Add HVX IEEE floating-point arithmetic instructions:
- vmpy_sf_sf, vmpy_sf_hf, vmpy_hf_hf: multiply operations
- vdmpy_sf_hf: dot-product multiply
- vmpy_sf_hf_acc, vmpy_hf_hf_acc, vdmpy_sf_hf_acc: multiply-accumulate
- vadd_sf_sf, vsub_sf_sf, vadd_sf_hf, vsub_sf_hf: add/sub with sf output
- vadd_hf_hf, vsub_hf_hf: add/sub with hf output
Reviewed-by: Taylor Simpson <ltaylorsimpson@gmail.com>
Signed-off-by: Matheus Tavares Bernardino <matheus.bernardino@oss.qualcomm.com>
---
target/hexagon/cpu.h | 1 +
target/hexagon/mmvec/hvx_ieee_fp.h | 18 ++++
target/hexagon/mmvec/macros.h | 1 +
target/hexagon/mmvec/mmvec.h | 2 +
target/hexagon/attribs_def.h.inc | 4 +
target/hexagon/arch.c | 8 ++
target/hexagon/cpu.c | 3 +
target/hexagon/mmvec/hvx_ieee_fp.c | 21 ++++
target/hexagon/hex_common.py | 1 +
target/hexagon/imported/mmvec/encode_ext.def | 18 ++++
target/hexagon/imported/mmvec/ext.idef | 101 +++++++++++++++++++
target/hexagon/meson.build | 1 +
12 files changed, 179 insertions(+)
create mode 100644 target/hexagon/mmvec/hvx_ieee_fp.h
create mode 100644 target/hexagon/mmvec/hvx_ieee_fp.c
diff --git a/target/hexagon/cpu.h b/target/hexagon/cpu.h
index d28beaa92f..5a008d1949 100644
--- a/target/hexagon/cpu.h
+++ b/target/hexagon/cpu.h
@@ -87,6 +87,7 @@ typedef struct CPUArchState {
MemLog mem_log_stores[STORES_MAX];
float_status fp_status;
+ float_status hvx_fp_status;
target_ulong llsc_addr;
target_ulong llsc_val;
diff --git a/target/hexagon/mmvec/hvx_ieee_fp.h b/target/hexagon/mmvec/hvx_ieee_fp.h
new file mode 100644
index 0000000000..75008deb3b
--- /dev/null
+++ b/target/hexagon/mmvec/hvx_ieee_fp.h
@@ -0,0 +1,18 @@
+/*
+ * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries.
+ *
+ * SPDX-License-Identifier: GPL-2.0-or-later
+ */
+
+#ifndef HEXAGON_HVX_IEEE_H
+#define HEXAGON_HVX_IEEE_H
+
+#include "fpu/softfloat.h"
+
+#define f16_to_f32(A) float16_to_float32((A), true, &env->hvx_fp_status)
+
+float32 fp_mult_sf_hf(float16 a1, float16 a2, float_status *fp_status);
+float32 fp_vdmpy(float16 a1, float16 a2, float16 a3, float16 a4,
+ float_status *fp_status);
+
+#endif
diff --git a/target/hexagon/mmvec/macros.h b/target/hexagon/mmvec/macros.h
index c7840fbf2e..ac709d8993 100644
--- a/target/hexagon/mmvec/macros.h
+++ b/target/hexagon/mmvec/macros.h
@@ -23,6 +23,7 @@
#include "mmvec/system_ext_mmvec.h"
#include "accel/tcg/getpc.h"
#include "accel/tcg/probe.h"
+#include "mmvec/hvx_ieee_fp.h"
#ifndef QEMU_GENERATE
#define VdV (*(MMVector *restrict)(VdV_void))
diff --git a/target/hexagon/mmvec/mmvec.h b/target/hexagon/mmvec/mmvec.h
index 52d470709c..31909303b5 100644
--- a/target/hexagon/mmvec/mmvec.h
+++ b/target/hexagon/mmvec/mmvec.h
@@ -38,6 +38,8 @@ typedef union {
int16_t h[MAX_VEC_SIZE_BYTES / 2];
uint8_t ub[MAX_VEC_SIZE_BYTES / 1];
int8_t b[MAX_VEC_SIZE_BYTES / 1];
+ float32 sf[MAX_VEC_SIZE_BYTES / 4];
+ float16 hf[MAX_VEC_SIZE_BYTES / 2];
} MMVector;
typedef union {
diff --git a/target/hexagon/attribs_def.h.inc b/target/hexagon/attribs_def.h.inc
index c85cd5d17c..d3c4bf6301 100644
--- a/target/hexagon/attribs_def.h.inc
+++ b/target/hexagon/attribs_def.h.inc
@@ -175,6 +175,10 @@ DEF_ATTRIB(RESTRICT_LATEPRED, "Predicate can not be used as a .new.", "", "")
/* HVX IEEE FP extension attributes */
DEF_ATTRIB(HVX_IEEE_FP, "HVX IEEE FP extension instruction", "", "")
+DEF_ATTRIB(HVX_IEEE_FP_ACC, "HVX IEEE FP accumulate instruction", "", "")
+DEF_ATTRIB(HVX_IEEE_FP_OUT_16, "HVX IEEE FP 16-bit output", "", "")
+DEF_ATTRIB(HVX_IEEE_FP_OUT_32, "HVX IEEE FP 32-bit output", "", "")
+DEF_ATTRIB(CVI_VX_NO_TMP_LD, "HVX multiply without tmp load", "", "")
/* Keep this as the last attribute: */
DEF_ATTRIB(ZZ_LASTATTRIB, "Last attribute in the file", "", "")
diff --git a/target/hexagon/arch.c b/target/hexagon/arch.c
index e17e714a6a..358aa71e03 100644
--- a/target/hexagon/arch.c
+++ b/target/hexagon/arch.c
@@ -199,6 +199,10 @@ void arch_fpop_start(CPUHexagonState *env)
set_float_rounding_mode(
softfloat_roundingmodes[fREAD_REG_FIELD(USR, USR_FPRND)],
&env->fp_status);
+ /*
+ * No need to check env->hvx_fp_status, these instructions don't
+ * raise exceptions nor interact with usr fields.
+ */
}
#ifdef CONFIG_USER_ONLY
@@ -232,6 +236,10 @@ void arch_fpop_end(CPUHexagonState *env, bool pkt_need_commit)
SOFTFLOAT_TEST_FLAG(float_flag_overflow, FPOVFF, FPOVFE);
SOFTFLOAT_TEST_FLAG(float_flag_underflow, FPUNFF, FPUNFE);
}
+ /*
+ * No need to check env->hvx_fp_status, these instructions don't
+ * raise exceptions nor interact with usr fields.
+ */
}
int arch_sf_recip_common(float32 *Rs, float32 *Rt, float32 *Rd, int *adjust,
diff --git a/target/hexagon/cpu.c b/target/hexagon/cpu.c
index d7f4df5f96..d6ca51f175 100644
--- a/target/hexagon/cpu.c
+++ b/target/hexagon/cpu.c
@@ -300,6 +300,9 @@ static void hexagon_cpu_reset_hold(Object *obj, ResetType type)
set_float_detect_tininess(float_tininess_before_rounding, &env->fp_status);
/* Default NaN value: sign bit set, all frac bits set */
set_float_default_nan_pattern(0b11111111, &env->fp_status);
+
+ set_default_nan_mode(1, &env->hvx_fp_status);
+ set_float_default_nan_pattern(0b01111111, &env->hvx_fp_status);
}
static void hexagon_cpu_disas_set_info(const CPUState *cs,
diff --git a/target/hexagon/mmvec/hvx_ieee_fp.c b/target/hexagon/mmvec/hvx_ieee_fp.c
new file mode 100644
index 0000000000..3367226998
--- /dev/null
+++ b/target/hexagon/mmvec/hvx_ieee_fp.c
@@ -0,0 +1,21 @@
+/*
+ * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries.
+ *
+ * SPDX-License-Identifier: GPL-2.0-or-later
+ */
+
+#include "qemu/osdep.h"
+#include "hvx_ieee_fp.h"
+
+float32 fp_mult_sf_hf(float16 a1, float16 a2, float_status *fp_status)
+{
+ return float32_mul(float16_to_float32(a1, true, fp_status),
+ float16_to_float32(a2, true, fp_status), fp_status);
+}
+
+float32 fp_vdmpy(float16 a1, float16 a2, float16 a3, float16 a4,
+ float_status *fp_status)
+{
+ return float32_add(fp_mult_sf_hf(a1, a3, fp_status),
+ fp_mult_sf_hf(a2, a4, fp_status), fp_status);
+}
diff --git a/target/hexagon/hex_common.py b/target/hexagon/hex_common.py
index e82a3da1e4..9819201b50 100755
--- a/target/hexagon/hex_common.py
+++ b/target/hexagon/hex_common.py
@@ -215,6 +215,7 @@ def need_env(tag):
"A_LOAD" in attribdict[tag] or
"A_CVI_GATHER" in attribdict[tag] or
"A_CVI_SCATTER" in attribdict[tag] or
+ "A_HVX_IEEE_FP" in attribdict[tag] or
"A_IMPLICIT_WRITES_USR" in attribdict[tag])
diff --git a/target/hexagon/imported/mmvec/encode_ext.def b/target/hexagon/imported/mmvec/encode_ext.def
index 6d70086b5f..4ce87d09fd 100644
--- a/target/hexagon/imported/mmvec/encode_ext.def
+++ b/target/hexagon/imported/mmvec/encode_ext.def
@@ -804,5 +804,23 @@ DEF_ENC(V6_vmpyewuh, ICLASS_CJ" 1 111 111 vvvvv PP 0 uuuuu 101 ddddd")
DEF_ENC(V6_vmpyowh, ICLASS_CJ" 1 111 111 vvvvv PP 0 uuuuu 111 ddddd")
DEF_ENC(V6_vmpyuhvs,"00011111110vvvvvPP1uuuuu111ddddd")
+/* IEEE FP multiply instructions */
+DEF_ENC(V6_vmpy_sf_sf,"00011111100vvvvvPP1uuuuu001ddddd")
+DEF_ENC(V6_vmpy_sf_hf,"00011111100vvvvvPP1uuuuu010ddddd")
+DEF_ENC(V6_vmpy_hf_hf,"00011111100vvvvvPP1uuuuu011ddddd")
+DEF_ENC(V6_vdmpy_sf_hf,"00011111101vvvvvPP1uuuuu110ddddd")
+
+/* IEEE FP multiply-accumulate instructions */
+DEF_ENC(V6_vmpy_sf_hf_acc,"00011100010vvvvvPP1uuuuu001xxxxx")
+DEF_ENC(V6_vmpy_hf_hf_acc,"00011100010vvvvvPP1uuuuu010xxxxx")
+DEF_ENC(V6_vdmpy_sf_hf_acc,"00011100010vvvvvPP1uuuuu011xxxxx")
+
+/* IEEE FP add/sub instructions */
+DEF_ENC(V6_vadd_sf_sf,"00011111100vvvvvPP1uuuuu110ddddd")
+DEF_ENC(V6_vsub_sf_sf,"00011111100vvvvvPP1uuuuu111ddddd")
+DEF_ENC(V6_vadd_sf_hf,"00011111100vvvvvPP1uuuuu100ddddd")
+DEF_ENC(V6_vsub_sf_hf,"00011111100vvvvvPP1uuuuu101ddddd")
+DEF_ENC(V6_vadd_hf_hf,"00011111101vvvvvPP1uuuuu111ddddd")
+DEF_ENC(V6_vsub_hf_hf,"00011111011vvvvvPP1uuuuu000ddddd")
#endif /* NO MMVEC */
diff --git a/target/hexagon/imported/mmvec/ext.idef b/target/hexagon/imported/mmvec/ext.idef
index 03d31f6181..14df8e4790 100644
--- a/target/hexagon/imported/mmvec/ext.idef
+++ b/target/hexagon/imported/mmvec/ext.idef
@@ -2895,9 +2895,110 @@ EXTINSN(V6_vprefixqw,"Vd32.w=prefixsum(Qv4)", ATTRIBS(A_EXTENSION,A_CVI,A_CVI_
}
} )
+/* KVX - IEEE FP Instructions */
+/* Single pipe, 32-bit output */
+#define ITERATOR_INSN_IEEE_FP_32(WIDTH,TAG,SYNTAX,DESCR,CODE) \
+EXTINSN(V6_##TAG, SYNTAX, \
+ATTRIBS(A_EXTENSION,A_HVX_IEEE_FP,A_CVI,A_CVI_VX,A_HVX_IEEE_FP_OUT_32), \
+DESCR, DO_FOR_EACH_CODE(WIDTH, CODE))
+/* Single pipe, 16-bit output */
+#define ITERATOR_INSN_IEEE_FP_16(WIDTH,TAG,SYNTAX,DESCR,CODE) \
+EXTINSN(V6_##TAG, SYNTAX, \
+ATTRIBS(A_EXTENSION,A_HVX_IEEE_FP,A_CVI,A_CVI_VX,A_HVX_IEEE_FP_OUT_16), \
+DESCR, DO_FOR_EACH_CODE(WIDTH, CODE))
+/* Two pipes: P2 & P3, single output: P2, 32-bit output */
+#define ITERATOR_INSN_IEEE_FP_DOUBLE_SINGLE_32(WIDTH,TAG,SYNTAX,DESCR,CODE) \
+EXTINSN(V6_##TAG, SYNTAX, \
+ATTRIBS(A_EXTENSION,A_HVX_IEEE_FP,A_CVI,A_CVI_VX_DV,A_HVX_IEEE_FP_OUT_32), \
+DESCR, DO_FOR_EACH_CODE(WIDTH, CODE))
+
+/* Two pipes: P2 & P3, two outputs, 32-bit output */
+#define ITERATOR_INSN_IEEE_FP_DOUBLE_32(WIDTH,TAG,SYNTAX,DESCR,CODE) \
+EXTINSN(V6_##TAG, SYNTAX, \
+ATTRIBS(A_EXTENSION,A_HVX_IEEE_FP,A_CVI,A_CVI_VX_DV,A_HVX_IEEE_FP_OUT_32), \
+DESCR, DO_FOR_EACH_CODE(WIDTH, CODE))
+
+/*
+ * single pipe, accumulate instruction, produces 16-bit output, requires 16-bit
+ * accumulate input
+ */
+#define ITERATOR_INSN_IEEE_FP_ACC_16(WIDTH,TAG,SYNTAX,DESCR,CODE) \
+EXTINSN(V6_##TAG, SYNTAX, \
+ATTRIBS(A_EXTENSION,A_HVX_IEEE_FP,A_CVI,A_CVI_VX,A_HVX_IEEE_FP_ACC,A_HVX_IEEE_FP_OUT_16,A_CVI_VX_NO_TMP_LD), \
+DESCR, DO_FOR_EACH_CODE(WIDTH, CODE))
+
+/*
+ * single pipe, accumulate instruction, produces 32-bit output, requires 32-bit
+ * accumulate input
+ */
+#define ITERATOR_INSN_IEEE_FP_ACC_32(WIDTH,TAG,SYNTAX,DESCR,CODE) \
+EXTINSN(V6_##TAG, SYNTAX, \
+ATTRIBS(A_EXTENSION,A_HVX_IEEE_FP,A_CVI,A_CVI_VX,A_HVX_IEEE_FP_ACC,A_HVX_IEEE_FP_OUT_32,A_CVI_VX_NO_TMP_LD), \
+DESCR, DO_FOR_EACH_CODE(WIDTH, CODE))
+
+/* IEEE FP multiply instructions */
+ITERATOR_INSN_IEEE_FP_DOUBLE_SINGLE_32(32, vmpy_sf_sf,
+ "Vd32.sf=vmpy(Vu32.sf,Vv32.sf)", "Vector IEEE mul: sf",
+ VdV.sf[i] = float32_mul(VuV.sf[i], VvV.sf[i], &env->hvx_fp_status))
+ITERATOR_INSN_IEEE_FP_DOUBLE_32(32, vmpy_sf_hf,
+ "Vdd32.sf=vmpy(Vu32.hf,Vv32.hf)", "Vector IEEE mul: hf widen to sf",
+ VddV.v[0].sf[i] = fp_mult_sf_hf(VuV.hf[2*i], VvV.hf[2*i], &env->hvx_fp_status);
+ VddV.v[1].sf[i] = fp_mult_sf_hf(VuV.hf[2*i+1], VvV.hf[2*i+1], &env->hvx_fp_status))
+ITERATOR_INSN_IEEE_FP_16(16, vmpy_hf_hf, "Vd32.hf=vmpy(Vu32.hf,Vv32.hf)",
+ "Vector IEEE mul: hf",
+ VdV.hf[i] = float16_mul(VuV.hf[i], VvV.hf[i], &env->hvx_fp_status))
+ITERATOR_INSN_IEEE_FP_32(32, vdmpy_sf_hf, "Vd32.sf=vdmpy(Vu32.hf,Vv32.hf)",
+ "Vector IEEE mul reduction: hf widen to sf",
+ VdV.sf[i] = fp_vdmpy(VuV.hf[2*i+1], VuV.hf[2*i], VvV.hf[2*i+1],
+ VvV.hf[2*i], &env->hvx_fp_status))
+
+/* IEEE FP multiply-accumulate instructions */
+ITERATOR_INSN_IEEE_FP_DOUBLE_32(32, vmpy_sf_hf_acc,
+ "Vxx32.sf+=vmpy(Vu32.hf,Vv32.hf)", "Vector IEEE fma: hf widen to sf",
+ VxxV.v[0].sf[i] = float32_muladd(f16_to_f32(VuV.hf[2*i]),
+ f16_to_f32(VvV.hf[2*i]),
+ VxxV.v[0].sf[i], 0, &env->hvx_fp_status);
+ VxxV.v[1].sf[i] = float32_muladd(f16_to_f32(VuV.hf[2*i+1]),
+ f16_to_f32(VvV.hf[2*i+1]),
+ VxxV.v[1].sf[i], 0, &env->hvx_fp_status))
+ITERATOR_INSN_IEEE_FP_ACC_16(16, vmpy_hf_hf_acc,
+ "Vx32.hf+=vmpy(Vu32.hf,Vv32.hf)", "Vector IEEE fma: hf",
+ VxV.hf[i] = float16_muladd(VuV.hf[i], VvV.hf[i], VxV.hf[i], 0, &env->hvx_fp_status))
+ITERATOR_INSN_IEEE_FP_ACC_32(32, vdmpy_sf_hf_acc,
+ "Vx32.sf+=vdmpy(Vu32.hf,Vv32.hf)", "Vector IEEE fma reduce: hf widen to sf",
+ VxV.sf[i] = float32_add(fp_vdmpy(VuV.hf[2*i+1], VuV.hf[2*i],
+ VvV.hf[2*i+1], VvV.hf[2*i],
+ &env->hvx_fp_status),
+ VxV.sf[i], &env->hvx_fp_status))
+
+/* IEEE FP add/sub instructions */
+ITERATOR_INSN_IEEE_FP_32(32, vadd_sf_sf, "Vd32.sf=vadd(Vu32.sf,Vv32.sf)",
+ "Vector IEEE add: sf",
+ VdV.sf[i] = float32_add(VuV.sf[i], VvV.sf[i], &env->hvx_fp_status))
+ITERATOR_INSN_IEEE_FP_32(32, vsub_sf_sf, "Vd32.sf=vsub(Vu32.sf,Vv32.sf)",
+ "Vector IEEE sub: sf",
+ VdV.sf[i] = float32_sub(VuV.sf[i], VvV.sf[i], &env->hvx_fp_status))
+ITERATOR_INSN_IEEE_FP_16(16, vadd_hf_hf, "Vd32.hf=vadd(Vu32.hf,Vv32.hf)",
+ "Vector IEEE add: hf",
+ VdV.hf[i] = float16_add(VuV.hf[i], VvV.hf[i], &env->hvx_fp_status))
+ITERATOR_INSN_IEEE_FP_16(16, vsub_hf_hf, "Vd32.hf=vsub(Vu32.hf,Vv32.hf)",
+ "Vector IEEE sub: hf",
+ VdV.hf[i] = float16_sub(VuV.hf[i], VvV.hf[i], &env->hvx_fp_status))
+ITERATOR_INSN_IEEE_FP_DOUBLE_32(32, vadd_sf_hf,
+ "Vdd32.sf=vadd(Vu32.hf,Vv32.hf)", "Vector IEEE add: hf widen to sf",
+ VddV.v[0].sf[i] = float32_add(f16_to_f32(VuV.hf[2*i]),
+ f16_to_f32(VvV.hf[2*i]), &env->hvx_fp_status);
+ VddV.v[1].sf[i] = float32_add(f16_to_f32(VuV.hf[2*i+1]),
+ f16_to_f32(VvV.hf[2*i+1]), &env->hvx_fp_status))
+ITERATOR_INSN_IEEE_FP_DOUBLE_32(32, vsub_sf_hf,
+ "Vdd32.sf=vsub(Vu32.hf,Vv32.hf)", "Vector IEEE sub: hf widen to sf",
+ VddV.v[0].sf[i] = float32_sub(f16_to_f32(VuV.hf[2*i]),
+ f16_to_f32(VvV.hf[2*i]), &env->hvx_fp_status);
+ VddV.v[1].sf[i] = float32_sub(f16_to_f32(VuV.hf[2*i+1]),
+ f16_to_f32(VvV.hf[2*i+1]), &env->hvx_fp_status))
/******************************************************************************
DEBUG Vector/Register Printing
diff --git a/target/hexagon/meson.build b/target/hexagon/meson.build
index d169cf71b2..9195014821 100644
--- a/target/hexagon/meson.build
+++ b/target/hexagon/meson.build
@@ -250,6 +250,7 @@ hexagon_ss.add(files(
'fma_emu.c',
'mmvec/decode_ext_mmvec.c',
'mmvec/system_ext_mmvec.c',
+ 'mmvec/hvx_ieee_fp.c',
))
#
--
2.37.2© 2016 - 2026 Red Hat, Inc.