converted SSE/MMX ops to TCG

git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@4441 c046a42c-6fe2-441c-8c8c-71466251a162

target-i386/helper.c

@@ -4462,3 +4462,32 @@ void vmexit(uint64_t exit_code, uint64_t exit_info_1)
 }
 
 #endif
+
+/* MMX/SSE */
+/* XXX: optimize by storing fptt and fptags in the static cpu state */
+void helper_enter_mmx(void)
+{
+    env->fpstt = 0;
+    *(uint32_t *)(env->fptags) = 0;
+    *(uint32_t *)(env->fptags + 4) = 0;
+}
+
+void helper_emms(void)
+{
+    /* set to empty state */
+    *(uint32_t *)(env->fptags) = 0x01010101;
+    *(uint32_t *)(env->fptags + 4) = 0x01010101;
+}
+
+/* XXX: suppress */
+void helper_movq(uint64_t *d, uint64_t *s)
+{
+    *d = *s;
+}
+
+#define SHIFT 0
+#include "ops_sse.h"
+
+#define SHIFT 1
+#include "ops_sse.h"
+

target-i386/helper.h

@@ -2,3 +2,12 @@
 
 void TCG_HELPER_PROTO helper_divl_EAX_T0(target_ulong t0);
 void TCG_HELPER_PROTO helper_idivl_EAX_T0(target_ulong t0);
+void TCG_HELPER_PROTO helper_enter_mmx(void);
+void TCG_HELPER_PROTO helper_emms(void);
+void TCG_HELPER_PROTO helper_movq(uint64_t *d, uint64_t *s);
+
+#define SHIFT 0
+#include "ops_sse_header.h"
+#define SHIFT 1
+#include "ops_sse_header.h"
+

target-i386/op.c

@@ -2144,44 +2144,9 @@ void OPPROTO op_unlock(void)
 }
 
 /* SSE support */
-static inline void memcpy16(void *d, void *s)
+void OPPROTO op_com_dummy(void)
 {
-    ((uint32_t *)d)[0] = ((uint32_t *)s)[0];
-    ((uint32_t *)d)[1] = ((uint32_t *)s)[1];
-    ((uint32_t *)d)[2] = ((uint32_t *)s)[2];
-    ((uint32_t *)d)[3] = ((uint32_t *)s)[3];
-}
-
-void OPPROTO op_movo(void)
-{
-    /* XXX: badly generated code */
-    XMMReg *d, *s;
-    d = (XMMReg *)((char *)env + PARAM1);
-    s = (XMMReg *)((char *)env + PARAM2);
-    memcpy16(d, s);
-}
-
-void OPPROTO op_movq(void)
-{
-    uint64_t *d, *s;
-    d = (uint64_t *)((char *)env + PARAM1);
-    s = (uint64_t *)((char *)env + PARAM2);
-    *d = *s;
-}
-
-void OPPROTO op_movl(void)
-{
-    uint32_t *d, *s;
-    d = (uint32_t *)((char *)env + PARAM1);
-    s = (uint32_t *)((char *)env + PARAM2);
-    *d = *s;
-}
-
-void OPPROTO op_movq_env_0(void)
-{
-    uint64_t *d;
-    d = (uint64_t *)((char *)env + PARAM1);
-    *d = 0;
+    T0 = 0;
 }
 
 void OPPROTO op_fxsave_A0(void)
@@ -2194,27 +2159,6 @@ void OPPROTO op_fxrstor_A0(void)
     helper_fxrstor(A0, PARAM1);
 }
 
-/* XXX: optimize by storing fptt and fptags in the static cpu state */
-void OPPROTO op_enter_mmx(void)
-{
-    env->fpstt = 0;
-    *(uint32_t *)(env->fptags) = 0;
-    *(uint32_t *)(env->fptags + 4) = 0;
-}
-
-void OPPROTO op_emms(void)
-{
-    /* set to empty state */
-    *(uint32_t *)(env->fptags) = 0x01010101;
-    *(uint32_t *)(env->fptags + 4) = 0x01010101;
-}
-
-#define SHIFT 0
-#include "ops_sse.h"
-
-#define SHIFT 1
-#include "ops_sse.h"
-
 /* Secure Virtual Machine ops */
 
 void OPPROTO op_vmrun(void)

target-i386/ops_sse_header.h

+/*
+ *  MMX/3DNow!/SSE/SSE2/SSE3/PNI support
+ *
+ *  Copyright (c) 2005 Fabrice Bellard
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2 of the License, or (at your option) any later version.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ */
+#if SHIFT == 0
+#define Reg MMXReg
+#define SUFFIX _mmx
+#else
+#define Reg XMMReg
+#define SUFFIX _xmm
+#endif
+
+void glue(helper_psrlw, SUFFIX)(Reg *d, Reg *s);
+void glue(helper_psraw, SUFFIX)(Reg *d, Reg *s);
+void glue(helper_psllw, SUFFIX)(Reg *d, Reg *s);
+void glue(helper_psrld, SUFFIX)(Reg *d, Reg *s);
+void glue(helper_psrad, SUFFIX)(Reg *d, Reg *s);
+void glue(helper_pslld, SUFFIX)(Reg *d, Reg *s);
+void glue(helper_psrlq, SUFFIX)(Reg *d, Reg *s);
+void glue(helper_psllq, SUFFIX)(Reg *d, Reg *s);
+
+#if SHIFT == 1
+void glue(helper_psrldq, SUFFIX)(Reg *d, Reg *s);
+void glue(helper_pslldq, SUFFIX)(Reg *d, Reg *s);
+#endif
+
+#define SSE_HELPER_B(name, F)\
+    void glue(name, SUFFIX) (Reg *d, Reg *s);
+
+#define SSE_HELPER_W(name, F)\
+    void glue(name, SUFFIX) (Reg *d, Reg *s);
+
+#define SSE_HELPER_L(name, F)\
+    void glue(name, SUFFIX) (Reg *d, Reg *s);
+
+#define SSE_HELPER_Q(name, F)\
+    void glue(name, SUFFIX) (Reg *d, Reg *s);
+
+SSE_HELPER_B(helper_paddb, FADD);
+SSE_HELPER_W(helper_paddw, FADD);
+SSE_HELPER_L(helper_paddl, FADD);
+SSE_HELPER_Q(helper_paddq, FADD);
+
+SSE_HELPER_B(helper_psubb, FSUB);
+SSE_HELPER_W(helper_psubw, FSUB);
+SSE_HELPER_L(helper_psubl, FSUB);
+SSE_HELPER_Q(helper_psubq, FSUB);
+
+SSE_HELPER_B(helper_paddusb, FADDUB);
+SSE_HELPER_B(helper_paddsb, FADDSB);
+SSE_HELPER_B(helper_psubusb, FSUBUB);
+SSE_HELPER_B(helper_psubsb, FSUBSB);
+
+SSE_HELPER_W(helper_paddusw, FADDUW);
+SSE_HELPER_W(helper_paddsw, FADDSW);
+SSE_HELPER_W(helper_psubusw, FSUBUW);
+SSE_HELPER_W(helper_psubsw, FSUBSW);
+
+SSE_HELPER_B(helper_pminub, FMINUB);
+SSE_HELPER_B(helper_pmaxub, FMAXUB);
+
+SSE_HELPER_W(helper_pminsw, FMINSW);
+SSE_HELPER_W(helper_pmaxsw, FMAXSW);
+
+SSE_HELPER_Q(helper_pand, FAND);
+SSE_HELPER_Q(helper_pandn, FANDN);
+SSE_HELPER_Q(helper_por, FOR);
+SSE_HELPER_Q(helper_pxor, FXOR);
+
+SSE_HELPER_B(helper_pcmpgtb, FCMPGTB);
+SSE_HELPER_W(helper_pcmpgtw, FCMPGTW);
+SSE_HELPER_L(helper_pcmpgtl, FCMPGTL);
+
+SSE_HELPER_B(helper_pcmpeqb, FCMPEQ);
+SSE_HELPER_W(helper_pcmpeqw, FCMPEQ);
+SSE_HELPER_L(helper_pcmpeql, FCMPEQ);
+
+SSE_HELPER_W(helper_pmullw, FMULLW);
+#if SHIFT == 0
+SSE_HELPER_W(helper_pmulhrw, FMULHRW);
+#endif
+SSE_HELPER_W(helper_pmulhuw, FMULHUW);
+SSE_HELPER_W(helper_pmulhw, FMULHW);
+
+SSE_HELPER_B(helper_pavgb, FAVG);
+SSE_HELPER_W(helper_pavgw, FAVG);
+
+void glue(helper_pmuludq, SUFFIX) (Reg *d, Reg *s);
+void glue(helper_pmaddwd, SUFFIX) (Reg *d, Reg *s);
+
+void glue(helper_psadbw, SUFFIX) (Reg *d, Reg *s);
+void glue(helper_maskmov, SUFFIX) (Reg *d, Reg *s);
+void glue(helper_movl_mm_T0, SUFFIX) (Reg *d, uint32_t val);
+#ifdef TARGET_X86_64
+void glue(helper_movq_mm_T0, SUFFIX) (Reg *d, uint64_t val);
+#endif
+
+#if SHIFT == 0
+void glue(helper_pshufw, SUFFIX) (Reg *d, Reg *s, int order);
+#else
+void helper_shufps(Reg *d, Reg *s, int order);
+void helper_shufpd(Reg *d, Reg *s, int order);
+void glue(helper_pshufd, SUFFIX) (Reg *d, Reg *s, int order);
+void glue(helper_pshuflw, SUFFIX) (Reg *d, Reg *s, int order);
+void glue(helper_pshufhw, SUFFIX) (Reg *d, Reg *s, int order);
+#endif
+
+#if SHIFT == 1
+/* FPU ops */
+/* XXX: not accurate */
+
+#define SSE_HELPER_S(name, F)\
+    void helper_ ## name ## ps (Reg *d, Reg *s);        \
+    void helper_ ## name ## ss (Reg *d, Reg *s);        \
+    void helper_ ## name ## pd (Reg *d, Reg *s);        \
+    void helper_ ## name ## sd (Reg *d, Reg *s);
+
+SSE_HELPER_S(add, FPU_ADD);
+SSE_HELPER_S(sub, FPU_SUB);
+SSE_HELPER_S(mul, FPU_MUL);
+SSE_HELPER_S(div, FPU_DIV);
+SSE_HELPER_S(min, FPU_MIN);
+SSE_HELPER_S(max, FPU_MAX);
+SSE_HELPER_S(sqrt, FPU_SQRT);
+
+
+void helper_cvtps2pd(Reg *d, Reg *s);
+void helper_cvtpd2ps(Reg *d, Reg *s);
+void helper_cvtss2sd(Reg *d, Reg *s);
+void helper_cvtsd2ss(Reg *d, Reg *s);
+void helper_cvtdq2ps(Reg *d, Reg *s);
+void helper_cvtdq2pd(Reg *d, Reg *s);
+void helper_cvtpi2ps(XMMReg *d, MMXReg *s);
+void helper_cvtpi2pd(XMMReg *d, MMXReg *s);
+void helper_cvtsi2ss(XMMReg *d, uint32_t val);
+void helper_cvtsi2sd(XMMReg *d, uint32_t val);
+
+#ifdef TARGET_X86_64
+void helper_cvtsq2ss(XMMReg *d, uint64_t val);
+void helper_cvtsq2sd(XMMReg *d, uint64_t val);
+#endif
+
+void helper_cvtps2dq(XMMReg *d, XMMReg *s);
+void helper_cvtpd2dq(XMMReg *d, XMMReg *s);
+void helper_cvtps2pi(MMXReg *d, XMMReg *s);
+void helper_cvtpd2pi(MMXReg *d, XMMReg *s);
+int32_t helper_cvtss2si(XMMReg *s);
+int32_t helper_cvtsd2si(XMMReg *s);
+#ifdef TARGET_X86_64
+int64_t helper_cvtss2sq(XMMReg *s);
+int64_t helper_cvtsd2sq(XMMReg *s);
+#endif
+
+void helper_cvttps2dq(XMMReg *d, XMMReg *s);
+void helper_cvttpd2dq(XMMReg *d, XMMReg *s);
+void helper_cvttps2pi(MMXReg *d, XMMReg *s);
+void helper_cvttpd2pi(MMXReg *d, XMMReg *s);
+int32_t helper_cvttss2si(XMMReg *s);
+int32_t helper_cvttsd2si(XMMReg *s);
+#ifdef TARGET_X86_64
+int64_t helper_cvttss2sq(XMMReg *s);
+int64_t helper_cvttsd2sq(XMMReg *s);
+#endif
+
+void helper_rsqrtps(XMMReg *d, XMMReg *s);
+void helper_rsqrtss(XMMReg *d, XMMReg *s);
+void helper_rcpps(XMMReg *d, XMMReg *s);
+void helper_rcpss(XMMReg *d, XMMReg *s);
+void helper_haddps(XMMReg *d, XMMReg *s);
+void helper_haddpd(XMMReg *d, XMMReg *s);
+void helper_hsubps(XMMReg *d, XMMReg *s);
+void helper_hsubpd(XMMReg *d, XMMReg *s);
+void helper_addsubps(XMMReg *d, XMMReg *s);
+void helper_addsubpd(XMMReg *d, XMMReg *s);
+
+#define SSE_HELPER_CMP(name, F)\
+    void helper_ ## name ## ps (Reg *d, Reg *s);        \
+    void helper_ ## name ## ss (Reg *d, Reg *s);        \
+    void helper_ ## name ## pd (Reg *d, Reg *s);        \
+    void helper_ ## name ## sd (Reg *d, Reg *s);
+
+SSE_HELPER_CMP(cmpeq, FPU_CMPEQ);
+SSE_HELPER_CMP(cmplt, FPU_CMPLT);
+SSE_HELPER_CMP(cmple, FPU_CMPLE);
+SSE_HELPER_CMP(cmpunord, FPU_CMPUNORD);
+SSE_HELPER_CMP(cmpneq, FPU_CMPNEQ);
+SSE_HELPER_CMP(cmpnlt, FPU_CMPNLT);
+SSE_HELPER_CMP(cmpnle, FPU_CMPNLE);
+SSE_HELPER_CMP(cmpord, FPU_CMPORD);
+
+void helper_ucomiss(Reg *d, Reg *s);
+void helper_comiss(Reg *d, Reg *s);
+void helper_ucomisd(Reg *d, Reg *s);
+void helper_comisd(Reg *d, Reg *s);
+uint32_t helper_movmskps(Reg *s);
+uint32_t helper_movmskpd(Reg *s);
+#endif
+
+uint32_t glue(helper_pmovmskb, SUFFIX)(Reg *s);
+void glue(helper_packsswb, SUFFIX) (Reg *d, Reg *s);
+void glue(helper_packuswb, SUFFIX) (Reg *d, Reg *s);
+void glue(helper_packssdw, SUFFIX) (Reg *d, Reg *s);
+#define UNPCK_OP(base_name, base)                               \
+    void glue(helper_punpck ## base_name ## bw, SUFFIX) (Reg *d, Reg *s); \
+    void glue(helper_punpck ## base_name ## wd, SUFFIX) (Reg *d, Reg *s); \
+    void glue(helper_punpck ## base_name ## dq, SUFFIX) (Reg *d, Reg *s);
+
+UNPCK_OP(l, 0);
+UNPCK_OP(h, 1);
+
+#if SHIFT == 1
+void glue(helper_punpcklqdq, SUFFIX) (Reg *d, Reg *s);
+void glue(helper_punpckhqdq, SUFFIX) (Reg *d, Reg *s);
+#endif
+
+/* 3DNow! float ops */
+#if SHIFT == 0
+void helper_pi2fd(MMXReg *d, MMXReg *s);
+void helper_pi2fw(MMXReg *d, MMXReg *s);
+void helper_pf2id(MMXReg *d, MMXReg *s);
+void helper_pf2iw(MMXReg *d, MMXReg *s);
+void helper_pfacc(MMXReg *d, MMXReg *s);
+void helper_pfadd(MMXReg *d, MMXReg *s);
+void helper_pfcmpeq(MMXReg *d, MMXReg *s);
+void helper_pfcmpge(MMXReg *d, MMXReg *s);
+void helper_pfcmpgt(MMXReg *d, MMXReg *s);
+void helper_pfmax(MMXReg *d, MMXReg *s);
+void helper_pfmin(MMXReg *d, MMXReg *s);
+void helper_pfmul(MMXReg *d, MMXReg *s);
+void helper_pfnacc(MMXReg *d, MMXReg *s);
+void helper_pfpnacc(MMXReg *d, MMXReg *s);
+void helper_pfrcp(MMXReg *d, MMXReg *s);
+void helper_pfrsqrt(MMXReg *d, MMXReg *s);
+void helper_pfsub(MMXReg *d, MMXReg *s);
+void helper_pfsubr(MMXReg *d, MMXReg *s);
+void helper_pswapd(MMXReg *d, MMXReg *s);
+#endif
+
+#undef SHIFT
+#undef Reg
+#undef SUFFIX
+
+#undef SSE_HELPER_B
+#undef SSE_HELPER_W
+#undef SSE_HELPER_L
+#undef SSE_HELPER_Q
+#undef SSE_HELPER_S
+#undef SSE_HELPER_CMP
+#undef UNPCK_OP