From: Richard Henderson <richard.henderson@linaro.org>
To: qemu-devel@nongnu.org
Cc: mark.cave-ayland@ilande.co.uk, qemu-ppc@nongnu.org,
david@gibson.dropbear.id.au
Subject: [Qemu-devel] [PATCH 07/34] tcg: Add opcodes for vector minmax arithmetic
Date: Mon, 17 Dec 2018 22:38:44 -0800 [thread overview]
Message-ID: <20181218063911.2112-8-richard.henderson@linaro.org> (raw)
In-Reply-To: <20181218063911.2112-1-richard.henderson@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
accel/tcg/tcg-runtime.h | 20 ++++
tcg/aarch64/tcg-target.h | 1 +
tcg/i386/tcg-target.h | 1 +
tcg/tcg-op-gvec.h | 10 ++
tcg/tcg-op.h | 4 +
tcg/tcg-opc.h | 4 +
tcg/tcg.h | 1 +
accel/tcg/tcg-runtime-gvec.c | 224 +++++++++++++++++++++++++++++++++++
tcg/tcg-op-gvec.c | 108 +++++++++++++++++
tcg/tcg-op-vec.c | 20 ++++
tcg/tcg.c | 5 +
11 files changed, 398 insertions(+)
diff --git a/accel/tcg/tcg-runtime.h b/accel/tcg/tcg-runtime.h
index 835ddfebb2..dfe325625c 100644
--- a/accel/tcg/tcg-runtime.h
+++ b/accel/tcg/tcg-runtime.h
@@ -200,6 +200,26 @@ DEF_HELPER_FLAGS_4(gvec_ussub16, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32)
DEF_HELPER_FLAGS_4(gvec_ussub32, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32)
DEF_HELPER_FLAGS_4(gvec_ussub64, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32)
+DEF_HELPER_FLAGS_4(gvec_smin8, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32)
+DEF_HELPER_FLAGS_4(gvec_smin16, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32)
+DEF_HELPER_FLAGS_4(gvec_smin32, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32)
+DEF_HELPER_FLAGS_4(gvec_smin64, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32)
+
+DEF_HELPER_FLAGS_4(gvec_smax8, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32)
+DEF_HELPER_FLAGS_4(gvec_smax16, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32)
+DEF_HELPER_FLAGS_4(gvec_smax32, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32)
+DEF_HELPER_FLAGS_4(gvec_smax64, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32)
+
+DEF_HELPER_FLAGS_4(gvec_umin8, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32)
+DEF_HELPER_FLAGS_4(gvec_umin16, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32)
+DEF_HELPER_FLAGS_4(gvec_umin32, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32)
+DEF_HELPER_FLAGS_4(gvec_umin64, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32)
+
+DEF_HELPER_FLAGS_4(gvec_umax8, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32)
+DEF_HELPER_FLAGS_4(gvec_umax16, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32)
+DEF_HELPER_FLAGS_4(gvec_umax32, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32)
+DEF_HELPER_FLAGS_4(gvec_umax64, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32)
+
DEF_HELPER_FLAGS_3(gvec_neg8, TCG_CALL_NO_RWG, void, ptr, ptr, i32)
DEF_HELPER_FLAGS_3(gvec_neg16, TCG_CALL_NO_RWG, void, ptr, ptr, i32)
DEF_HELPER_FLAGS_3(gvec_neg32, TCG_CALL_NO_RWG, void, ptr, ptr, i32)
diff --git a/tcg/aarch64/tcg-target.h b/tcg/aarch64/tcg-target.h
index 98556bcf22..545a6eec75 100644
--- a/tcg/aarch64/tcg-target.h
+++ b/tcg/aarch64/tcg-target.h
@@ -136,6 +136,7 @@ typedef enum {
#define TCG_TARGET_HAS_cmp_vec 1
#define TCG_TARGET_HAS_mul_vec 1
#define TCG_TARGET_HAS_sat_vec 0
+#define TCG_TARGET_HAS_minmax_vec 0
#define TCG_TARGET_DEFAULT_MO (0)
#define TCG_TARGET_HAS_MEMORY_BSWAP 1
diff --git a/tcg/i386/tcg-target.h b/tcg/i386/tcg-target.h
index f50234d97b..efbd5a6fc9 100644
--- a/tcg/i386/tcg-target.h
+++ b/tcg/i386/tcg-target.h
@@ -186,6 +186,7 @@ extern bool have_avx2;
#define TCG_TARGET_HAS_cmp_vec 1
#define TCG_TARGET_HAS_mul_vec 1
#define TCG_TARGET_HAS_sat_vec 1
+#define TCG_TARGET_HAS_minmax_vec 0
#define TCG_TARGET_deposit_i32_valid(ofs, len) \
(((ofs) == 0 && (len) == 8) || ((ofs) == 8 && (len) == 8) || \
diff --git a/tcg/tcg-op-gvec.h b/tcg/tcg-op-gvec.h
index 2cb447112e..4734eef7de 100644
--- a/tcg/tcg-op-gvec.h
+++ b/tcg/tcg-op-gvec.h
@@ -234,6 +234,16 @@ void tcg_gen_gvec_usadd(unsigned vece, uint32_t dofs, uint32_t aofs,
void tcg_gen_gvec_ussub(unsigned vece, uint32_t dofs, uint32_t aofs,
uint32_t bofs, uint32_t oprsz, uint32_t maxsz);
+/* Min/max. */
+void tcg_gen_gvec_smin(unsigned vece, uint32_t dofs, uint32_t aofs,
+ uint32_t bofs, uint32_t oprsz, uint32_t maxsz);
+void tcg_gen_gvec_umin(unsigned vece, uint32_t dofs, uint32_t aofs,
+ uint32_t bofs, uint32_t oprsz, uint32_t maxsz);
+void tcg_gen_gvec_smax(unsigned vece, uint32_t dofs, uint32_t aofs,
+ uint32_t bofs, uint32_t oprsz, uint32_t maxsz);
+void tcg_gen_gvec_umax(unsigned vece, uint32_t dofs, uint32_t aofs,
+ uint32_t bofs, uint32_t oprsz, uint32_t maxsz);
+
void tcg_gen_gvec_and(unsigned vece, uint32_t dofs, uint32_t aofs,
uint32_t bofs, uint32_t oprsz, uint32_t maxsz);
void tcg_gen_gvec_or(unsigned vece, uint32_t dofs, uint32_t aofs,
diff --git a/tcg/tcg-op.h b/tcg/tcg-op.h
index 90b3193bf3..042c45e807 100644
--- a/tcg/tcg-op.h
+++ b/tcg/tcg-op.h
@@ -970,6 +970,10 @@ void tcg_gen_ssadd_vec(unsigned vece, TCGv_vec r, TCGv_vec a, TCGv_vec b);
void tcg_gen_usadd_vec(unsigned vece, TCGv_vec r, TCGv_vec a, TCGv_vec b);
void tcg_gen_sssub_vec(unsigned vece, TCGv_vec r, TCGv_vec a, TCGv_vec b);
void tcg_gen_ussub_vec(unsigned vece, TCGv_vec r, TCGv_vec a, TCGv_vec b);
+void tcg_gen_smin_vec(unsigned vece, TCGv_vec r, TCGv_vec a, TCGv_vec b);
+void tcg_gen_umin_vec(unsigned vece, TCGv_vec r, TCGv_vec a, TCGv_vec b);
+void tcg_gen_smax_vec(unsigned vece, TCGv_vec r, TCGv_vec a, TCGv_vec b);
+void tcg_gen_umax_vec(unsigned vece, TCGv_vec r, TCGv_vec a, TCGv_vec b);
void tcg_gen_shli_vec(unsigned vece, TCGv_vec r, TCGv_vec a, int64_t i);
void tcg_gen_shri_vec(unsigned vece, TCGv_vec r, TCGv_vec a, int64_t i);
diff --git a/tcg/tcg-opc.h b/tcg/tcg-opc.h
index 94691e849b..691eddebdf 100644
--- a/tcg/tcg-opc.h
+++ b/tcg/tcg-opc.h
@@ -225,6 +225,10 @@ DEF(ssadd_vec, 1, 2, 0, IMPLVEC | IMPL(TCG_TARGET_HAS_sat_vec))
DEF(usadd_vec, 1, 2, 0, IMPLVEC | IMPL(TCG_TARGET_HAS_sat_vec))
DEF(sssub_vec, 1, 2, 0, IMPLVEC | IMPL(TCG_TARGET_HAS_sat_vec))
DEF(ussub_vec, 1, 2, 0, IMPLVEC | IMPL(TCG_TARGET_HAS_sat_vec))
+DEF(smin_vec, 1, 2, 0, IMPLVEC | IMPL(TCG_TARGET_HAS_minmax_vec))
+DEF(umin_vec, 1, 2, 0, IMPLVEC | IMPL(TCG_TARGET_HAS_minmax_vec))
+DEF(smax_vec, 1, 2, 0, IMPLVEC | IMPL(TCG_TARGET_HAS_minmax_vec))
+DEF(umax_vec, 1, 2, 0, IMPLVEC | IMPL(TCG_TARGET_HAS_minmax_vec))
DEF(and_vec, 1, 2, 0, IMPLVEC)
DEF(or_vec, 1, 2, 0, IMPLVEC)
diff --git a/tcg/tcg.h b/tcg/tcg.h
index c90f65a387..b5bec3abf8 100644
--- a/tcg/tcg.h
+++ b/tcg/tcg.h
@@ -184,6 +184,7 @@ typedef uint64_t TCGRegSet;
#define TCG_TARGET_HAS_shv_vec 0
#define TCG_TARGET_HAS_mul_vec 0
#define TCG_TARGET_HAS_sat_vec 0
+#define TCG_TARGET_HAS_minmax_vec 0
#else
#define TCG_TARGET_MAYBE_vec 1
#endif
diff --git a/accel/tcg/tcg-runtime-gvec.c b/accel/tcg/tcg-runtime-gvec.c
index d1802467d5..9358749741 100644
--- a/accel/tcg/tcg-runtime-gvec.c
+++ b/accel/tcg/tcg-runtime-gvec.c
@@ -1028,3 +1028,227 @@ void HELPER(gvec_ussub64)(void *d, void *a, void *b, uint32_t desc)
}
clear_high(d, oprsz, desc);
}
+
+void HELPER(gvec_smin8)(void *d, void *a, void *b, uint32_t desc)
+{
+ intptr_t oprsz = simd_oprsz(desc);
+ intptr_t i;
+
+ for (i = 0; i < oprsz; i += sizeof(int8_t)) {
+ int8_t aa = *(int8_t *)(a + i);
+ int8_t bb = *(int8_t *)(b + i);
+ int8_t dd = aa < bb ? aa : bb;
+ *(int8_t *)(d + i) = dd;
+ }
+ clear_high(d, oprsz, desc);
+}
+
+void HELPER(gvec_smin16)(void *d, void *a, void *b, uint32_t desc)
+{
+ intptr_t oprsz = simd_oprsz(desc);
+ intptr_t i;
+
+ for (i = 0; i < oprsz; i += sizeof(int16_t)) {
+ int16_t aa = *(int16_t *)(a + i);
+ int16_t bb = *(int16_t *)(b + i);
+ int16_t dd = aa < bb ? aa : bb;
+ *(int16_t *)(d + i) = dd;
+ }
+ clear_high(d, oprsz, desc);
+}
+
+void HELPER(gvec_smin32)(void *d, void *a, void *b, uint32_t desc)
+{
+ intptr_t oprsz = simd_oprsz(desc);
+ intptr_t i;
+
+ for (i = 0; i < oprsz; i += sizeof(int32_t)) {
+ int32_t aa = *(int32_t *)(a + i);
+ int32_t bb = *(int32_t *)(b + i);
+ int32_t dd = aa < bb ? aa : bb;
+ *(int32_t *)(d + i) = dd;
+ }
+ clear_high(d, oprsz, desc);
+}
+
+void HELPER(gvec_smin64)(void *d, void *a, void *b, uint32_t desc)
+{
+ intptr_t oprsz = simd_oprsz(desc);
+ intptr_t i;
+
+ for (i = 0; i < oprsz; i += sizeof(int64_t)) {
+ int64_t aa = *(int64_t *)(a + i);
+ int64_t bb = *(int64_t *)(b + i);
+ int64_t dd = aa < bb ? aa : bb;
+ *(int64_t *)(d + i) = dd;
+ }
+ clear_high(d, oprsz, desc);
+}
+
+void HELPER(gvec_smax8)(void *d, void *a, void *b, uint32_t desc)
+{
+ intptr_t oprsz = simd_oprsz(desc);
+ intptr_t i;
+
+ for (i = 0; i < oprsz; i += sizeof(int8_t)) {
+ int8_t aa = *(int8_t *)(a + i);
+ int8_t bb = *(int8_t *)(b + i);
+ int8_t dd = aa > bb ? aa : bb;
+ *(int8_t *)(d + i) = dd;
+ }
+ clear_high(d, oprsz, desc);
+}
+
+void HELPER(gvec_smax16)(void *d, void *a, void *b, uint32_t desc)
+{
+ intptr_t oprsz = simd_oprsz(desc);
+ intptr_t i;
+
+ for (i = 0; i < oprsz; i += sizeof(int16_t)) {
+ int16_t aa = *(int16_t *)(a + i);
+ int16_t bb = *(int16_t *)(b + i);
+ int16_t dd = aa > bb ? aa : bb;
+ *(int16_t *)(d + i) = dd;
+ }
+ clear_high(d, oprsz, desc);
+}
+
+void HELPER(gvec_smax32)(void *d, void *a, void *b, uint32_t desc)
+{
+ intptr_t oprsz = simd_oprsz(desc);
+ intptr_t i;
+
+ for (i = 0; i < oprsz; i += sizeof(int32_t)) {
+ int32_t aa = *(int32_t *)(a + i);
+ int32_t bb = *(int32_t *)(b + i);
+ int32_t dd = aa > bb ? aa : bb;
+ *(int32_t *)(d + i) = dd;
+ }
+ clear_high(d, oprsz, desc);
+}
+
+void HELPER(gvec_smax64)(void *d, void *a, void *b, uint32_t desc)
+{
+ intptr_t oprsz = simd_oprsz(desc);
+ intptr_t i;
+
+ for (i = 0; i < oprsz; i += sizeof(int64_t)) {
+ int64_t aa = *(int64_t *)(a + i);
+ int64_t bb = *(int64_t *)(b + i);
+ int64_t dd = aa > bb ? aa : bb;
+ *(int64_t *)(d + i) = dd;
+ }
+ clear_high(d, oprsz, desc);
+}
+
+void HELPER(gvec_umin8)(void *d, void *a, void *b, uint32_t desc)
+{
+ intptr_t oprsz = simd_oprsz(desc);
+ intptr_t i;
+
+ for (i = 0; i < oprsz; i += sizeof(uint8_t)) {
+ uint8_t aa = *(uint8_t *)(a + i);
+ uint8_t bb = *(uint8_t *)(b + i);
+ uint8_t dd = aa < bb ? aa : bb;
+ *(uint8_t *)(d + i) = dd;
+ }
+ clear_high(d, oprsz, desc);
+}
+
+void HELPER(gvec_umin16)(void *d, void *a, void *b, uint32_t desc)
+{
+ intptr_t oprsz = simd_oprsz(desc);
+ intptr_t i;
+
+ for (i = 0; i < oprsz; i += sizeof(uint16_t)) {
+ uint16_t aa = *(uint16_t *)(a + i);
+ uint16_t bb = *(uint16_t *)(b + i);
+ uint16_t dd = aa < bb ? aa : bb;
+ *(uint16_t *)(d + i) = dd;
+ }
+ clear_high(d, oprsz, desc);
+}
+
+void HELPER(gvec_umin32)(void *d, void *a, void *b, uint32_t desc)
+{
+ intptr_t oprsz = simd_oprsz(desc);
+ intptr_t i;
+
+ for (i = 0; i < oprsz; i += sizeof(uint32_t)) {
+ uint32_t aa = *(uint32_t *)(a + i);
+ uint32_t bb = *(uint32_t *)(b + i);
+ uint32_t dd = aa < bb ? aa : bb;
+ *(uint32_t *)(d + i) = dd;
+ }
+ clear_high(d, oprsz, desc);
+}
+
+void HELPER(gvec_umin64)(void *d, void *a, void *b, uint32_t desc)
+{
+ intptr_t oprsz = simd_oprsz(desc);
+ intptr_t i;
+
+ for (i = 0; i < oprsz; i += sizeof(uint64_t)) {
+ uint64_t aa = *(uint64_t *)(a + i);
+ uint64_t bb = *(uint64_t *)(b + i);
+ uint64_t dd = aa < bb ? aa : bb;
+ *(uint64_t *)(d + i) = dd;
+ }
+ clear_high(d, oprsz, desc);
+}
+
+void HELPER(gvec_umax8)(void *d, void *a, void *b, uint32_t desc)
+{
+ intptr_t oprsz = simd_oprsz(desc);
+ intptr_t i;
+
+ for (i = 0; i < oprsz; i += sizeof(uint8_t)) {
+ uint8_t aa = *(uint8_t *)(a + i);
+ uint8_t bb = *(uint8_t *)(b + i);
+ uint8_t dd = aa > bb ? aa : bb;
+ *(uint8_t *)(d + i) = dd;
+ }
+ clear_high(d, oprsz, desc);
+}
+
+void HELPER(gvec_umax16)(void *d, void *a, void *b, uint32_t desc)
+{
+ intptr_t oprsz = simd_oprsz(desc);
+ intptr_t i;
+
+ for (i = 0; i < oprsz; i += sizeof(uint16_t)) {
+ uint16_t aa = *(uint16_t *)(a + i);
+ uint16_t bb = *(uint16_t *)(b + i);
+ uint16_t dd = aa > bb ? aa : bb;
+ *(uint16_t *)(d + i) = dd;
+ }
+ clear_high(d, oprsz, desc);
+}
+
+void HELPER(gvec_umax32)(void *d, void *a, void *b, uint32_t desc)
+{
+ intptr_t oprsz = simd_oprsz(desc);
+ intptr_t i;
+
+ for (i = 0; i < oprsz; i += sizeof(uint32_t)) {
+ uint32_t aa = *(uint32_t *)(a + i);
+ uint32_t bb = *(uint32_t *)(b + i);
+ uint32_t dd = aa > bb ? aa : bb;
+ *(uint32_t *)(d + i) = dd;
+ }
+ clear_high(d, oprsz, desc);
+}
+
+void HELPER(gvec_umax64)(void *d, void *a, void *b, uint32_t desc)
+{
+ intptr_t oprsz = simd_oprsz(desc);
+ intptr_t i;
+
+ for (i = 0; i < oprsz; i += sizeof(uint64_t)) {
+ uint64_t aa = *(uint64_t *)(a + i);
+ uint64_t bb = *(uint64_t *)(b + i);
+ uint64_t dd = aa > bb ? aa : bb;
+ *(uint64_t *)(d + i) = dd;
+ }
+ clear_high(d, oprsz, desc);
+}
diff --git a/tcg/tcg-op-gvec.c b/tcg/tcg-op-gvec.c
index 0a33f51065..3ee44fcb75 100644
--- a/tcg/tcg-op-gvec.c
+++ b/tcg/tcg-op-gvec.c
@@ -1810,6 +1810,114 @@ void tcg_gen_gvec_ussub(unsigned vece, uint32_t dofs, uint32_t aofs,
tcg_gen_gvec_3(dofs, aofs, bofs, oprsz, maxsz, &g[vece]);
}
+void tcg_gen_gvec_smin(unsigned vece, uint32_t dofs, uint32_t aofs,
+ uint32_t bofs, uint32_t oprsz, uint32_t maxsz)
+{
+ static const GVecGen3 g[4] = {
+ { .fniv = tcg_gen_smin_vec,
+ .fno = gen_helper_gvec_smin8,
+ .opc = INDEX_op_smin_vec,
+ .vece = MO_8 },
+ { .fniv = tcg_gen_smin_vec,
+ .fno = gen_helper_gvec_smin16,
+ .opc = INDEX_op_smin_vec,
+ .vece = MO_16 },
+ { .fni4 = tcg_gen_smin_i32,
+ .fniv = tcg_gen_smin_vec,
+ .fno = gen_helper_gvec_smin32,
+ .opc = INDEX_op_smin_vec,
+ .vece = MO_32 },
+ { .fni8 = tcg_gen_smin_i64,
+ .fniv = tcg_gen_smin_vec,
+ .fno = gen_helper_gvec_smin64,
+ .opc = INDEX_op_smin_vec,
+ .vece = MO_64 }
+ };
+ tcg_debug_assert(vece <= MO_64);
+ tcg_gen_gvec_3(dofs, aofs, bofs, oprsz, maxsz, &g[vece]);
+}
+
+void tcg_gen_gvec_umin(unsigned vece, uint32_t dofs, uint32_t aofs,
+ uint32_t bofs, uint32_t oprsz, uint32_t maxsz)
+{
+ static const GVecGen3 g[4] = {
+ { .fniv = tcg_gen_umin_vec,
+ .fno = gen_helper_gvec_umin8,
+ .opc = INDEX_op_umin_vec,
+ .vece = MO_8 },
+ { .fniv = tcg_gen_umin_vec,
+ .fno = gen_helper_gvec_umin16,
+ .opc = INDEX_op_umin_vec,
+ .vece = MO_16 },
+ { .fni4 = tcg_gen_umin_i32,
+ .fniv = tcg_gen_umin_vec,
+ .fno = gen_helper_gvec_umin32,
+ .opc = INDEX_op_umin_vec,
+ .vece = MO_32 },
+ { .fni8 = tcg_gen_umin_i64,
+ .fniv = tcg_gen_umin_vec,
+ .fno = gen_helper_gvec_umin64,
+ .opc = INDEX_op_umin_vec,
+ .vece = MO_64 }
+ };
+ tcg_debug_assert(vece <= MO_64);
+ tcg_gen_gvec_3(dofs, aofs, bofs, oprsz, maxsz, &g[vece]);
+}
+
+void tcg_gen_gvec_smax(unsigned vece, uint32_t dofs, uint32_t aofs,
+ uint32_t bofs, uint32_t oprsz, uint32_t maxsz)
+{
+ static const GVecGen3 g[4] = {
+ { .fniv = tcg_gen_smax_vec,
+ .fno = gen_helper_gvec_smax8,
+ .opc = INDEX_op_smax_vec,
+ .vece = MO_8 },
+ { .fniv = tcg_gen_smax_vec,
+ .fno = gen_helper_gvec_smax16,
+ .opc = INDEX_op_smax_vec,
+ .vece = MO_16 },
+ { .fni4 = tcg_gen_smax_i32,
+ .fniv = tcg_gen_smax_vec,
+ .fno = gen_helper_gvec_smax32,
+ .opc = INDEX_op_smax_vec,
+ .vece = MO_32 },
+ { .fni8 = tcg_gen_smax_i64,
+ .fniv = tcg_gen_smax_vec,
+ .fno = gen_helper_gvec_smax64,
+ .opc = INDEX_op_smax_vec,
+ .vece = MO_64 }
+ };
+ tcg_debug_assert(vece <= MO_64);
+ tcg_gen_gvec_3(dofs, aofs, bofs, oprsz, maxsz, &g[vece]);
+}
+
+void tcg_gen_gvec_umax(unsigned vece, uint32_t dofs, uint32_t aofs,
+ uint32_t bofs, uint32_t oprsz, uint32_t maxsz)
+{
+ static const GVecGen3 g[4] = {
+ { .fniv = tcg_gen_umax_vec,
+ .fno = gen_helper_gvec_umax8,
+ .opc = INDEX_op_umax_vec,
+ .vece = MO_8 },
+ { .fniv = tcg_gen_umax_vec,
+ .fno = gen_helper_gvec_umax16,
+ .opc = INDEX_op_umax_vec,
+ .vece = MO_16 },
+ { .fni4 = tcg_gen_umax_i32,
+ .fniv = tcg_gen_umax_vec,
+ .fno = gen_helper_gvec_umax32,
+ .opc = INDEX_op_umax_vec,
+ .vece = MO_32 },
+ { .fni8 = tcg_gen_umax_i64,
+ .fniv = tcg_gen_umax_vec,
+ .fno = gen_helper_gvec_umax64,
+ .opc = INDEX_op_umax_vec,
+ .vece = MO_64 }
+ };
+ tcg_debug_assert(vece <= MO_64);
+ tcg_gen_gvec_3(dofs, aofs, bofs, oprsz, maxsz, &g[vece]);
+}
+
/* Perform a vector negation using normal negation and a mask.
Compare gen_subv_mask above. */
static void gen_negv_mask(TCGv_i64 d, TCGv_i64 b, TCGv_i64 m)
diff --git a/tcg/tcg-op-vec.c b/tcg/tcg-op-vec.c
index 675aa09258..36f35022ac 100644
--- a/tcg/tcg-op-vec.c
+++ b/tcg/tcg-op-vec.c
@@ -433,3 +433,23 @@ void tcg_gen_ussub_vec(unsigned vece, TCGv_vec r, TCGv_vec a, TCGv_vec b)
{
do_op3(vece, r, a, b, INDEX_op_ussub_vec);
}
+
+void tcg_gen_smin_vec(unsigned vece, TCGv_vec r, TCGv_vec a, TCGv_vec b)
+{
+ do_op3(vece, r, a, b, INDEX_op_smin_vec);
+}
+
+void tcg_gen_umin_vec(unsigned vece, TCGv_vec r, TCGv_vec a, TCGv_vec b)
+{
+ do_op3(vece, r, a, b, INDEX_op_umin_vec);
+}
+
+void tcg_gen_smax_vec(unsigned vece, TCGv_vec r, TCGv_vec a, TCGv_vec b)
+{
+ do_op3(vece, r, a, b, INDEX_op_smax_vec);
+}
+
+void tcg_gen_umax_vec(unsigned vece, TCGv_vec r, TCGv_vec a, TCGv_vec b)
+{
+ do_op3(vece, r, a, b, INDEX_op_umax_vec);
+}
diff --git a/tcg/tcg.c b/tcg/tcg.c
index f2cf60425b..2ee031fcf7 100644
--- a/tcg/tcg.c
+++ b/tcg/tcg.c
@@ -1612,6 +1612,11 @@ bool tcg_op_supported(TCGOpcode op)
case INDEX_op_sssub_vec:
case INDEX_op_ussub_vec:
return have_vec && TCG_TARGET_HAS_sat_vec;
+ case INDEX_op_smin_vec:
+ case INDEX_op_umin_vec:
+ case INDEX_op_smax_vec:
+ case INDEX_op_umax_vec:
+ return have_vec && TCG_TARGET_HAS_minmax_vec;
default:
tcg_debug_assert(op > INDEX_op_last_generic && op < NB_OPS);
--
2.17.2
next prev parent reply other threads:[~2018-12-18 6:40 UTC|newest]
Thread overview: 75+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-12-18 6:38 [Qemu-devel] [PATCH 00/34] tcg, target/ppc vector improvements Richard Henderson
2018-12-18 6:38 ` [Qemu-devel] [PATCH 01/34] tcg: Add logical simplifications during gvec expand Richard Henderson
2018-12-19 5:36 ` David Gibson
2018-12-18 6:38 ` [Qemu-devel] [PATCH 02/34] target/arm: Rely on optimization within tcg_gen_gvec_or Richard Henderson
2018-12-19 5:37 ` David Gibson
2018-12-18 6:38 ` [Qemu-devel] [PATCH 03/34] tcg: Add gvec expanders for nand, nor, eqv Richard Henderson
2018-12-19 5:39 ` David Gibson
2018-12-18 6:38 ` [Qemu-devel] [PATCH 04/34] tcg: Add write_aofs to GVecGen4 Richard Henderson
2018-12-18 6:38 ` [Qemu-devel] [PATCH 05/34] tcg: Add opcodes for vector saturated arithmetic Richard Henderson
2018-12-18 6:38 ` [Qemu-devel] [PATCH 06/34] tcg/i386: Implement vector saturating arithmetic Richard Henderson
2018-12-18 6:38 ` Richard Henderson [this message]
2018-12-18 6:38 ` [Qemu-devel] [PATCH 08/34] tcg/i386: Implement vector minmax arithmetic Richard Henderson
2018-12-18 6:38 ` [Qemu-devel] [PATCH 09/34] target/arm: Use vector minmax expanders for aarch64 Richard Henderson
2018-12-18 6:38 ` [Qemu-devel] [PATCH 10/34] target/arm: Use vector minmax expanders for aarch32 Richard Henderson
2018-12-18 6:38 ` [Qemu-devel] [PATCH 11/34] target/ppc: introduce get_fpr() and set_fpr() helpers for FP register access Richard Henderson
2018-12-19 6:15 ` David Gibson
2018-12-19 12:29 ` Mark Cave-Ayland
2018-12-20 16:52 ` Mark Cave-Ayland
2018-12-18 6:38 ` [Qemu-devel] [PATCH 12/34] target/ppc: introduce get_avr64() and set_avr64() helpers for VMX " Richard Henderson
2018-12-19 6:15 ` David Gibson
2018-12-18 6:38 ` [Qemu-devel] [PATCH 13/34] target/ppc: introduce get_cpu_vsr{l, h}() and set_cpu_vsr{l, h}() helpers for VSR " Richard Henderson
2018-12-19 6:17 ` David Gibson
2018-12-18 6:38 ` [Qemu-devel] [PATCH 14/34] target/ppc: switch FPR, VMX and VSX helpers to access data directly from cpu_env Richard Henderson
2018-12-19 6:20 ` David Gibson
2018-12-18 6:38 ` [Qemu-devel] [PATCH 15/34] target/ppc: merge ppc_vsr_t and ppc_avr_t union types Richard Henderson
2018-12-19 6:21 ` David Gibson
2018-12-18 6:38 ` [Qemu-devel] [PATCH 16/34] target/ppc: move FP and VMX registers into aligned vsr register array Richard Henderson
2018-12-19 6:27 ` David Gibson
2018-12-18 6:38 ` [Qemu-devel] [PATCH 17/34] target/ppc: convert VMX logical instructions to use vector operations Richard Henderson
2018-12-19 6:29 ` David Gibson
2018-12-18 6:38 ` [Qemu-devel] [PATCH 18/34] target/ppc: convert vaddu[b, h, w, d] and vsubu[b, h, w, d] over " Richard Henderson
2018-12-19 6:29 ` David Gibson
2018-12-18 6:38 ` [Qemu-devel] [PATCH 19/34] target/ppc: convert vspltis[bhw] " Richard Henderson
2018-12-19 6:31 ` David Gibson
2018-12-18 6:38 ` [Qemu-devel] [PATCH 20/34] target/ppc: convert vsplt[bhw] " Richard Henderson
2018-12-19 6:32 ` David Gibson
2018-12-18 6:38 ` [Qemu-devel] [PATCH 21/34] target/ppc: nand, nor, eqv are now generic " Richard Henderson
2018-12-19 6:32 ` David Gibson
2018-12-18 6:38 ` [Qemu-devel] [PATCH 22/34] target/ppc: convert VSX logical operations to " Richard Henderson
2018-12-19 6:33 ` David Gibson
2018-12-18 6:39 ` [Qemu-devel] [PATCH 23/34] target/ppc: convert xxspltib " Richard Henderson
2018-12-19 6:34 ` David Gibson
2018-12-18 6:39 ` [Qemu-devel] [PATCH 24/34] target/ppc: convert xxspltw " Richard Henderson
2018-12-19 6:35 ` David Gibson
2018-12-18 6:39 ` [Qemu-devel] [PATCH 25/34] target/ppc: convert xxsel " Richard Henderson
2018-12-19 6:35 ` David Gibson
2018-12-18 6:39 ` [Qemu-devel] [PATCH 26/34] target/ppc: Pass integer to helper_mtvscr Richard Henderson
2018-12-19 6:37 ` David Gibson
2018-12-18 6:39 ` [Qemu-devel] [PATCH 27/34] target/ppc: Use helper_mtvscr for reset and gdb Richard Henderson
2018-12-19 6:38 ` David Gibson
2018-12-18 6:39 ` [Qemu-devel] [PATCH 28/34] target/ppc: Remove vscr_nj and vscr_sat Richard Henderson
2018-12-19 6:38 ` David Gibson
2018-12-18 6:39 ` [Qemu-devel] [PATCH 29/34] target/ppc: Add helper_mfvscr Richard Henderson
2018-12-19 6:39 ` David Gibson
2018-12-18 6:39 ` [Qemu-devel] [PATCH 30/34] target/ppc: Use mtvscr/mfvscr for vmstate Richard Henderson
2018-12-19 6:40 ` David Gibson
2018-12-18 6:39 ` [Qemu-devel] [PATCH 31/34] target/ppc: Add set_vscr_sat Richard Henderson
2018-12-19 6:40 ` David Gibson
2018-12-18 6:39 ` [Qemu-devel] [PATCH 32/34] target/ppc: Split out VSCR_SAT to a vector field Richard Henderson
2018-12-19 6:41 ` David Gibson
2018-12-18 6:39 ` [Qemu-devel] [PATCH 33/34] target/ppc: convert vadd*s and vsub*s to vector operations Richard Henderson
2018-12-19 6:42 ` David Gibson
2018-12-18 6:39 ` [Qemu-devel] [PATCH 34/34] target/ppc: convert vmin* and vmax* " Richard Henderson
2018-12-19 6:42 ` David Gibson
2018-12-18 9:49 ` [Qemu-devel] [PATCH 00/34] tcg, target/ppc vector improvements Mark Cave-Ayland
2018-12-18 14:51 ` Mark Cave-Ayland
2018-12-18 15:07 ` Richard Henderson
2018-12-18 15:22 ` Mark Cave-Ayland
2018-12-18 15:05 ` Mark Cave-Ayland
2018-12-18 15:17 ` Richard Henderson
2018-12-18 15:26 ` Mark Cave-Ayland
2018-12-18 16:16 ` Richard Henderson
2019-01-03 14:58 ` Mark Cave-Ayland
2019-01-03 18:31 ` Mark Cave-Ayland
2019-01-04 22:33 ` Richard Henderson
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20181218063911.2112-8-richard.henderson@linaro.org \
--to=richard.henderson@linaro.org \
--cc=david@gibson.dropbear.id.au \
--cc=mark.cave-ayland@ilande.co.uk \
--cc=qemu-devel@nongnu.org \
--cc=qemu-ppc@nongnu.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).