Menu
▾
▴
[Valgrind-developers] vex: r2388: More AVX insns:
|
From: <sv...@va...> - 2012-06-18 15:01:45
|
sewardj 2012-06-18 16:01:30 +0100 (Mon, 18 Jun 2012)
New Revision: 2388
Log:
More AVX insns:
VMOVUPD ymm1, ymm2/m256 = VEX.256.66.0F.WIG 11 /r
VMOVUPS ymm1, ymm2/m256 = VEX.256.0F.WIG 11 /r
VCOMISD xmm2/m64, xmm1 = VEX.LIG.66.0F.WIG 2F /r
VPCMPGTD r/m, rV, r ::: r = rV `>s-by-32s` r/m
VPMOVSXBD xmm2/m32, xmm1
VPMOVZXBD xmm2/m32, xmm1
VDPPD xmm3/m128,xmm2,xmm1 = VEX.NDS.128.66.0F3A.WIG 41 /r ib
and common up duplication in implementation of PINSRW/VPINSRW.
Modified files:
trunk/priv/guest_amd64_toIR.c
Modified: trunk/priv/guest_amd64_toIR.c (+256 -188)
===================================================================
--- trunk/priv/guest_amd64_toIR.c 2012-06-18 15:05:52 +01:00 (rev 2387)
+++ trunk/priv/guest_amd64_toIR.c 2012-06-18 16:01:30 +01:00 (rev 2388)
@@ -1467,12 +1467,6 @@
stmt( IRStmt_Put( xmmGuestRegLane32offset(xmmreg,laneno), e ) );
}
-static void putXMMRegLane16 ( UInt xmmreg, Int laneno, IRExpr* e )
-{
- vassert(typeOfIRExpr(irsb->tyenv,e) == Ity_I16);
- stmt( IRStmt_Put( xmmGuestRegLane16offset(xmmreg,laneno), e ) );
-}
-
static IRExpr* getYMMReg ( UInt xmmreg )
{
return IRExpr_Get( ymmGuestRegOffset(xmmreg), Ity_V256 );
@@ -10321,6 +10315,32 @@
}
+static IRTemp math_PINSRW_128 ( IRTemp v128, IRTemp u16, UInt imm8 )
+{
+ vassert(imm8 >= 0 && imm8 <= 7);
+
+ // Create a V128 value which has the selected word in the
+ // specified lane, and zeroes everywhere else.
+ IRTemp tmp128 = newTemp(Ity_V128);
+ IRTemp halfshift = newTemp(Ity_I64);
+ assign(halfshift, binop(Iop_Shl64,
+ unop(Iop_16Uto64, mkexpr(u16)),
+ mkU8(16 * (imm8 & 3))));
+ if (imm8 < 4) {
+ assign(tmp128, binop(Iop_64HLtoV128, mkU64(0), mkexpr(halfshift)));
+ } else {
+ assign(tmp128, binop(Iop_64HLtoV128, mkexpr(halfshift), mkU64(0)));
+ }
+
+ UShort mask = ~(3 << (imm8 * 2));
+ IRTemp res = newTemp(Ity_V128);
+ assign( res, binop(Iop_OrV128,
+ mkexpr(tmp128),
+ binop(Iop_AndV128, mkexpr(v128), mkV128(mask))) );
+ return res;
+}
+
+
/* Note, this also handles SSE(1) insns. */
__attribute__((noinline))
static
@@ -11180,14 +11200,14 @@
case 0x2E:
case 0x2F:
+ /* 66 0F 2E = UCOMISD -- 64F0x2 comparison G,E, and set ZCP */
/* 66 0F 2F = COMISD -- 64F0x2 comparison G,E, and set ZCP */
- /* 66 0F 2E = UCOMISD -- 64F0x2 comparison G,E, and set ZCP */
if (have66noF2noF3(pfx) && sz == 2) {
delta = dis_COMISD( vbi, pfx, delta, False/*!isAvx*/, opc );
goto decode_success;
}
+ /* 0F 2E = UCOMISS -- 32F0x4 comparison G,E, and set ZCP */
/* 0F 2F = COMISS -- 32F0x4 comparison G,E, and set ZCP */
- /* 0F 2E = UCOMISS -- 32F0x4 comparison G,E, and set ZCP */
if (haveNo66noF2noF3(pfx) && sz == 4) {
delta = dis_COMISS( vbi, pfx, delta, False/*!isAvx*/, opc );
goto decode_success;
@@ -12482,26 +12502,27 @@
Int lane;
t4 = newTemp(Ity_I16);
modrm = getUChar(delta);
-
+ UInt rG = gregOfRexRM(pfx,modrm);
if (epartIsReg(modrm)) {
- assign(t4, getIReg16(eregOfRexRM(pfx,modrm)));
+ UInt rE = eregOfRexRM(pfx,modrm);
+ assign(t4, getIReg16(rE));
delta += 1+1;
lane = getUChar(delta-1);
- DIP("pinsrw $%d,%s,%s\n", (Int)lane,
- nameIReg16(eregOfRexRM(pfx,modrm)),
- nameXMMReg(gregOfRexRM(pfx,modrm)));
+ DIP("pinsrw $%d,%s,%s\n",
+ (Int)lane, nameIReg16(rE), nameXMMReg(rG));
} else {
addr = disAMode ( &alen, vbi, pfx, delta, dis_buf,
1/*byte after the amode*/ );
delta += 1+alen;
lane = getUChar(delta-1);
assign(t4, loadLE(Ity_I16, mkexpr(addr)));
- DIP("pinsrw $%d,%s,%s\n", (Int)lane,
- dis_buf,
- nameXMMReg(gregOfRexRM(pfx,modrm)));
- }
-
- putXMMRegLane16( gregOfRexRM(pfx,modrm), lane & 7, mkexpr(t4) );
+ DIP("pinsrw $%d,%s,%s\n",
+ (Int)lane, dis_buf, nameXMMReg(rG));
+ }
+ IRTemp src_vec = newTemp(Ity_V128);
+ assign(src_vec, getXMMReg(rG));
+ IRTemp res_vec = math_PINSRW_128( src_vec, t4, lane & 7);
+ putXMMReg(rG, mkexpr(res_vec));
goto decode_success;
}
break;
@@ -14862,6 +14883,49 @@
}
+/* Handles 128 bit versions of PMOVZXBD and PMOVSXBD. */
+static Long dis_PMOVxXBD_128 ( VexAbiInfo* vbi, Prefix pfx,
+ Long delta, Bool isAvx, Bool xIsZ )
+{
+ IRTemp addr = IRTemp_INVALID;
+ Int alen = 0;
+ HChar dis_buf[50];
+ IRTemp srcVec = newTemp(Ity_V128);
+ UChar modrm = getUChar(delta);
+ UChar* mbV = isAvx ? "v" : "";
+ UChar how = xIsZ ? 'z' : 's';
+ UInt rG = gregOfRexRM(pfx, modrm);
+ if ( epartIsReg(modrm) ) {
+ UInt rE = eregOfRexRM(pfx, modrm);
+ assign( srcVec, getXMMReg(rE) );
+ delta += 1;
+ DIP( "%spmov%cxbd %s,%s\n", mbV, how, nameXMMReg(rE), nameXMMReg(rG) );
+ } else {
+ addr = disAMode( &alen, vbi, pfx, delta, dis_buf, 0 );
+ assign( srcVec,
+ unop( Iop_32UtoV128, loadLE( Ity_I32, mkexpr(addr) ) ) );
+ delta += alen;
+ DIP( "%spmov%cxbd %s,%s\n", mbV, how, dis_buf, nameXMMReg(rG) );
+ }
+
+ IRTemp zeroVec = newTemp(Ity_V128);
+ assign( zeroVec, IRExpr_Const( IRConst_V128(0) ) );
+
+ IRExpr* res
+ = binop(Iop_InterleaveLO8x16,
+ mkexpr(zeroVec),
+ binop(Iop_InterleaveLO8x16,
+ mkexpr(zeroVec), mkexpr(srcVec)));
+ if (!xIsZ)
+ res = binop(Iop_SarN32x4,
+ binop(Iop_ShlN32x4, res, mkU8(24)), mkU8(24));
+
+ (isAvx ? putYMMRegLoAndZU : putXMMReg) ( rG, res );
+
+ return delta;
+}
+
+
static Long dis_PHMINPOSUW_128 ( VexAbiInfo* vbi, Prefix pfx,
Long delta, Bool isAvx )
{
@@ -15109,39 +15173,8 @@
/* 66 0F 38 21 /r = PMOVSXBD xmm1, xmm2/m32
Packed Move with Sign Extend from Byte to DWord (XMM) */
if (have66noF2noF3(pfx) && sz == 2) {
-
- modrm = getUChar(delta);
-
- IRTemp srcVec = newTemp(Ity_V128);
-
- if ( epartIsReg( modrm ) ) {
- assign( srcVec, getXMMReg( eregOfRexRM(pfx, modrm) ) );
- delta += 1;
- DIP( "pmovsxbd %s,%s\n",
- nameXMMReg( eregOfRexRM(pfx, modrm) ),
- nameXMMReg( gregOfRexRM(pfx, modrm) ) );
- } else {
- addr = disAMode( &alen, vbi, pfx, delta, dis_buf, 0 );
- assign( srcVec,
- unop( Iop_32UtoV128, loadLE( Ity_I32, mkexpr(addr) ) ) );
- delta += alen;
- DIP( "pmovsxbd %s,%s\n",
- dis_buf, nameXMMReg( gregOfRexRM(pfx, modrm) ) );
- }
-
- IRTemp zeroVec = newTemp(Ity_V128);
- assign( zeroVec, IRExpr_Const( IRConst_V128(0) ) );
-
- putXMMReg( gregOfRexRM(pfx, modrm),
- binop( Iop_SarN32x4,
- binop( Iop_ShlN32x4,
- binop( Iop_InterleaveLO8x16,
- mkexpr(zeroVec),
- binop( Iop_InterleaveLO8x16,
- mkexpr(zeroVec),
- mkexpr(srcVec) ) ),
- mkU8(24) ), mkU8(24) ) );
-
+ delta = dis_PMOVxXBD_128( vbi, pfx, delta,
+ False/*!isAvx*/, False/*!xIsZ*/ );
goto decode_success;
}
break;
@@ -15336,35 +15369,8 @@
/* 66 0F 38 31 /r = PMOVZXBD xmm1, xmm2/m32
Packed Move with Zero Extend from Byte to DWord (XMM) */
if (have66noF2noF3(pfx) && sz == 2) {
-
- modrm = getUChar(delta);
-
- IRTemp srcVec = newTemp(Ity_V128);
-
- if ( epartIsReg(modrm) ) {
- assign( srcVec, getXMMReg( eregOfRexRM(pfx, modrm) ) );
- delta += 1;
- DIP( "pmovzxbd %s,%s\n",
- nameXMMReg( eregOfRexRM(pfx, modrm) ),
- nameXMMReg( gregOfRexRM(pfx, modrm) ) );
- } else {
- addr = disAMode( &alen, vbi, pfx, delta, dis_buf, 0 );
- assign( srcVec,
- unop( Iop_32UtoV128, loadLE( Ity_I32, mkexpr(addr) ) ) );
- delta += alen;
- DIP( "pmovzxbd %s,%s\n",
- dis_buf, nameXMMReg( gregOfRexRM(pfx, modrm) ) );
- }
-
- IRTemp zeroVec = newTemp(Ity_V128);
- assign( zeroVec, IRExpr_Const( IRConst_V128(0) ) );
-
- putXMMReg( gregOfRexRM( pfx, modrm ),
- binop( Iop_InterleaveLO8x16,
- mkexpr(zeroVec),
- binop( Iop_InterleaveLO8x16,
- mkexpr(zeroVec), mkexpr(srcVec) ) ) );
-
+ delta = dis_PMOVxXBD_128( vbi, pfx, delta,
+ False/*!isAvx*/, True/*xIsZ*/ );
goto decode_success;
}
break;
@@ -16003,13 +16009,15 @@
static IRTemp math_PINSRB_128 ( IRTemp v128, IRTemp u8, UInt imm8 )
{
+ vassert(imm8 >= 0 && imm8 <= 15);
+
// Create a V128 value which has the selected byte in the
// specified lane, and zeroes everywhere else.
- IRTemp tmp128 = newTemp(Ity_V128);
+ IRTemp tmp128 = newTemp(Ity_V128);
IRTemp halfshift = newTemp(Ity_I64);
assign(halfshift, binop(Iop_Shl64,
- mkexpr(u8), mkU8(8 * (imm8 & 7))));
- vassert(imm8 >= 0 && imm8 <= 15);
+ unop(Iop_8Uto64, mkexpr(u8)),
+ mkU8(8 * (imm8 & 7))));
if (imm8 < 8) {
assign(tmp128, binop(Iop_64HLtoV128, mkU64(0), mkexpr(halfshift)));
} else {
@@ -16017,11 +16025,10 @@
}
UShort mask = ~(1 << imm8);
-
- IRTemp res = newTemp(Ity_V128);
- assign(res, binop( Iop_OrV128, mkexpr(tmp128),
- binop( Iop_AndV128,
- mkexpr(v128), mkV128(mask) ) ) );
+ IRTemp res = newTemp(Ity_V128);
+ assign( res, binop(Iop_OrV128,
+ mkexpr(tmp128),
+ binop(Iop_AndV128, mkexpr(v128), mkV128(mask))) );
return res;
}
@@ -16156,6 +16163,31 @@
}
+static IRTemp math_DPPD_128 ( IRTemp src_vec, IRTemp dst_vec, UInt imm8 )
+{
+ vassert(imm8 < 256);
+ UShort imm8_perms[4] = { 0x0000, 0x00FF, 0xFF00, 0xFFFF };
+ IRTemp and_vec = newTemp(Ity_V128);
+ IRTemp sum_vec = newTemp(Ity_V128);
+ assign( and_vec, binop( Iop_AndV128,
+ binop( Iop_Mul64Fx2,
+ mkexpr(dst_vec), mkexpr(src_vec) ),
+ mkV128( imm8_perms[ ((imm8 >> 4) & 3) ] ) ) );
+
+ assign( sum_vec, binop( Iop_Add64F0x2,
+ binop( Iop_InterleaveHI64x2,
+ mkexpr(and_vec), mkexpr(and_vec) ),
+ binop( Iop_InterleaveLO64x2,
+ mkexpr(and_vec), mkexpr(and_vec) ) ) );
+ IRTemp res = newTemp(Ity_V128);
+ assign(res, binop( Iop_AndV128,
+ binop( Iop_InterleaveLO64x2,
+ mkexpr(sum_vec), mkexpr(sum_vec) ),
+ mkV128( imm8_perms[ (imm8 & 3) ] ) ) );
+ return res;
+}
+
+
__attribute__((noinline))
static
Long dis_ESC_0F3A__SSE4 ( Bool* decode_OK,
@@ -16600,35 +16632,29 @@
/* 66 0F 3A 20 /r ib = PINSRB xmm1, r32/m8, imm8
Extract byte from r32/m8 and insert into xmm1 */
if (have66noF2noF3(pfx) && sz == 2) {
-
Int imm8;
- IRTemp new8 = newTemp(Ity_I64);
-
+ IRTemp new8 = newTemp(Ity_I8);
modrm = getUChar(delta);
UInt rG = gregOfRexRM(pfx, modrm);
-
if ( epartIsReg( modrm ) ) {
+ UInt rE = eregOfRexRM(pfx,modrm);
imm8 = (Int)(getUChar(delta+1) & 0xF);
- assign( new8, binop(Iop_And64,
- unop(Iop_32Uto64,
- getIReg32(eregOfRexRM(pfx,modrm))),
- mkU64(0xFF)));
+ assign( new8, unop(Iop_32to8, getIReg32(rE)) );
delta += 1+1;
DIP( "pinsrb $%d,%s,%s\n", imm8,
- nameIReg32( eregOfRexRM(pfx, modrm) ),
- nameXMMReg( gregOfRexRM(pfx, modrm) ) );
+ nameIReg32(rE), nameXMMReg(rG) );
} else {
addr = disAMode( &alen, vbi, pfx, delta, dis_buf, 1 );
imm8 = (Int)(getUChar(delta+alen) & 0xF);
- assign( new8, unop(Iop_8Uto64, loadLE( Ity_I8, mkexpr(addr) )));
+ assign( new8, loadLE( Ity_I8, mkexpr(addr) ) );
delta += alen+1;
DIP( "pinsrb $%d,%s,%s\n",
- imm8, dis_buf, nameXMMReg( gregOfRexRM(pfx, modrm) ) );
+ imm8, dis_buf, nameXMMReg(rG) );
}
-
IRTemp src_vec = newTemp(Ity_V128);
- assign(src_vec, getXMMReg( rG ));
- putXMMReg( rG, mkexpr(math_PINSRB_128( src_vec, new8, imm8 )) );
+ assign(src_vec, getXMMReg( gregOfRexRM(pfx, modrm) ));
+ IRTemp res = math_PINSRB_128( src_vec, new8, imm8 );
+ putXMMReg( rG, mkexpr(res) );
goto decode_success;
}
break;
@@ -16809,24 +16835,19 @@
/* 66 0F 3A 41 /r ib = DPPD xmm1, xmm2/m128, imm8
Dot Product of Packed Double Precision Floating-Point Values (XMM) */
if (have66noF2noF3(pfx) && sz == 2) {
-
- Int imm8;
+ modrm = getUChar(delta);
+ Int imm8;
IRTemp src_vec = newTemp(Ity_V128);
IRTemp dst_vec = newTemp(Ity_V128);
- IRTemp and_vec = newTemp(Ity_V128);
- IRTemp sum_vec = newTemp(Ity_V128);
-
- modrm = getUChar(delta);
-
- assign( dst_vec, getXMMReg( gregOfRexRM(pfx, modrm) ) );
-
+ UInt rG = gregOfRexRM(pfx, modrm);
+ assign( dst_vec, getXMMReg( rG ) );
if ( epartIsReg( modrm ) ) {
+ UInt rE = eregOfRexRM(pfx, modrm);
imm8 = (Int)getUChar(delta+1);
- assign( src_vec, getXMMReg( eregOfRexRM(pfx, modrm) ) );
+ assign( src_vec, getXMMReg(rE) );
delta += 1+1;
- DIP( "dppd $%d, %s,%s\n", imm8,
- nameXMMReg( eregOfRexRM(pfx, modrm) ),
- nameXMMReg( gregOfRexRM(pfx, modrm) ) );
+ DIP( "dppd $%d, %s,%s\n",
+ imm8, nameXMMReg(rE), nameXMMReg(rG) );
} else {
addr = disAMode( &alen, vbi, pfx, delta, dis_buf,
1/* imm8 is 1 byte after the amode */ );
@@ -16835,28 +16856,10 @@
imm8 = (Int)getUChar(delta+alen);
delta += alen+1;
DIP( "dppd $%d, %s,%s\n",
- imm8, dis_buf, nameXMMReg( gregOfRexRM(pfx, modrm) ) );
+ imm8, dis_buf, nameXMMReg(rG) );
}
-
- UShort imm8_perms[4] = { 0x0000, 0x00FF, 0xFF00, 0xFFFF };
-
- assign( and_vec, binop( Iop_AndV128,
- binop( Iop_Mul64Fx2,
- mkexpr(dst_vec), mkexpr(src_vec) ),
- mkV128( imm8_perms[ ((imm8 >> 4) & 3) ] ) ) );
-
- assign( sum_vec, binop( Iop_Add64F0x2,
- binop( Iop_InterleaveHI64x2,
- mkexpr(and_vec), mkexpr(and_vec) ),
- binop( Iop_InterleaveLO64x2,
- mkexpr(and_vec), mkexpr(and_vec) ) ) );
-
- putXMMReg( gregOfRexRM( pfx, modrm ),
- binop( Iop_AndV128,
- binop( Iop_InterleaveLO64x2,
- mkexpr(sum_vec), mkexpr(sum_vec) ),
- mkV128( imm8_perms[ (imm8 & 3) ] ) ) );
-
+ IRTemp res = math_DPPD_128( src_vec, dst_vec, imm8 );
+ putXMMReg( rG, mkexpr(res) );
goto decode_success;
}
break;
@@ -20345,6 +20348,23 @@
}
goto decode_success;
}
+ /* VMOVUPD ymm1, ymm2/m256 = VEX.256.66.0F.WIG 11 /r */
+ if (have66noF2noF3(pfx) && 1==getVexL(pfx)/*256*/) {
+ UChar modrm = getUChar(delta);
+ UInt rG = gregOfRexRM(pfx,modrm);
+ if (epartIsReg(modrm)) {
+ UInt rE = eregOfRexRM(pfx,modrm);
+ putYMMReg( rE, getYMMReg(rG) );
+ DIP("vmovupd %s,%s\n", nameYMMReg(rG), nameYMMReg(rE));
+ delta += 1;
+ } else {
+ addr = disAMode ( &alen, vbi, pfx, delta, dis_buf, 0 );
+ storeLE( mkexpr(addr), getYMMReg(rG) );
+ DIP("vmovupd %s,%s\n", nameYMMReg(rG), dis_buf);
+ delta += alen;
+ }
+ goto decode_success;
+ }
/* VMOVUPS xmm1, xmm2/m128 = VEX.128.0F.WIG 11 /r */
if (haveNo66noF2noF3(pfx) && 0==getVexL(pfx)/*128*/) {
UChar modrm = getUChar(delta);
@@ -20362,19 +20382,19 @@
}
goto decode_success;
}
- /* VMOVUPD ymm1, ymm2/m256 = VEX.256.66.0F.WIG 11 /r */
- if (have66noF2noF3(pfx) && 1==getVexL(pfx)/*256*/) {
+ /* VMOVUPS ymm1, ymm2/m256 = VEX.256.0F.WIG 11 /r */
+ if (haveNo66noF2noF3(pfx) && 1==getVexL(pfx)/*256*/) {
UChar modrm = getUChar(delta);
UInt rG = gregOfRexRM(pfx,modrm);
if (epartIsReg(modrm)) {
UInt rE = eregOfRexRM(pfx,modrm);
putYMMReg( rE, getYMMReg(rG) );
- DIP("vmovupd %s,%s\n", nameYMMReg(rG), nameYMMReg(rE));
+ DIP("vmovups %s,%s\n", nameYMMReg(rG), nameYMMReg(rE));
delta += 1;
} else {
addr = disAMode ( &alen, vbi, pfx, delta, dis_buf, 0 );
storeLE( mkexpr(addr), getYMMReg(rG) );
- DIP("vmovupd %s,%s\n", nameYMMReg(rG), dis_buf);
+ DIP("vmovups %s,%s\n", nameYMMReg(rG), dis_buf);
delta += alen;
}
goto decode_success;
@@ -20909,12 +20929,15 @@
break;
case 0x2E:
+ case 0x2F:
/* VUCOMISD xmm2/m64, xmm1 = VEX.LIG.66.0F.WIG 2E /r */
+ /* VCOMISD xmm2/m64, xmm1 = VEX.LIG.66.0F.WIG 2F /r */
if (have66noF2noF3(pfx)) {
delta = dis_COMISD( vbi, pfx, delta, True/*isAvx*/, opc );
goto decode_success;
}
/* VUCOMISS xmm2/m32, xmm1 = VEX.LIG.0F.WIG 2E /r */
+ /* VCOMISS xmm2/m32, xmm1 = VEX.LIG.0F.WIG 2F /r */
if (haveNo66noF2noF3(pfx)) {
delta = dis_COMISS( vbi, pfx, delta, True/*isAvx*/, opc );
goto decode_success;
@@ -21370,6 +21393,16 @@
}
break;
+ case 0x66:
+ /* VPCMPGTD r/m, rV, r ::: r = rV `>s-by-32s` r/m */
+ /* VPCMPGTD = VEX.NDS.128.66.0F.WIG 66 /r */
+ if (have66noF2noF3(pfx) && 0==getVexL(pfx)/*128*/) {
+ delta = dis_VEX_NDS_128_AnySimdPfx_0F_WIG_simple(
+ uses_vvvv, vbi, pfx, delta, "vpcmpgtd", Iop_CmpGT32Sx4 );
+ goto decode_success;
+ }
+ break;
+
case 0x67:
/* VPACKUSWB r/m, rV, r ::: r = QNarrowBin16Sto8Ux16(rV, r/m) */
/* VPACKUSWB = VEX.NDS.128.66.0F.WIG 67 /r */
@@ -21883,49 +21916,32 @@
UInt rG = gregOfRexRM(pfx, modrm);
UInt rV = getVexNvvvv(pfx);
Int imm8;
- IRTemp new16 = newTemp(Ity_I64);
+ IRTemp new16 = newTemp(Ity_I16);
if ( epartIsReg( modrm ) ) {
imm8 = (Int)(getUChar(delta+1) & 7);
- assign( new16, binop(Iop_And64,
- unop(Iop_32Uto64,
- getIReg32(eregOfRexRM(pfx,modrm))),
- mkU64(0xFFFF)));
+ assign( new16, unop(Iop_32to16,
+ getIReg32(eregOfRexRM(pfx,modrm))) );
delta += 1+1;
DIP( "vpinsrw $%d,%s,%s\n", imm8,
nameIReg32( eregOfRexRM(pfx, modrm) ), nameXMMReg(rG) );
} else {
addr = disAMode( &alen, vbi, pfx, delta, dis_buf, 1 );
imm8 = (Int)(getUChar(delta+alen) & 7);
- assign( new16, unop(Iop_16Uto64, loadLE( Ity_I16, mkexpr(addr) )));
+ assign( new16, loadLE( Ity_I16, mkexpr(addr) ));
delta += alen+1;
DIP( "vpinsrw $%d,%s,%s\n",
imm8, dis_buf, nameXMMReg(rG) );
}
- // Create a V128 value which has the selected word in the
- // specified lane, and zeroes everywhere else.
- IRTemp tmp128 = newTemp(Ity_V128);
- IRTemp halfshift = newTemp(Ity_I64);
- assign(halfshift, binop(Iop_Shl64,
- mkexpr(new16), mkU8(16 * (imm8 & 3))));
- if (imm8 < 4) {
- assign(tmp128, binop(Iop_64HLtoV128, mkU64(0), mkexpr(halfshift)));
- } else {
- assign(tmp128, binop(Iop_64HLtoV128, mkexpr(halfshift), mkU64(0)));
- }
-
- UShort mask = ~(3 << (imm8 * 2));
-
- putYMMRegLoAndZU( rG,
- binop( Iop_OrV128,
- mkexpr(tmp128),
- binop( Iop_AndV128,
- getXMMReg( rV ),
- mkV128(mask) ) ) );
+ IRTemp src_vec = newTemp(Ity_V128);
+ assign(src_vec, getXMMReg( rV ));
+ IRTemp res_vec = math_PINSRW_128( src_vec, new16, imm8 );
+ putYMMRegLoAndZU( rG, mkexpr(res_vec) );
*uses_vvvv = True;
goto decode_success;
}
+ break;
case 0xC5:
/* VPEXTRW imm8, xmm1, reg32 = VEX.128.66.0F.W0 C5 /r ib */
@@ -22483,6 +22499,16 @@
}
break;
+ case 0x21:
+ /* VPMOVSXBD xmm2/m32, xmm1 */
+ /* VPMOVSXBD = VEX.128.66.0F38.WIG 21 /r */
+ if (have66noF2noF3(pfx) && 0==getVexL(pfx)/*128*/) {
+ delta = dis_PMOVxXBD_128( vbi, pfx, delta,
+ True/*isAvx*/, False/*!xIsZ*/ );
+ goto decode_success;
+ }
+ break;
+
case 0x23:
/* VPMOVSXWD xmm2/m64, xmm1 = VEX.128.66.0F38.WIG 23 /r */
if (have66noF2noF3(pfx) && 0==getVexL(pfx)/*128*/) {
@@ -22521,6 +22547,16 @@
}
break;
+ case 0x31:
+ /* VPMOVZXBD xmm2/m32, xmm1 */
+ /* VPMOVZXBD = VEX.128.66.0F38.WIG 31 /r */
+ if (have66noF2noF3(pfx) && 0==getVexL(pfx)/*128*/) {
+ delta = dis_PMOVxXBD_128( vbi, pfx, delta,
+ True/*isAvx*/, True/*xIsZ*/ );
+ goto decode_success;
+ }
+ break;
+
case 0x33:
/* VPMOVZXWD xmm2/m64, xmm1 */
/* VPMOVZXWD = VEX.128.66.0F38.WIG 33 /r */
@@ -23036,7 +23072,6 @@
}
break;
-
case 0x0F:
/* VPALIGNR imm8, xmm3/m128, xmm2, xmm1 */
/* VPALIGNR = VEX.NDS.128.66.0F3A.WIG 0F /r ib */
@@ -23168,39 +23203,39 @@
break;
case 0x20:
- /* VPINSRB r32/m8, xmm2, xmm1 = VEX.NDS.128.66.0F3A.WIG C4 /r ib */
- if (have66noF2noF3(pfx) && 0==getVexL(pfx)/*128*/) {
- UChar modrm = getUChar(delta);
- UInt rG = gregOfRexRM(pfx, modrm);
- UInt rV = getVexNvvvv(pfx);
+ /* VPINSRB r32/m8, xmm2, xmm1 = VEX.NDS.128.66.0F3A.W0 20 /r ib */
+ if (have66noF2noF3(pfx)
+ && 0==getVexL(pfx)/*128*/ && 0==getRexW(pfx)/*W0*/) {
+ UChar modrm = getUChar(delta);
+ UInt rG = gregOfRexRM(pfx, modrm);
+ UInt rV = getVexNvvvv(pfx);
Int imm8;
- IRTemp new8 = newTemp(Ity_I64);
+ IRTemp src_u8 = newTemp(Ity_I8);
if ( epartIsReg( modrm ) ) {
- imm8 = (Int)(getUChar(delta+1) & 0xF);
- assign( new8, binop(Iop_And64,
- unop(Iop_32Uto64,
- getIReg32(eregOfRexRM(pfx,modrm))),
- mkU64(0xFF)));
+ UInt rE = eregOfRexRM(pfx,modrm);
+ imm8 = (Int)(getUChar(delta+1) & 15);
+ assign( src_u8, unop(Iop_32to8, getIReg32( rE )) );
delta += 1+1;
- DIP( "vpinsrb $%d,%s,%s\n", imm8,
- nameIReg32( eregOfRexRM(pfx, modrm) ), nameXMMReg(rG) );
+ DIP( "vpinsrb $%d,%s,%s,%s\n",
+ imm8, nameIReg32(rE), nameXMMReg(rV), nameXMMReg(rG) );
} else {
addr = disAMode( &alen, vbi, pfx, delta, dis_buf, 1 );
- imm8 = (Int)(getUChar(delta+alen) & 0xF);
- assign( new8, unop(Iop_8Uto64, loadLE( Ity_I8, mkexpr(addr) )));
+ imm8 = (Int)(getUChar(delta+alen) & 15);
+ assign( src_u8, loadLE( Ity_I8, mkexpr(addr) ) );
delta += alen+1;
- DIP( "vpinsrb $%d,%s,%s\n",
- imm8, dis_buf, nameXMMReg(rG) );
+ DIP( "vpinsrb $%d,%s,%s,%s\n",
+ imm8, dis_buf, nameXMMReg(rV), nameXMMReg(rG) );
}
IRTemp src_vec = newTemp(Ity_V128);
assign(src_vec, getXMMReg( rV ));
- putYMMRegLoAndZU( rG,
- mkexpr(math_PINSRB_128( src_vec, new8, imm8 )) );
+ IRTemp res_vec = math_PINSRB_128( src_vec, src_u8, imm8 );
+ putYMMRegLoAndZU( rG, mkexpr(res_vec) );
*uses_vvvv = True;
goto decode_success;
}
+ break;
case 0x21:
/* VINSERTPS imm8, xmm3/m32, xmm2, xmm1
@@ -23309,6 +23344,39 @@
}
break;
+ case 0x41:
+ /* VDPPD xmm3/m128,xmm2,xmm1 = VEX.NDS.128.66.0F3A.WIG 41 /r ib */
+ if (have66noF2noF3(pfx) && 0==getVexL(pfx)/*128*/) {
+ UChar modrm = getUChar(delta);
+ UInt rG = gregOfRexRM(pfx, modrm);
+ UInt rV = getVexNvvvv(pfx);
+ IRTemp dst_vec = newTemp(Ity_V128);
+ Int imm8;
+ if (epartIsReg( modrm )) {
+ UInt rE = eregOfRexRM(pfx,modrm);
+ imm8 = (Int)getUChar(delta+1);
+ assign( dst_vec, getXMMReg( rE ) );
+ delta += 1+1;
+ DIP( "vdppd $%d,%s,%s,%s\n",
+ imm8, nameXMMReg(rE), nameXMMReg(rV), nameXMMReg(rG) );
+ } else {
+ addr = disAMode( &alen, vbi, pfx, delta, dis_buf, 1 );
+ imm8 = (Int)getUChar(delta+alen);
+ assign( dst_vec, loadLE( Ity_V128, mkexpr(addr) ) );
+ delta += alen+1;
+ DIP( "vdppd $%d,%s,%s,%s\n",
+ imm8, dis_buf, nameXMMReg(rV), nameXMMReg(rG) );
+ }
+
+ IRTemp src_vec = newTemp(Ity_V128);
+ assign(src_vec, getXMMReg( rV ));
+ IRTemp res_vec = math_DPPD_128( src_vec, dst_vec, imm8 );
+ putYMMRegLoAndZU( rG, mkexpr(res_vec) );
+ *uses_vvvv = True;
+ goto decode_success;
+ }
+ break;
+
case 0x4C:
/* VPBLENDVB xmmG, xmmE/memE, xmmV, xmmIS4
::: xmmG:V128 = PBLEND(xmmE, xmmV, xmmIS4) (RMVR) */
|
