summaryrefslogtreecommitdiff
path: root/src/arch
diff options
context:
space:
mode:
authorKorey Sewell <ksewell@umich.edu>2009-05-12 15:01:15 -0400
committerKorey Sewell <ksewell@umich.edu>2009-05-12 15:01:15 -0400
commit6211fe5d2ee00dae9cd72d9cccdc900241a8f9a2 (patch)
treeee0ac43983c31c9c22b7ea28e69725d033287ef4 /src/arch
parent3603dd25efb70ec400e8ea1e76137e8e288e533a (diff)
downloadgem5-6211fe5d2ee00dae9cd72d9cccdc900241a8f9a2.tar.xz
inorder-float: Fix storage of FP results
inorder was incorrectly storing FP values and confusing the integer/fp storage view of floating point operations. A big issue was knowing trying to infer when were doing single or double precision access because this lets you know the size of value to store (32-64 bits). This isnt exactly straightforward since alpha uses all 64-bit regs while mips/sparc uses a dual-reg view. by getting this value from the actual floating point register file, the model can figure out what it needs to store
Diffstat (limited to 'src/arch')
-rw-r--r--src/arch/alpha/floatregfile.hh2
1 files changed, 2 insertions, 0 deletions
diff --git a/src/arch/alpha/floatregfile.hh b/src/arch/alpha/floatregfile.hh
index a5a3a7861..f6abb1f86 100644
--- a/src/arch/alpha/floatregfile.hh
+++ b/src/arch/alpha/floatregfile.hh
@@ -52,6 +52,8 @@ const int QuadBytes = QuadWidth / 4;
class FloatRegFile
{
public:
+ static const int regWidth = DoubleWidth;
+
union {
uint64_t q[NumFloatRegs]; // integer qword view
double d[NumFloatRegs]; // double-precision floating point view