Chromium Code Reviews
chromiumcodereview-hr@appspot.gserviceaccount.com (chromiumcodereview-hr) | Please choose your nickname with Settings | Help | Chromium Project | Gerrit Changes | Sign out
(17)

Unified Diff: src/arm/assembler-arm.h

Issue 2711863002: Implement remaining Boolean SIMD operations on ARM. (Closed)
Patch Set: Fix macro assembler test. Created 3 years, 10 months ago
Use n/p to move between diff chunks; N/P to move between comments. Draft comments are only viewable by you.
Jump to:
View side-by-side diff with in-line comments
Download patch
« no previous file with comments | « no previous file | src/arm/assembler-arm.cc » ('j') | no next file with comments »
Expand Comments ('e') | Collapse Comments ('c') | Show Comments Hide Comments ('s')
Index: src/arm/assembler-arm.h
diff --git a/src/arm/assembler-arm.h b/src/arm/assembler-arm.h
index e9424f0244cd15364cc248d93cb372b2727ccc1b..420d195f3d4e7c7e9ada4a86841453b44f38829e 100644
--- a/src/arm/assembler-arm.h
+++ b/src/arm/assembler-arm.h
@@ -418,7 +418,8 @@ const QwNeonRegister q15 = { 15 };
// compilation unit that includes this header doesn't use the variables.
#define kFirstCalleeSavedDoubleReg d8
#define kLastCalleeSavedDoubleReg d15
-// kDoubleRegZero and kScratchDoubleReg must pair to form kScratchQuadReg.
+// kDoubleRegZero and kScratchDoubleReg must pair to form kScratchQuadReg. SIMD
+// code depends on kDoubleRegZero before kScratchDoubleReg.
#define kDoubleRegZero d14
#define kScratchDoubleReg d15
// After using kScratchQuadReg, kDoubleRegZero must be reset to 0.
@@ -1359,6 +1360,10 @@ class Assembler : public AssemblerBase {
void vmax(QwNeonRegister dst, QwNeonRegister src1, QwNeonRegister src2);
void vmax(NeonDataType dt, QwNeonRegister dst,
QwNeonRegister src1, QwNeonRegister src2);
+ void vpmin(NeonDataType dt, DwVfpRegister dst, DwVfpRegister src1,
+ DwVfpRegister src2);
+ void vpmax(NeonDataType dt, DwVfpRegister dst, DwVfpRegister src1,
+ DwVfpRegister src2);
void vshl(NeonDataType dt, QwNeonRegister dst, QwNeonRegister src, int shift);
void vshr(NeonDataType dt, QwNeonRegister dst, QwNeonRegister src, int shift);
// vrecpe and vrsqrte only support floating point lanes.
« no previous file with comments | « no previous file | src/arm/assembler-arm.cc » ('j') | no next file with comments »

Powered by Google App Engine
This is Rietveld 408576698