commit b45a435e25d69592f27084a615b351760f040875
author Shalini Salomi Bodapati <shalini.salomi.bodapati@intel.com> Wed Jul 10 16:09:41 2019 +0530
committer Vladimir Marko <vmarko@google.com> Wed Jul 17 11:12:22 2019 +0000
tree 6920830856dbe3cceb648b365b0d057ea3ea93ec
parent 5132e0d1433b0962f9609dbbd68427e1fb5a4b9d

Add AVX support for packed mul/div instructions.

This is a follow up for the below patch:
https://android-review.googlesource.com/c/platform/build/+/830841

Test: ./test.py --host --64, test-art-host-gtest
Change-Id: Id2aa473035556ee230e66addeb69707df8530e75
Signed-off-by: Shalini Salomi Bodapati <shalini.salomi.bodapati@intel.com>

compiler/utils/x86/assembler_x86.h

diff --git a/compiler/utils/x86/assembler_x86.h b/compiler/utils/x86/assembler_x86.h
index 17039f0..dce546e 100644
--- a/compiler/utils/x86/assembler_x86.h
+++ b/compiler/utils/x86/assembler_x86.h
@@ -417,6 +417,11 @@
   void mulps(XmmRegister dst, XmmRegister src);
   void divps(XmmRegister dst, XmmRegister src);
 
+  void vmulps(XmmRegister dst, XmmRegister src1, XmmRegister src2);
+  void vmulpd(XmmRegister dst, XmmRegister src1, XmmRegister src2);
+  void vdivps(XmmRegister dst, XmmRegister src1, XmmRegister src2);
+  void vdivpd(XmmRegister dst, XmmRegister src1, XmmRegister src2);
+
   void vaddps(XmmRegister dst, XmmRegister add_left, XmmRegister add_right);
   void vsubps(XmmRegister dst, XmmRegister add_left, XmmRegister add_right);
   void vsubpd(XmmRegister dst, XmmRegister add_left, XmmRegister add_right);
@@ -476,6 +481,7 @@
   void paddw(XmmRegister dst, XmmRegister src);
   void psubw(XmmRegister dst, XmmRegister src);
   void pmullw(XmmRegister dst, XmmRegister src);
+  void vpmullw(XmmRegister dst, XmmRegister src1, XmmRegister src2);
 
   void vpsubb(XmmRegister dst, XmmRegister src1, XmmRegister src2);
   void vpsubw(XmmRegister dst, XmmRegister src1, XmmRegister src2);
@@ -485,6 +491,8 @@
   void psubd(XmmRegister dst, XmmRegister src);
   void pmulld(XmmRegister dst, XmmRegister src);
 
+  void vpmulld(XmmRegister dst, XmmRegister src1, XmmRegister src2);
+
   void vpaddd(XmmRegister dst, XmmRegister src1, XmmRegister src2);
 
   void paddq(XmmRegister dst, XmmRegister src);