* [PATCH v5 0/8] target/riscv: support Zfh, Zfhmin extension v0.1
@ 2021-10-21 16:29 frank.chang
2021-10-21 16:29 ` frank.chang
` (8 more replies)
0 siblings, 9 replies; 23+ messages in thread
From: frank.chang @ 2021-10-21 16:29 UTC (permalink / raw)
To: qemu-riscv; +Cc: Frank Chang
From: Frank Chang <frank.chang@sifive.com>
Zfh - Half width floating point
Zfhmin - Subset of half width floating point
Zfh, Zfhmin v0.1 is now in public review period and is required by
RVV extension:
https://groups.google.com/a/groups.riscv.org/g/isa-dev/c/63gDCinXTwE/m/871Wm9XIBQAJ
Zfh, Zfhmin can be enabled with -cpu option: Zfh=true and Zfhmin=true
respectively.
The port is available at:
https://github.com/sifive/qemu/tree/zfh-upstream-v5
Note: This patchset depends on another patchset listed in Based-on
section below so it is not able to be built unless the patchset
is applied.
Changelog:
v5:
* Rebase on riscv-to-apply.next.
v4:
* Spilt Zfh, Zfhmin cpu properties related changes into individual
patches.
v3:
* Use the renamed softfloat min/max APIs: *_minimum_number()
and *_maximum_number().
* Pick softfloat min/max APIs based on CPU privilege spec version.
* Add braces for if statements in REQUIRE_ZFH() and
REQUIRE_ZFH_OR_ZFHMIN().
* Rearrange the positions of Zfh and Zfhmin cpu properties.
v2:
* Use {get,dest}_gpr APIs.
* Add Zfhmin extension.
Based-on: <20211021160847.2748577-1-frank.chang@sifive.com>
Frank Chang (3):
target/riscv: zfh: add Zfh cpu property
target/riscv: zfh: implement zfhmin extension
target/riscv: zfh: add Zfhmin cpu property
Kito Cheng (5):
target/riscv: zfh: half-precision load and store
target/riscv: zfh: half-precision computational
target/riscv: zfh: half-precision convert and move
target/riscv: zfh: half-precision floating-point compare
target/riscv: zfh: half-precision floating-point classify
target/riscv/cpu.c | 2 +
target/riscv/cpu.h | 2 +
target/riscv/fpu_helper.c | 180 ++++++++
target/riscv/helper.h | 29 ++
target/riscv/insn32.decode | 38 ++
target/riscv/insn_trans/trans_rvzfh.c.inc | 537 ++++++++++++++++++++++
target/riscv/internals.h | 16 +
target/riscv/translate.c | 20 +
8 files changed, 824 insertions(+)
create mode 100644 target/riscv/insn_trans/trans_rvzfh.c.inc
--
2.25.1
^ permalink raw reply [flat|nested] 23+ messages in thread
* [PATCH v5 1/8] target/riscv: zfh: half-precision load and store
2021-10-21 16:29 [PATCH v5 0/8] target/riscv: support Zfh, Zfhmin extension v0.1 frank.chang
@ 2021-10-21 16:29 ` frank.chang
2021-10-21 16:29 ` frank.chang
` (7 subsequent siblings)
8 siblings, 0 replies; 23+ messages in thread
From: frank.chang @ 2021-10-21 16:29 UTC (permalink / raw)
To: qemu-riscv
Cc: Frank Chang, Bin Meng, Richard Henderson, qemu-devel,
Chih-Min Chao, Palmer Dabbelt, Alistair Francis, Kito Cheng
From: Kito Cheng <kito.cheng@sifive.com>
Signed-off-by: Kito Cheng <kito.cheng@sifive.com>
Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com>
Signed-off-by: Frank Chang <frank.chang@sifive.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
---
target/riscv/cpu.h | 1 +
target/riscv/insn32.decode | 4 ++
target/riscv/insn_trans/trans_rvzfh.c.inc | 65 +++++++++++++++++++++++
target/riscv/translate.c | 8 +++
4 files changed, 78 insertions(+)
create mode 100644 target/riscv/insn_trans/trans_rvzfh.c.inc
diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h
index a33dc30be81..da3c436987c 100644
--- a/target/riscv/cpu.h
+++ b/target/riscv/cpu.h
@@ -299,6 +299,7 @@ struct RISCVCPU {
bool ext_counters;
bool ext_ifencei;
bool ext_icsr;
+ bool ext_zfh;
char *priv_spec;
char *user_spec;
diff --git a/target/riscv/insn32.decode b/target/riscv/insn32.decode
index 2f251dac1bb..b36a3d8dbf8 100644
--- a/target/riscv/insn32.decode
+++ b/target/riscv/insn32.decode
@@ -726,3 +726,7 @@ binv 0110100 .......... 001 ..... 0110011 @r
binvi 01101. ........... 001 ..... 0010011 @sh
bset 0010100 .......... 001 ..... 0110011 @r
bseti 00101. ........... 001 ..... 0010011 @sh
+
+# *** RV32 Zfh Extension ***
+flh ............ ..... 001 ..... 0000111 @i
+fsh ....... ..... ..... 001 ..... 0100111 @s
diff --git a/target/riscv/insn_trans/trans_rvzfh.c.inc b/target/riscv/insn_trans/trans_rvzfh.c.inc
new file mode 100644
index 00000000000..dad1d703d72
--- /dev/null
+++ b/target/riscv/insn_trans/trans_rvzfh.c.inc
@@ -0,0 +1,65 @@
+/*
+ * RISC-V translation routines for the RV64Zfh Standard Extension.
+ *
+ * Copyright (c) 2020 Chih-Min Chao, chihmin.chao@sifive.com
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2 or later, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#define REQUIRE_ZFH(ctx) do { \
+ if (!ctx->ext_zfh) { \
+ return false; \
+ } \
+} while (0)
+
+static bool trans_flh(DisasContext *ctx, arg_flh *a)
+{
+ TCGv_i64 dest;
+ TCGv t0;
+
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ t0 = get_gpr(ctx, a->rs1, EXT_NONE);
+ if (a->imm) {
+ TCGv temp = temp_new(ctx);
+ tcg_gen_addi_tl(temp, t0, a->imm);
+ t0 = temp;
+ }
+
+ dest = cpu_fpr[a->rd];
+ tcg_gen_qemu_ld_i64(dest, t0, ctx->mem_idx, MO_TEUW);
+ gen_nanbox_h(dest, dest);
+
+ mark_fs_dirty(ctx);
+ return true;
+}
+
+static bool trans_fsh(DisasContext *ctx, arg_fsh *a)
+{
+ TCGv t0;
+
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ t0 = get_gpr(ctx, a->rs1, EXT_NONE);
+ if (a->imm) {
+ TCGv temp = tcg_temp_new();
+ tcg_gen_addi_tl(temp, t0, a->imm);
+ t0 = temp;
+ }
+
+ tcg_gen_qemu_st_i64(cpu_fpr[a->rs2], t0, ctx->mem_idx, MO_TEUW);
+
+ return true;
+}
diff --git a/target/riscv/translate.c b/target/riscv/translate.c
index d38f87d7188..ad963d1898f 100644
--- a/target/riscv/translate.c
+++ b/target/riscv/translate.c
@@ -70,6 +70,7 @@ typedef struct DisasContext {
RISCVMXL ol;
bool virt_enabled;
bool ext_ifencei;
+ bool ext_zfh;
bool hlsx;
/* vector extension */
bool vill;
@@ -127,6 +128,11 @@ static void gen_nanbox_s(TCGv_i64 out, TCGv_i64 in)
tcg_gen_ori_i64(out, in, MAKE_64BIT_MASK(32, 32));
}
+static void gen_nanbox_h(TCGv_i64 out, TCGv_i64 in)
+{
+ tcg_gen_ori_i64(out, in, MAKE_64BIT_MASK(16, 48));
+}
+
/*
* A narrow n-bit operation, where n < FLEN, checks that input operands
* are correctly Nan-boxed, i.e., all upper FLEN - n bits are 1.
@@ -550,6 +556,7 @@ static uint32_t opcode_at(DisasContextBase *dcbase, target_ulong pc)
#include "insn_trans/trans_rvh.c.inc"
#include "insn_trans/trans_rvv.c.inc"
#include "insn_trans/trans_rvb.c.inc"
+#include "insn_trans/trans_rvzfh.c.inc"
#include "insn_trans/trans_privileged.c.inc"
/* Include the auto-generated decoder for 16 bit insn */
@@ -602,6 +609,7 @@ static void riscv_tr_init_disas_context(DisasContextBase *dcbase, CPUState *cs)
ctx->misa_ext = env->misa_ext;
ctx->frm = -1; /* unknown rounding mode */
ctx->ext_ifencei = cpu->cfg.ext_ifencei;
+ ctx->ext_zfh = cpu->cfg.ext_zfh;
ctx->vlen = cpu->cfg.vlen;
ctx->mstatus_hs_fs = FIELD_EX32(tb_flags, TB_FLAGS, MSTATUS_HS_FS);
ctx->hlsx = FIELD_EX32(tb_flags, TB_FLAGS, HLSX);
--
2.25.1
^ permalink raw reply related [flat|nested] 23+ messages in thread
* [PATCH v5 1/8] target/riscv: zfh: half-precision load and store
@ 2021-10-21 16:29 ` frank.chang
0 siblings, 0 replies; 23+ messages in thread
From: frank.chang @ 2021-10-21 16:29 UTC (permalink / raw)
To: qemu-riscv
Cc: Kito Cheng, Chih-Min Chao, Frank Chang, Richard Henderson,
Palmer Dabbelt, Alistair Francis, Bin Meng, qemu-devel
From: Kito Cheng <kito.cheng@sifive.com>
Signed-off-by: Kito Cheng <kito.cheng@sifive.com>
Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com>
Signed-off-by: Frank Chang <frank.chang@sifive.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
---
target/riscv/cpu.h | 1 +
target/riscv/insn32.decode | 4 ++
target/riscv/insn_trans/trans_rvzfh.c.inc | 65 +++++++++++++++++++++++
target/riscv/translate.c | 8 +++
4 files changed, 78 insertions(+)
create mode 100644 target/riscv/insn_trans/trans_rvzfh.c.inc
diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h
index a33dc30be81..da3c436987c 100644
--- a/target/riscv/cpu.h
+++ b/target/riscv/cpu.h
@@ -299,6 +299,7 @@ struct RISCVCPU {
bool ext_counters;
bool ext_ifencei;
bool ext_icsr;
+ bool ext_zfh;
char *priv_spec;
char *user_spec;
diff --git a/target/riscv/insn32.decode b/target/riscv/insn32.decode
index 2f251dac1bb..b36a3d8dbf8 100644
--- a/target/riscv/insn32.decode
+++ b/target/riscv/insn32.decode
@@ -726,3 +726,7 @@ binv 0110100 .......... 001 ..... 0110011 @r
binvi 01101. ........... 001 ..... 0010011 @sh
bset 0010100 .......... 001 ..... 0110011 @r
bseti 00101. ........... 001 ..... 0010011 @sh
+
+# *** RV32 Zfh Extension ***
+flh ............ ..... 001 ..... 0000111 @i
+fsh ....... ..... ..... 001 ..... 0100111 @s
diff --git a/target/riscv/insn_trans/trans_rvzfh.c.inc b/target/riscv/insn_trans/trans_rvzfh.c.inc
new file mode 100644
index 00000000000..dad1d703d72
--- /dev/null
+++ b/target/riscv/insn_trans/trans_rvzfh.c.inc
@@ -0,0 +1,65 @@
+/*
+ * RISC-V translation routines for the RV64Zfh Standard Extension.
+ *
+ * Copyright (c) 2020 Chih-Min Chao, chihmin.chao@sifive.com
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2 or later, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#define REQUIRE_ZFH(ctx) do { \
+ if (!ctx->ext_zfh) { \
+ return false; \
+ } \
+} while (0)
+
+static bool trans_flh(DisasContext *ctx, arg_flh *a)
+{
+ TCGv_i64 dest;
+ TCGv t0;
+
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ t0 = get_gpr(ctx, a->rs1, EXT_NONE);
+ if (a->imm) {
+ TCGv temp = temp_new(ctx);
+ tcg_gen_addi_tl(temp, t0, a->imm);
+ t0 = temp;
+ }
+
+ dest = cpu_fpr[a->rd];
+ tcg_gen_qemu_ld_i64(dest, t0, ctx->mem_idx, MO_TEUW);
+ gen_nanbox_h(dest, dest);
+
+ mark_fs_dirty(ctx);
+ return true;
+}
+
+static bool trans_fsh(DisasContext *ctx, arg_fsh *a)
+{
+ TCGv t0;
+
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ t0 = get_gpr(ctx, a->rs1, EXT_NONE);
+ if (a->imm) {
+ TCGv temp = tcg_temp_new();
+ tcg_gen_addi_tl(temp, t0, a->imm);
+ t0 = temp;
+ }
+
+ tcg_gen_qemu_st_i64(cpu_fpr[a->rs2], t0, ctx->mem_idx, MO_TEUW);
+
+ return true;
+}
diff --git a/target/riscv/translate.c b/target/riscv/translate.c
index d38f87d7188..ad963d1898f 100644
--- a/target/riscv/translate.c
+++ b/target/riscv/translate.c
@@ -70,6 +70,7 @@ typedef struct DisasContext {
RISCVMXL ol;
bool virt_enabled;
bool ext_ifencei;
+ bool ext_zfh;
bool hlsx;
/* vector extension */
bool vill;
@@ -127,6 +128,11 @@ static void gen_nanbox_s(TCGv_i64 out, TCGv_i64 in)
tcg_gen_ori_i64(out, in, MAKE_64BIT_MASK(32, 32));
}
+static void gen_nanbox_h(TCGv_i64 out, TCGv_i64 in)
+{
+ tcg_gen_ori_i64(out, in, MAKE_64BIT_MASK(16, 48));
+}
+
/*
* A narrow n-bit operation, where n < FLEN, checks that input operands
* are correctly Nan-boxed, i.e., all upper FLEN - n bits are 1.
@@ -550,6 +556,7 @@ static uint32_t opcode_at(DisasContextBase *dcbase, target_ulong pc)
#include "insn_trans/trans_rvh.c.inc"
#include "insn_trans/trans_rvv.c.inc"
#include "insn_trans/trans_rvb.c.inc"
+#include "insn_trans/trans_rvzfh.c.inc"
#include "insn_trans/trans_privileged.c.inc"
/* Include the auto-generated decoder for 16 bit insn */
@@ -602,6 +609,7 @@ static void riscv_tr_init_disas_context(DisasContextBase *dcbase, CPUState *cs)
ctx->misa_ext = env->misa_ext;
ctx->frm = -1; /* unknown rounding mode */
ctx->ext_ifencei = cpu->cfg.ext_ifencei;
+ ctx->ext_zfh = cpu->cfg.ext_zfh;
ctx->vlen = cpu->cfg.vlen;
ctx->mstatus_hs_fs = FIELD_EX32(tb_flags, TB_FLAGS, MSTATUS_HS_FS);
ctx->hlsx = FIELD_EX32(tb_flags, TB_FLAGS, HLSX);
--
2.25.1
^ permalink raw reply related [flat|nested] 23+ messages in thread
* [PATCH v5 2/8] target/riscv: zfh: half-precision computational
2021-10-21 16:29 [PATCH v5 0/8] target/riscv: support Zfh, Zfhmin extension v0.1 frank.chang
@ 2021-10-21 16:29 ` frank.chang
2021-10-21 16:29 ` frank.chang
` (7 subsequent siblings)
8 siblings, 0 replies; 23+ messages in thread
From: frank.chang @ 2021-10-21 16:29 UTC (permalink / raw)
To: qemu-riscv
Cc: Frank Chang, Bin Meng, Richard Henderson, qemu-devel,
Chih-Min Chao, Palmer Dabbelt, Alistair Francis, Kito Cheng
From: Kito Cheng <kito.cheng@sifive.com>
Signed-off-by: Kito Cheng <kito.cheng@sifive.com>
Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com>
Signed-off-by: Frank Chang <frank.chang@sifive.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
---
target/riscv/fpu_helper.c | 86 +++++++++++++++
target/riscv/helper.h | 13 +++
target/riscv/insn32.decode | 11 ++
target/riscv/insn_trans/trans_rvzfh.c.inc | 129 ++++++++++++++++++++++
target/riscv/internals.h | 16 +++
5 files changed, 255 insertions(+)
diff --git a/target/riscv/fpu_helper.c b/target/riscv/fpu_helper.c
index d62f4709002..20bb89ad14f 100644
--- a/target/riscv/fpu_helper.c
+++ b/target/riscv/fpu_helper.c
@@ -81,6 +81,15 @@ void helper_set_rounding_mode(CPURISCVState *env, uint32_t rm)
set_float_rounding_mode(softrm, &env->fp_status);
}
+static uint64_t do_fmadd_h(CPURISCVState *env, uint64_t rs1, uint64_t rs2,
+ uint64_t rs3, int flags)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ float16 frs2 = check_nanbox_h(rs2);
+ float16 frs3 = check_nanbox_h(rs3);
+ return nanbox_h(float16_muladd(frs1, frs2, frs3, flags, &env->fp_status));
+}
+
static uint64_t do_fmadd_s(CPURISCVState *env, uint64_t rs1, uint64_t rs2,
uint64_t rs3, int flags)
{
@@ -102,6 +111,12 @@ uint64_t helper_fmadd_d(CPURISCVState *env, uint64_t frs1, uint64_t frs2,
return float64_muladd(frs1, frs2, frs3, 0, &env->fp_status);
}
+uint64_t helper_fmadd_h(CPURISCVState *env, uint64_t frs1, uint64_t frs2,
+ uint64_t frs3)
+{
+ return do_fmadd_h(env, frs1, frs2, frs3, 0);
+}
+
uint64_t helper_fmsub_s(CPURISCVState *env, uint64_t frs1, uint64_t frs2,
uint64_t frs3)
{
@@ -115,6 +130,12 @@ uint64_t helper_fmsub_d(CPURISCVState *env, uint64_t frs1, uint64_t frs2,
&env->fp_status);
}
+uint64_t helper_fmsub_h(CPURISCVState *env, uint64_t frs1, uint64_t frs2,
+ uint64_t frs3)
+{
+ return do_fmadd_h(env, frs1, frs2, frs3, float_muladd_negate_c);
+}
+
uint64_t helper_fnmsub_s(CPURISCVState *env, uint64_t frs1, uint64_t frs2,
uint64_t frs3)
{
@@ -128,6 +149,12 @@ uint64_t helper_fnmsub_d(CPURISCVState *env, uint64_t frs1, uint64_t frs2,
&env->fp_status);
}
+uint64_t helper_fnmsub_h(CPURISCVState *env, uint64_t frs1, uint64_t frs2,
+ uint64_t frs3)
+{
+ return do_fmadd_h(env, frs1, frs2, frs3, float_muladd_negate_product);
+}
+
uint64_t helper_fnmadd_s(CPURISCVState *env, uint64_t frs1, uint64_t frs2,
uint64_t frs3)
{
@@ -142,6 +169,13 @@ uint64_t helper_fnmadd_d(CPURISCVState *env, uint64_t frs1, uint64_t frs2,
float_muladd_negate_product, &env->fp_status);
}
+uint64_t helper_fnmadd_h(CPURISCVState *env, uint64_t frs1, uint64_t frs2,
+ uint64_t frs3)
+{
+ return do_fmadd_h(env, frs1, frs2, frs3,
+ float_muladd_negate_c | float_muladd_negate_product);
+}
+
uint64_t helper_fadd_s(CPURISCVState *env, uint64_t rs1, uint64_t rs2)
{
float32 frs1 = check_nanbox_s(rs1);
@@ -374,3 +408,55 @@ target_ulong helper_fclass_d(uint64_t frs1)
{
return fclass_d(frs1);
}
+
+uint64_t helper_fadd_h(CPURISCVState *env, uint64_t rs1, uint64_t rs2)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ float16 frs2 = check_nanbox_h(rs2);
+ return nanbox_h(float16_add(frs1, frs2, &env->fp_status));
+}
+
+uint64_t helper_fsub_h(CPURISCVState *env, uint64_t rs1, uint64_t rs2)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ float16 frs2 = check_nanbox_h(rs2);
+ return nanbox_h(float16_sub(frs1, frs2, &env->fp_status));
+}
+
+uint64_t helper_fmul_h(CPURISCVState *env, uint64_t rs1, uint64_t rs2)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ float16 frs2 = check_nanbox_h(rs2);
+ return nanbox_h(float16_mul(frs1, frs2, &env->fp_status));
+}
+
+uint64_t helper_fdiv_h(CPURISCVState *env, uint64_t rs1, uint64_t rs2)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ float16 frs2 = check_nanbox_h(rs2);
+ return nanbox_h(float16_div(frs1, frs2, &env->fp_status));
+}
+
+uint64_t helper_fmin_h(CPURISCVState *env, uint64_t rs1, uint64_t rs2)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ float16 frs2 = check_nanbox_h(rs2);
+ return nanbox_h(env->priv_ver < PRIV_VERSION_1_11_0 ?
+ float16_minnum(frs1, frs2, &env->fp_status) :
+ float16_minimum_number(frs1, frs2, &env->fp_status));
+}
+
+uint64_t helper_fmax_h(CPURISCVState *env, uint64_t rs1, uint64_t rs2)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ float16 frs2 = check_nanbox_h(rs2);
+ return nanbox_h(env->priv_ver < PRIV_VERSION_1_11_0 ?
+ float16_maxnum(frs1, frs2, &env->fp_status) :
+ float16_maximum_number(frs1, frs2, &env->fp_status));
+}
+
+uint64_t helper_fsqrt_h(CPURISCVState *env, uint64_t rs1)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ return nanbox_h(float16_sqrt(frs1, &env->fp_status));
+}
diff --git a/target/riscv/helper.h b/target/riscv/helper.h
index c7a53762277..c6c0323fafc 100644
--- a/target/riscv/helper.h
+++ b/target/riscv/helper.h
@@ -7,12 +7,16 @@ DEF_HELPER_FLAGS_2(set_rounding_mode, TCG_CALL_NO_WG, void, env, i32)
/* Floating Point - fused */
DEF_HELPER_FLAGS_4(fmadd_s, TCG_CALL_NO_RWG, i64, env, i64, i64, i64)
DEF_HELPER_FLAGS_4(fmadd_d, TCG_CALL_NO_RWG, i64, env, i64, i64, i64)
+DEF_HELPER_FLAGS_4(fmadd_h, TCG_CALL_NO_RWG, i64, env, i64, i64, i64)
DEF_HELPER_FLAGS_4(fmsub_s, TCG_CALL_NO_RWG, i64, env, i64, i64, i64)
DEF_HELPER_FLAGS_4(fmsub_d, TCG_CALL_NO_RWG, i64, env, i64, i64, i64)
+DEF_HELPER_FLAGS_4(fmsub_h, TCG_CALL_NO_RWG, i64, env, i64, i64, i64)
DEF_HELPER_FLAGS_4(fnmsub_s, TCG_CALL_NO_RWG, i64, env, i64, i64, i64)
DEF_HELPER_FLAGS_4(fnmsub_d, TCG_CALL_NO_RWG, i64, env, i64, i64, i64)
+DEF_HELPER_FLAGS_4(fnmsub_h, TCG_CALL_NO_RWG, i64, env, i64, i64, i64)
DEF_HELPER_FLAGS_4(fnmadd_s, TCG_CALL_NO_RWG, i64, env, i64, i64, i64)
DEF_HELPER_FLAGS_4(fnmadd_d, TCG_CALL_NO_RWG, i64, env, i64, i64, i64)
+DEF_HELPER_FLAGS_4(fnmadd_h, TCG_CALL_NO_RWG, i64, env, i64, i64, i64)
/* Floating Point - Single Precision */
DEF_HELPER_FLAGS_3(fadd_s, TCG_CALL_NO_RWG, i64, env, i64, i64)
@@ -62,6 +66,15 @@ DEF_HELPER_FLAGS_1(fclass_d, TCG_CALL_NO_RWG_SE, tl, i64)
DEF_HELPER_FLAGS_2(clmul, TCG_CALL_NO_RWG_SE, tl, tl, tl)
DEF_HELPER_FLAGS_2(clmulr, TCG_CALL_NO_RWG_SE, tl, tl, tl)
+/* Floating Point - Half Precision */
+DEF_HELPER_FLAGS_3(fadd_h, TCG_CALL_NO_RWG, i64, env, i64, i64)
+DEF_HELPER_FLAGS_3(fsub_h, TCG_CALL_NO_RWG, i64, env, i64, i64)
+DEF_HELPER_FLAGS_3(fmul_h, TCG_CALL_NO_RWG, i64, env, i64, i64)
+DEF_HELPER_FLAGS_3(fdiv_h, TCG_CALL_NO_RWG, i64, env, i64, i64)
+DEF_HELPER_FLAGS_3(fmin_h, TCG_CALL_NO_RWG, i64, env, i64, i64)
+DEF_HELPER_FLAGS_3(fmax_h, TCG_CALL_NO_RWG, i64, env, i64, i64)
+DEF_HELPER_FLAGS_2(fsqrt_h, TCG_CALL_NO_RWG, i64, env, i64)
+
/* Special functions */
DEF_HELPER_2(csrr, tl, env, int)
DEF_HELPER_3(csrw, void, env, int, tl)
diff --git a/target/riscv/insn32.decode b/target/riscv/insn32.decode
index b36a3d8dbf8..66c231a3010 100644
--- a/target/riscv/insn32.decode
+++ b/target/riscv/insn32.decode
@@ -730,3 +730,14 @@ bseti 00101. ........... 001 ..... 0010011 @sh
# *** RV32 Zfh Extension ***
flh ............ ..... 001 ..... 0000111 @i
fsh ....... ..... ..... 001 ..... 0100111 @s
+fmadd_h ..... 10 ..... ..... ... ..... 1000011 @r4_rm
+fmsub_h ..... 10 ..... ..... ... ..... 1000111 @r4_rm
+fnmsub_h ..... 10 ..... ..... ... ..... 1001011 @r4_rm
+fnmadd_h ..... 10 ..... ..... ... ..... 1001111 @r4_rm
+fadd_h 0000010 ..... ..... ... ..... 1010011 @r_rm
+fsub_h 0000110 ..... ..... ... ..... 1010011 @r_rm
+fmul_h 0001010 ..... ..... ... ..... 1010011 @r_rm
+fdiv_h 0001110 ..... ..... ... ..... 1010011 @r_rm
+fsqrt_h 0101110 00000 ..... ... ..... 1010011 @r2_rm
+fmin_h 0010110 ..... ..... 000 ..... 1010011 @r
+fmax_h 0010110 ..... ..... 001 ..... 1010011 @r
diff --git a/target/riscv/insn_trans/trans_rvzfh.c.inc b/target/riscv/insn_trans/trans_rvzfh.c.inc
index dad1d703d72..9764d76f8bc 100644
--- a/target/riscv/insn_trans/trans_rvzfh.c.inc
+++ b/target/riscv/insn_trans/trans_rvzfh.c.inc
@@ -63,3 +63,132 @@ static bool trans_fsh(DisasContext *ctx, arg_fsh *a)
return true;
}
+
+static bool trans_fmadd_h(DisasContext *ctx, arg_fmadd_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fmadd_h(cpu_fpr[a->rd], cpu_env, cpu_fpr[a->rs1],
+ cpu_fpr[a->rs2], cpu_fpr[a->rs3]);
+ mark_fs_dirty(ctx);
+ return true;
+}
+
+static bool trans_fmsub_h(DisasContext *ctx, arg_fmsub_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fmsub_h(cpu_fpr[a->rd], cpu_env, cpu_fpr[a->rs1],
+ cpu_fpr[a->rs2], cpu_fpr[a->rs3]);
+ mark_fs_dirty(ctx);
+ return true;
+}
+
+static bool trans_fnmsub_h(DisasContext *ctx, arg_fnmsub_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fnmsub_h(cpu_fpr[a->rd], cpu_env, cpu_fpr[a->rs1],
+ cpu_fpr[a->rs2], cpu_fpr[a->rs3]);
+ mark_fs_dirty(ctx);
+ return true;
+}
+
+static bool trans_fnmadd_h(DisasContext *ctx, arg_fnmadd_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fnmadd_h(cpu_fpr[a->rd], cpu_env, cpu_fpr[a->rs1],
+ cpu_fpr[a->rs2], cpu_fpr[a->rs3]);
+ mark_fs_dirty(ctx);
+ return true;
+}
+
+static bool trans_fadd_h(DisasContext *ctx, arg_fadd_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fadd_h(cpu_fpr[a->rd], cpu_env,
+ cpu_fpr[a->rs1], cpu_fpr[a->rs2]);
+ mark_fs_dirty(ctx);
+ return true;
+}
+
+static bool trans_fsub_h(DisasContext *ctx, arg_fsub_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fsub_h(cpu_fpr[a->rd], cpu_env,
+ cpu_fpr[a->rs1], cpu_fpr[a->rs2]);
+ mark_fs_dirty(ctx);
+ return true;
+}
+
+static bool trans_fmul_h(DisasContext *ctx, arg_fmul_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fmul_h(cpu_fpr[a->rd], cpu_env,
+ cpu_fpr[a->rs1], cpu_fpr[a->rs2]);
+ mark_fs_dirty(ctx);
+ return true;
+}
+
+static bool trans_fdiv_h(DisasContext *ctx, arg_fdiv_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fdiv_h(cpu_fpr[a->rd], cpu_env,
+ cpu_fpr[a->rs1], cpu_fpr[a->rs2]);
+ mark_fs_dirty(ctx);
+ return true;
+}
+
+static bool trans_fsqrt_h(DisasContext *ctx, arg_fsqrt_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fsqrt_h(cpu_fpr[a->rd], cpu_env, cpu_fpr[a->rs1]);
+ mark_fs_dirty(ctx);
+ return true;
+}
+
+static bool trans_fmin_h(DisasContext *ctx, arg_fmin_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ gen_helper_fmin_h(cpu_fpr[a->rd], cpu_env, cpu_fpr[a->rs1],
+ cpu_fpr[a->rs2]);
+ mark_fs_dirty(ctx);
+ return true;
+}
+
+static bool trans_fmax_h(DisasContext *ctx, arg_fmax_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ gen_helper_fmax_h(cpu_fpr[a->rd], cpu_env, cpu_fpr[a->rs1],
+ cpu_fpr[a->rs2]);
+ mark_fs_dirty(ctx);
+ return true;
+}
diff --git a/target/riscv/internals.h b/target/riscv/internals.h
index b15ad394bb9..bce91da11a4 100644
--- a/target/riscv/internals.h
+++ b/target/riscv/internals.h
@@ -58,4 +58,20 @@ static inline float32 check_nanbox_s(uint64_t f)
}
}
+static inline uint64_t nanbox_h(float16 f)
+{
+ return f | MAKE_64BIT_MASK(16, 48);
+}
+
+static inline float16 check_nanbox_h(uint64_t f)
+{
+ uint64_t mask = MAKE_64BIT_MASK(16, 48);
+
+ if (likely((f & mask) == mask)) {
+ return (uint16_t)f;
+ } else {
+ return 0x7E00u; /* default qnan */
+ }
+}
+
#endif
--
2.25.1
^ permalink raw reply related [flat|nested] 23+ messages in thread
* [PATCH v5 2/8] target/riscv: zfh: half-precision computational
@ 2021-10-21 16:29 ` frank.chang
0 siblings, 0 replies; 23+ messages in thread
From: frank.chang @ 2021-10-21 16:29 UTC (permalink / raw)
To: qemu-riscv
Cc: Kito Cheng, Chih-Min Chao, Frank Chang, Richard Henderson,
Alistair Francis, Palmer Dabbelt, Bin Meng, qemu-devel
From: Kito Cheng <kito.cheng@sifive.com>
Signed-off-by: Kito Cheng <kito.cheng@sifive.com>
Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com>
Signed-off-by: Frank Chang <frank.chang@sifive.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
---
target/riscv/fpu_helper.c | 86 +++++++++++++++
target/riscv/helper.h | 13 +++
target/riscv/insn32.decode | 11 ++
target/riscv/insn_trans/trans_rvzfh.c.inc | 129 ++++++++++++++++++++++
target/riscv/internals.h | 16 +++
5 files changed, 255 insertions(+)
diff --git a/target/riscv/fpu_helper.c b/target/riscv/fpu_helper.c
index d62f4709002..20bb89ad14f 100644
--- a/target/riscv/fpu_helper.c
+++ b/target/riscv/fpu_helper.c
@@ -81,6 +81,15 @@ void helper_set_rounding_mode(CPURISCVState *env, uint32_t rm)
set_float_rounding_mode(softrm, &env->fp_status);
}
+static uint64_t do_fmadd_h(CPURISCVState *env, uint64_t rs1, uint64_t rs2,
+ uint64_t rs3, int flags)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ float16 frs2 = check_nanbox_h(rs2);
+ float16 frs3 = check_nanbox_h(rs3);
+ return nanbox_h(float16_muladd(frs1, frs2, frs3, flags, &env->fp_status));
+}
+
static uint64_t do_fmadd_s(CPURISCVState *env, uint64_t rs1, uint64_t rs2,
uint64_t rs3, int flags)
{
@@ -102,6 +111,12 @@ uint64_t helper_fmadd_d(CPURISCVState *env, uint64_t frs1, uint64_t frs2,
return float64_muladd(frs1, frs2, frs3, 0, &env->fp_status);
}
+uint64_t helper_fmadd_h(CPURISCVState *env, uint64_t frs1, uint64_t frs2,
+ uint64_t frs3)
+{
+ return do_fmadd_h(env, frs1, frs2, frs3, 0);
+}
+
uint64_t helper_fmsub_s(CPURISCVState *env, uint64_t frs1, uint64_t frs2,
uint64_t frs3)
{
@@ -115,6 +130,12 @@ uint64_t helper_fmsub_d(CPURISCVState *env, uint64_t frs1, uint64_t frs2,
&env->fp_status);
}
+uint64_t helper_fmsub_h(CPURISCVState *env, uint64_t frs1, uint64_t frs2,
+ uint64_t frs3)
+{
+ return do_fmadd_h(env, frs1, frs2, frs3, float_muladd_negate_c);
+}
+
uint64_t helper_fnmsub_s(CPURISCVState *env, uint64_t frs1, uint64_t frs2,
uint64_t frs3)
{
@@ -128,6 +149,12 @@ uint64_t helper_fnmsub_d(CPURISCVState *env, uint64_t frs1, uint64_t frs2,
&env->fp_status);
}
+uint64_t helper_fnmsub_h(CPURISCVState *env, uint64_t frs1, uint64_t frs2,
+ uint64_t frs3)
+{
+ return do_fmadd_h(env, frs1, frs2, frs3, float_muladd_negate_product);
+}
+
uint64_t helper_fnmadd_s(CPURISCVState *env, uint64_t frs1, uint64_t frs2,
uint64_t frs3)
{
@@ -142,6 +169,13 @@ uint64_t helper_fnmadd_d(CPURISCVState *env, uint64_t frs1, uint64_t frs2,
float_muladd_negate_product, &env->fp_status);
}
+uint64_t helper_fnmadd_h(CPURISCVState *env, uint64_t frs1, uint64_t frs2,
+ uint64_t frs3)
+{
+ return do_fmadd_h(env, frs1, frs2, frs3,
+ float_muladd_negate_c | float_muladd_negate_product);
+}
+
uint64_t helper_fadd_s(CPURISCVState *env, uint64_t rs1, uint64_t rs2)
{
float32 frs1 = check_nanbox_s(rs1);
@@ -374,3 +408,55 @@ target_ulong helper_fclass_d(uint64_t frs1)
{
return fclass_d(frs1);
}
+
+uint64_t helper_fadd_h(CPURISCVState *env, uint64_t rs1, uint64_t rs2)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ float16 frs2 = check_nanbox_h(rs2);
+ return nanbox_h(float16_add(frs1, frs2, &env->fp_status));
+}
+
+uint64_t helper_fsub_h(CPURISCVState *env, uint64_t rs1, uint64_t rs2)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ float16 frs2 = check_nanbox_h(rs2);
+ return nanbox_h(float16_sub(frs1, frs2, &env->fp_status));
+}
+
+uint64_t helper_fmul_h(CPURISCVState *env, uint64_t rs1, uint64_t rs2)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ float16 frs2 = check_nanbox_h(rs2);
+ return nanbox_h(float16_mul(frs1, frs2, &env->fp_status));
+}
+
+uint64_t helper_fdiv_h(CPURISCVState *env, uint64_t rs1, uint64_t rs2)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ float16 frs2 = check_nanbox_h(rs2);
+ return nanbox_h(float16_div(frs1, frs2, &env->fp_status));
+}
+
+uint64_t helper_fmin_h(CPURISCVState *env, uint64_t rs1, uint64_t rs2)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ float16 frs2 = check_nanbox_h(rs2);
+ return nanbox_h(env->priv_ver < PRIV_VERSION_1_11_0 ?
+ float16_minnum(frs1, frs2, &env->fp_status) :
+ float16_minimum_number(frs1, frs2, &env->fp_status));
+}
+
+uint64_t helper_fmax_h(CPURISCVState *env, uint64_t rs1, uint64_t rs2)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ float16 frs2 = check_nanbox_h(rs2);
+ return nanbox_h(env->priv_ver < PRIV_VERSION_1_11_0 ?
+ float16_maxnum(frs1, frs2, &env->fp_status) :
+ float16_maximum_number(frs1, frs2, &env->fp_status));
+}
+
+uint64_t helper_fsqrt_h(CPURISCVState *env, uint64_t rs1)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ return nanbox_h(float16_sqrt(frs1, &env->fp_status));
+}
diff --git a/target/riscv/helper.h b/target/riscv/helper.h
index c7a53762277..c6c0323fafc 100644
--- a/target/riscv/helper.h
+++ b/target/riscv/helper.h
@@ -7,12 +7,16 @@ DEF_HELPER_FLAGS_2(set_rounding_mode, TCG_CALL_NO_WG, void, env, i32)
/* Floating Point - fused */
DEF_HELPER_FLAGS_4(fmadd_s, TCG_CALL_NO_RWG, i64, env, i64, i64, i64)
DEF_HELPER_FLAGS_4(fmadd_d, TCG_CALL_NO_RWG, i64, env, i64, i64, i64)
+DEF_HELPER_FLAGS_4(fmadd_h, TCG_CALL_NO_RWG, i64, env, i64, i64, i64)
DEF_HELPER_FLAGS_4(fmsub_s, TCG_CALL_NO_RWG, i64, env, i64, i64, i64)
DEF_HELPER_FLAGS_4(fmsub_d, TCG_CALL_NO_RWG, i64, env, i64, i64, i64)
+DEF_HELPER_FLAGS_4(fmsub_h, TCG_CALL_NO_RWG, i64, env, i64, i64, i64)
DEF_HELPER_FLAGS_4(fnmsub_s, TCG_CALL_NO_RWG, i64, env, i64, i64, i64)
DEF_HELPER_FLAGS_4(fnmsub_d, TCG_CALL_NO_RWG, i64, env, i64, i64, i64)
+DEF_HELPER_FLAGS_4(fnmsub_h, TCG_CALL_NO_RWG, i64, env, i64, i64, i64)
DEF_HELPER_FLAGS_4(fnmadd_s, TCG_CALL_NO_RWG, i64, env, i64, i64, i64)
DEF_HELPER_FLAGS_4(fnmadd_d, TCG_CALL_NO_RWG, i64, env, i64, i64, i64)
+DEF_HELPER_FLAGS_4(fnmadd_h, TCG_CALL_NO_RWG, i64, env, i64, i64, i64)
/* Floating Point - Single Precision */
DEF_HELPER_FLAGS_3(fadd_s, TCG_CALL_NO_RWG, i64, env, i64, i64)
@@ -62,6 +66,15 @@ DEF_HELPER_FLAGS_1(fclass_d, TCG_CALL_NO_RWG_SE, tl, i64)
DEF_HELPER_FLAGS_2(clmul, TCG_CALL_NO_RWG_SE, tl, tl, tl)
DEF_HELPER_FLAGS_2(clmulr, TCG_CALL_NO_RWG_SE, tl, tl, tl)
+/* Floating Point - Half Precision */
+DEF_HELPER_FLAGS_3(fadd_h, TCG_CALL_NO_RWG, i64, env, i64, i64)
+DEF_HELPER_FLAGS_3(fsub_h, TCG_CALL_NO_RWG, i64, env, i64, i64)
+DEF_HELPER_FLAGS_3(fmul_h, TCG_CALL_NO_RWG, i64, env, i64, i64)
+DEF_HELPER_FLAGS_3(fdiv_h, TCG_CALL_NO_RWG, i64, env, i64, i64)
+DEF_HELPER_FLAGS_3(fmin_h, TCG_CALL_NO_RWG, i64, env, i64, i64)
+DEF_HELPER_FLAGS_3(fmax_h, TCG_CALL_NO_RWG, i64, env, i64, i64)
+DEF_HELPER_FLAGS_2(fsqrt_h, TCG_CALL_NO_RWG, i64, env, i64)
+
/* Special functions */
DEF_HELPER_2(csrr, tl, env, int)
DEF_HELPER_3(csrw, void, env, int, tl)
diff --git a/target/riscv/insn32.decode b/target/riscv/insn32.decode
index b36a3d8dbf8..66c231a3010 100644
--- a/target/riscv/insn32.decode
+++ b/target/riscv/insn32.decode
@@ -730,3 +730,14 @@ bseti 00101. ........... 001 ..... 0010011 @sh
# *** RV32 Zfh Extension ***
flh ............ ..... 001 ..... 0000111 @i
fsh ....... ..... ..... 001 ..... 0100111 @s
+fmadd_h ..... 10 ..... ..... ... ..... 1000011 @r4_rm
+fmsub_h ..... 10 ..... ..... ... ..... 1000111 @r4_rm
+fnmsub_h ..... 10 ..... ..... ... ..... 1001011 @r4_rm
+fnmadd_h ..... 10 ..... ..... ... ..... 1001111 @r4_rm
+fadd_h 0000010 ..... ..... ... ..... 1010011 @r_rm
+fsub_h 0000110 ..... ..... ... ..... 1010011 @r_rm
+fmul_h 0001010 ..... ..... ... ..... 1010011 @r_rm
+fdiv_h 0001110 ..... ..... ... ..... 1010011 @r_rm
+fsqrt_h 0101110 00000 ..... ... ..... 1010011 @r2_rm
+fmin_h 0010110 ..... ..... 000 ..... 1010011 @r
+fmax_h 0010110 ..... ..... 001 ..... 1010011 @r
diff --git a/target/riscv/insn_trans/trans_rvzfh.c.inc b/target/riscv/insn_trans/trans_rvzfh.c.inc
index dad1d703d72..9764d76f8bc 100644
--- a/target/riscv/insn_trans/trans_rvzfh.c.inc
+++ b/target/riscv/insn_trans/trans_rvzfh.c.inc
@@ -63,3 +63,132 @@ static bool trans_fsh(DisasContext *ctx, arg_fsh *a)
return true;
}
+
+static bool trans_fmadd_h(DisasContext *ctx, arg_fmadd_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fmadd_h(cpu_fpr[a->rd], cpu_env, cpu_fpr[a->rs1],
+ cpu_fpr[a->rs2], cpu_fpr[a->rs3]);
+ mark_fs_dirty(ctx);
+ return true;
+}
+
+static bool trans_fmsub_h(DisasContext *ctx, arg_fmsub_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fmsub_h(cpu_fpr[a->rd], cpu_env, cpu_fpr[a->rs1],
+ cpu_fpr[a->rs2], cpu_fpr[a->rs3]);
+ mark_fs_dirty(ctx);
+ return true;
+}
+
+static bool trans_fnmsub_h(DisasContext *ctx, arg_fnmsub_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fnmsub_h(cpu_fpr[a->rd], cpu_env, cpu_fpr[a->rs1],
+ cpu_fpr[a->rs2], cpu_fpr[a->rs3]);
+ mark_fs_dirty(ctx);
+ return true;
+}
+
+static bool trans_fnmadd_h(DisasContext *ctx, arg_fnmadd_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fnmadd_h(cpu_fpr[a->rd], cpu_env, cpu_fpr[a->rs1],
+ cpu_fpr[a->rs2], cpu_fpr[a->rs3]);
+ mark_fs_dirty(ctx);
+ return true;
+}
+
+static bool trans_fadd_h(DisasContext *ctx, arg_fadd_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fadd_h(cpu_fpr[a->rd], cpu_env,
+ cpu_fpr[a->rs1], cpu_fpr[a->rs2]);
+ mark_fs_dirty(ctx);
+ return true;
+}
+
+static bool trans_fsub_h(DisasContext *ctx, arg_fsub_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fsub_h(cpu_fpr[a->rd], cpu_env,
+ cpu_fpr[a->rs1], cpu_fpr[a->rs2]);
+ mark_fs_dirty(ctx);
+ return true;
+}
+
+static bool trans_fmul_h(DisasContext *ctx, arg_fmul_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fmul_h(cpu_fpr[a->rd], cpu_env,
+ cpu_fpr[a->rs1], cpu_fpr[a->rs2]);
+ mark_fs_dirty(ctx);
+ return true;
+}
+
+static bool trans_fdiv_h(DisasContext *ctx, arg_fdiv_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fdiv_h(cpu_fpr[a->rd], cpu_env,
+ cpu_fpr[a->rs1], cpu_fpr[a->rs2]);
+ mark_fs_dirty(ctx);
+ return true;
+}
+
+static bool trans_fsqrt_h(DisasContext *ctx, arg_fsqrt_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fsqrt_h(cpu_fpr[a->rd], cpu_env, cpu_fpr[a->rs1]);
+ mark_fs_dirty(ctx);
+ return true;
+}
+
+static bool trans_fmin_h(DisasContext *ctx, arg_fmin_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ gen_helper_fmin_h(cpu_fpr[a->rd], cpu_env, cpu_fpr[a->rs1],
+ cpu_fpr[a->rs2]);
+ mark_fs_dirty(ctx);
+ return true;
+}
+
+static bool trans_fmax_h(DisasContext *ctx, arg_fmax_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ gen_helper_fmax_h(cpu_fpr[a->rd], cpu_env, cpu_fpr[a->rs1],
+ cpu_fpr[a->rs2]);
+ mark_fs_dirty(ctx);
+ return true;
+}
diff --git a/target/riscv/internals.h b/target/riscv/internals.h
index b15ad394bb9..bce91da11a4 100644
--- a/target/riscv/internals.h
+++ b/target/riscv/internals.h
@@ -58,4 +58,20 @@ static inline float32 check_nanbox_s(uint64_t f)
}
}
+static inline uint64_t nanbox_h(float16 f)
+{
+ return f | MAKE_64BIT_MASK(16, 48);
+}
+
+static inline float16 check_nanbox_h(uint64_t f)
+{
+ uint64_t mask = MAKE_64BIT_MASK(16, 48);
+
+ if (likely((f & mask) == mask)) {
+ return (uint16_t)f;
+ } else {
+ return 0x7E00u; /* default qnan */
+ }
+}
+
#endif
--
2.25.1
^ permalink raw reply related [flat|nested] 23+ messages in thread
* [PATCH v5 3/8] target/riscv: zfh: half-precision convert and move
2021-10-21 16:29 [PATCH v5 0/8] target/riscv: support Zfh, Zfhmin extension v0.1 frank.chang
@ 2021-10-21 16:29 ` frank.chang
2021-10-21 16:29 ` frank.chang
` (7 subsequent siblings)
8 siblings, 0 replies; 23+ messages in thread
From: frank.chang @ 2021-10-21 16:29 UTC (permalink / raw)
To: qemu-riscv
Cc: Frank Chang, Bin Meng, Richard Henderson, qemu-devel,
Chih-Min Chao, Palmer Dabbelt, Alistair Francis, Kito Cheng
From: Kito Cheng <kito.cheng@sifive.com>
Signed-off-by: Kito Cheng <kito.cheng@sifive.com>
Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com>
Signed-off-by: Frank Chang <frank.chang@sifive.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Acked-by: Alistair Francis <alistair.francis@wdc.com>
---
target/riscv/fpu_helper.c | 67 +++++
target/riscv/helper.h | 12 +
target/riscv/insn32.decode | 19 ++
target/riscv/insn_trans/trans_rvzfh.c.inc | 288 ++++++++++++++++++++++
target/riscv/translate.c | 10 +
5 files changed, 396 insertions(+)
diff --git a/target/riscv/fpu_helper.c b/target/riscv/fpu_helper.c
index 20bb89ad14f..2ed9b03193c 100644
--- a/target/riscv/fpu_helper.c
+++ b/target/riscv/fpu_helper.c
@@ -460,3 +460,70 @@ uint64_t helper_fsqrt_h(CPURISCVState *env, uint64_t rs1)
float16 frs1 = check_nanbox_h(rs1);
return nanbox_h(float16_sqrt(frs1, &env->fp_status));
}
+
+target_ulong helper_fcvt_w_h(CPURISCVState *env, uint64_t rs1)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ return float16_to_int32(frs1, &env->fp_status);
+}
+
+target_ulong helper_fcvt_wu_h(CPURISCVState *env, uint64_t rs1)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ return (int32_t)float16_to_uint32(frs1, &env->fp_status);
+}
+
+target_ulong helper_fcvt_l_h(CPURISCVState *env, uint64_t rs1)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ return float16_to_int64(frs1, &env->fp_status);
+}
+
+target_ulong helper_fcvt_lu_h(CPURISCVState *env, uint64_t rs1)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ return float16_to_uint64(frs1, &env->fp_status);
+}
+
+uint64_t helper_fcvt_h_w(CPURISCVState *env, target_ulong rs1)
+{
+ return nanbox_h(int32_to_float16((int32_t)rs1, &env->fp_status));
+}
+
+uint64_t helper_fcvt_h_wu(CPURISCVState *env, target_ulong rs1)
+{
+ return nanbox_h(uint32_to_float16((uint32_t)rs1, &env->fp_status));
+}
+
+uint64_t helper_fcvt_h_l(CPURISCVState *env, target_ulong rs1)
+{
+ return nanbox_h(int64_to_float16(rs1, &env->fp_status));
+}
+
+uint64_t helper_fcvt_h_lu(CPURISCVState *env, target_ulong rs1)
+{
+ return nanbox_h(uint64_to_float16(rs1, &env->fp_status));
+}
+
+uint64_t helper_fcvt_h_s(CPURISCVState *env, uint64_t rs1)
+{
+ float32 frs1 = check_nanbox_s(rs1);
+ return nanbox_h(float32_to_float16(frs1, true, &env->fp_status));
+}
+
+uint64_t helper_fcvt_s_h(CPURISCVState *env, uint64_t rs1)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ return nanbox_s(float16_to_float32(frs1, true, &env->fp_status));
+}
+
+uint64_t helper_fcvt_h_d(CPURISCVState *env, uint64_t rs1)
+{
+ return nanbox_h(float64_to_float16(rs1, true, &env->fp_status));
+}
+
+uint64_t helper_fcvt_d_h(CPURISCVState *env, uint64_t rs1)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ return float16_to_float64(frs1, true, &env->fp_status);
+}
diff --git a/target/riscv/helper.h b/target/riscv/helper.h
index c6c0323fafc..b50672d1684 100644
--- a/target/riscv/helper.h
+++ b/target/riscv/helper.h
@@ -74,6 +74,18 @@ DEF_HELPER_FLAGS_3(fdiv_h, TCG_CALL_NO_RWG, i64, env, i64, i64)
DEF_HELPER_FLAGS_3(fmin_h, TCG_CALL_NO_RWG, i64, env, i64, i64)
DEF_HELPER_FLAGS_3(fmax_h, TCG_CALL_NO_RWG, i64, env, i64, i64)
DEF_HELPER_FLAGS_2(fsqrt_h, TCG_CALL_NO_RWG, i64, env, i64)
+DEF_HELPER_FLAGS_2(fcvt_s_h, TCG_CALL_NO_RWG, i64, env, i64)
+DEF_HELPER_FLAGS_2(fcvt_h_s, TCG_CALL_NO_RWG, i64, env, i64)
+DEF_HELPER_FLAGS_2(fcvt_d_h, TCG_CALL_NO_RWG, i64, env, i64)
+DEF_HELPER_FLAGS_2(fcvt_h_d, TCG_CALL_NO_RWG, i64, env, i64)
+DEF_HELPER_FLAGS_2(fcvt_w_h, TCG_CALL_NO_RWG, tl, env, i64)
+DEF_HELPER_FLAGS_2(fcvt_wu_h, TCG_CALL_NO_RWG, tl, env, i64)
+DEF_HELPER_FLAGS_2(fcvt_l_h, TCG_CALL_NO_RWG, tl, env, i64)
+DEF_HELPER_FLAGS_2(fcvt_lu_h, TCG_CALL_NO_RWG, tl, env, i64)
+DEF_HELPER_FLAGS_2(fcvt_h_w, TCG_CALL_NO_RWG, i64, env, tl)
+DEF_HELPER_FLAGS_2(fcvt_h_wu, TCG_CALL_NO_RWG, i64, env, tl)
+DEF_HELPER_FLAGS_2(fcvt_h_l, TCG_CALL_NO_RWG, i64, env, tl)
+DEF_HELPER_FLAGS_2(fcvt_h_lu, TCG_CALL_NO_RWG, i64, env, tl)
/* Special functions */
DEF_HELPER_2(csrr, tl, env, int)
diff --git a/target/riscv/insn32.decode b/target/riscv/insn32.decode
index 66c231a3010..ba40f3e7f89 100644
--- a/target/riscv/insn32.decode
+++ b/target/riscv/insn32.decode
@@ -739,5 +739,24 @@ fsub_h 0000110 ..... ..... ... ..... 1010011 @r_rm
fmul_h 0001010 ..... ..... ... ..... 1010011 @r_rm
fdiv_h 0001110 ..... ..... ... ..... 1010011 @r_rm
fsqrt_h 0101110 00000 ..... ... ..... 1010011 @r2_rm
+fsgnj_h 0010010 ..... ..... 000 ..... 1010011 @r
+fsgnjn_h 0010010 ..... ..... 001 ..... 1010011 @r
+fsgnjx_h 0010010 ..... ..... 010 ..... 1010011 @r
fmin_h 0010110 ..... ..... 000 ..... 1010011 @r
fmax_h 0010110 ..... ..... 001 ..... 1010011 @r
+fcvt_h_s 0100010 00000 ..... ... ..... 1010011 @r2_rm
+fcvt_s_h 0100000 00010 ..... ... ..... 1010011 @r2_rm
+fcvt_h_d 0100010 00001 ..... ... ..... 1010011 @r2_rm
+fcvt_d_h 0100001 00010 ..... ... ..... 1010011 @r2_rm
+fcvt_w_h 1100010 00000 ..... ... ..... 1010011 @r2_rm
+fcvt_wu_h 1100010 00001 ..... ... ..... 1010011 @r2_rm
+fmv_x_h 1110010 00000 ..... 000 ..... 1010011 @r2
+fcvt_h_w 1101010 00000 ..... ... ..... 1010011 @r2_rm
+fcvt_h_wu 1101010 00001 ..... ... ..... 1010011 @r2_rm
+fmv_h_x 1111010 00000 ..... 000 ..... 1010011 @r2
+
+# *** RV64 Zfh Extension (in addition to RV32 Zfh) ***
+fcvt_l_h 1100010 00010 ..... ... ..... 1010011 @r2_rm
+fcvt_lu_h 1100010 00011 ..... ... ..... 1010011 @r2_rm
+fcvt_h_l 1101010 00010 ..... ... ..... 1010011 @r2_rm
+fcvt_h_lu 1101010 00011 ..... ... ..... 1010011 @r2_rm
diff --git a/target/riscv/insn_trans/trans_rvzfh.c.inc b/target/riscv/insn_trans/trans_rvzfh.c.inc
index 9764d76f8bc..d1250257666 100644
--- a/target/riscv/insn_trans/trans_rvzfh.c.inc
+++ b/target/riscv/insn_trans/trans_rvzfh.c.inc
@@ -171,6 +171,93 @@ static bool trans_fsqrt_h(DisasContext *ctx, arg_fsqrt_h *a)
return true;
}
+static bool trans_fsgnj_h(DisasContext *ctx, arg_fsgnj_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ if (a->rs1 == a->rs2) { /* FMOV */
+ gen_check_nanbox_h(cpu_fpr[a->rd], cpu_fpr[a->rs1]);
+ } else {
+ TCGv_i64 rs1 = tcg_temp_new_i64();
+ TCGv_i64 rs2 = tcg_temp_new_i64();
+
+ gen_check_nanbox_h(rs1, cpu_fpr[a->rs1]);
+ gen_check_nanbox_h(rs2, cpu_fpr[a->rs2]);
+
+ /* This formulation retains the nanboxing of rs2. */
+ tcg_gen_deposit_i64(cpu_fpr[a->rd], rs2, rs1, 0, 15);
+ tcg_temp_free_i64(rs1);
+ tcg_temp_free_i64(rs2);
+ }
+
+ mark_fs_dirty(ctx);
+ return true;
+}
+
+static bool trans_fsgnjn_h(DisasContext *ctx, arg_fsgnjn_h *a)
+{
+ TCGv_i64 rs1, rs2, mask;
+
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ rs1 = tcg_temp_new_i64();
+ gen_check_nanbox_h(rs1, cpu_fpr[a->rs1]);
+
+ if (a->rs1 == a->rs2) { /* FNEG */
+ tcg_gen_xori_i64(cpu_fpr[a->rd], rs1, MAKE_64BIT_MASK(15, 1));
+ } else {
+ rs2 = tcg_temp_new_i64();
+ gen_check_nanbox_h(rs2, cpu_fpr[a->rs2]);
+
+ /*
+ * Replace bit 15 in rs1 with inverse in rs2.
+ * This formulation retains the nanboxing of rs1.
+ */
+ mask = tcg_const_i64(~MAKE_64BIT_MASK(15, 1));
+ tcg_gen_not_i64(rs2, rs2);
+ tcg_gen_andc_i64(rs2, rs2, mask);
+ tcg_gen_and_i64(rs1, mask, rs1);
+ tcg_gen_or_i64(cpu_fpr[a->rd], rs1, rs2);
+
+ tcg_temp_free_i64(mask);
+ tcg_temp_free_i64(rs2);
+ }
+ mark_fs_dirty(ctx);
+ return true;
+}
+
+static bool trans_fsgnjx_h(DisasContext *ctx, arg_fsgnjx_h *a)
+{
+ TCGv_i64 rs1, rs2;
+
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ rs1 = tcg_temp_new_i64();
+ gen_check_nanbox_s(rs1, cpu_fpr[a->rs1]);
+
+ if (a->rs1 == a->rs2) { /* FABS */
+ tcg_gen_andi_i64(cpu_fpr[a->rd], rs1, ~MAKE_64BIT_MASK(15, 1));
+ } else {
+ rs2 = tcg_temp_new_i64();
+ gen_check_nanbox_s(rs2, cpu_fpr[a->rs2]);
+
+ /*
+ * Xor bit 15 in rs1 with that in rs2.
+ * This formulation retains the nanboxing of rs1.
+ */
+ tcg_gen_andi_i64(rs2, rs2, MAKE_64BIT_MASK(15, 1));
+ tcg_gen_xor_i64(cpu_fpr[a->rd], rs1, rs2);
+
+ tcg_temp_free_i64(rs2);
+ }
+
+ mark_fs_dirty(ctx);
+ return true;
+}
+
static bool trans_fmin_h(DisasContext *ctx, arg_fmin_h *a)
{
REQUIRE_FPU;
@@ -192,3 +279,204 @@ static bool trans_fmax_h(DisasContext *ctx, arg_fmax_h *a)
mark_fs_dirty(ctx);
return true;
}
+
+static bool trans_fcvt_s_h(DisasContext *ctx, arg_fcvt_s_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fcvt_s_h(cpu_fpr[a->rd], cpu_env, cpu_fpr[a->rs1]);
+
+ mark_fs_dirty(ctx);
+
+ return true;
+}
+
+static bool trans_fcvt_d_h(DisasContext *ctx, arg_fcvt_d_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+ REQUIRE_EXT(ctx, RVD);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fcvt_d_h(cpu_fpr[a->rd], cpu_env, cpu_fpr[a->rs1]);
+
+ mark_fs_dirty(ctx);
+
+
+ return true;
+}
+
+static bool trans_fcvt_h_s(DisasContext *ctx, arg_fcvt_h_s *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fcvt_h_s(cpu_fpr[a->rd], cpu_env, cpu_fpr[a->rs1]);
+
+ mark_fs_dirty(ctx);
+
+ return true;
+}
+
+static bool trans_fcvt_h_d(DisasContext *ctx, arg_fcvt_h_d *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+ REQUIRE_EXT(ctx, RVD);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fcvt_h_d(cpu_fpr[a->rd], cpu_env, cpu_fpr[a->rs1]);
+
+ mark_fs_dirty(ctx);
+
+ return true;
+}
+
+static bool trans_fcvt_w_h(DisasContext *ctx, arg_fcvt_w_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ TCGv dest = dest_gpr(ctx, a->rd);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fcvt_w_h(dest, cpu_env, cpu_fpr[a->rs1]);
+ gen_set_gpr(ctx, a->rd, dest);
+ return true;
+}
+
+static bool trans_fcvt_wu_h(DisasContext *ctx, arg_fcvt_wu_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ TCGv dest = dest_gpr(ctx, a->rd);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fcvt_wu_h(dest, cpu_env, cpu_fpr[a->rs1]);
+ gen_set_gpr(ctx, a->rd, dest);
+ return true;
+}
+
+static bool trans_fcvt_h_w(DisasContext *ctx, arg_fcvt_h_w *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ TCGv t0 = get_gpr(ctx, a->rs1, EXT_SIGN);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fcvt_h_w(cpu_fpr[a->rd], cpu_env, t0);
+
+ mark_fs_dirty(ctx);
+ return true;
+}
+
+static bool trans_fcvt_h_wu(DisasContext *ctx, arg_fcvt_h_wu *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ TCGv t0 = get_gpr(ctx, a->rs1, EXT_SIGN);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fcvt_h_wu(cpu_fpr[a->rd], cpu_env, t0);
+
+ mark_fs_dirty(ctx);
+ return true;
+}
+
+static bool trans_fmv_x_h(DisasContext *ctx, arg_fmv_x_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ TCGv dest = dest_gpr(ctx, a->rd);
+
+#if defined(TARGET_RISCV64)
+ /* 16 bits -> 64 bits */
+ tcg_gen_ext16s_tl(dest, cpu_fpr[a->rs1]);
+#else
+ /* 16 bits -> 32 bits */
+ tcg_gen_extrl_i64_i32(dest, cpu_fpr[a->rs1]);
+ tcg_gen_ext16s_tl(dest, dest);
+#endif
+
+ gen_set_gpr(ctx, a->rd, dest);
+ return true;
+}
+
+static bool trans_fmv_h_x(DisasContext *ctx, arg_fmv_h_x *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ TCGv t0 = get_gpr(ctx, a->rs1, EXT_ZERO);
+
+ tcg_gen_extu_tl_i64(cpu_fpr[a->rd], t0);
+ gen_nanbox_h(cpu_fpr[a->rd], cpu_fpr[a->rd]);
+
+ mark_fs_dirty(ctx);
+ return true;
+}
+
+static bool trans_fcvt_l_h(DisasContext *ctx, arg_fcvt_l_h *a)
+{
+ REQUIRE_64BIT(ctx);
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ TCGv dest = dest_gpr(ctx, a->rd);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fcvt_l_h(dest, cpu_env, cpu_fpr[a->rs1]);
+ gen_set_gpr(ctx, a->rd, dest);
+ return true;
+}
+
+static bool trans_fcvt_lu_h(DisasContext *ctx, arg_fcvt_lu_h *a)
+{
+ REQUIRE_64BIT(ctx);
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ TCGv dest = dest_gpr(ctx, a->rd);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fcvt_lu_h(dest, cpu_env, cpu_fpr[a->rs1]);
+ gen_set_gpr(ctx, a->rd, dest);
+ return true;
+}
+
+static bool trans_fcvt_h_l(DisasContext *ctx, arg_fcvt_h_l *a)
+{
+ REQUIRE_64BIT(ctx);
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ TCGv t0 = get_gpr(ctx, a->rs1, EXT_SIGN);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fcvt_h_l(cpu_fpr[a->rd], cpu_env, t0);
+
+ mark_fs_dirty(ctx);
+ return true;
+}
+
+static bool trans_fcvt_h_lu(DisasContext *ctx, arg_fcvt_h_lu *a)
+{
+ REQUIRE_64BIT(ctx);
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ TCGv t0 = get_gpr(ctx, a->rs1, EXT_SIGN);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fcvt_h_lu(cpu_fpr[a->rd], cpu_env, t0);
+
+ mark_fs_dirty(ctx);
+ return true;
+}
diff --git a/target/riscv/translate.c b/target/riscv/translate.c
index ad963d1898f..e6643a7d777 100644
--- a/target/riscv/translate.c
+++ b/target/riscv/translate.c
@@ -141,6 +141,16 @@ static void gen_nanbox_h(TCGv_i64 out, TCGv_i64 in)
*
* Here, the result is always nan-boxed, even the canonical nan.
*/
+static void gen_check_nanbox_h(TCGv_i64 out, TCGv_i64 in)
+{
+ TCGv_i64 t_max = tcg_const_i64(0xffffffffffff0000ull);
+ TCGv_i64 t_nan = tcg_const_i64(0xffffffffffff7e00ull);
+
+ tcg_gen_movcond_i64(TCG_COND_GEU, out, in, t_max, in, t_nan);
+ tcg_temp_free_i64(t_max);
+ tcg_temp_free_i64(t_nan);
+}
+
static void gen_check_nanbox_s(TCGv_i64 out, TCGv_i64 in)
{
TCGv_i64 t_max = tcg_constant_i64(0xffffffff00000000ull);
--
2.25.1
^ permalink raw reply related [flat|nested] 23+ messages in thread
* [PATCH v5 3/8] target/riscv: zfh: half-precision convert and move
@ 2021-10-21 16:29 ` frank.chang
0 siblings, 0 replies; 23+ messages in thread
From: frank.chang @ 2021-10-21 16:29 UTC (permalink / raw)
To: qemu-riscv
Cc: Kito Cheng, Chih-Min Chao, Frank Chang, Richard Henderson,
Alistair Francis, Palmer Dabbelt, Bin Meng, qemu-devel
From: Kito Cheng <kito.cheng@sifive.com>
Signed-off-by: Kito Cheng <kito.cheng@sifive.com>
Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com>
Signed-off-by: Frank Chang <frank.chang@sifive.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Acked-by: Alistair Francis <alistair.francis@wdc.com>
---
target/riscv/fpu_helper.c | 67 +++++
target/riscv/helper.h | 12 +
target/riscv/insn32.decode | 19 ++
target/riscv/insn_trans/trans_rvzfh.c.inc | 288 ++++++++++++++++++++++
target/riscv/translate.c | 10 +
5 files changed, 396 insertions(+)
diff --git a/target/riscv/fpu_helper.c b/target/riscv/fpu_helper.c
index 20bb89ad14f..2ed9b03193c 100644
--- a/target/riscv/fpu_helper.c
+++ b/target/riscv/fpu_helper.c
@@ -460,3 +460,70 @@ uint64_t helper_fsqrt_h(CPURISCVState *env, uint64_t rs1)
float16 frs1 = check_nanbox_h(rs1);
return nanbox_h(float16_sqrt(frs1, &env->fp_status));
}
+
+target_ulong helper_fcvt_w_h(CPURISCVState *env, uint64_t rs1)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ return float16_to_int32(frs1, &env->fp_status);
+}
+
+target_ulong helper_fcvt_wu_h(CPURISCVState *env, uint64_t rs1)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ return (int32_t)float16_to_uint32(frs1, &env->fp_status);
+}
+
+target_ulong helper_fcvt_l_h(CPURISCVState *env, uint64_t rs1)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ return float16_to_int64(frs1, &env->fp_status);
+}
+
+target_ulong helper_fcvt_lu_h(CPURISCVState *env, uint64_t rs1)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ return float16_to_uint64(frs1, &env->fp_status);
+}
+
+uint64_t helper_fcvt_h_w(CPURISCVState *env, target_ulong rs1)
+{
+ return nanbox_h(int32_to_float16((int32_t)rs1, &env->fp_status));
+}
+
+uint64_t helper_fcvt_h_wu(CPURISCVState *env, target_ulong rs1)
+{
+ return nanbox_h(uint32_to_float16((uint32_t)rs1, &env->fp_status));
+}
+
+uint64_t helper_fcvt_h_l(CPURISCVState *env, target_ulong rs1)
+{
+ return nanbox_h(int64_to_float16(rs1, &env->fp_status));
+}
+
+uint64_t helper_fcvt_h_lu(CPURISCVState *env, target_ulong rs1)
+{
+ return nanbox_h(uint64_to_float16(rs1, &env->fp_status));
+}
+
+uint64_t helper_fcvt_h_s(CPURISCVState *env, uint64_t rs1)
+{
+ float32 frs1 = check_nanbox_s(rs1);
+ return nanbox_h(float32_to_float16(frs1, true, &env->fp_status));
+}
+
+uint64_t helper_fcvt_s_h(CPURISCVState *env, uint64_t rs1)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ return nanbox_s(float16_to_float32(frs1, true, &env->fp_status));
+}
+
+uint64_t helper_fcvt_h_d(CPURISCVState *env, uint64_t rs1)
+{
+ return nanbox_h(float64_to_float16(rs1, true, &env->fp_status));
+}
+
+uint64_t helper_fcvt_d_h(CPURISCVState *env, uint64_t rs1)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ return float16_to_float64(frs1, true, &env->fp_status);
+}
diff --git a/target/riscv/helper.h b/target/riscv/helper.h
index c6c0323fafc..b50672d1684 100644
--- a/target/riscv/helper.h
+++ b/target/riscv/helper.h
@@ -74,6 +74,18 @@ DEF_HELPER_FLAGS_3(fdiv_h, TCG_CALL_NO_RWG, i64, env, i64, i64)
DEF_HELPER_FLAGS_3(fmin_h, TCG_CALL_NO_RWG, i64, env, i64, i64)
DEF_HELPER_FLAGS_3(fmax_h, TCG_CALL_NO_RWG, i64, env, i64, i64)
DEF_HELPER_FLAGS_2(fsqrt_h, TCG_CALL_NO_RWG, i64, env, i64)
+DEF_HELPER_FLAGS_2(fcvt_s_h, TCG_CALL_NO_RWG, i64, env, i64)
+DEF_HELPER_FLAGS_2(fcvt_h_s, TCG_CALL_NO_RWG, i64, env, i64)
+DEF_HELPER_FLAGS_2(fcvt_d_h, TCG_CALL_NO_RWG, i64, env, i64)
+DEF_HELPER_FLAGS_2(fcvt_h_d, TCG_CALL_NO_RWG, i64, env, i64)
+DEF_HELPER_FLAGS_2(fcvt_w_h, TCG_CALL_NO_RWG, tl, env, i64)
+DEF_HELPER_FLAGS_2(fcvt_wu_h, TCG_CALL_NO_RWG, tl, env, i64)
+DEF_HELPER_FLAGS_2(fcvt_l_h, TCG_CALL_NO_RWG, tl, env, i64)
+DEF_HELPER_FLAGS_2(fcvt_lu_h, TCG_CALL_NO_RWG, tl, env, i64)
+DEF_HELPER_FLAGS_2(fcvt_h_w, TCG_CALL_NO_RWG, i64, env, tl)
+DEF_HELPER_FLAGS_2(fcvt_h_wu, TCG_CALL_NO_RWG, i64, env, tl)
+DEF_HELPER_FLAGS_2(fcvt_h_l, TCG_CALL_NO_RWG, i64, env, tl)
+DEF_HELPER_FLAGS_2(fcvt_h_lu, TCG_CALL_NO_RWG, i64, env, tl)
/* Special functions */
DEF_HELPER_2(csrr, tl, env, int)
diff --git a/target/riscv/insn32.decode b/target/riscv/insn32.decode
index 66c231a3010..ba40f3e7f89 100644
--- a/target/riscv/insn32.decode
+++ b/target/riscv/insn32.decode
@@ -739,5 +739,24 @@ fsub_h 0000110 ..... ..... ... ..... 1010011 @r_rm
fmul_h 0001010 ..... ..... ... ..... 1010011 @r_rm
fdiv_h 0001110 ..... ..... ... ..... 1010011 @r_rm
fsqrt_h 0101110 00000 ..... ... ..... 1010011 @r2_rm
+fsgnj_h 0010010 ..... ..... 000 ..... 1010011 @r
+fsgnjn_h 0010010 ..... ..... 001 ..... 1010011 @r
+fsgnjx_h 0010010 ..... ..... 010 ..... 1010011 @r
fmin_h 0010110 ..... ..... 000 ..... 1010011 @r
fmax_h 0010110 ..... ..... 001 ..... 1010011 @r
+fcvt_h_s 0100010 00000 ..... ... ..... 1010011 @r2_rm
+fcvt_s_h 0100000 00010 ..... ... ..... 1010011 @r2_rm
+fcvt_h_d 0100010 00001 ..... ... ..... 1010011 @r2_rm
+fcvt_d_h 0100001 00010 ..... ... ..... 1010011 @r2_rm
+fcvt_w_h 1100010 00000 ..... ... ..... 1010011 @r2_rm
+fcvt_wu_h 1100010 00001 ..... ... ..... 1010011 @r2_rm
+fmv_x_h 1110010 00000 ..... 000 ..... 1010011 @r2
+fcvt_h_w 1101010 00000 ..... ... ..... 1010011 @r2_rm
+fcvt_h_wu 1101010 00001 ..... ... ..... 1010011 @r2_rm
+fmv_h_x 1111010 00000 ..... 000 ..... 1010011 @r2
+
+# *** RV64 Zfh Extension (in addition to RV32 Zfh) ***
+fcvt_l_h 1100010 00010 ..... ... ..... 1010011 @r2_rm
+fcvt_lu_h 1100010 00011 ..... ... ..... 1010011 @r2_rm
+fcvt_h_l 1101010 00010 ..... ... ..... 1010011 @r2_rm
+fcvt_h_lu 1101010 00011 ..... ... ..... 1010011 @r2_rm
diff --git a/target/riscv/insn_trans/trans_rvzfh.c.inc b/target/riscv/insn_trans/trans_rvzfh.c.inc
index 9764d76f8bc..d1250257666 100644
--- a/target/riscv/insn_trans/trans_rvzfh.c.inc
+++ b/target/riscv/insn_trans/trans_rvzfh.c.inc
@@ -171,6 +171,93 @@ static bool trans_fsqrt_h(DisasContext *ctx, arg_fsqrt_h *a)
return true;
}
+static bool trans_fsgnj_h(DisasContext *ctx, arg_fsgnj_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ if (a->rs1 == a->rs2) { /* FMOV */
+ gen_check_nanbox_h(cpu_fpr[a->rd], cpu_fpr[a->rs1]);
+ } else {
+ TCGv_i64 rs1 = tcg_temp_new_i64();
+ TCGv_i64 rs2 = tcg_temp_new_i64();
+
+ gen_check_nanbox_h(rs1, cpu_fpr[a->rs1]);
+ gen_check_nanbox_h(rs2, cpu_fpr[a->rs2]);
+
+ /* This formulation retains the nanboxing of rs2. */
+ tcg_gen_deposit_i64(cpu_fpr[a->rd], rs2, rs1, 0, 15);
+ tcg_temp_free_i64(rs1);
+ tcg_temp_free_i64(rs2);
+ }
+
+ mark_fs_dirty(ctx);
+ return true;
+}
+
+static bool trans_fsgnjn_h(DisasContext *ctx, arg_fsgnjn_h *a)
+{
+ TCGv_i64 rs1, rs2, mask;
+
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ rs1 = tcg_temp_new_i64();
+ gen_check_nanbox_h(rs1, cpu_fpr[a->rs1]);
+
+ if (a->rs1 == a->rs2) { /* FNEG */
+ tcg_gen_xori_i64(cpu_fpr[a->rd], rs1, MAKE_64BIT_MASK(15, 1));
+ } else {
+ rs2 = tcg_temp_new_i64();
+ gen_check_nanbox_h(rs2, cpu_fpr[a->rs2]);
+
+ /*
+ * Replace bit 15 in rs1 with inverse in rs2.
+ * This formulation retains the nanboxing of rs1.
+ */
+ mask = tcg_const_i64(~MAKE_64BIT_MASK(15, 1));
+ tcg_gen_not_i64(rs2, rs2);
+ tcg_gen_andc_i64(rs2, rs2, mask);
+ tcg_gen_and_i64(rs1, mask, rs1);
+ tcg_gen_or_i64(cpu_fpr[a->rd], rs1, rs2);
+
+ tcg_temp_free_i64(mask);
+ tcg_temp_free_i64(rs2);
+ }
+ mark_fs_dirty(ctx);
+ return true;
+}
+
+static bool trans_fsgnjx_h(DisasContext *ctx, arg_fsgnjx_h *a)
+{
+ TCGv_i64 rs1, rs2;
+
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ rs1 = tcg_temp_new_i64();
+ gen_check_nanbox_s(rs1, cpu_fpr[a->rs1]);
+
+ if (a->rs1 == a->rs2) { /* FABS */
+ tcg_gen_andi_i64(cpu_fpr[a->rd], rs1, ~MAKE_64BIT_MASK(15, 1));
+ } else {
+ rs2 = tcg_temp_new_i64();
+ gen_check_nanbox_s(rs2, cpu_fpr[a->rs2]);
+
+ /*
+ * Xor bit 15 in rs1 with that in rs2.
+ * This formulation retains the nanboxing of rs1.
+ */
+ tcg_gen_andi_i64(rs2, rs2, MAKE_64BIT_MASK(15, 1));
+ tcg_gen_xor_i64(cpu_fpr[a->rd], rs1, rs2);
+
+ tcg_temp_free_i64(rs2);
+ }
+
+ mark_fs_dirty(ctx);
+ return true;
+}
+
static bool trans_fmin_h(DisasContext *ctx, arg_fmin_h *a)
{
REQUIRE_FPU;
@@ -192,3 +279,204 @@ static bool trans_fmax_h(DisasContext *ctx, arg_fmax_h *a)
mark_fs_dirty(ctx);
return true;
}
+
+static bool trans_fcvt_s_h(DisasContext *ctx, arg_fcvt_s_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fcvt_s_h(cpu_fpr[a->rd], cpu_env, cpu_fpr[a->rs1]);
+
+ mark_fs_dirty(ctx);
+
+ return true;
+}
+
+static bool trans_fcvt_d_h(DisasContext *ctx, arg_fcvt_d_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+ REQUIRE_EXT(ctx, RVD);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fcvt_d_h(cpu_fpr[a->rd], cpu_env, cpu_fpr[a->rs1]);
+
+ mark_fs_dirty(ctx);
+
+
+ return true;
+}
+
+static bool trans_fcvt_h_s(DisasContext *ctx, arg_fcvt_h_s *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fcvt_h_s(cpu_fpr[a->rd], cpu_env, cpu_fpr[a->rs1]);
+
+ mark_fs_dirty(ctx);
+
+ return true;
+}
+
+static bool trans_fcvt_h_d(DisasContext *ctx, arg_fcvt_h_d *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+ REQUIRE_EXT(ctx, RVD);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fcvt_h_d(cpu_fpr[a->rd], cpu_env, cpu_fpr[a->rs1]);
+
+ mark_fs_dirty(ctx);
+
+ return true;
+}
+
+static bool trans_fcvt_w_h(DisasContext *ctx, arg_fcvt_w_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ TCGv dest = dest_gpr(ctx, a->rd);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fcvt_w_h(dest, cpu_env, cpu_fpr[a->rs1]);
+ gen_set_gpr(ctx, a->rd, dest);
+ return true;
+}
+
+static bool trans_fcvt_wu_h(DisasContext *ctx, arg_fcvt_wu_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ TCGv dest = dest_gpr(ctx, a->rd);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fcvt_wu_h(dest, cpu_env, cpu_fpr[a->rs1]);
+ gen_set_gpr(ctx, a->rd, dest);
+ return true;
+}
+
+static bool trans_fcvt_h_w(DisasContext *ctx, arg_fcvt_h_w *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ TCGv t0 = get_gpr(ctx, a->rs1, EXT_SIGN);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fcvt_h_w(cpu_fpr[a->rd], cpu_env, t0);
+
+ mark_fs_dirty(ctx);
+ return true;
+}
+
+static bool trans_fcvt_h_wu(DisasContext *ctx, arg_fcvt_h_wu *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ TCGv t0 = get_gpr(ctx, a->rs1, EXT_SIGN);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fcvt_h_wu(cpu_fpr[a->rd], cpu_env, t0);
+
+ mark_fs_dirty(ctx);
+ return true;
+}
+
+static bool trans_fmv_x_h(DisasContext *ctx, arg_fmv_x_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ TCGv dest = dest_gpr(ctx, a->rd);
+
+#if defined(TARGET_RISCV64)
+ /* 16 bits -> 64 bits */
+ tcg_gen_ext16s_tl(dest, cpu_fpr[a->rs1]);
+#else
+ /* 16 bits -> 32 bits */
+ tcg_gen_extrl_i64_i32(dest, cpu_fpr[a->rs1]);
+ tcg_gen_ext16s_tl(dest, dest);
+#endif
+
+ gen_set_gpr(ctx, a->rd, dest);
+ return true;
+}
+
+static bool trans_fmv_h_x(DisasContext *ctx, arg_fmv_h_x *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ TCGv t0 = get_gpr(ctx, a->rs1, EXT_ZERO);
+
+ tcg_gen_extu_tl_i64(cpu_fpr[a->rd], t0);
+ gen_nanbox_h(cpu_fpr[a->rd], cpu_fpr[a->rd]);
+
+ mark_fs_dirty(ctx);
+ return true;
+}
+
+static bool trans_fcvt_l_h(DisasContext *ctx, arg_fcvt_l_h *a)
+{
+ REQUIRE_64BIT(ctx);
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ TCGv dest = dest_gpr(ctx, a->rd);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fcvt_l_h(dest, cpu_env, cpu_fpr[a->rs1]);
+ gen_set_gpr(ctx, a->rd, dest);
+ return true;
+}
+
+static bool trans_fcvt_lu_h(DisasContext *ctx, arg_fcvt_lu_h *a)
+{
+ REQUIRE_64BIT(ctx);
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ TCGv dest = dest_gpr(ctx, a->rd);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fcvt_lu_h(dest, cpu_env, cpu_fpr[a->rs1]);
+ gen_set_gpr(ctx, a->rd, dest);
+ return true;
+}
+
+static bool trans_fcvt_h_l(DisasContext *ctx, arg_fcvt_h_l *a)
+{
+ REQUIRE_64BIT(ctx);
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ TCGv t0 = get_gpr(ctx, a->rs1, EXT_SIGN);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fcvt_h_l(cpu_fpr[a->rd], cpu_env, t0);
+
+ mark_fs_dirty(ctx);
+ return true;
+}
+
+static bool trans_fcvt_h_lu(DisasContext *ctx, arg_fcvt_h_lu *a)
+{
+ REQUIRE_64BIT(ctx);
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ TCGv t0 = get_gpr(ctx, a->rs1, EXT_SIGN);
+
+ gen_set_rm(ctx, a->rm);
+ gen_helper_fcvt_h_lu(cpu_fpr[a->rd], cpu_env, t0);
+
+ mark_fs_dirty(ctx);
+ return true;
+}
diff --git a/target/riscv/translate.c b/target/riscv/translate.c
index ad963d1898f..e6643a7d777 100644
--- a/target/riscv/translate.c
+++ b/target/riscv/translate.c
@@ -141,6 +141,16 @@ static void gen_nanbox_h(TCGv_i64 out, TCGv_i64 in)
*
* Here, the result is always nan-boxed, even the canonical nan.
*/
+static void gen_check_nanbox_h(TCGv_i64 out, TCGv_i64 in)
+{
+ TCGv_i64 t_max = tcg_const_i64(0xffffffffffff0000ull);
+ TCGv_i64 t_nan = tcg_const_i64(0xffffffffffff7e00ull);
+
+ tcg_gen_movcond_i64(TCG_COND_GEU, out, in, t_max, in, t_nan);
+ tcg_temp_free_i64(t_max);
+ tcg_temp_free_i64(t_nan);
+}
+
static void gen_check_nanbox_s(TCGv_i64 out, TCGv_i64 in)
{
TCGv_i64 t_max = tcg_constant_i64(0xffffffff00000000ull);
--
2.25.1
^ permalink raw reply related [flat|nested] 23+ messages in thread
* [PATCH v5 4/8] target/riscv: zfh: half-precision floating-point compare
2021-10-21 16:29 [PATCH v5 0/8] target/riscv: support Zfh, Zfhmin extension v0.1 frank.chang
@ 2021-10-21 16:29 ` frank.chang
2021-10-21 16:29 ` frank.chang
` (7 subsequent siblings)
8 siblings, 0 replies; 23+ messages in thread
From: frank.chang @ 2021-10-21 16:29 UTC (permalink / raw)
To: qemu-riscv
Cc: Frank Chang, Bin Meng, Richard Henderson, qemu-devel,
Chih-Min Chao, Palmer Dabbelt, Alistair Francis, Kito Cheng
From: Kito Cheng <kito.cheng@sifive.com>
Signed-off-by: Kito Cheng <kito.cheng@sifive.com>
Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com>
Signed-off-by: Frank Chang <frank.chang@sifive.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
---
target/riscv/fpu_helper.c | 21 +++++++++++++
target/riscv/helper.h | 3 ++
target/riscv/insn32.decode | 3 ++
target/riscv/insn_trans/trans_rvzfh.c.inc | 37 +++++++++++++++++++++++
4 files changed, 64 insertions(+)
diff --git a/target/riscv/fpu_helper.c b/target/riscv/fpu_helper.c
index 2ed9b03193c..ec2009ee65b 100644
--- a/target/riscv/fpu_helper.c
+++ b/target/riscv/fpu_helper.c
@@ -461,6 +461,27 @@ uint64_t helper_fsqrt_h(CPURISCVState *env, uint64_t rs1)
return nanbox_h(float16_sqrt(frs1, &env->fp_status));
}
+target_ulong helper_fle_h(CPURISCVState *env, uint64_t rs1, uint64_t rs2)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ float16 frs2 = check_nanbox_h(rs2);
+ return float16_le(frs1, frs2, &env->fp_status);
+}
+
+target_ulong helper_flt_h(CPURISCVState *env, uint64_t rs1, uint64_t rs2)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ float16 frs2 = check_nanbox_h(rs2);
+ return float16_lt(frs1, frs2, &env->fp_status);
+}
+
+target_ulong helper_feq_h(CPURISCVState *env, uint64_t rs1, uint64_t rs2)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ float16 frs2 = check_nanbox_h(rs2);
+ return float16_eq_quiet(frs1, frs2, &env->fp_status);
+}
+
target_ulong helper_fcvt_w_h(CPURISCVState *env, uint64_t rs1)
{
float16 frs1 = check_nanbox_h(rs1);
diff --git a/target/riscv/helper.h b/target/riscv/helper.h
index b50672d1684..9c89521d4ad 100644
--- a/target/riscv/helper.h
+++ b/target/riscv/helper.h
@@ -74,6 +74,9 @@ DEF_HELPER_FLAGS_3(fdiv_h, TCG_CALL_NO_RWG, i64, env, i64, i64)
DEF_HELPER_FLAGS_3(fmin_h, TCG_CALL_NO_RWG, i64, env, i64, i64)
DEF_HELPER_FLAGS_3(fmax_h, TCG_CALL_NO_RWG, i64, env, i64, i64)
DEF_HELPER_FLAGS_2(fsqrt_h, TCG_CALL_NO_RWG, i64, env, i64)
+DEF_HELPER_FLAGS_3(fle_h, TCG_CALL_NO_RWG, tl, env, i64, i64)
+DEF_HELPER_FLAGS_3(flt_h, TCG_CALL_NO_RWG, tl, env, i64, i64)
+DEF_HELPER_FLAGS_3(feq_h, TCG_CALL_NO_RWG, tl, env, i64, i64)
DEF_HELPER_FLAGS_2(fcvt_s_h, TCG_CALL_NO_RWG, i64, env, i64)
DEF_HELPER_FLAGS_2(fcvt_h_s, TCG_CALL_NO_RWG, i64, env, i64)
DEF_HELPER_FLAGS_2(fcvt_d_h, TCG_CALL_NO_RWG, i64, env, i64)
diff --git a/target/riscv/insn32.decode b/target/riscv/insn32.decode
index ba40f3e7f89..3906c9fb201 100644
--- a/target/riscv/insn32.decode
+++ b/target/riscv/insn32.decode
@@ -751,6 +751,9 @@ fcvt_d_h 0100001 00010 ..... ... ..... 1010011 @r2_rm
fcvt_w_h 1100010 00000 ..... ... ..... 1010011 @r2_rm
fcvt_wu_h 1100010 00001 ..... ... ..... 1010011 @r2_rm
fmv_x_h 1110010 00000 ..... 000 ..... 1010011 @r2
+feq_h 1010010 ..... ..... 010 ..... 1010011 @r
+flt_h 1010010 ..... ..... 001 ..... 1010011 @r
+fle_h 1010010 ..... ..... 000 ..... 1010011 @r
fcvt_h_w 1101010 00000 ..... ... ..... 1010011 @r2_rm
fcvt_h_wu 1101010 00001 ..... ... ..... 1010011 @r2_rm
fmv_h_x 1111010 00000 ..... 000 ..... 1010011 @r2
diff --git a/target/riscv/insn_trans/trans_rvzfh.c.inc b/target/riscv/insn_trans/trans_rvzfh.c.inc
index d1250257666..8d0959a6671 100644
--- a/target/riscv/insn_trans/trans_rvzfh.c.inc
+++ b/target/riscv/insn_trans/trans_rvzfh.c.inc
@@ -335,6 +335,43 @@ static bool trans_fcvt_h_d(DisasContext *ctx, arg_fcvt_h_d *a)
return true;
}
+static bool trans_feq_h(DisasContext *ctx, arg_feq_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ TCGv dest = dest_gpr(ctx, a->rd);
+
+ gen_helper_feq_h(dest, cpu_env, cpu_fpr[a->rs1], cpu_fpr[a->rs2]);
+ gen_set_gpr(ctx, a->rd, dest);
+ return true;
+}
+
+static bool trans_flt_h(DisasContext *ctx, arg_flt_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ TCGv dest = dest_gpr(ctx, a->rd);
+
+ gen_helper_flt_h(dest, cpu_env, cpu_fpr[a->rs1], cpu_fpr[a->rs2]);
+ gen_set_gpr(ctx, a->rd, dest);
+
+ return true;
+}
+
+static bool trans_fle_h(DisasContext *ctx, arg_fle_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ TCGv dest = dest_gpr(ctx, a->rd);
+
+ gen_helper_fle_h(dest, cpu_env, cpu_fpr[a->rs1], cpu_fpr[a->rs2]);
+ gen_set_gpr(ctx, a->rd, dest);
+ return true;
+}
+
static bool trans_fcvt_w_h(DisasContext *ctx, arg_fcvt_w_h *a)
{
REQUIRE_FPU;
--
2.25.1
^ permalink raw reply related [flat|nested] 23+ messages in thread
* [PATCH v5 4/8] target/riscv: zfh: half-precision floating-point compare
@ 2021-10-21 16:29 ` frank.chang
0 siblings, 0 replies; 23+ messages in thread
From: frank.chang @ 2021-10-21 16:29 UTC (permalink / raw)
To: qemu-riscv
Cc: Kito Cheng, Chih-Min Chao, Frank Chang, Richard Henderson,
Alistair Francis, Palmer Dabbelt, Bin Meng, qemu-devel
From: Kito Cheng <kito.cheng@sifive.com>
Signed-off-by: Kito Cheng <kito.cheng@sifive.com>
Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com>
Signed-off-by: Frank Chang <frank.chang@sifive.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
---
target/riscv/fpu_helper.c | 21 +++++++++++++
target/riscv/helper.h | 3 ++
target/riscv/insn32.decode | 3 ++
target/riscv/insn_trans/trans_rvzfh.c.inc | 37 +++++++++++++++++++++++
4 files changed, 64 insertions(+)
diff --git a/target/riscv/fpu_helper.c b/target/riscv/fpu_helper.c
index 2ed9b03193c..ec2009ee65b 100644
--- a/target/riscv/fpu_helper.c
+++ b/target/riscv/fpu_helper.c
@@ -461,6 +461,27 @@ uint64_t helper_fsqrt_h(CPURISCVState *env, uint64_t rs1)
return nanbox_h(float16_sqrt(frs1, &env->fp_status));
}
+target_ulong helper_fle_h(CPURISCVState *env, uint64_t rs1, uint64_t rs2)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ float16 frs2 = check_nanbox_h(rs2);
+ return float16_le(frs1, frs2, &env->fp_status);
+}
+
+target_ulong helper_flt_h(CPURISCVState *env, uint64_t rs1, uint64_t rs2)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ float16 frs2 = check_nanbox_h(rs2);
+ return float16_lt(frs1, frs2, &env->fp_status);
+}
+
+target_ulong helper_feq_h(CPURISCVState *env, uint64_t rs1, uint64_t rs2)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ float16 frs2 = check_nanbox_h(rs2);
+ return float16_eq_quiet(frs1, frs2, &env->fp_status);
+}
+
target_ulong helper_fcvt_w_h(CPURISCVState *env, uint64_t rs1)
{
float16 frs1 = check_nanbox_h(rs1);
diff --git a/target/riscv/helper.h b/target/riscv/helper.h
index b50672d1684..9c89521d4ad 100644
--- a/target/riscv/helper.h
+++ b/target/riscv/helper.h
@@ -74,6 +74,9 @@ DEF_HELPER_FLAGS_3(fdiv_h, TCG_CALL_NO_RWG, i64, env, i64, i64)
DEF_HELPER_FLAGS_3(fmin_h, TCG_CALL_NO_RWG, i64, env, i64, i64)
DEF_HELPER_FLAGS_3(fmax_h, TCG_CALL_NO_RWG, i64, env, i64, i64)
DEF_HELPER_FLAGS_2(fsqrt_h, TCG_CALL_NO_RWG, i64, env, i64)
+DEF_HELPER_FLAGS_3(fle_h, TCG_CALL_NO_RWG, tl, env, i64, i64)
+DEF_HELPER_FLAGS_3(flt_h, TCG_CALL_NO_RWG, tl, env, i64, i64)
+DEF_HELPER_FLAGS_3(feq_h, TCG_CALL_NO_RWG, tl, env, i64, i64)
DEF_HELPER_FLAGS_2(fcvt_s_h, TCG_CALL_NO_RWG, i64, env, i64)
DEF_HELPER_FLAGS_2(fcvt_h_s, TCG_CALL_NO_RWG, i64, env, i64)
DEF_HELPER_FLAGS_2(fcvt_d_h, TCG_CALL_NO_RWG, i64, env, i64)
diff --git a/target/riscv/insn32.decode b/target/riscv/insn32.decode
index ba40f3e7f89..3906c9fb201 100644
--- a/target/riscv/insn32.decode
+++ b/target/riscv/insn32.decode
@@ -751,6 +751,9 @@ fcvt_d_h 0100001 00010 ..... ... ..... 1010011 @r2_rm
fcvt_w_h 1100010 00000 ..... ... ..... 1010011 @r2_rm
fcvt_wu_h 1100010 00001 ..... ... ..... 1010011 @r2_rm
fmv_x_h 1110010 00000 ..... 000 ..... 1010011 @r2
+feq_h 1010010 ..... ..... 010 ..... 1010011 @r
+flt_h 1010010 ..... ..... 001 ..... 1010011 @r
+fle_h 1010010 ..... ..... 000 ..... 1010011 @r
fcvt_h_w 1101010 00000 ..... ... ..... 1010011 @r2_rm
fcvt_h_wu 1101010 00001 ..... ... ..... 1010011 @r2_rm
fmv_h_x 1111010 00000 ..... 000 ..... 1010011 @r2
diff --git a/target/riscv/insn_trans/trans_rvzfh.c.inc b/target/riscv/insn_trans/trans_rvzfh.c.inc
index d1250257666..8d0959a6671 100644
--- a/target/riscv/insn_trans/trans_rvzfh.c.inc
+++ b/target/riscv/insn_trans/trans_rvzfh.c.inc
@@ -335,6 +335,43 @@ static bool trans_fcvt_h_d(DisasContext *ctx, arg_fcvt_h_d *a)
return true;
}
+static bool trans_feq_h(DisasContext *ctx, arg_feq_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ TCGv dest = dest_gpr(ctx, a->rd);
+
+ gen_helper_feq_h(dest, cpu_env, cpu_fpr[a->rs1], cpu_fpr[a->rs2]);
+ gen_set_gpr(ctx, a->rd, dest);
+ return true;
+}
+
+static bool trans_flt_h(DisasContext *ctx, arg_flt_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ TCGv dest = dest_gpr(ctx, a->rd);
+
+ gen_helper_flt_h(dest, cpu_env, cpu_fpr[a->rs1], cpu_fpr[a->rs2]);
+ gen_set_gpr(ctx, a->rd, dest);
+
+ return true;
+}
+
+static bool trans_fle_h(DisasContext *ctx, arg_fle_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ TCGv dest = dest_gpr(ctx, a->rd);
+
+ gen_helper_fle_h(dest, cpu_env, cpu_fpr[a->rs1], cpu_fpr[a->rs2]);
+ gen_set_gpr(ctx, a->rd, dest);
+ return true;
+}
+
static bool trans_fcvt_w_h(DisasContext *ctx, arg_fcvt_w_h *a)
{
REQUIRE_FPU;
--
2.25.1
^ permalink raw reply related [flat|nested] 23+ messages in thread
* [PATCH v5 5/8] target/riscv: zfh: half-precision floating-point classify
2021-10-21 16:29 [PATCH v5 0/8] target/riscv: support Zfh, Zfhmin extension v0.1 frank.chang
@ 2021-10-21 16:29 ` frank.chang
2021-10-21 16:29 ` frank.chang
` (7 subsequent siblings)
8 siblings, 0 replies; 23+ messages in thread
From: frank.chang @ 2021-10-21 16:29 UTC (permalink / raw)
To: qemu-riscv
Cc: Frank Chang, Bin Meng, Richard Henderson, qemu-devel,
Chih-Min Chao, Palmer Dabbelt, Alistair Francis, Kito Cheng
From: Kito Cheng <kito.cheng@sifive.com>
Signed-off-by: Kito Cheng <kito.cheng@sifive.com>
Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com>
Signed-off-by: Frank Chang <frank.chang@sifive.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
---
target/riscv/fpu_helper.c | 6 ++++++
target/riscv/helper.h | 1 +
target/riscv/insn32.decode | 1 +
target/riscv/insn_trans/trans_rvzfh.c.inc | 12 ++++++++++++
4 files changed, 20 insertions(+)
diff --git a/target/riscv/fpu_helper.c b/target/riscv/fpu_helper.c
index ec2009ee65b..388e23ca670 100644
--- a/target/riscv/fpu_helper.c
+++ b/target/riscv/fpu_helper.c
@@ -482,6 +482,12 @@ target_ulong helper_feq_h(CPURISCVState *env, uint64_t rs1, uint64_t rs2)
return float16_eq_quiet(frs1, frs2, &env->fp_status);
}
+target_ulong helper_fclass_h(uint64_t rs1)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ return fclass_h(frs1);
+}
+
target_ulong helper_fcvt_w_h(CPURISCVState *env, uint64_t rs1)
{
float16 frs1 = check_nanbox_h(rs1);
diff --git a/target/riscv/helper.h b/target/riscv/helper.h
index 9c89521d4ad..d25cf725c57 100644
--- a/target/riscv/helper.h
+++ b/target/riscv/helper.h
@@ -89,6 +89,7 @@ DEF_HELPER_FLAGS_2(fcvt_h_w, TCG_CALL_NO_RWG, i64, env, tl)
DEF_HELPER_FLAGS_2(fcvt_h_wu, TCG_CALL_NO_RWG, i64, env, tl)
DEF_HELPER_FLAGS_2(fcvt_h_l, TCG_CALL_NO_RWG, i64, env, tl)
DEF_HELPER_FLAGS_2(fcvt_h_lu, TCG_CALL_NO_RWG, i64, env, tl)
+DEF_HELPER_FLAGS_1(fclass_h, TCG_CALL_NO_RWG_SE, tl, i64)
/* Special functions */
DEF_HELPER_2(csrr, tl, env, int)
diff --git a/target/riscv/insn32.decode b/target/riscv/insn32.decode
index 3906c9fb201..6c4cde216bc 100644
--- a/target/riscv/insn32.decode
+++ b/target/riscv/insn32.decode
@@ -754,6 +754,7 @@ fmv_x_h 1110010 00000 ..... 000 ..... 1010011 @r2
feq_h 1010010 ..... ..... 010 ..... 1010011 @r
flt_h 1010010 ..... ..... 001 ..... 1010011 @r
fle_h 1010010 ..... ..... 000 ..... 1010011 @r
+fclass_h 1110010 00000 ..... 001 ..... 1010011 @r2
fcvt_h_w 1101010 00000 ..... ... ..... 1010011 @r2_rm
fcvt_h_wu 1101010 00001 ..... ... ..... 1010011 @r2_rm
fmv_h_x 1111010 00000 ..... 000 ..... 1010011 @r2
diff --git a/target/riscv/insn_trans/trans_rvzfh.c.inc b/target/riscv/insn_trans/trans_rvzfh.c.inc
index 8d0959a6671..0549e25fb45 100644
--- a/target/riscv/insn_trans/trans_rvzfh.c.inc
+++ b/target/riscv/insn_trans/trans_rvzfh.c.inc
@@ -372,6 +372,18 @@ static bool trans_fle_h(DisasContext *ctx, arg_fle_h *a)
return true;
}
+static bool trans_fclass_h(DisasContext *ctx, arg_fclass_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ TCGv dest = dest_gpr(ctx, a->rd);
+
+ gen_helper_fclass_h(dest, cpu_fpr[a->rs1]);
+ gen_set_gpr(ctx, a->rd, dest);
+ return true;
+}
+
static bool trans_fcvt_w_h(DisasContext *ctx, arg_fcvt_w_h *a)
{
REQUIRE_FPU;
--
2.25.1
^ permalink raw reply related [flat|nested] 23+ messages in thread
* [PATCH v5 5/8] target/riscv: zfh: half-precision floating-point classify
@ 2021-10-21 16:29 ` frank.chang
0 siblings, 0 replies; 23+ messages in thread
From: frank.chang @ 2021-10-21 16:29 UTC (permalink / raw)
To: qemu-riscv
Cc: Kito Cheng, Chih-Min Chao, Frank Chang, Richard Henderson,
Alistair Francis, Palmer Dabbelt, Bin Meng, qemu-devel
From: Kito Cheng <kito.cheng@sifive.com>
Signed-off-by: Kito Cheng <kito.cheng@sifive.com>
Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com>
Signed-off-by: Frank Chang <frank.chang@sifive.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
---
target/riscv/fpu_helper.c | 6 ++++++
target/riscv/helper.h | 1 +
target/riscv/insn32.decode | 1 +
target/riscv/insn_trans/trans_rvzfh.c.inc | 12 ++++++++++++
4 files changed, 20 insertions(+)
diff --git a/target/riscv/fpu_helper.c b/target/riscv/fpu_helper.c
index ec2009ee65b..388e23ca670 100644
--- a/target/riscv/fpu_helper.c
+++ b/target/riscv/fpu_helper.c
@@ -482,6 +482,12 @@ target_ulong helper_feq_h(CPURISCVState *env, uint64_t rs1, uint64_t rs2)
return float16_eq_quiet(frs1, frs2, &env->fp_status);
}
+target_ulong helper_fclass_h(uint64_t rs1)
+{
+ float16 frs1 = check_nanbox_h(rs1);
+ return fclass_h(frs1);
+}
+
target_ulong helper_fcvt_w_h(CPURISCVState *env, uint64_t rs1)
{
float16 frs1 = check_nanbox_h(rs1);
diff --git a/target/riscv/helper.h b/target/riscv/helper.h
index 9c89521d4ad..d25cf725c57 100644
--- a/target/riscv/helper.h
+++ b/target/riscv/helper.h
@@ -89,6 +89,7 @@ DEF_HELPER_FLAGS_2(fcvt_h_w, TCG_CALL_NO_RWG, i64, env, tl)
DEF_HELPER_FLAGS_2(fcvt_h_wu, TCG_CALL_NO_RWG, i64, env, tl)
DEF_HELPER_FLAGS_2(fcvt_h_l, TCG_CALL_NO_RWG, i64, env, tl)
DEF_HELPER_FLAGS_2(fcvt_h_lu, TCG_CALL_NO_RWG, i64, env, tl)
+DEF_HELPER_FLAGS_1(fclass_h, TCG_CALL_NO_RWG_SE, tl, i64)
/* Special functions */
DEF_HELPER_2(csrr, tl, env, int)
diff --git a/target/riscv/insn32.decode b/target/riscv/insn32.decode
index 3906c9fb201..6c4cde216bc 100644
--- a/target/riscv/insn32.decode
+++ b/target/riscv/insn32.decode
@@ -754,6 +754,7 @@ fmv_x_h 1110010 00000 ..... 000 ..... 1010011 @r2
feq_h 1010010 ..... ..... 010 ..... 1010011 @r
flt_h 1010010 ..... ..... 001 ..... 1010011 @r
fle_h 1010010 ..... ..... 000 ..... 1010011 @r
+fclass_h 1110010 00000 ..... 001 ..... 1010011 @r2
fcvt_h_w 1101010 00000 ..... ... ..... 1010011 @r2_rm
fcvt_h_wu 1101010 00001 ..... ... ..... 1010011 @r2_rm
fmv_h_x 1111010 00000 ..... 000 ..... 1010011 @r2
diff --git a/target/riscv/insn_trans/trans_rvzfh.c.inc b/target/riscv/insn_trans/trans_rvzfh.c.inc
index 8d0959a6671..0549e25fb45 100644
--- a/target/riscv/insn_trans/trans_rvzfh.c.inc
+++ b/target/riscv/insn_trans/trans_rvzfh.c.inc
@@ -372,6 +372,18 @@ static bool trans_fle_h(DisasContext *ctx, arg_fle_h *a)
return true;
}
+static bool trans_fclass_h(DisasContext *ctx, arg_fclass_h *a)
+{
+ REQUIRE_FPU;
+ REQUIRE_ZFH(ctx);
+
+ TCGv dest = dest_gpr(ctx, a->rd);
+
+ gen_helper_fclass_h(dest, cpu_fpr[a->rs1]);
+ gen_set_gpr(ctx, a->rd, dest);
+ return true;
+}
+
static bool trans_fcvt_w_h(DisasContext *ctx, arg_fcvt_w_h *a)
{
REQUIRE_FPU;
--
2.25.1
^ permalink raw reply related [flat|nested] 23+ messages in thread
* [PATCH v5 6/8] target/riscv: zfh: add Zfh cpu property
2021-10-21 16:29 [PATCH v5 0/8] target/riscv: support Zfh, Zfhmin extension v0.1 frank.chang
@ 2021-10-21 16:29 ` frank.chang
2021-10-21 16:29 ` frank.chang
` (7 subsequent siblings)
8 siblings, 0 replies; 23+ messages in thread
From: frank.chang @ 2021-10-21 16:29 UTC (permalink / raw)
To: qemu-riscv
Cc: Frank Chang, Palmer Dabbelt, Bin Meng, Alistair Francis, qemu-devel
From: Frank Chang <frank.chang@sifive.com>
Signed-off-by: Frank Chang <frank.chang@sifive.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
---
target/riscv/cpu.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
index 788fa0b11c0..1b3a7784b06 100644
--- a/target/riscv/cpu.c
+++ b/target/riscv/cpu.c
@@ -617,6 +617,7 @@ static Property riscv_cpu_properties[] = {
DEFINE_PROP_BOOL("Counters", RISCVCPU, cfg.ext_counters, true),
DEFINE_PROP_BOOL("Zifencei", RISCVCPU, cfg.ext_ifencei, true),
DEFINE_PROP_BOOL("Zicsr", RISCVCPU, cfg.ext_icsr, true),
+ DEFINE_PROP_BOOL("Zfh", RISCVCPU, cfg.ext_zfh, false),
DEFINE_PROP_BOOL("mmu", RISCVCPU, cfg.mmu, true),
DEFINE_PROP_BOOL("pmp", RISCVCPU, cfg.pmp, true),
--
2.25.1
^ permalink raw reply related [flat|nested] 23+ messages in thread
* [PATCH v5 6/8] target/riscv: zfh: add Zfh cpu property
@ 2021-10-21 16:29 ` frank.chang
0 siblings, 0 replies; 23+ messages in thread
From: frank.chang @ 2021-10-21 16:29 UTC (permalink / raw)
To: qemu-riscv
Cc: Frank Chang, Alistair Francis, Palmer Dabbelt, Bin Meng, qemu-devel
From: Frank Chang <frank.chang@sifive.com>
Signed-off-by: Frank Chang <frank.chang@sifive.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
---
target/riscv/cpu.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
index 788fa0b11c0..1b3a7784b06 100644
--- a/target/riscv/cpu.c
+++ b/target/riscv/cpu.c
@@ -617,6 +617,7 @@ static Property riscv_cpu_properties[] = {
DEFINE_PROP_BOOL("Counters", RISCVCPU, cfg.ext_counters, true),
DEFINE_PROP_BOOL("Zifencei", RISCVCPU, cfg.ext_ifencei, true),
DEFINE_PROP_BOOL("Zicsr", RISCVCPU, cfg.ext_icsr, true),
+ DEFINE_PROP_BOOL("Zfh", RISCVCPU, cfg.ext_zfh, false),
DEFINE_PROP_BOOL("mmu", RISCVCPU, cfg.mmu, true),
DEFINE_PROP_BOOL("pmp", RISCVCPU, cfg.pmp, true),
--
2.25.1
^ permalink raw reply related [flat|nested] 23+ messages in thread
* [PATCH v5 7/8] target/riscv: zfh: implement zfhmin extension
2021-10-21 16:29 [PATCH v5 0/8] target/riscv: support Zfh, Zfhmin extension v0.1 frank.chang
@ 2021-10-21 16:29 ` frank.chang
2021-10-21 16:29 ` frank.chang
` (7 subsequent siblings)
8 siblings, 0 replies; 23+ messages in thread
From: frank.chang @ 2021-10-21 16:29 UTC (permalink / raw)
To: qemu-riscv
Cc: Frank Chang, Bin Meng, Richard Henderson, qemu-devel,
Chih-Min Chao, Alistair Francis, Palmer Dabbelt, Kito Cheng
From: Frank Chang <frank.chang@sifive.com>
Zfhmin extension is a subset of Zfh extension, consisting only of data
transfer and conversion instructions.
If enabled, only the following instructions from Zfh extension are
included:
* flh, fsh, fmv.x.h, fmv.h.x, fcvt.s.h, fcvt.h.s
* If D extension is present: fcvt.d.h, fcvt.h.d
Signed-off-by: Frank Chang <frank.chang@sifive.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
---
target/riscv/cpu.h | 1 +
target/riscv/insn_trans/trans_rvzfh.c.inc | 22 ++++++++++++++--------
target/riscv/translate.c | 2 ++
3 files changed, 17 insertions(+), 8 deletions(-)
diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h
index da3c436987c..8b93e25a270 100644
--- a/target/riscv/cpu.h
+++ b/target/riscv/cpu.h
@@ -300,6 +300,7 @@ struct RISCVCPU {
bool ext_ifencei;
bool ext_icsr;
bool ext_zfh;
+ bool ext_zfhmin;
char *priv_spec;
char *user_spec;
diff --git a/target/riscv/insn_trans/trans_rvzfh.c.inc b/target/riscv/insn_trans/trans_rvzfh.c.inc
index 0549e25fb45..5a7cac89585 100644
--- a/target/riscv/insn_trans/trans_rvzfh.c.inc
+++ b/target/riscv/insn_trans/trans_rvzfh.c.inc
@@ -22,13 +22,19 @@
} \
} while (0)
+#define REQUIRE_ZFH_OR_ZFHMIN(ctx) do { \
+ if (!(ctx->ext_zfh || ctx->ext_zfhmin)) { \
+ return false; \
+ } \
+} while (0)
+
static bool trans_flh(DisasContext *ctx, arg_flh *a)
{
TCGv_i64 dest;
TCGv t0;
REQUIRE_FPU;
- REQUIRE_ZFH(ctx);
+ REQUIRE_ZFH_OR_ZFHMIN(ctx);
t0 = get_gpr(ctx, a->rs1, EXT_NONE);
if (a->imm) {
@@ -50,7 +56,7 @@ static bool trans_fsh(DisasContext *ctx, arg_fsh *a)
TCGv t0;
REQUIRE_FPU;
- REQUIRE_ZFH(ctx);
+ REQUIRE_ZFH_OR_ZFHMIN(ctx);
t0 = get_gpr(ctx, a->rs1, EXT_NONE);
if (a->imm) {
@@ -283,7 +289,7 @@ static bool trans_fmax_h(DisasContext *ctx, arg_fmax_h *a)
static bool trans_fcvt_s_h(DisasContext *ctx, arg_fcvt_s_h *a)
{
REQUIRE_FPU;
- REQUIRE_ZFH(ctx);
+ REQUIRE_ZFH_OR_ZFHMIN(ctx);
gen_set_rm(ctx, a->rm);
gen_helper_fcvt_s_h(cpu_fpr[a->rd], cpu_env, cpu_fpr[a->rs1]);
@@ -296,7 +302,7 @@ static bool trans_fcvt_s_h(DisasContext *ctx, arg_fcvt_s_h *a)
static bool trans_fcvt_d_h(DisasContext *ctx, arg_fcvt_d_h *a)
{
REQUIRE_FPU;
- REQUIRE_ZFH(ctx);
+ REQUIRE_ZFH_OR_ZFHMIN(ctx);
REQUIRE_EXT(ctx, RVD);
gen_set_rm(ctx, a->rm);
@@ -311,7 +317,7 @@ static bool trans_fcvt_d_h(DisasContext *ctx, arg_fcvt_d_h *a)
static bool trans_fcvt_h_s(DisasContext *ctx, arg_fcvt_h_s *a)
{
REQUIRE_FPU;
- REQUIRE_ZFH(ctx);
+ REQUIRE_ZFH_OR_ZFHMIN(ctx);
gen_set_rm(ctx, a->rm);
gen_helper_fcvt_h_s(cpu_fpr[a->rd], cpu_env, cpu_fpr[a->rs1]);
@@ -324,7 +330,7 @@ static bool trans_fcvt_h_s(DisasContext *ctx, arg_fcvt_h_s *a)
static bool trans_fcvt_h_d(DisasContext *ctx, arg_fcvt_h_d *a)
{
REQUIRE_FPU;
- REQUIRE_ZFH(ctx);
+ REQUIRE_ZFH_OR_ZFHMIN(ctx);
REQUIRE_EXT(ctx, RVD);
gen_set_rm(ctx, a->rm);
@@ -441,7 +447,7 @@ static bool trans_fcvt_h_wu(DisasContext *ctx, arg_fcvt_h_wu *a)
static bool trans_fmv_x_h(DisasContext *ctx, arg_fmv_x_h *a)
{
REQUIRE_FPU;
- REQUIRE_ZFH(ctx);
+ REQUIRE_ZFH_OR_ZFHMIN(ctx);
TCGv dest = dest_gpr(ctx, a->rd);
@@ -461,7 +467,7 @@ static bool trans_fmv_x_h(DisasContext *ctx, arg_fmv_x_h *a)
static bool trans_fmv_h_x(DisasContext *ctx, arg_fmv_h_x *a)
{
REQUIRE_FPU;
- REQUIRE_ZFH(ctx);
+ REQUIRE_ZFH_OR_ZFHMIN(ctx);
TCGv t0 = get_gpr(ctx, a->rs1, EXT_ZERO);
diff --git a/target/riscv/translate.c b/target/riscv/translate.c
index e6643a7d777..66709091ca1 100644
--- a/target/riscv/translate.c
+++ b/target/riscv/translate.c
@@ -71,6 +71,7 @@ typedef struct DisasContext {
bool virt_enabled;
bool ext_ifencei;
bool ext_zfh;
+ bool ext_zfhmin;
bool hlsx;
/* vector extension */
bool vill;
@@ -620,6 +621,7 @@ static void riscv_tr_init_disas_context(DisasContextBase *dcbase, CPUState *cs)
ctx->frm = -1; /* unknown rounding mode */
ctx->ext_ifencei = cpu->cfg.ext_ifencei;
ctx->ext_zfh = cpu->cfg.ext_zfh;
+ ctx->ext_zfhmin = cpu->cfg.ext_zfhmin;
ctx->vlen = cpu->cfg.vlen;
ctx->mstatus_hs_fs = FIELD_EX32(tb_flags, TB_FLAGS, MSTATUS_HS_FS);
ctx->hlsx = FIELD_EX32(tb_flags, TB_FLAGS, HLSX);
--
2.25.1
^ permalink raw reply related [flat|nested] 23+ messages in thread
* [PATCH v5 7/8] target/riscv: zfh: implement zfhmin extension
@ 2021-10-21 16:29 ` frank.chang
0 siblings, 0 replies; 23+ messages in thread
From: frank.chang @ 2021-10-21 16:29 UTC (permalink / raw)
To: qemu-riscv
Cc: Frank Chang, Richard Henderson, Alistair Francis, Palmer Dabbelt,
Bin Meng, Chih-Min Chao, Kito Cheng, qemu-devel
From: Frank Chang <frank.chang@sifive.com>
Zfhmin extension is a subset of Zfh extension, consisting only of data
transfer and conversion instructions.
If enabled, only the following instructions from Zfh extension are
included:
* flh, fsh, fmv.x.h, fmv.h.x, fcvt.s.h, fcvt.h.s
* If D extension is present: fcvt.d.h, fcvt.h.d
Signed-off-by: Frank Chang <frank.chang@sifive.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
---
target/riscv/cpu.h | 1 +
target/riscv/insn_trans/trans_rvzfh.c.inc | 22 ++++++++++++++--------
target/riscv/translate.c | 2 ++
3 files changed, 17 insertions(+), 8 deletions(-)
diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h
index da3c436987c..8b93e25a270 100644
--- a/target/riscv/cpu.h
+++ b/target/riscv/cpu.h
@@ -300,6 +300,7 @@ struct RISCVCPU {
bool ext_ifencei;
bool ext_icsr;
bool ext_zfh;
+ bool ext_zfhmin;
char *priv_spec;
char *user_spec;
diff --git a/target/riscv/insn_trans/trans_rvzfh.c.inc b/target/riscv/insn_trans/trans_rvzfh.c.inc
index 0549e25fb45..5a7cac89585 100644
--- a/target/riscv/insn_trans/trans_rvzfh.c.inc
+++ b/target/riscv/insn_trans/trans_rvzfh.c.inc
@@ -22,13 +22,19 @@
} \
} while (0)
+#define REQUIRE_ZFH_OR_ZFHMIN(ctx) do { \
+ if (!(ctx->ext_zfh || ctx->ext_zfhmin)) { \
+ return false; \
+ } \
+} while (0)
+
static bool trans_flh(DisasContext *ctx, arg_flh *a)
{
TCGv_i64 dest;
TCGv t0;
REQUIRE_FPU;
- REQUIRE_ZFH(ctx);
+ REQUIRE_ZFH_OR_ZFHMIN(ctx);
t0 = get_gpr(ctx, a->rs1, EXT_NONE);
if (a->imm) {
@@ -50,7 +56,7 @@ static bool trans_fsh(DisasContext *ctx, arg_fsh *a)
TCGv t0;
REQUIRE_FPU;
- REQUIRE_ZFH(ctx);
+ REQUIRE_ZFH_OR_ZFHMIN(ctx);
t0 = get_gpr(ctx, a->rs1, EXT_NONE);
if (a->imm) {
@@ -283,7 +289,7 @@ static bool trans_fmax_h(DisasContext *ctx, arg_fmax_h *a)
static bool trans_fcvt_s_h(DisasContext *ctx, arg_fcvt_s_h *a)
{
REQUIRE_FPU;
- REQUIRE_ZFH(ctx);
+ REQUIRE_ZFH_OR_ZFHMIN(ctx);
gen_set_rm(ctx, a->rm);
gen_helper_fcvt_s_h(cpu_fpr[a->rd], cpu_env, cpu_fpr[a->rs1]);
@@ -296,7 +302,7 @@ static bool trans_fcvt_s_h(DisasContext *ctx, arg_fcvt_s_h *a)
static bool trans_fcvt_d_h(DisasContext *ctx, arg_fcvt_d_h *a)
{
REQUIRE_FPU;
- REQUIRE_ZFH(ctx);
+ REQUIRE_ZFH_OR_ZFHMIN(ctx);
REQUIRE_EXT(ctx, RVD);
gen_set_rm(ctx, a->rm);
@@ -311,7 +317,7 @@ static bool trans_fcvt_d_h(DisasContext *ctx, arg_fcvt_d_h *a)
static bool trans_fcvt_h_s(DisasContext *ctx, arg_fcvt_h_s *a)
{
REQUIRE_FPU;
- REQUIRE_ZFH(ctx);
+ REQUIRE_ZFH_OR_ZFHMIN(ctx);
gen_set_rm(ctx, a->rm);
gen_helper_fcvt_h_s(cpu_fpr[a->rd], cpu_env, cpu_fpr[a->rs1]);
@@ -324,7 +330,7 @@ static bool trans_fcvt_h_s(DisasContext *ctx, arg_fcvt_h_s *a)
static bool trans_fcvt_h_d(DisasContext *ctx, arg_fcvt_h_d *a)
{
REQUIRE_FPU;
- REQUIRE_ZFH(ctx);
+ REQUIRE_ZFH_OR_ZFHMIN(ctx);
REQUIRE_EXT(ctx, RVD);
gen_set_rm(ctx, a->rm);
@@ -441,7 +447,7 @@ static bool trans_fcvt_h_wu(DisasContext *ctx, arg_fcvt_h_wu *a)
static bool trans_fmv_x_h(DisasContext *ctx, arg_fmv_x_h *a)
{
REQUIRE_FPU;
- REQUIRE_ZFH(ctx);
+ REQUIRE_ZFH_OR_ZFHMIN(ctx);
TCGv dest = dest_gpr(ctx, a->rd);
@@ -461,7 +467,7 @@ static bool trans_fmv_x_h(DisasContext *ctx, arg_fmv_x_h *a)
static bool trans_fmv_h_x(DisasContext *ctx, arg_fmv_h_x *a)
{
REQUIRE_FPU;
- REQUIRE_ZFH(ctx);
+ REQUIRE_ZFH_OR_ZFHMIN(ctx);
TCGv t0 = get_gpr(ctx, a->rs1, EXT_ZERO);
diff --git a/target/riscv/translate.c b/target/riscv/translate.c
index e6643a7d777..66709091ca1 100644
--- a/target/riscv/translate.c
+++ b/target/riscv/translate.c
@@ -71,6 +71,7 @@ typedef struct DisasContext {
bool virt_enabled;
bool ext_ifencei;
bool ext_zfh;
+ bool ext_zfhmin;
bool hlsx;
/* vector extension */
bool vill;
@@ -620,6 +621,7 @@ static void riscv_tr_init_disas_context(DisasContextBase *dcbase, CPUState *cs)
ctx->frm = -1; /* unknown rounding mode */
ctx->ext_ifencei = cpu->cfg.ext_ifencei;
ctx->ext_zfh = cpu->cfg.ext_zfh;
+ ctx->ext_zfhmin = cpu->cfg.ext_zfhmin;
ctx->vlen = cpu->cfg.vlen;
ctx->mstatus_hs_fs = FIELD_EX32(tb_flags, TB_FLAGS, MSTATUS_HS_FS);
ctx->hlsx = FIELD_EX32(tb_flags, TB_FLAGS, HLSX);
--
2.25.1
^ permalink raw reply related [flat|nested] 23+ messages in thread
* [PATCH v5 8/8] target/riscv: zfh: add Zfhmin cpu property
2021-10-21 16:29 [PATCH v5 0/8] target/riscv: support Zfh, Zfhmin extension v0.1 frank.chang
@ 2021-10-21 16:29 ` frank.chang
2021-10-21 16:29 ` frank.chang
` (7 subsequent siblings)
8 siblings, 0 replies; 23+ messages in thread
From: frank.chang @ 2021-10-21 16:29 UTC (permalink / raw)
To: qemu-riscv
Cc: Frank Chang, Palmer Dabbelt, Bin Meng, Alistair Francis, qemu-devel
From: Frank Chang <frank.chang@sifive.com>
Signed-off-by: Frank Chang <frank.chang@sifive.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
---
target/riscv/cpu.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
index 1b3a7784b06..e6e3ef183ae 100644
--- a/target/riscv/cpu.c
+++ b/target/riscv/cpu.c
@@ -618,6 +618,7 @@ static Property riscv_cpu_properties[] = {
DEFINE_PROP_BOOL("Zifencei", RISCVCPU, cfg.ext_ifencei, true),
DEFINE_PROP_BOOL("Zicsr", RISCVCPU, cfg.ext_icsr, true),
DEFINE_PROP_BOOL("Zfh", RISCVCPU, cfg.ext_zfh, false),
+ DEFINE_PROP_BOOL("Zfhmin", RISCVCPU, cfg.ext_zfhmin, false),
DEFINE_PROP_BOOL("mmu", RISCVCPU, cfg.mmu, true),
DEFINE_PROP_BOOL("pmp", RISCVCPU, cfg.pmp, true),
--
2.25.1
^ permalink raw reply related [flat|nested] 23+ messages in thread
* [PATCH v5 8/8] target/riscv: zfh: add Zfhmin cpu property
@ 2021-10-21 16:29 ` frank.chang
0 siblings, 0 replies; 23+ messages in thread
From: frank.chang @ 2021-10-21 16:29 UTC (permalink / raw)
To: qemu-riscv
Cc: Frank Chang, Alistair Francis, Palmer Dabbelt, Bin Meng, qemu-devel
From: Frank Chang <frank.chang@sifive.com>
Signed-off-by: Frank Chang <frank.chang@sifive.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
---
target/riscv/cpu.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
index 1b3a7784b06..e6e3ef183ae 100644
--- a/target/riscv/cpu.c
+++ b/target/riscv/cpu.c
@@ -618,6 +618,7 @@ static Property riscv_cpu_properties[] = {
DEFINE_PROP_BOOL("Zifencei", RISCVCPU, cfg.ext_ifencei, true),
DEFINE_PROP_BOOL("Zicsr", RISCVCPU, cfg.ext_icsr, true),
DEFINE_PROP_BOOL("Zfh", RISCVCPU, cfg.ext_zfh, false),
+ DEFINE_PROP_BOOL("Zfhmin", RISCVCPU, cfg.ext_zfhmin, false),
DEFINE_PROP_BOOL("mmu", RISCVCPU, cfg.mmu, true),
DEFINE_PROP_BOOL("pmp", RISCVCPU, cfg.pmp, true),
--
2.25.1
^ permalink raw reply related [flat|nested] 23+ messages in thread
* Re: [PATCH v5 1/8] target/riscv: zfh: half-precision load and store
2021-10-21 16:29 ` frank.chang
(?)
@ 2021-10-21 22:01 ` Alistair Francis
-1 siblings, 0 replies; 23+ messages in thread
From: Alistair Francis @ 2021-10-21 22:01 UTC (permalink / raw)
To: Frank Chang
Cc: open list:RISC-V, Bin Meng, Richard Henderson,
qemu-devel@nongnu.org Developers, Chih-Min Chao, Palmer Dabbelt,
Alistair Francis, Kito Cheng
On Fri, Oct 22, 2021 at 2:30 AM <frank.chang@sifive.com> wrote:
>
> From: Kito Cheng <kito.cheng@sifive.com>
>
> Signed-off-by: Kito Cheng <kito.cheng@sifive.com>
> Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com>
> Signed-off-by: Frank Chang <frank.chang@sifive.com>
> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Alistair
> ---
> target/riscv/cpu.h | 1 +
> target/riscv/insn32.decode | 4 ++
> target/riscv/insn_trans/trans_rvzfh.c.inc | 65 +++++++++++++++++++++++
> target/riscv/translate.c | 8 +++
> 4 files changed, 78 insertions(+)
> create mode 100644 target/riscv/insn_trans/trans_rvzfh.c.inc
>
> diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h
> index a33dc30be81..da3c436987c 100644
> --- a/target/riscv/cpu.h
> +++ b/target/riscv/cpu.h
> @@ -299,6 +299,7 @@ struct RISCVCPU {
> bool ext_counters;
> bool ext_ifencei;
> bool ext_icsr;
> + bool ext_zfh;
>
> char *priv_spec;
> char *user_spec;
> diff --git a/target/riscv/insn32.decode b/target/riscv/insn32.decode
> index 2f251dac1bb..b36a3d8dbf8 100644
> --- a/target/riscv/insn32.decode
> +++ b/target/riscv/insn32.decode
> @@ -726,3 +726,7 @@ binv 0110100 .......... 001 ..... 0110011 @r
> binvi 01101. ........... 001 ..... 0010011 @sh
> bset 0010100 .......... 001 ..... 0110011 @r
> bseti 00101. ........... 001 ..... 0010011 @sh
> +
> +# *** RV32 Zfh Extension ***
> +flh ............ ..... 001 ..... 0000111 @i
> +fsh ....... ..... ..... 001 ..... 0100111 @s
> diff --git a/target/riscv/insn_trans/trans_rvzfh.c.inc b/target/riscv/insn_trans/trans_rvzfh.c.inc
> new file mode 100644
> index 00000000000..dad1d703d72
> --- /dev/null
> +++ b/target/riscv/insn_trans/trans_rvzfh.c.inc
> @@ -0,0 +1,65 @@
> +/*
> + * RISC-V translation routines for the RV64Zfh Standard Extension.
> + *
> + * Copyright (c) 2020 Chih-Min Chao, chihmin.chao@sifive.com
> + *
> + * This program is free software; you can redistribute it and/or modify it
> + * under the terms and conditions of the GNU General Public License,
> + * version 2 or later, as published by the Free Software Foundation.
> + *
> + * This program is distributed in the hope it will be useful, but WITHOUT
> + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
> + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
> + * more details.
> + *
> + * You should have received a copy of the GNU General Public License along with
> + * this program. If not, see <http://www.gnu.org/licenses/>.
> + */
> +
> +#define REQUIRE_ZFH(ctx) do { \
> + if (!ctx->ext_zfh) { \
> + return false; \
> + } \
> +} while (0)
> +
> +static bool trans_flh(DisasContext *ctx, arg_flh *a)
> +{
> + TCGv_i64 dest;
> + TCGv t0;
> +
> + REQUIRE_FPU;
> + REQUIRE_ZFH(ctx);
> +
> + t0 = get_gpr(ctx, a->rs1, EXT_NONE);
> + if (a->imm) {
> + TCGv temp = temp_new(ctx);
> + tcg_gen_addi_tl(temp, t0, a->imm);
> + t0 = temp;
> + }
> +
> + dest = cpu_fpr[a->rd];
> + tcg_gen_qemu_ld_i64(dest, t0, ctx->mem_idx, MO_TEUW);
> + gen_nanbox_h(dest, dest);
> +
> + mark_fs_dirty(ctx);
> + return true;
> +}
> +
> +static bool trans_fsh(DisasContext *ctx, arg_fsh *a)
> +{
> + TCGv t0;
> +
> + REQUIRE_FPU;
> + REQUIRE_ZFH(ctx);
> +
> + t0 = get_gpr(ctx, a->rs1, EXT_NONE);
> + if (a->imm) {
> + TCGv temp = tcg_temp_new();
> + tcg_gen_addi_tl(temp, t0, a->imm);
> + t0 = temp;
> + }
> +
> + tcg_gen_qemu_st_i64(cpu_fpr[a->rs2], t0, ctx->mem_idx, MO_TEUW);
> +
> + return true;
> +}
> diff --git a/target/riscv/translate.c b/target/riscv/translate.c
> index d38f87d7188..ad963d1898f 100644
> --- a/target/riscv/translate.c
> +++ b/target/riscv/translate.c
> @@ -70,6 +70,7 @@ typedef struct DisasContext {
> RISCVMXL ol;
> bool virt_enabled;
> bool ext_ifencei;
> + bool ext_zfh;
> bool hlsx;
> /* vector extension */
> bool vill;
> @@ -127,6 +128,11 @@ static void gen_nanbox_s(TCGv_i64 out, TCGv_i64 in)
> tcg_gen_ori_i64(out, in, MAKE_64BIT_MASK(32, 32));
> }
>
> +static void gen_nanbox_h(TCGv_i64 out, TCGv_i64 in)
> +{
> + tcg_gen_ori_i64(out, in, MAKE_64BIT_MASK(16, 48));
> +}
> +
> /*
> * A narrow n-bit operation, where n < FLEN, checks that input operands
> * are correctly Nan-boxed, i.e., all upper FLEN - n bits are 1.
> @@ -550,6 +556,7 @@ static uint32_t opcode_at(DisasContextBase *dcbase, target_ulong pc)
> #include "insn_trans/trans_rvh.c.inc"
> #include "insn_trans/trans_rvv.c.inc"
> #include "insn_trans/trans_rvb.c.inc"
> +#include "insn_trans/trans_rvzfh.c.inc"
> #include "insn_trans/trans_privileged.c.inc"
>
> /* Include the auto-generated decoder for 16 bit insn */
> @@ -602,6 +609,7 @@ static void riscv_tr_init_disas_context(DisasContextBase *dcbase, CPUState *cs)
> ctx->misa_ext = env->misa_ext;
> ctx->frm = -1; /* unknown rounding mode */
> ctx->ext_ifencei = cpu->cfg.ext_ifencei;
> + ctx->ext_zfh = cpu->cfg.ext_zfh;
> ctx->vlen = cpu->cfg.vlen;
> ctx->mstatus_hs_fs = FIELD_EX32(tb_flags, TB_FLAGS, MSTATUS_HS_FS);
> ctx->hlsx = FIELD_EX32(tb_flags, TB_FLAGS, HLSX);
> --
> 2.25.1
>
>
^ permalink raw reply [flat|nested] 23+ messages in thread
* Re: [PATCH v5 1/8] target/riscv: zfh: half-precision load and store
2021-10-21 16:29 ` frank.chang
(?)
(?)
@ 2021-10-21 22:31 ` Alistair Francis
2021-10-21 23:28 ` Richard Henderson
-1 siblings, 1 reply; 23+ messages in thread
From: Alistair Francis @ 2021-10-21 22:31 UTC (permalink / raw)
To: Frank Chang
Cc: open list:RISC-V, Bin Meng, Richard Henderson,
qemu-devel@nongnu.org Developers, Chih-Min Chao, Palmer Dabbelt,
Alistair Francis, Kito Cheng
On Fri, Oct 22, 2021 at 2:30 AM <frank.chang@sifive.com> wrote:
>
> From: Kito Cheng <kito.cheng@sifive.com>
>
> Signed-off-by: Kito Cheng <kito.cheng@sifive.com>
> Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com>
> Signed-off-by: Frank Chang <frank.chang@sifive.com>
> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
It doesn't look like this made it through to the list. I only see v3
on patchew: https://patchew.org/QEMU/20211016090742.3034669-1-frank.chang@sifive.com/
Can you rebase this on
https://github.com/alistair23/qemu/tree/riscv-to-apply.next when you
re-send it?
Alistair
> ---
> target/riscv/cpu.h | 1 +
> target/riscv/insn32.decode | 4 ++
> target/riscv/insn_trans/trans_rvzfh.c.inc | 65 +++++++++++++++++++++++
> target/riscv/translate.c | 8 +++
> 4 files changed, 78 insertions(+)
> create mode 100644 target/riscv/insn_trans/trans_rvzfh.c.inc
>
> diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h
> index a33dc30be81..da3c436987c 100644
> --- a/target/riscv/cpu.h
> +++ b/target/riscv/cpu.h
> @@ -299,6 +299,7 @@ struct RISCVCPU {
> bool ext_counters;
> bool ext_ifencei;
> bool ext_icsr;
> + bool ext_zfh;
>
> char *priv_spec;
> char *user_spec;
> diff --git a/target/riscv/insn32.decode b/target/riscv/insn32.decode
> index 2f251dac1bb..b36a3d8dbf8 100644
> --- a/target/riscv/insn32.decode
> +++ b/target/riscv/insn32.decode
> @@ -726,3 +726,7 @@ binv 0110100 .......... 001 ..... 0110011 @r
> binvi 01101. ........... 001 ..... 0010011 @sh
> bset 0010100 .......... 001 ..... 0110011 @r
> bseti 00101. ........... 001 ..... 0010011 @sh
> +
> +# *** RV32 Zfh Extension ***
> +flh ............ ..... 001 ..... 0000111 @i
> +fsh ....... ..... ..... 001 ..... 0100111 @s
> diff --git a/target/riscv/insn_trans/trans_rvzfh.c.inc b/target/riscv/insn_trans/trans_rvzfh.c.inc
> new file mode 100644
> index 00000000000..dad1d703d72
> --- /dev/null
> +++ b/target/riscv/insn_trans/trans_rvzfh.c.inc
> @@ -0,0 +1,65 @@
> +/*
> + * RISC-V translation routines for the RV64Zfh Standard Extension.
> + *
> + * Copyright (c) 2020 Chih-Min Chao, chihmin.chao@sifive.com
> + *
> + * This program is free software; you can redistribute it and/or modify it
> + * under the terms and conditions of the GNU General Public License,
> + * version 2 or later, as published by the Free Software Foundation.
> + *
> + * This program is distributed in the hope it will be useful, but WITHOUT
> + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
> + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
> + * more details.
> + *
> + * You should have received a copy of the GNU General Public License along with
> + * this program. If not, see <http://www.gnu.org/licenses/>.
> + */
> +
> +#define REQUIRE_ZFH(ctx) do { \
> + if (!ctx->ext_zfh) { \
> + return false; \
> + } \
> +} while (0)
> +
> +static bool trans_flh(DisasContext *ctx, arg_flh *a)
> +{
> + TCGv_i64 dest;
> + TCGv t0;
> +
> + REQUIRE_FPU;
> + REQUIRE_ZFH(ctx);
> +
> + t0 = get_gpr(ctx, a->rs1, EXT_NONE);
> + if (a->imm) {
> + TCGv temp = temp_new(ctx);
> + tcg_gen_addi_tl(temp, t0, a->imm);
> + t0 = temp;
> + }
> +
> + dest = cpu_fpr[a->rd];
> + tcg_gen_qemu_ld_i64(dest, t0, ctx->mem_idx, MO_TEUW);
> + gen_nanbox_h(dest, dest);
> +
> + mark_fs_dirty(ctx);
> + return true;
> +}
> +
> +static bool trans_fsh(DisasContext *ctx, arg_fsh *a)
> +{
> + TCGv t0;
> +
> + REQUIRE_FPU;
> + REQUIRE_ZFH(ctx);
> +
> + t0 = get_gpr(ctx, a->rs1, EXT_NONE);
> + if (a->imm) {
> + TCGv temp = tcg_temp_new();
> + tcg_gen_addi_tl(temp, t0, a->imm);
> + t0 = temp;
> + }
> +
> + tcg_gen_qemu_st_i64(cpu_fpr[a->rs2], t0, ctx->mem_idx, MO_TEUW);
> +
> + return true;
> +}
> diff --git a/target/riscv/translate.c b/target/riscv/translate.c
> index d38f87d7188..ad963d1898f 100644
> --- a/target/riscv/translate.c
> +++ b/target/riscv/translate.c
> @@ -70,6 +70,7 @@ typedef struct DisasContext {
> RISCVMXL ol;
> bool virt_enabled;
> bool ext_ifencei;
> + bool ext_zfh;
> bool hlsx;
> /* vector extension */
> bool vill;
> @@ -127,6 +128,11 @@ static void gen_nanbox_s(TCGv_i64 out, TCGv_i64 in)
> tcg_gen_ori_i64(out, in, MAKE_64BIT_MASK(32, 32));
> }
>
> +static void gen_nanbox_h(TCGv_i64 out, TCGv_i64 in)
> +{
> + tcg_gen_ori_i64(out, in, MAKE_64BIT_MASK(16, 48));
> +}
> +
> /*
> * A narrow n-bit operation, where n < FLEN, checks that input operands
> * are correctly Nan-boxed, i.e., all upper FLEN - n bits are 1.
> @@ -550,6 +556,7 @@ static uint32_t opcode_at(DisasContextBase *dcbase, target_ulong pc)
> #include "insn_trans/trans_rvh.c.inc"
> #include "insn_trans/trans_rvv.c.inc"
> #include "insn_trans/trans_rvb.c.inc"
> +#include "insn_trans/trans_rvzfh.c.inc"
> #include "insn_trans/trans_privileged.c.inc"
>
> /* Include the auto-generated decoder for 16 bit insn */
> @@ -602,6 +609,7 @@ static void riscv_tr_init_disas_context(DisasContextBase *dcbase, CPUState *cs)
> ctx->misa_ext = env->misa_ext;
> ctx->frm = -1; /* unknown rounding mode */
> ctx->ext_ifencei = cpu->cfg.ext_ifencei;
> + ctx->ext_zfh = cpu->cfg.ext_zfh;
> ctx->vlen = cpu->cfg.vlen;
> ctx->mstatus_hs_fs = FIELD_EX32(tb_flags, TB_FLAGS, MSTATUS_HS_FS);
> ctx->hlsx = FIELD_EX32(tb_flags, TB_FLAGS, HLSX);
> --
> 2.25.1
>
>
^ permalink raw reply [flat|nested] 23+ messages in thread
* Re: [PATCH v5 1/8] target/riscv: zfh: half-precision load and store
2021-10-21 22:31 ` Alistair Francis
@ 2021-10-21 23:28 ` Richard Henderson
2021-10-22 3:25 ` Frank Chang
0 siblings, 1 reply; 23+ messages in thread
From: Richard Henderson @ 2021-10-21 23:28 UTC (permalink / raw)
To: Alistair Francis, Frank Chang
Cc: open list:RISC-V, Bin Meng, qemu-devel@nongnu.org Developers,
Chih-Min Chao, Palmer Dabbelt, Alistair Francis, Kito Cheng
On 10/21/21 3:31 PM, Alistair Francis wrote:
> On Fri, Oct 22, 2021 at 2:30 AM <frank.chang@sifive.com> wrote:
>>
>> From: Kito Cheng <kito.cheng@sifive.com>
>>
>> Signed-off-by: Kito Cheng <kito.cheng@sifive.com>
>> Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com>
>> Signed-off-by: Frank Chang <frank.chang@sifive.com>
>> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
>
> It doesn't look like this made it through to the list. I only see v3
> on patchew: https://patchew.org/QEMU/20211016090742.3034669-1-frank.chang@sifive.com/
It's just the cover-letter that got lost.
https://lore.kernel.org/qemu-devel/20211021162956.2772656-2-frank.chang@sifive.com/
> Can you rebase this on
> https://github.com/alistair23/qemu/tree/riscv-to-apply.next when you
> re-send it?
But a rebase is probably required anyway.
r~
^ permalink raw reply [flat|nested] 23+ messages in thread
* Re: [PATCH v5 1/8] target/riscv: zfh: half-precision load and store
2021-10-21 23:28 ` Richard Henderson
@ 2021-10-22 3:25 ` Frank Chang
0 siblings, 0 replies; 23+ messages in thread
From: Frank Chang @ 2021-10-22 3:25 UTC (permalink / raw)
To: Richard Henderson
Cc: Alistair Francis, open list:RISC-V, Bin Meng,
qemu-devel@nongnu.org Developers, Chih-Min Chao, Palmer Dabbelt,
Alistair Francis, Kito Cheng
[-- Attachment #1: Type: text/plain, Size: 1553 bytes --]
On Fri, Oct 22, 2021 at 7:28 AM Richard Henderson <
richard.henderson@linaro.org> wrote:
> On 10/21/21 3:31 PM, Alistair Francis wrote:
> > On Fri, Oct 22, 2021 at 2:30 AM <frank.chang@sifive.com> wrote:
> >>
> >> From: Kito Cheng <kito.cheng@sifive.com>
> >>
> >> Signed-off-by: Kito Cheng <kito.cheng@sifive.com>
> >> Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com>
> >> Signed-off-by: Frank Chang <frank.chang@sifive.com>
> >> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
> >
> > It doesn't look like this made it through to the list. I only see v3
> > on patchew:
> https://patchew.org/QEMU/20211016090742.3034669-1-frank.chang@sifive.com/
>
> It's just the cover-letter that got lost.
>
>
> https://lore.kernel.org/qemu-devel/20211021162956.2772656-2-frank.chang@sifive.com/
>
>
> > Can you rebase this on
> > https://github.com/alistair23/qemu/tree/riscv-to-apply.next when you
> > re-send it?
>
> But a rebase is probably required anyway.
>
Hi Alistair and Richard,
This patchset is already rebased on riscv-to-apply.next.
The Zfh and Zfhmin cpu properties are already moved to
/* Defaults for standard extensions */ section,
along with other Zi* extensions.
BTW, I found out my patchset won't appear at Patchew
if cover letter is only sent to qemu-riscv@nongnu.org.
(I used: "git send-email --to 'qemu-riscv@nongnu.org' ...",
instead of: "git send-email --to 'qemu-devel@nongnu.org,
qemu-riscv@nongnu.org' ...")
I will send it to qemu-devel@nongnu.org as well in my future patchset.
Regards,
Frank Chang
>
> r~
>
[-- Attachment #2: Type: text/html, Size: 3326 bytes --]
^ permalink raw reply [flat|nested] 23+ messages in thread
* Re: [PATCH v5 1/8] target/riscv: zfh: half-precision load and store
@ 2021-10-22 3:25 ` Frank Chang
0 siblings, 0 replies; 23+ messages in thread
From: Frank Chang @ 2021-10-22 3:25 UTC (permalink / raw)
To: Richard Henderson
Cc: Alistair Francis, open list:RISC-V, Bin Meng,
qemu-devel@nongnu.org Developers, Chih-Min Chao, Palmer Dabbelt,
Alistair Francis, Kito Cheng
[-- Attachment #1: Type: text/plain, Size: 1553 bytes --]
On Fri, Oct 22, 2021 at 7:28 AM Richard Henderson <
richard.henderson@linaro.org> wrote:
> On 10/21/21 3:31 PM, Alistair Francis wrote:
> > On Fri, Oct 22, 2021 at 2:30 AM <frank.chang@sifive.com> wrote:
> >>
> >> From: Kito Cheng <kito.cheng@sifive.com>
> >>
> >> Signed-off-by: Kito Cheng <kito.cheng@sifive.com>
> >> Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com>
> >> Signed-off-by: Frank Chang <frank.chang@sifive.com>
> >> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
> >
> > It doesn't look like this made it through to the list. I only see v3
> > on patchew:
> https://patchew.org/QEMU/20211016090742.3034669-1-frank.chang@sifive.com/
>
> It's just the cover-letter that got lost.
>
>
> https://lore.kernel.org/qemu-devel/20211021162956.2772656-2-frank.chang@sifive.com/
>
>
> > Can you rebase this on
> > https://github.com/alistair23/qemu/tree/riscv-to-apply.next when you
> > re-send it?
>
> But a rebase is probably required anyway.
>
Hi Alistair and Richard,
This patchset is already rebased on riscv-to-apply.next.
The Zfh and Zfhmin cpu properties are already moved to
/* Defaults for standard extensions */ section,
along with other Zi* extensions.
BTW, I found out my patchset won't appear at Patchew
if cover letter is only sent to qemu-riscv@nongnu.org.
(I used: "git send-email --to 'qemu-riscv@nongnu.org' ...",
instead of: "git send-email --to 'qemu-devel@nongnu.org,
qemu-riscv@nongnu.org' ...")
I will send it to qemu-devel@nongnu.org as well in my future patchset.
Regards,
Frank Chang
>
> r~
>
[-- Attachment #2: Type: text/html, Size: 3326 bytes --]
^ permalink raw reply [flat|nested] 23+ messages in thread
* Re: [PATCH v5 0/8] target/riscv: support Zfh, Zfhmin extension v0.1
2021-10-21 16:29 [PATCH v5 0/8] target/riscv: support Zfh, Zfhmin extension v0.1 frank.chang
` (7 preceding siblings ...)
2021-10-21 16:29 ` frank.chang
@ 2021-11-02 14:07 ` Frank Chang
8 siblings, 0 replies; 23+ messages in thread
From: Frank Chang @ 2021-11-02 14:07 UTC (permalink / raw)
To: open list:RISC-V, qemu-devel@nongnu.org Developers, Alistair Francis
[-- Attachment #1: Type: text/plain, Size: 2537 bytes --]
On Fri, Oct 22, 2021 at 12:30 AM <frank.chang@sifive.com> wrote:
> From: Frank Chang <frank.chang@sifive.com>
>
> Zfh - Half width floating point
> Zfhmin - Subset of half width floating point
>
> Zfh, Zfhmin v0.1 is now in public review period and is required by
> RVV extension:
>
> https://groups.google.com/a/groups.riscv.org/g/isa-dev/c/63gDCinXTwE/m/871Wm9XIBQAJ
>
> Zfh, Zfhmin can be enabled with -cpu option: Zfh=true and Zfhmin=true
> respectively.
>
> The port is available at:
> https://github.com/sifive/qemu/tree/zfh-upstream-v5
>
> Note: This patchset depends on another patchset listed in Based-on
> section below so it is not able to be built unless the patchset
> is applied.
>
> Changelog:
>
> v5:
> * Rebase on riscv-to-apply.next.
>
> v4:
> * Spilt Zfh, Zfhmin cpu properties related changes into individual
> patches.
>
> v3:
> * Use the renamed softfloat min/max APIs: *_minimum_number()
> and *_maximum_number().
> * Pick softfloat min/max APIs based on CPU privilege spec version.
> * Add braces for if statements in REQUIRE_ZFH() and
> REQUIRE_ZFH_OR_ZFHMIN().
> * Rearrange the positions of Zfh and Zfhmin cpu properties.
>
> v2:
> * Use {get,dest}_gpr APIs.
> * Add Zfhmin extension.
>
> Based-on: <20211021160847.2748577-1-frank.chang@sifive.com>
>
> Frank Chang (3):
> target/riscv: zfh: add Zfh cpu property
> target/riscv: zfh: implement zfhmin extension
> target/riscv: zfh: add Zfhmin cpu property
>
> Kito Cheng (5):
> target/riscv: zfh: half-precision load and store
> target/riscv: zfh: half-precision computational
> target/riscv: zfh: half-precision convert and move
> target/riscv: zfh: half-precision floating-point compare
> target/riscv: zfh: half-precision floating-point classify
>
> target/riscv/cpu.c | 2 +
> target/riscv/cpu.h | 2 +
> target/riscv/fpu_helper.c | 180 ++++++++
> target/riscv/helper.h | 29 ++
> target/riscv/insn32.decode | 38 ++
> target/riscv/insn_trans/trans_rvzfh.c.inc | 537 ++++++++++++++++++++++
> target/riscv/internals.h | 16 +
> target/riscv/translate.c | 20 +
> 8 files changed, 824 insertions(+)
> create mode 100644 target/riscv/insn_trans/trans_rvzfh.c.inc
>
> --
> 2.25.1
>
>
Hi Alistair,
Is this one able to be merged to v6.2.0 as the soft feature freeze is due
today (2021-11-02).
I think this series has gotten all Reviewed-by.
Thanks,
Frank Chang
[-- Attachment #2: Type: text/html, Size: 3630 bytes --]
^ permalink raw reply [flat|nested] 23+ messages in thread
end of thread, other threads:[~2021-11-02 14:41 UTC | newest]
Thread overview: 23+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-10-21 16:29 [PATCH v5 0/8] target/riscv: support Zfh, Zfhmin extension v0.1 frank.chang
2021-10-21 16:29 ` [PATCH v5 1/8] target/riscv: zfh: half-precision load and store frank.chang
2021-10-21 16:29 ` frank.chang
2021-10-21 22:01 ` Alistair Francis
2021-10-21 22:31 ` Alistair Francis
2021-10-21 23:28 ` Richard Henderson
2021-10-22 3:25 ` Frank Chang
2021-10-22 3:25 ` Frank Chang
2021-10-21 16:29 ` [PATCH v5 2/8] target/riscv: zfh: half-precision computational frank.chang
2021-10-21 16:29 ` frank.chang
2021-10-21 16:29 ` [PATCH v5 3/8] target/riscv: zfh: half-precision convert and move frank.chang
2021-10-21 16:29 ` frank.chang
2021-10-21 16:29 ` [PATCH v5 4/8] target/riscv: zfh: half-precision floating-point compare frank.chang
2021-10-21 16:29 ` frank.chang
2021-10-21 16:29 ` [PATCH v5 5/8] target/riscv: zfh: half-precision floating-point classify frank.chang
2021-10-21 16:29 ` frank.chang
2021-10-21 16:29 ` [PATCH v5 6/8] target/riscv: zfh: add Zfh cpu property frank.chang
2021-10-21 16:29 ` frank.chang
2021-10-21 16:29 ` [PATCH v5 7/8] target/riscv: zfh: implement zfhmin extension frank.chang
2021-10-21 16:29 ` frank.chang
2021-10-21 16:29 ` [PATCH v5 8/8] target/riscv: zfh: add Zfhmin cpu property frank.chang
2021-10-21 16:29 ` frank.chang
2021-11-02 14:07 ` [PATCH v5 0/8] target/riscv: support Zfh, Zfhmin extension v0.1 Frank Chang
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.