason.ekstrand@intel.com>2014-07-21 16:46:39 -0700
committerJason Ekstrand <jason.ekstrand@intel.com>2014-07-24 12:44:56 -0700
commit989d2e370993c87d1bbda4950657bfcc5b0a58dd (patch)
tree1b9a048a63e291fb68d7c1a2bf00e08d69d7e672
parent2a33510f1649f2ef5c5b2d693aa89ef0efc5dcfb (diff)
Add an accelerated version of F_TO_I for x86_64
According to a quick micro-benchmark, this new version is 20% faster on my Haswell laptop. v2: Removed the XXX note about x86_64 from the comment v3: Use an intrinsic instead of an __asm__ block. This should give us MSVC support for free. v4: Enable it for all x86_64 builds, not just with USE_X86_64_ASM Signed-off-by: Jason Ekstrand <jason.ekstrand@intel.com> Reviewed-by: Matt Turner <mattst88@gmail.com>
-rw-r--r--src/mesa/main/imports.h6
1 files changed, 5 insertions, 1 deletions
diff --git a/src/mesa/main/imports.h b/src/mesa/main/imports.h
index af780b2498f..09e55ebf0ff 100644
--- a/src/mesa/main/imports.h
+++ b/src/mesa/main/imports.h
@@ -274,10 +274,12 @@ static inline int IROUND_POS(float f)
return (int) (f + 0.5F);
}
+#ifdef __x86_64__
+# include <xmmintrin.h>
+#endif
/**
* Convert float to int using a fast method. The rounding mode may vary.
- * XXX We could use an x86-64/SSE2 version here.
*/
static inline int F_TO_I(float f)
{
@@ -292,6 +294,8 @@ static inline int F_TO_I(float f)
fistp r
}
return r;
+#elif defined(__x86_64__)
+ return _mm_cvt_ss2si(_mm_load_ss(&f));
#else
return IROUND(f);
#endif