Subzero, MIPS32: Fix floating point comparison crosstest

Floating point comparison crosstest was failing in filetype=obj mode because of missing breaks in load encoding functions. With this patch, crosstest generator, with vector tests disabled, for parameters --filetype=obj --include=test_fcmp,mips32,native,Om1,base returns TotalTests=123904 Passes=123904 Failures=0 R=stichnot@chromium.org Review URL: https://codereview.chromium.org/2355413008 . Patch from Srdjan Obucina <Srdjan.Obucina@imgtec.com>.

Subzero, MIPS32: Fix floating point comparison crosstest
6ee373fc · Srdjan Obucina · Jim Stichnoth · d27ce3d9 · 6ee373fc
Commit 6ee373fc authored Sep 22, 2016 by Srdjan Obucina Committed by Jim Stichnoth Sep 22, 2016
Hide whitespace changes
Inline Side-by-side

Showing with 3 additions and 0 deletions

IceAssemblerMIPS32.cpp src/IceAssemblerMIPS32.cpp +3 -0

No files found.
--- a/src/IceAssemblerMIPS32.cpp
+++ b/src/IceAssemblerMIPS32.cpp
@@ -528,10 +528,12 @@ void AssemblerMIPS32::lw(const Operand *OpRt, const Operand *OpBase,
  case IceType_i8: {
    static constexpr IValueT Opcode = 0x80000000;
    emitRtRsImm16(Opcode, OpRt, OpBase, Offset, "lb");
+    break;
  }
  case IceType_i16: {
    static constexpr IValueT Opcode = 0x84000000;
    emitRtRsImm16(Opcode, OpRt, OpBase, Offset, "lh");
+    break;
  }
  case IceType_i32: {
    static constexpr IValueT Opcode = 0x8C000000;
@@ -541,6 +543,7 @@ void AssemblerMIPS32::lw(const Operand *OpRt, const Operand *OpBase,
  case IceType_f32: {
    static constexpr IValueT Opcode = 0xC4000000;
    emitFtRsImm16(Opcode, OpRt, OpBase, Offset, "lwc1");
+    break;
  }
  case IceType_f64: {
    static constexpr IValueT Opcode = 0xD4000000;