[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Qemu-devel] [PULL 14/35] s390x/tcg: Implement VECTOR FP ADD
From: |
Cornelia Huck |
Subject: |
[Qemu-devel] [PULL 14/35] s390x/tcg: Implement VECTOR FP ADD |
Date: |
Fri, 7 Jun 2019 11:52:16 +0200 |
From: David Hildenbrand <address@hidden>
1. We'll reuse op_vfa() for similar instructions later, prepare for
that.
2. We'll reuse vop64_3() for other instructions later.
3. Take care of modifying the vector register only if no trap happened.
- on traps, flags are not updated and no elements are modified
- traps don't modify the fpc flags
- without traps, all exceptions of all elements are merged
4. We'll reuse check_ieee_exc() later when we need the XxC flag.
We have to check for exceptions after processing each element.
Provide separate handlers for single/all element processing. We'll do
the same for all applicable FP instructions.
Reviewed-by: Richard Henderson <address@hidden>
Signed-off-by: David Hildenbrand <address@hidden>
---
target/s390x/Makefile.objs | 1 +
target/s390x/helper.h | 4 ++
target/s390x/insn-data.def | 5 ++
target/s390x/translate_vx.inc.c | 29 ++++++++
target/s390x/vec_fpu_helper.c | 119 ++++++++++++++++++++++++++++++++
5 files changed, 158 insertions(+)
create mode 100644 target/s390x/vec_fpu_helper.c
diff --git a/target/s390x/Makefile.objs b/target/s390x/Makefile.objs
index ffdd484ef05e..3e2745594a29 100644
--- a/target/s390x/Makefile.objs
+++ b/target/s390x/Makefile.objs
@@ -2,6 +2,7 @@ obj-y += cpu.o cpu_models.o cpu_features.o gdbstub.o
interrupt.o helper.o
obj-$(CONFIG_TCG) += translate.o cc_helper.o excp_helper.o fpu_helper.o
obj-$(CONFIG_TCG) += int_helper.o mem_helper.o misc_helper.o crypto_helper.o
obj-$(CONFIG_TCG) += vec_helper.o vec_int_helper.o vec_string_helper.o
+obj-$(CONFIG_TCG) += vec_fpu_helper.o
obj-$(CONFIG_SOFTMMU) += machine.o ioinst.o arch_dump.o mmu_helper.o diag.o
obj-$(CONFIG_SOFTMMU) += sigp.o
obj-$(CONFIG_KVM) += kvm.o
diff --git a/target/s390x/helper.h b/target/s390x/helper.h
index 5db67779d3c6..21658a277190 100644
--- a/target/s390x/helper.h
+++ b/target/s390x/helper.h
@@ -249,6 +249,10 @@ DEF_HELPER_6(gvec_vstrc_cc_rt8, void, ptr, cptr, cptr,
cptr, env, i32)
DEF_HELPER_6(gvec_vstrc_cc_rt16, void, ptr, cptr, cptr, cptr, env, i32)
DEF_HELPER_6(gvec_vstrc_cc_rt32, void, ptr, cptr, cptr, cptr, env, i32)
+/* === Vector Floating-Point Instructions */
+DEF_HELPER_FLAGS_5(gvec_vfa64, TCG_CALL_NO_WG, void, ptr, cptr, cptr, env, i32)
+DEF_HELPER_FLAGS_5(gvec_vfa64s, TCG_CALL_NO_WG, void, ptr, cptr, cptr, env,
i32)
+
#ifndef CONFIG_USER_ONLY
DEF_HELPER_3(servc, i32, env, i64, i64)
DEF_HELPER_4(diag, void, env, i32, i32, i32)
diff --git a/target/s390x/insn-data.def b/target/s390x/insn-data.def
index a2969fab5884..79892f6042f0 100644
--- a/target/s390x/insn-data.def
+++ b/target/s390x/insn-data.def
@@ -1204,6 +1204,11 @@
/* VECTOR STRING RANGE COMPARE */
F(0xe78a, VSTRC, VRR_d, V, 0, 0, 0, 0, vstrc, 0, IF_VEC)
+/* === Vector Floating-Point Instructions */
+
+/* VECTOR FP ADD */
+ F(0xe7e3, VFA, VRR_c, V, 0, 0, 0, 0, vfa, 0, IF_VEC)
+
#ifndef CONFIG_USER_ONLY
/* COMPARE AND SWAP AND PURGE */
E(0xb250, CSP, RRE, Z, r1_32u, ra2, r1_P, 0, csp, 0, MO_TEUL,
IF_PRIV)
diff --git a/target/s390x/translate_vx.inc.c b/target/s390x/translate_vx.inc.c
index f26ffa28957a..44da9f26457b 100644
--- a/target/s390x/translate_vx.inc.c
+++ b/target/s390x/translate_vx.inc.c
@@ -52,6 +52,11 @@
#define ES_64 MO_64
#define ES_128 4
+/* Floating-Point Format */
+#define FPF_SHORT 2
+#define FPF_LONG 3
+#define FPF_EXT 4
+
static inline bool valid_vec_element(uint8_t enr, TCGMemOp es)
{
return !(enr & ~(NUM_VEC_ELEMENTS(es) - 1));
@@ -2538,3 +2543,27 @@ static DisasJumpType op_vstrc(DisasContext *s, DisasOps
*o)
}
return DISAS_NEXT;
}
+
+static DisasJumpType op_vfa(DisasContext *s, DisasOps *o)
+{
+ const uint8_t fpf = get_field(s->fields, m4);
+ const uint8_t m5 = get_field(s->fields, m5);
+ const bool se = extract32(m5, 3, 1);
+ gen_helper_gvec_3_ptr *fn;
+
+ if (fpf != FPF_LONG || extract32(m5, 0, 3)) {
+ gen_program_exception(s, PGM_SPECIFICATION);
+ return DISAS_NORETURN;
+ }
+
+ switch (s->fields->op2) {
+ case 0xe3:
+ fn = se ? gen_helper_gvec_vfa64s : gen_helper_gvec_vfa64;
+ break;
+ default:
+ g_assert_not_reached();
+ }
+ gen_gvec_3_ptr(get_field(s->fields, v1), get_field(s->fields, v2),
+ get_field(s->fields, v3), cpu_env, 0, fn);
+ return DISAS_NEXT;
+}
diff --git a/target/s390x/vec_fpu_helper.c b/target/s390x/vec_fpu_helper.c
new file mode 100644
index 000000000000..c7db0791d7ca
--- /dev/null
+++ b/target/s390x/vec_fpu_helper.c
@@ -0,0 +1,119 @@
+/*
+ * QEMU TCG support -- s390x vector floating point instruction support
+ *
+ * Copyright (C) 2019 Red Hat Inc
+ *
+ * Authors:
+ * David Hildenbrand <address@hidden>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
+ * See the COPYING file in the top-level directory.
+ */
+#include "qemu/osdep.h"
+#include "qemu-common.h"
+#include "cpu.h"
+#include "internal.h"
+#include "vec.h"
+#include "tcg_s390x.h"
+#include "tcg/tcg-gvec-desc.h"
+#include "exec/exec-all.h"
+#include "exec/helper-proto.h"
+#include "fpu/softfloat.h"
+
+#define VIC_INVALID 0x1
+#define VIC_DIVBYZERO 0x2
+#define VIC_OVERFLOW 0x3
+#define VIC_UNDERFLOW 0x4
+#define VIC_INEXACT 0x5
+
+/* returns the VEX. If the VEX is 0, there is no trap */
+static uint8_t check_ieee_exc(CPUS390XState *env, uint8_t enr, bool XxC,
+ uint8_t *vec_exc)
+{
+ uint8_t vece_exc = 0, trap_exc;
+ unsigned qemu_exc;
+
+ /* Retrieve and clear the softfloat exceptions */
+ qemu_exc = env->fpu_status.float_exception_flags;
+ if (qemu_exc == 0) {
+ return 0;
+ }
+ env->fpu_status.float_exception_flags = 0;
+
+ vece_exc = s390_softfloat_exc_to_ieee(qemu_exc);
+
+ /* Add them to the vector-wide s390x exception bits */
+ *vec_exc |= vece_exc;
+
+ /* Check for traps and construct the VXC */
+ trap_exc = vece_exc & env->fpc >> 24;
+ if (trap_exc) {
+ if (trap_exc & S390_IEEE_MASK_INVALID) {
+ return enr << 4 | VIC_INVALID;
+ } else if (trap_exc & S390_IEEE_MASK_DIVBYZERO) {
+ return enr << 4 | VIC_DIVBYZERO;
+ } else if (trap_exc & S390_IEEE_MASK_OVERFLOW) {
+ return enr << 4 | VIC_OVERFLOW;
+ } else if (trap_exc & S390_IEEE_MASK_UNDERFLOW) {
+ return enr << 4 | VIC_UNDERFLOW;
+ } else if (!XxC) {
+ g_assert(trap_exc & S390_IEEE_MASK_INEXACT);
+ /* inexact has lowest priority on traps */
+ return enr << 4 | VIC_INEXACT;
+ }
+ }
+ return 0;
+}
+
+static void handle_ieee_exc(CPUS390XState *env, uint8_t vxc, uint8_t vec_exc,
+ uintptr_t retaddr)
+{
+ if (vxc) {
+ /* on traps, the fpc flags are not updated, instruction is suppressed
*/
+ tcg_s390_vector_exception(env, vxc, retaddr);
+ }
+ if (vec_exc) {
+ /* indicate exceptions for all elements combined */
+ env->fpc |= vec_exc << 16;
+ }
+}
+
+typedef uint64_t (*vop64_3_fn)(uint64_t a, uint64_t b, float_status *s);
+static void vop64_3(S390Vector *v1, const S390Vector *v2, const S390Vector *v3,
+ CPUS390XState *env, bool s, vop64_3_fn fn,
+ uintptr_t retaddr)
+{
+ uint8_t vxc, vec_exc = 0;
+ S390Vector tmp = {};
+ int i;
+
+ for (i = 0; i < 2; i++) {
+ const uint64_t a = s390_vec_read_element64(v2, i);
+ const uint64_t b = s390_vec_read_element64(v3, i);
+
+ s390_vec_write_element64(&tmp, i, fn(a, b, &env->fpu_status));
+ vxc = check_ieee_exc(env, i, false, &vec_exc);
+ if (s || vxc) {
+ break;
+ }
+ }
+ handle_ieee_exc(env, vxc, vec_exc, retaddr);
+ *v1 = tmp;
+}
+
+static uint64_t vfa64(uint64_t a, uint64_t b, float_status *s)
+{
+ return float64_add(a, b, s);
+}
+
+void HELPER(gvec_vfa64)(void *v1, const void *v2, const void *v3,
+ CPUS390XState *env, uint32_t desc)
+{
+ vop64_3(v1, v2, v3, env, false, vfa64, GETPC());
+}
+
+void HELPER(gvec_vfa64s)(void *v1, const void *v2, const void *v3,
+ CPUS390XState *env, uint32_t desc)
+{
+ vop64_3(v1, v2, v3, env, true, vfa64, GETPC());
+}
--
2.20.1
- [Qemu-devel] [PULL 01/35] MAINTAINERS: cover tests/migration/s390x/, (continued)
- [Qemu-devel] [PULL 01/35] MAINTAINERS: cover tests/migration/s390x/, Cornelia Huck, 2019/06/07
- [Qemu-devel] [PULL 08/35] s390x: Align vector registers to 16 bytes, Cornelia Huck, 2019/06/07
- [Qemu-devel] [PULL 06/35] s390x/tcg: Implement VECTOR ISOLATE STRING, Cornelia Huck, 2019/06/07
- [Qemu-devel] [PULL 07/35] s390x/tcg: Implement VECTOR STRING RANGE COMPARE, Cornelia Huck, 2019/06/07
- [Qemu-devel] [PULL 10/35] s390x/tcg: Fix max_byte detection for stfle, Cornelia Huck, 2019/06/07
- [Qemu-devel] [PULL 09/35] s390x: Use uint64_t for vector registers, Cornelia Huck, 2019/06/07
- [Qemu-devel] [PULL 11/35] s390x/tcg: Store only the necessary amount of doublewords for STFLE, Cornelia Huck, 2019/06/07
- [Qemu-devel] [PULL 12/35] s390x/tcg: Introduce tcg_s390_vector_exception(), Cornelia Huck, 2019/06/07
- [Qemu-devel] [PULL 13/35] s390x/tcg: Export float_comp_to_cc() and float(32|64|128)_dcmask(), Cornelia Huck, 2019/06/07
- [Qemu-devel] [PULL 15/35] s390x/tcg: Implement VECTOR FP COMPARE (AND SIGNAL) SCALAR, Cornelia Huck, 2019/06/07
- [Qemu-devel] [PULL 14/35] s390x/tcg: Implement VECTOR FP ADD,
Cornelia Huck <=
- [Qemu-devel] [PULL 16/35] s390x/tcg: Implement VECTOR FP COMPARE (EQUAL|HIGH|HIGH OR EQUAL), Cornelia Huck, 2019/06/07
- [Qemu-devel] [PULL 17/35] s390x/tcg: Implement VECTOR FP CONVERT FROM FIXED 64-BIT, Cornelia Huck, 2019/06/07
- [Qemu-devel] [PULL 18/35] s390x/tcg: Implement VECTOR FP CONVERT FROM LOGICAL 64-BIT, Cornelia Huck, 2019/06/07
- [Qemu-devel] [PULL 19/35] s390x/tcg: Implement VECTOR FP CONVERT TO FIXED 64-BIT, Cornelia Huck, 2019/06/07
- [Qemu-devel] [PULL 20/35] s390x/tcg: Implement VECTOR FP CONVERT TO LOGICAL 64-BIT, Cornelia Huck, 2019/06/07
- [Qemu-devel] [PULL 25/35] s390x/tcg: Implement VECTOR FP MULTIPLY, Cornelia Huck, 2019/06/07
- [Qemu-devel] [PULL 24/35] s390x/tcg: Implement VECTOR LOAD ROUNDED, Cornelia Huck, 2019/06/07
- [Qemu-devel] [PULL 21/35] s390x/tcg: Implement VECTOR FP DIVIDE, Cornelia Huck, 2019/06/07
- [Qemu-devel] [PULL 23/35] s390x/tcg: Implement VECTOR LOAD LENGTHENED, Cornelia Huck, 2019/06/07
- [Qemu-devel] [PULL 22/35] s390x/tcg: Implement VECTOR LOAD FP INTEGER, Cornelia Huck, 2019/06/07