[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[PATCH v2 19/42] i386: Rewrite blendv helpers
|
From: |
Paul Brook |
|
Subject: |
[PATCH v2 19/42] i386: Rewrite blendv helpers |
|
Date: |
Sun, 24 Apr 2022 23:01:41 +0100 |
Rewrite the blendv helpers so that they can easily be extended to support
the AVX encodings, which make all 4 arguments explicit.
No functional changes to the existing helpers
Signed-off-by: Paul Brook <paul@nowt.org>
---
target/i386/ops_sse.h | 119 +++++++++++++++++++++---------------------
1 file changed, 60 insertions(+), 59 deletions(-)
diff --git a/target/i386/ops_sse.h b/target/i386/ops_sse.h
index 3202c00572..9f388b02b9 100644
--- a/target/i386/ops_sse.h
+++ b/target/i386/ops_sse.h
@@ -2141,73 +2141,74 @@ void glue(helper_palignr, SUFFIX)(CPUX86State *env, Reg
*d, Reg *s,
}
}
-#define XMM0 (env->xmm_regs[0])
+#if SHIFT >= 1
+
+#define BLEND_V128(elem, num, F, b) do { \
+ d->elem(b + 0) = F(v->elem(b + 0), s->elem(b + 0), m->elem(b + 0)); \
+ d->elem(b + 1) = F(v->elem(b + 1), s->elem(b + 1), m->elem(b + 1)); \
+ if (num > 2) { \
+ d->elem(b + 2) = F(v->elem(b + 2), s->elem(b + 2), m->elem(b + 2)); \
+ d->elem(b + 3) = F(v->elem(b + 3), s->elem(b + 3), m->elem(b + 3)); \
+ } \
+ if (num > 4) { \
+ d->elem(b + 4) = F(v->elem(b + 4), s->elem(b + 4), m->elem(b + 4)); \
+ d->elem(b + 5) = F(v->elem(b + 5), s->elem(b + 5), m->elem(b + 5)); \
+ d->elem(b + 6) = F(v->elem(b + 6), s->elem(b + 6), m->elem(b + 6)); \
+ d->elem(b + 7) = F(v->elem(b + 7), s->elem(b + 7), m->elem(b + 7)); \
+ } \
+ if (num > 8) { \
+ d->elem(b + 8) = F(v->elem(b + 8), s->elem(b + 8), m->elem(b + 8)); \
+ d->elem(b + 9) = F(v->elem(b + 9), s->elem(b + 9), m->elem(b + 9)); \
+ d->elem(b + 10) = F(v->elem(b + 10), s->elem(b + 10), m->elem(b +
10));\
+ d->elem(b + 11) = F(v->elem(b + 11), s->elem(b + 11), m->elem(b +
11));\
+ d->elem(b + 12) = F(v->elem(b + 12), s->elem(b + 12), m->elem(b +
12));\
+ d->elem(b + 13) = F(v->elem(b + 13), s->elem(b + 13), m->elem(b +
13));\
+ d->elem(b + 14) = F(v->elem(b + 14), s->elem(b + 14), m->elem(b +
14));\
+ d->elem(b + 15) = F(v->elem(b + 15), s->elem(b + 15), m->elem(b +
15));\
+ } \
+ } while (0)
-#if SHIFT == 1
#define SSE_HELPER_V(name, elem, num, F) \
- void glue(name, SUFFIX)(CPUX86State *env, Reg *d, Reg *s) \
+ void glue(name, SUFFIX)(CPUX86State *env, Reg *d, Reg *s) \
{ \
- d->elem(0) = F(d->elem(0), s->elem(0), XMM0.elem(0)); \
- d->elem(1) = F(d->elem(1), s->elem(1), XMM0.elem(1)); \
- if (num > 2) { \
- d->elem(2) = F(d->elem(2), s->elem(2), XMM0.elem(2)); \
- d->elem(3) = F(d->elem(3), s->elem(3), XMM0.elem(3)); \
- if (num > 4) { \
- d->elem(4) = F(d->elem(4), s->elem(4), XMM0.elem(4)); \
- d->elem(5) = F(d->elem(5), s->elem(5), XMM0.elem(5)); \
- d->elem(6) = F(d->elem(6), s->elem(6), XMM0.elem(6)); \
- d->elem(7) = F(d->elem(7), s->elem(7), XMM0.elem(7)); \
- if (num > 8) { \
- d->elem(8) = F(d->elem(8), s->elem(8), XMM0.elem(8)); \
- d->elem(9) = F(d->elem(9), s->elem(9), XMM0.elem(9)); \
- d->elem(10) = F(d->elem(10), s->elem(10), XMM0.elem(10)); \
- d->elem(11) = F(d->elem(11), s->elem(11), XMM0.elem(11)); \
- d->elem(12) = F(d->elem(12), s->elem(12), XMM0.elem(12)); \
- d->elem(13) = F(d->elem(13), s->elem(13), XMM0.elem(13)); \
- d->elem(14) = F(d->elem(14), s->elem(14), XMM0.elem(14)); \
- d->elem(15) = F(d->elem(15), s->elem(15), XMM0.elem(15)); \
- } \
- } \
- } \
- }
+ Reg *v = d; \
+ Reg *m = &env->xmm_regs[0]; \
+ BLEND_V128(elem, num, F, 0); \
+ YMM_ONLY(BLEND_V128(elem, num, F, num);) \
+ }
+
+#define BLEND_I128(elem, num, F, b) do { \
+ d->elem(b + 0) = F(v->elem(b + 0), s->elem(b + 0), ((imm >> 0) & 1)); \
+ d->elem(b + 1) = F(v->elem(b + 1), s->elem(b + 1), ((imm >> 1) & 1)); \
+ if (num > 2) { \
+ d->elem(b + 2) = F(v->elem(b + 2), s->elem(b + 2), ((imm >> 2) & 1)); \
+ d->elem(b + 3) = F(v->elem(b + 3), s->elem(b + 3), ((imm >> 3) & 1)); \
+ } \
+ if (num > 4) { \
+ d->elem(b + 4) = F(v->elem(b + 4), s->elem(b + 4), ((imm >> 4) & 1)); \
+ d->elem(b + 5) = F(v->elem(b + 5), s->elem(b + 5), ((imm >> 5) & 1)); \
+ d->elem(b + 6) = F(v->elem(b + 6), s->elem(b + 6), ((imm >> 6) & 1)); \
+ d->elem(b + 7) = F(v->elem(b + 7), s->elem(b + 7), ((imm >> 7) & 1)); \
+ } \
+ } while (0)
#define SSE_HELPER_I(name, elem, num, F) \
- void glue(name, SUFFIX)(CPUX86State *env, Reg *d, Reg *s, uint32_t imm) \
+ void glue(name, SUFFIX)(CPUX86State *env, Reg *d, Reg *s, \
+ uint32_t imm) \
{ \
- d->elem(0) = F(d->elem(0), s->elem(0), ((imm >> 0) & 1)); \
- d->elem(1) = F(d->elem(1), s->elem(1), ((imm >> 1) & 1)); \
- if (num > 2) { \
- d->elem(2) = F(d->elem(2), s->elem(2), ((imm >> 2) & 1)); \
- d->elem(3) = F(d->elem(3), s->elem(3), ((imm >> 3) & 1)); \
- if (num > 4) { \
- d->elem(4) = F(d->elem(4), s->elem(4), ((imm >> 4) & 1)); \
- d->elem(5) = F(d->elem(5), s->elem(5), ((imm >> 5) & 1)); \
- d->elem(6) = F(d->elem(6), s->elem(6), ((imm >> 6) & 1)); \
- d->elem(7) = F(d->elem(7), s->elem(7), ((imm >> 7) & 1)); \
- if (num > 8) { \
- d->elem(8) = F(d->elem(8), s->elem(8), ((imm >> 8) & 1)); \
- d->elem(9) = F(d->elem(9), s->elem(9), ((imm >> 9) & 1)); \
- d->elem(10) = F(d->elem(10), s->elem(10), \
- ((imm >> 10) & 1)); \
- d->elem(11) = F(d->elem(11), s->elem(11), \
- ((imm >> 11) & 1)); \
- d->elem(12) = F(d->elem(12), s->elem(12), \
- ((imm >> 12) & 1)); \
- d->elem(13) = F(d->elem(13), s->elem(13), \
- ((imm >> 13) & 1)); \
- d->elem(14) = F(d->elem(14), s->elem(14), \
- ((imm >> 14) & 1)); \
- d->elem(15) = F(d->elem(15), s->elem(15), \
- ((imm >> 15) & 1)); \
- } \
- } \
- } \
+ Reg *v = d; \
+ BLEND_I128(elem, num, F, 0); \
+ YMM_ONLY( \
+ if (num < 8) \
+ imm >>= num; \
+ BLEND_I128(elem, num, F, num); \
+ ) \
}
/* SSE4.1 op helpers */
-#define FBLENDVB(d, s, m) ((m & 0x80) ? s : d)
-#define FBLENDVPS(d, s, m) ((m & 0x80000000) ? s : d)
-#define FBLENDVPD(d, s, m) ((m & 0x8000000000000000LL) ? s : d)
+#define FBLENDVB(v, s, m) ((m & 0x80) ? s : v)
+#define FBLENDVPS(v, s, m) ((m & 0x80000000) ? s : v)
+#define FBLENDVPD(v, s, m) ((m & 0x8000000000000000LL) ? s : v)
SSE_HELPER_V(helper_pblendvb, B, 16, FBLENDVB)
SSE_HELPER_V(helper_blendvps, L, 4, FBLENDVPS)
SSE_HELPER_V(helper_blendvpd, Q, 2, FBLENDVPD)
@@ -2521,7 +2522,7 @@ void glue(helper_roundsd, SUFFIX)(CPUX86State *env, Reg
*d, Reg *s,
}
#endif
-#define FBLENDP(d, s, m) (m ? s : d)
+#define FBLENDP(v, s, m) (m ? s : v)
SSE_HELPER_I(helper_blendps, L, 4, FBLENDP)
SSE_HELPER_I(helper_blendpd, Q, 2, FBLENDP)
SSE_HELPER_I(helper_pblendw, W, 8, FBLENDP)
--
2.36.0
- [PATCH v2 34/42] i386: Implement VGATHER, (continued)
- [PATCH v2 34/42] i386: Implement VGATHER, Paul Brook, 2022/04/24
- [PATCH v2 18/42] i386: Misc AVX helper prep, Paul Brook, 2022/04/24
- [PATCH v2 23/42] i386: AVX comparison helpers, Paul Brook, 2022/04/24
- [PATCH v2 12/42] i386: Misc integer AVX helper prep, Paul Brook, 2022/04/24
- [PATCH v2 21/42] i386: AVX+AES helpers, Paul Brook, 2022/04/24
- [PATCH v2 42/42] i386: Add sha512-avx test, Paul Brook, 2022/04/24
- [PATCH v2 31/42] i386: Implement AVX variable shifts, Paul Brook, 2022/04/24
- [PATCH v2 26/42] i386: Utility function for 128 bit AVX, Paul Brook, 2022/04/24
- [PATCH v2 36/42] i386: Implement VINSERT128/VEXTRACT128, Paul Brook, 2022/04/24
- [PATCH v2 27/42] i386: Translate 256 bit AVX instructions, Paul Brook, 2022/04/24
- [PATCH v2 19/42] i386: Rewrite blendv helpers,
Paul Brook <=
- [PATCH v2 35/42] i386: Implement VPERM, Paul Brook, 2022/04/24
- [PATCH v2 32/42] i386: Implement VTEST, Paul Brook, 2022/04/24
- [PATCH v2 15/42] i386: Floating point atithmetic helper AVX prep, Paul Brook, 2022/04/24