Menu
▾
▴
[Valgrind-developers] valgrind: r11819 - trunk/memcheck
|
From: <sv...@va...> - 2011-06-16 11:42:11
|
Author: sewardj
Date: 2011-06-16 12:37:21 +0100 (Thu, 16 Jun 2011)
New Revision: 11819
Log:
Track vex r2163 (en-mass renaming of vector narrowing and widening primops)
Modified:
trunk/memcheck/mc_translate.c
Modified: trunk/memcheck/mc_translate.c
===================================================================
--- trunk/memcheck/mc_translate.c 2011-06-15 21:30:55 UTC (rev 11818)
+++ trunk/memcheck/mc_translate.c 2011-06-16 11:37:21 UTC (rev 11819)
@@ -42,6 +42,13 @@
#include "mc_include.h"
+/* FIXMEs JRS 2011-June-16.
+
+ Check the interpretation for vector narrowing and widening ops,
+ particularly the saturating ones. I suspect they are either overly
+ pessimistic and/or wrong.
+*/
+
/* This file implements the Memcheck instrumentation, and in
particular contains the core of its undefined value detection
machinery. For a comprehensive background of the terminology,
@@ -1981,19 +1988,19 @@
primops.
*/
static
-IRAtom* vectorNarrowV128 ( MCEnv* mce, IROp narrow_op,
- IRAtom* vatom1, IRAtom* vatom2)
+IRAtom* vectorNarrowBinV128 ( MCEnv* mce, IROp narrow_op,
+ IRAtom* vatom1, IRAtom* vatom2)
{
IRAtom *at1, *at2, *at3;
IRAtom* (*pcast)( MCEnv*, IRAtom* );
switch (narrow_op) {
- case Iop_QNarrow32Sto16Sx8: pcast = mkPCast32x4; break;
- case Iop_QNarrow32Uto16Ux8: pcast = mkPCast32x4; break;
- case Iop_QNarrow32Sto16Ux8: pcast = mkPCast32x4; break;
- case Iop_QNarrow16Sto8Sx16: pcast = mkPCast16x8; break;
- case Iop_QNarrow16Uto8Ux16: pcast = mkPCast16x8; break;
- case Iop_QNarrow16Sto8Ux16: pcast = mkPCast16x8; break;
- default: VG_(tool_panic)("vectorNarrowV128");
+ case Iop_QNarrowBin32Sto16Sx8: pcast = mkPCast32x4; break;
+ case Iop_QNarrowBin32Uto16Ux8: pcast = mkPCast32x4; break;
+ case Iop_QNarrowBin32Sto16Ux8: pcast = mkPCast32x4; break;
+ case Iop_QNarrowBin16Sto8Sx16: pcast = mkPCast16x8; break;
+ case Iop_QNarrowBin16Uto8Ux16: pcast = mkPCast16x8; break;
+ case Iop_QNarrowBin16Sto8Ux16: pcast = mkPCast16x8; break;
+ default: VG_(tool_panic)("vectorNarrowBinV128");
}
tl_assert(isShadowAtom(mce,vatom1));
tl_assert(isShadowAtom(mce,vatom2));
@@ -2004,16 +2011,16 @@
}
static
-IRAtom* vectorNarrow64 ( MCEnv* mce, IROp narrow_op,
- IRAtom* vatom1, IRAtom* vatom2)
+IRAtom* vectorNarrowBin64 ( MCEnv* mce, IROp narrow_op,
+ IRAtom* vatom1, IRAtom* vatom2)
{
IRAtom *at1, *at2, *at3;
IRAtom* (*pcast)( MCEnv*, IRAtom* );
switch (narrow_op) {
- case Iop_QNarrow32Sto16Sx4: pcast = mkPCast32x2; break;
- case Iop_QNarrow16Sto8Sx8: pcast = mkPCast16x4; break;
- case Iop_QNarrow16Sto8Ux8: pcast = mkPCast16x4; break;
- default: VG_(tool_panic)("vectorNarrow64");
+ case Iop_QNarrowBin32Sto16Sx4: pcast = mkPCast32x2; break;
+ case Iop_QNarrowBin16Sto8Sx8: pcast = mkPCast16x4; break;
+ case Iop_QNarrowBin16Sto8Ux8: pcast = mkPCast16x4; break;
+ default: VG_(tool_panic)("vectorNarrowBin64");
}
tl_assert(isShadowAtom(mce,vatom1));
tl_assert(isShadowAtom(mce,vatom2));
@@ -2024,25 +2031,27 @@
}
static
-IRAtom* vectorShortenV128 ( MCEnv* mce, IROp shorten_op,
- IRAtom* vatom1)
+IRAtom* vectorNarrowUnV128 ( MCEnv* mce, IROp shorten_op,
+ IRAtom* vatom1)
{
IRAtom *at1, *at2;
IRAtom* (*pcast)( MCEnv*, IRAtom* );
switch (shorten_op) {
- case Iop_Shorten16x8: pcast = mkPCast16x8; break;
- case Iop_Shorten32x4: pcast = mkPCast32x4; break;
- case Iop_Shorten64x2: pcast = mkPCast64x2; break;
- case Iop_QShortenS16Sx8: pcast = mkPCast16x8; break;
- case Iop_QShortenU16Sx8: pcast = mkPCast16x8; break;
- case Iop_QShortenU16Ux8: pcast = mkPCast16x8; break;
- case Iop_QShortenS32Sx4: pcast = mkPCast32x4; break;
- case Iop_QShortenU32Sx4: pcast = mkPCast32x4; break;
- case Iop_QShortenU32Ux4: pcast = mkPCast32x4; break;
- case Iop_QShortenS64Sx2: pcast = mkPCast64x2; break;
- case Iop_QShortenU64Sx2: pcast = mkPCast64x2; break;
- case Iop_QShortenU64Ux2: pcast = mkPCast64x2; break;
- default: VG_(tool_panic)("vectorShortenV128");
+ /* FIXME: first 3 are too pessimistic; we can just
+ apply them directly to the V bits. */
+ case Iop_NarrowUn16to8x8: pcast = mkPCast16x8; break;
+ case Iop_NarrowUn32to16x4: pcast = mkPCast32x4; break;
+ case Iop_NarrowUn64to32x2: pcast = mkPCast64x2; break;
+ case Iop_QNarrowUn16Sto8Sx8: pcast = mkPCast16x8; break;
+ case Iop_QNarrowUn16Sto8Ux8: pcast = mkPCast16x8; break;
+ case Iop_QNarrowUn16Uto8Ux8: pcast = mkPCast16x8; break;
+ case Iop_QNarrowUn32Sto16Sx4: pcast = mkPCast32x4; break;
+ case Iop_QNarrowUn32Sto16Ux4: pcast = mkPCast32x4; break;
+ case Iop_QNarrowUn32Uto16Ux4: pcast = mkPCast32x4; break;
+ case Iop_QNarrowUn64Sto32Sx2: pcast = mkPCast64x2; break;
+ case Iop_QNarrowUn64Sto32Ux2: pcast = mkPCast64x2; break;
+ case Iop_QNarrowUn64Uto32Ux2: pcast = mkPCast64x2; break;
+ default: VG_(tool_panic)("vectorNarrowUnV128");
}
tl_assert(isShadowAtom(mce,vatom1));
at1 = assignNew('V', mce, Ity_V128, pcast(mce, vatom1));
@@ -2051,19 +2060,19 @@
}
static
-IRAtom* vectorLongenI64 ( MCEnv* mce, IROp longen_op,
- IRAtom* vatom1)
+IRAtom* vectorWidenI64 ( MCEnv* mce, IROp longen_op,
+ IRAtom* vatom1)
{
IRAtom *at1, *at2;
IRAtom* (*pcast)( MCEnv*, IRAtom* );
switch (longen_op) {
- case Iop_Longen8Ux8: pcast = mkPCast16x8; break;
- case Iop_Longen8Sx8: pcast = mkPCast16x8; break;
- case Iop_Longen16Ux4: pcast = mkPCast32x4; break;
- case Iop_Longen16Sx4: pcast = mkPCast32x4; break;
- case Iop_Longen32Ux2: pcast = mkPCast64x2; break;
- case Iop_Longen32Sx2: pcast = mkPCast64x2; break;
- default: VG_(tool_panic)("vectorLongenI64");
+ case Iop_Widen8Uto16x8: pcast = mkPCast16x8; break;
+ case Iop_Widen8Sto16x8: pcast = mkPCast16x8; break;
+ case Iop_Widen16Uto32x4: pcast = mkPCast32x4; break;
+ case Iop_Widen16Sto32x4: pcast = mkPCast32x4; break;
+ case Iop_Widen32Uto64x2: pcast = mkPCast64x2; break;
+ case Iop_Widen32Sto64x2: pcast = mkPCast64x2; break;
+ default: VG_(tool_panic)("vectorWidenI64");
}
tl_assert(isShadowAtom(mce,vatom1));
at1 = assignNew('V', mce, Ity_V128, unop(longen_op, vatom1));
@@ -2349,10 +2358,10 @@
complainIfUndefined(mce, atom2);
return assignNew('V', mce, Ity_I64, binop(op, vatom1, atom2));
- case Iop_QNarrow32Sto16Sx4:
- case Iop_QNarrow16Sto8Sx8:
- case Iop_QNarrow16Sto8Ux8:
- return vectorNarrow64(mce, op, vatom1, vatom2);
+ case Iop_QNarrowBin32Sto16Sx4:
+ case Iop_QNarrowBin16Sto8Sx8:
+ case Iop_QNarrowBin16Sto8Ux8:
+ return vectorNarrowBin64(mce, op, vatom1, vatom2);
case Iop_Min8Ux8:
case Iop_Min8Sx8:
@@ -2705,13 +2714,13 @@
case Iop_QSub64Sx2:
return binary64Ix2(mce, vatom1, vatom2);
- case Iop_QNarrow32Sto16Sx8:
- case Iop_QNarrow32Uto16Ux8:
- case Iop_QNarrow32Sto16Ux8:
- case Iop_QNarrow16Sto8Sx16:
- case Iop_QNarrow16Uto8Ux16:
- case Iop_QNarrow16Sto8Ux16:
- return vectorNarrowV128(mce, op, vatom1, vatom2);
+ case Iop_QNarrowBin32Sto16Sx8:
+ case Iop_QNarrowBin32Uto16Ux8:
+ case Iop_QNarrowBin32Sto16Ux8:
+ case Iop_QNarrowBin16Sto8Sx16:
+ case Iop_QNarrowBin16Uto8Ux16:
+ case Iop_QNarrowBin16Sto8Ux16:
+ return vectorNarrowBinV128(mce, op, vatom1, vatom2);
case Iop_Sub64Fx2:
case Iop_Mul64Fx2:
@@ -2804,20 +2813,20 @@
case Iop_Mull32Sx2:
case Iop_Mull32Ux2:
case Iop_QDMulLong32Sx2:
- return vectorLongenI64(mce, Iop_Longen32Sx2,
- mkUifU64(mce, vatom1, vatom2));
+ return vectorWidenI64(mce, Iop_Widen32Sto64x2,
+ mkUifU64(mce, vatom1, vatom2));
case Iop_Mull16Sx4:
case Iop_Mull16Ux4:
case Iop_QDMulLong16Sx4:
- return vectorLongenI64(mce, Iop_Longen16Sx4,
- mkUifU64(mce, vatom1, vatom2));
+ return vectorWidenI64(mce, Iop_Widen16Sto32x4,
+ mkUifU64(mce, vatom1, vatom2));
case Iop_Mull8Sx8:
case Iop_Mull8Ux8:
case Iop_PolynomialMull8x8:
- return vectorLongenI64(mce, Iop_Longen8Sx8,
- mkUifU64(mce, vatom1, vatom2));
+ return vectorWidenI64(mce, Iop_Widen8Sto16x8,
+ mkUifU64(mce, vatom1, vatom2));
case Iop_PwAdd32x4:
return mkPCast32x4(mce,
@@ -2913,8 +2922,8 @@
32x4 -> 16x8 laneage, discarding the upper half of each lane.
Simply apply same op to the V bits, since this really no more
than a data steering operation. */
- case Iop_Narrow32x4:
- case Iop_Narrow16x8:
+ case Iop_NarrowBin32to16x8:
+ case Iop_NarrowBin16to8x16:
return assignNew('V', mce, Ity_V128,
binop(op, vatom1, vatom2));
@@ -3410,27 +3419,27 @@
case Iop_CmpNEZ64x2:
return mkPCast64x2(mce, vatom);
- case Iop_Shorten16x8:
- case Iop_Shorten32x4:
- case Iop_Shorten64x2:
- case Iop_QShortenS16Sx8:
- case Iop_QShortenU16Sx8:
- case Iop_QShortenU16Ux8:
- case Iop_QShortenS32Sx4:
- case Iop_QShortenU32Sx4:
- case Iop_QShortenU32Ux4:
- case Iop_QShortenS64Sx2:
- case Iop_QShortenU64Sx2:
- case Iop_QShortenU64Ux2:
- return vectorShortenV128(mce, op, vatom);
+ case Iop_NarrowUn16to8x8:
+ case Iop_NarrowUn32to16x4:
+ case Iop_NarrowUn64to32x2:
+ case Iop_QNarrowUn16Sto8Sx8:
+ case Iop_QNarrowUn16Sto8Ux8:
+ case Iop_QNarrowUn16Uto8Ux8:
+ case Iop_QNarrowUn32Sto16Sx4:
+ case Iop_QNarrowUn32Sto16Ux4:
+ case Iop_QNarrowUn32Uto16Ux4:
+ case Iop_QNarrowUn64Sto32Sx2:
+ case Iop_QNarrowUn64Sto32Ux2:
+ case Iop_QNarrowUn64Uto32Ux2:
+ return vectorNarrowUnV128(mce, op, vatom);
- case Iop_Longen8Sx8:
- case Iop_Longen8Ux8:
- case Iop_Longen16Sx4:
- case Iop_Longen16Ux4:
- case Iop_Longen32Sx2:
- case Iop_Longen32Ux2:
- return vectorLongenI64(mce, op, vatom);
+ case Iop_Widen8Sto16x8:
+ case Iop_Widen8Uto16x8:
+ case Iop_Widen16Sto32x4:
+ case Iop_Widen16Uto32x4:
+ case Iop_Widen32Sto64x2:
+ case Iop_Widen32Uto64x2:
+ return vectorWidenI64(mce, op, vatom);
case Iop_PwAddL32Ux2:
case Iop_PwAddL32Sx2:
|