[Qemu-devel] [RFC PATCH v1 14/22] target/i386: reimplement (V)PADDS(B, W)

[Jan Bobek <jan.bobek@gmail.com>]

Use the gvec infrastructure to achieve the desired functionality.

Signed-off-by: Jan Bobek <jan.bobek@gmail.com>
---
 target/i386/ops_sse.h        |  4 ----
 target/i386/ops_sse_header.h |  2 --
 target/i386/translate.c      | 19 +++++++++++++++++--
 3 files changed, 17 insertions(+), 8 deletions(-)

diff --git a/target/i386/ops_sse.h b/target/i386/ops_sse.h
index 59935a65be..8829dcb781 100644
--- a/target/i386/ops_sse.h
+++ b/target/i386/ops_sse.h
@@ -339,8 +339,6 @@ static inline int satsw(int x)
 
 #define FADDUB(a, b) satub((a) + (b))
 #define FADDUW(a, b) satuw((a) + (b))
-#define FADDSB(a, b) satsb((int8_t)(a) + (int8_t)(b))
-#define FADDSW(a, b) satsw((int16_t)(a) + (int16_t)(b))
 
 #define FSUBUB(a, b) satub((a) - (b))
 #define FSUBUW(a, b) satuw((a) - (b))
@@ -365,12 +363,10 @@ static inline int satsw(int x)
 #endif
 
 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)
 
diff --git a/target/i386/ops_sse_header.h b/target/i386/ops_sse_header.h
index bcdbac99a0..78203e80a5 100644
--- a/target/i386/ops_sse_header.h
+++ b/target/i386/ops_sse_header.h
@@ -61,12 +61,10 @@ DEF_HELPER_3(glue(pslldq, SUFFIX), void, env, Reg, Reg)
     DEF_HELPER_3(glue(name, SUFFIX), void, env, Reg, Reg)
 
 SSE_HELPER_B(paddusb, FADDUB)
-SSE_HELPER_B(paddsb, FADDSB)
 SSE_HELPER_B(psubusb, FSUBUB)
 SSE_HELPER_B(psubsb, FSUBSB)
 
 SSE_HELPER_W(paddusw, FADDUW)
-SSE_HELPER_W(paddsw, FADDSW)
 SSE_HELPER_W(psubusw, FSUBUW)
 SSE_HELPER_W(psubsw, FSUBSW)
 
diff --git a/target/i386/translate.c b/target/i386/translate.c
index 6f4dfd06a1..5ea5014d99 100644
--- a/target/i386/translate.c
+++ b/target/i386/translate.c
@@ -2842,8 +2842,8 @@ static const SSEFunc_0_epp sse_op_table1[256][4] = {
     [0xe9] = MMX_OP2(psubsw),
     [0xea] = MMX_OP2(pminsw),
     [0xeb] = { SSE_TOMBSTONE, SSE_TOMBSTONE },
-    [0xec] = MMX_OP2(paddsb),
-    [0xed] = MMX_OP2(paddsw),
+    [0xec] = { SSE_TOMBSTONE, SSE_TOMBSTONE },
+    [0xed] = { SSE_TOMBSTONE, SSE_TOMBSTONE },
     [0xee] = MMX_OP2(pmaxsw),
     [0xef] = { SSE_TOMBSTONE, SSE_TOMBSTONE },
     [0xf0] = { NULL, NULL, NULL, SSE_SPECIAL }, /* lddqu */
@@ -3171,6 +3171,11 @@ static inline void gen_gvec_ld_modrm_3(CPUX86State *env, DisasContext *s,
 #define gen_vpadd_xmm(env, s, modrm, vece) gen_gvec_ld_modrm_vxmm((env), (s), (modrm), (vece), tcg_gen_gvec_add, 0123)
 #define gen_vpadd_ymm(env, s, modrm, vece) gen_gvec_ld_modrm_vymm((env), (s), (modrm), (vece), tcg_gen_gvec_add, 0123)
 
+#define gen_padds_mm(env, s, modrm, vece)   gen_gvec_ld_modrm_mm  ((env), (s), (modrm), (vece), tcg_gen_gvec_ssadd, 0112)
+#define gen_padds_xmm(env, s, modrm, vece)  gen_gvec_ld_modrm_xmm ((env), (s), (modrm), (vece), tcg_gen_gvec_ssadd, 0112)
+#define gen_vpadds_xmm(env, s, modrm, vece) gen_gvec_ld_modrm_vxmm((env), (s), (modrm), (vece), tcg_gen_gvec_ssadd, 0123)
+#define gen_vpadds_ymm(env, s, modrm, vece) gen_gvec_ld_modrm_vymm((env), (s), (modrm), (vece), tcg_gen_gvec_ssadd, 0123)
+
 #define gen_psub_mm(env, s, modrm, vece)   gen_gvec_ld_modrm_mm  ((env), (s), (modrm), (vece), tcg_gen_gvec_sub, 0112)
 #define gen_psub_xmm(env, s, modrm, vece)  gen_gvec_ld_modrm_xmm ((env), (s), (modrm), (vece), tcg_gen_gvec_sub, 0112)
 #define gen_vpsub_xmm(env, s, modrm, vece) gen_gvec_ld_modrm_vxmm((env), (s), (modrm), (vece), tcg_gen_gvec_sub, 0123)
@@ -3331,6 +3336,16 @@ static void gen_sse(CPUX86State *env, DisasContext *s, int b)
     case 0xd4 | M_0F | P_66 | VEX_128: gen_vpadd_xmm(env, s, modrm, MO_64); return;
     case 0xd4 | M_0F | P_66 | VEX_256: gen_vpadd_ymm(env, s, modrm, MO_64); return;
 
+    case 0xec | M_0F:                  gen_padds_mm(env, s, modrm, MO_8); return;
+    case 0xec | M_0F | P_66:           gen_padds_xmm(env, s, modrm, MO_8); return;
+    case 0xec | M_0F | P_66 | VEX_128: gen_vpadds_xmm(env, s, modrm, MO_8); return;
+    case 0xec | M_0F | P_66 | VEX_256: gen_vpadds_ymm(env, s, modrm, MO_8); return;
+
+    case 0xed | M_0F:                  gen_padds_mm(env, s, modrm, MO_16); return;
+    case 0xed | M_0F | P_66:           gen_padds_xmm(env, s, modrm, MO_16); return;
+    case 0xed | M_0F | P_66 | VEX_128: gen_vpadds_xmm(env, s, modrm, MO_16); return;
+    case 0xed | M_0F | P_66 | VEX_256: gen_vpadds_ymm(env, s, modrm, MO_16); return;
+
     case 0xf8 | M_0F:                  gen_psub_mm(env, s, modrm, MO_8); return;
     case 0xf8 | M_0F | P_66:           gen_psub_xmm(env, s, modrm, MO_8); return;
     case 0xf8 | M_0F | P_66 | VEX_128: gen_vpsub_xmm(env, s, modrm, MO_8); return;
-- 
2.20.1