[PATCH v1] RISC-V: Support FP ADD/SUB/MUL/DIV autovec for VLS mode

public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed

* [PATCH v1] RISC-V: Support FP ADD/SUB/MUL/DIV autovec for VLS mode
@ 2023-09-01  3:33 pan2.li
  2023-09-01  7:28 ` 钟居哲
  0 siblings, 1 reply; 3+ messages in thread
From: pan2.li @ 2023-09-01  3:33 UTC (permalink / raw)
  To: gcc-patches; +Cc: juzhe.zhong, pan2.li, yanzhang.wang, kito.cheng

From: Pan Li <pan2.li@intel.com>

This patch would like to allow the VLS mode autovec for the
floating-point binary operation ADD/SUB/MUL/DIV.

Given below code example:

test (float *out, float *in1, float *in2)
{
  for (int i = 0; i < 128; i++)
    out[i] = in1[i] + in2[i];
}

Before this patch:
test:
  csrr a4,vlenb
  slli a4,a4,1
  li   a5,128
  bleu a5,a4,.L38
  mv   a5,a4
.L38:
  vsetvli  zero,a5,e32,m8,ta,ma
  vle32.v  v16,0(a1)
  vsetvli  a4,zero,e32,m8,ta,ma
  vmv.v.i  v8,0
  vsetvli  zero,a5,e32,m8,tu,ma
  vle32.v  v24,0(a2)
  vfadd.vv v8,v24,v16
  vse32.v  v8,0(a0)
  ret

After this patch:
test:
  li       a5,128
  vsetvli  zero,a5,e32,m1,ta,ma
  vle32.v  v1,0(a2)
  vle32.v  v2,0(a1)
  vfadd.vv v1,v1,v2
  vse32.v  v1,0(a0)
  ret

Please note this patch also fix the execution failure of below
vect test cases.

* vect-alias-check-10.c
* vect-alias-check-11.c
* vect-alias-check-12.c
* vect-alias-check-14.c

Signed-off-by: Pan Li <pan2.li@intel.com>

gcc/ChangeLog:

	* config/riscv/autovec-vls.md (<optab><mode>3): New pattern for
	vls floating-point autovec.
	* config/riscv/vector-iterators.md: New iterator for
	floating-point V and VLS.
	* config/riscv/vector.md: Add VLS to floating-point binop.

gcc/testsuite/ChangeLog:

	* gcc.target/riscv/rvv/autovec/vls/def.h:
	* gcc.target/riscv/rvv/autovec/vls/floating-point-add-1.c: New test.
	* gcc.target/riscv/rvv/autovec/vls/floating-point-add-2.c: New test.
	* gcc.target/riscv/rvv/autovec/vls/floating-point-add-3.c: New test.
	* gcc.target/riscv/rvv/autovec/vls/floating-point-div-1.c: New test.
	* gcc.target/riscv/rvv/autovec/vls/floating-point-div-2.c: New test.
	* gcc.target/riscv/rvv/autovec/vls/floating-point-div-3.c: New test.
	* gcc.target/riscv/rvv/autovec/vls/floating-point-mul-1.c: New test.
	* gcc.target/riscv/rvv/autovec/vls/floating-point-mul-2.c: New test.
	* gcc.target/riscv/rvv/autovec/vls/floating-point-mul-3.c: New test.
	* gcc.target/riscv/rvv/autovec/vls/floating-point-sub-1.c: New test.
	* gcc.target/riscv/rvv/autovec/vls/floating-point-sub-2.c: New test.
	* gcc.target/riscv/rvv/autovec/vls/floating-point-sub-3.c: New test.
---
 gcc/config/riscv/autovec-vls.md               | 24 ++++++
 gcc/config/riscv/vector-iterators.md          | 80 +++++++++++++++++++
 gcc/config/riscv/vector.md                    | 12 +--
 .../gcc.target/riscv/rvv/autovec/vls/def.h    |  8 ++
 .../rvv/autovec/vls/floating-point-add-1.c    | 43 ++++++++++
 .../rvv/autovec/vls/floating-point-add-2.c    | 43 ++++++++++
 .../rvv/autovec/vls/floating-point-add-3.c    | 43 ++++++++++
 .../rvv/autovec/vls/floating-point-div-1.c    | 43 ++++++++++
 .../rvv/autovec/vls/floating-point-div-2.c    | 43 ++++++++++
 .../rvv/autovec/vls/floating-point-div-3.c    | 43 ++++++++++
 .../rvv/autovec/vls/floating-point-mul-1.c    | 43 ++++++++++
 .../rvv/autovec/vls/floating-point-mul-2.c    | 43 ++++++++++
 .../rvv/autovec/vls/floating-point-mul-3.c    | 43 ++++++++++
 .../rvv/autovec/vls/floating-point-sub-1.c    | 43 ++++++++++
 .../rvv/autovec/vls/floating-point-sub-2.c    | 43 ++++++++++
 .../rvv/autovec/vls/floating-point-sub-3.c    | 43 ++++++++++
 16 files changed, 634 insertions(+), 6 deletions(-)
 create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-1.c
 create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-2.c
 create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-3.c
 create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-1.c
 create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-2.c
 create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-3.c
 create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-1.c
 create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-2.c
 create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-3.c
 create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-1.c
 create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-2.c
 create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-3.c

diff --git a/gcc/config/riscv/autovec-vls.md b/gcc/config/riscv/autovec-vls.md
index 503ad691b9a..4ca640c11e2 100644
--- a/gcc/config/riscv/autovec-vls.md
+++ b/gcc/config/riscv/autovec-vls.md
@@ -208,6 +208,30 @@ (define_insn_and_split "<optab><mode>3"
 [(set_attr "type" "vector")]
 )
 
+;; -------------------------------------------------------------------------
+;; ---- [FP] Binary operations
+;; -------------------------------------------------------------------------
+;; Includes:
+;; - vfadd.vv/vfsub.vv/vfmul.vv/vfdiv.vv
+;; - vfadd.vf/vfsub.vf/vfmul.vf/vfdiv.vf
+;; -------------------------------------------------------------------------
+(define_insn_and_split "<optab><mode>3"
+  [(set (match_operand:VLSF 0 "register_operand")
+    (any_float_binop:VLSF
+     (match_operand:VLSF 1 "<binop_rhs1_predicate>")
+     (match_operand:VLSF 2 "<binop_rhs2_predicate>")))]
+  "TARGET_VECTOR && can_create_pseudo_p ()"
+  "#"
+  "&& 1"
+  [(const_int 0)]
+{
+  riscv_vector::emit_vlmax_insn (code_for_pred (<CODE>, <MODE>mode),
+				 riscv_vector::BINARY_OP_FRM_DYN, operands);
+  DONE;
+}
+[(set_attr "type" "vector")]
+)
+
 ;; -------------------------------------------------------------------------------
 ;; ---- [INT] Unary operations
 ;; -------------------------------------------------------------------------------
diff --git a/gcc/config/riscv/vector-iterators.md b/gcc/config/riscv/vector-iterators.md
index 4023a038fe9..09fa787c0fc 100644
--- a/gcc/config/riscv/vector-iterators.md
+++ b/gcc/config/riscv/vector-iterators.md
@@ -445,6 +445,50 @@ (define_mode_iterator V_VLSI [
   (V512DI "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_64 && TARGET_MIN_VLEN >= 4096")
 ])
 
+(define_mode_iterator V_VLSF [
+  (RVVM8HF "TARGET_ZVFH") (RVVM4HF "TARGET_ZVFH") (RVVM2HF "TARGET_ZVFH")
+  (RVVM1HF "TARGET_ZVFH") (RVVMF2HF "TARGET_ZVFH")
+  (RVVMF4HF "TARGET_ZVFH && TARGET_MIN_VLEN > 32")
+  (RVVM8SF "TARGET_VECTOR_ELEN_FP_32") (RVVM4SF "TARGET_VECTOR_ELEN_FP_32") (RVVM2SF "TARGET_VECTOR_ELEN_FP_32")
+  (RVVM1SF "TARGET_VECTOR_ELEN_FP_32") (RVVMF2SF "TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN > 32")
+  (RVVM8DF "TARGET_VECTOR_ELEN_FP_64") (RVVM4DF "TARGET_VECTOR_ELEN_FP_64")
+  (RVVM2DF "TARGET_VECTOR_ELEN_FP_64") (RVVM1DF "TARGET_VECTOR_ELEN_FP_64")
+
+  (V1HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16")
+  (V2HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16")
+  (V4HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16")
+  (V8HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16")
+  (V16HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16")
+  (V32HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 64")
+  (V64HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 128")
+  (V128HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 256")
+  (V256HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 512")
+  (V512HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 1024")
+  (V1024HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 2048")
+  (V2048HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 4096")
+  (V1SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32")
+  (V2SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32")
+  (V4SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32")
+  (V8SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32")
+  (V16SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 64")
+  (V32SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 128")
+  (V64SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 256")
+  (V128SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 512")
+  (V256SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 1024")
+  (V512SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 2048")
+  (V1024SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 4096")
+  (V1DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64")
+  (V2DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64")
+  (V4DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64")
+  (V8DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 64")
+  (V16DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 128")
+  (V32DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 256")
+  (V64DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 512")
+  (V128DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 1024")
+  (V256DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 2048")
+  (V512DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 4096")
+])
+
 (define_mode_iterator VF_ZVFHMIN [
   (RVVM8HF "TARGET_VECTOR_ELEN_FP_16") (RVVM4HF "TARGET_VECTOR_ELEN_FP_16") (RVVM2HF "TARGET_VECTOR_ELEN_FP_16")
   (RVVM1HF "TARGET_VECTOR_ELEN_FP_16") (RVVMF2HF "TARGET_VECTOR_ELEN_FP_16")
@@ -2494,3 +2538,39 @@ (define_mode_iterator VLSI [
   (V128DI "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_64 && TARGET_MIN_VLEN >= 1024")
   (V256DI "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_64 && TARGET_MIN_VLEN >= 2048")
   (V512DI "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_64 && TARGET_MIN_VLEN >= 4096")])
+
+(define_mode_iterator VLSF [
+  (V1HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16")
+  (V2HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16")
+  (V4HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16")
+  (V8HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16")
+  (V16HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16")
+  (V32HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 64")
+  (V64HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 128")
+  (V128HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 256")
+  (V256HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 512")
+  (V512HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 1024")
+  (V1024HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 2048")
+  (V2048HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 4096")
+  (V1SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32")
+  (V2SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32")
+  (V4SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32")
+  (V8SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32")
+  (V16SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 64")
+  (V32SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 128")
+  (V64SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 256")
+  (V128SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 512")
+  (V256SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 1024")
+  (V512SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 2048")
+  (V1024SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 4096")
+  (V1DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64")
+  (V2DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64")
+  (V4DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64")
+  (V8DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 64")
+  (V16DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 128")
+  (V32DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 256")
+  (V64DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 512")
+  (V128DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 1024")
+  (V256DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 2048")
+  (V512DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 4096")
+])
diff --git a/gcc/config/riscv/vector.md b/gcc/config/riscv/vector.md
index 451138a1b45..6fe750ca8a4 100644
--- a/gcc/config/riscv/vector.md
+++ b/gcc/config/riscv/vector.md
@@ -6026,8 +6026,8 @@ (define_insn "@pred_series<mode>"
 ;; -------------------------------------------------------------------------------
 
 (define_insn "@pred_<optab><mode>"
-  [(set (match_operand:VF 0 "register_operand"           "=vd, vd, vr, vr")
-	(if_then_else:VF
+  [(set (match_operand:V_VLSF 0 "register_operand"           "=vd, vd, vr, vr")
+	(if_then_else:V_VLSF
 	  (unspec:<VM>
 	    [(match_operand:<VM> 1 "vector_mask_operand" " vm, vm,Wc1,Wc1")
 	     (match_operand 5 "vector_length_operand"    " rK, rK, rK, rK")
@@ -6038,10 +6038,10 @@ (define_insn "@pred_<optab><mode>"
 	     (reg:SI VL_REGNUM)
 	     (reg:SI VTYPE_REGNUM)
 	     (reg:SI FRM_REGNUM)] UNSPEC_VPREDICATE)
-	  (any_float_binop:VF
-	    (match_operand:VF 3 "register_operand"       " vr, vr, vr, vr")
-	    (match_operand:VF 4 "register_operand"       " vr, vr, vr, vr"))
-	  (match_operand:VF 2 "vector_merge_operand"     " vu,  0, vu,  0")))]
+	  (any_float_binop:V_VLSF
+	    (match_operand:V_VLSF 3 "register_operand"       " vr, vr, vr, vr")
+	    (match_operand:V_VLSF 4 "register_operand"       " vr, vr, vr, vr"))
+	  (match_operand:V_VLSF 2 "vector_merge_operand"     " vu,  0, vu,  0")))]
   "TARGET_VECTOR"
   "vf<insn>.vv\t%0,%3,%4%p1"
   [(set_attr "type" "<float_insn_type>")
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/def.h b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/def.h
index 00a8a8d2849..476f966c427 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/def.h
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/def.h
@@ -95,6 +95,14 @@ typedef double v512df __attribute__ ((vector_size (4096)));
       a[i] = b[i] OP c[i];                                                     \
   }
 
+#define DEF_OP_VX(PREFIX, NUM, TYPE, OP)                                       \
+  void __attribute__ ((noinline, noclone))                                     \
+  PREFIX##_##TYPE##NUM (TYPE *restrict a, TYPE *restrict b, TYPE c)            \
+  {                                                                            \
+    for (int i = 0; i < NUM; ++i)                                              \
+      a[i] = b[i] OP c;                                                        \
+  }
+
 #define DEF_OP_VI_M16(PREFIX, NUM, TYPE, OP)                                   \
   void __attribute__ ((noinline, noclone))                                     \
   PREFIX##_##TYPE##NUM (TYPE *restrict a, TYPE *restrict b, TYPE *restrict c)  \
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-1.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-1.c
new file mode 100644
index 00000000000..5c2da4df24e
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-1.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VV (add, 1, _Float16, +)
+DEF_OP_VV (add, 2, _Float16, +)
+DEF_OP_VV (add, 4, _Float16, +)
+DEF_OP_VV (add, 8, _Float16, +)
+DEF_OP_VV (add, 16, _Float16, +)
+DEF_OP_VV (add, 32, _Float16, +)
+DEF_OP_VV (add, 64, _Float16, +)
+DEF_OP_VV (add, 128, _Float16, +)
+DEF_OP_VV (add, 256, _Float16, +)
+DEF_OP_VV (add, 512, _Float16, +)
+DEF_OP_VV (add, 1024, _Float16, +)
+DEF_OP_VV (add, 2048, _Float16, +)
+
+DEF_OP_VV (add, 1, float, +)
+DEF_OP_VV (add, 2, float, +)
+DEF_OP_VV (add, 4, float, +)
+DEF_OP_VV (add, 8, float, +)
+DEF_OP_VV (add, 16, float, +)
+DEF_OP_VV (add, 32, float, +)
+DEF_OP_VV (add, 64, float, +)
+DEF_OP_VV (add, 128, float, +)
+DEF_OP_VV (add, 256, float, +)
+DEF_OP_VV (add, 512, float, +)
+DEF_OP_VV (add, 1024, float, +)
+
+DEF_OP_VV (add, 1, double, +)
+DEF_OP_VV (add, 2, double, +)
+DEF_OP_VV (add, 4, double, +)
+DEF_OP_VV (add, 8, double, +)
+DEF_OP_VV (add, 16, double, +)
+DEF_OP_VV (add, 32, double, +)
+DEF_OP_VV (add, 64, double, +)
+DEF_OP_VV (add, 128, double, +)
+DEF_OP_VV (add, 256, double, +)
+DEF_OP_VV (add, 512, double, +)
+
+/* { dg-final { scan-assembler-times {vfadd\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-2.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-2.c
new file mode 100644
index 00000000000..73a355bc085
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-2.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VX (add, 1, _Float16, +)
+DEF_OP_VX (add, 2, _Float16, +)
+DEF_OP_VX (add, 4, _Float16, +)
+DEF_OP_VX (add, 8, _Float16, +)
+DEF_OP_VX (add, 16, _Float16, +)
+DEF_OP_VX (add, 32, _Float16, +)
+DEF_OP_VX (add, 64, _Float16, +)
+DEF_OP_VX (add, 128, _Float16, +)
+DEF_OP_VX (add, 256, _Float16, +)
+DEF_OP_VX (add, 512, _Float16, +)
+DEF_OP_VX (add, 1024, _Float16, +)
+DEF_OP_VX (add, 2048, _Float16, +)
+
+DEF_OP_VX (add, 1, float, +)
+DEF_OP_VX (add, 2, float, +)
+DEF_OP_VX (add, 4, float, +)
+DEF_OP_VX (add, 8, float, +)
+DEF_OP_VX (add, 16, float, +)
+DEF_OP_VX (add, 32, float, +)
+DEF_OP_VX (add, 64, float, +)
+DEF_OP_VX (add, 128, float, +)
+DEF_OP_VX (add, 256, float, +)
+DEF_OP_VX (add, 512, float, +)
+DEF_OP_VX (add, 1024, float, +)
+
+DEF_OP_VX (add, 1, double, +)
+DEF_OP_VX (add, 2, double, +)
+DEF_OP_VX (add, 4, double, +)
+DEF_OP_VX (add, 8, double, +)
+DEF_OP_VX (add, 16, double, +)
+DEF_OP_VX (add, 32, double, +)
+DEF_OP_VX (add, 64, double, +)
+DEF_OP_VX (add, 128, double, +)
+DEF_OP_VX (add, 256, double, +)
+DEF_OP_VX (add, 512, double, +)
+
+/* { dg-final { scan-assembler-times {vfadd\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-3.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-3.c
new file mode 100644
index 00000000000..42925e5143a
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-3.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VI_15 (add, 1, _Float16, +)
+DEF_OP_VI_15 (add, 2, _Float16, +)
+DEF_OP_VI_15 (add, 4, _Float16, +)
+DEF_OP_VI_15 (add, 8, _Float16, +)
+DEF_OP_VI_15 (add, 16, _Float16, +)
+DEF_OP_VI_15 (add, 32, _Float16, +)
+DEF_OP_VI_15 (add, 64, _Float16, +)
+DEF_OP_VI_15 (add, 128, _Float16, +)
+DEF_OP_VI_15 (add, 256, _Float16, +)
+DEF_OP_VI_15 (add, 512, _Float16, +)
+DEF_OP_VI_15 (add, 1024, _Float16, +)
+DEF_OP_VI_15 (add, 2048, _Float16, +)
+
+DEF_OP_VI_15 (add, 1, float, +)
+DEF_OP_VI_15 (add, 2, float, +)
+DEF_OP_VI_15 (add, 4, float, +)
+DEF_OP_VI_15 (add, 8, float, +)
+DEF_OP_VI_15 (add, 16, float, +)
+DEF_OP_VI_15 (add, 32, float, +)
+DEF_OP_VI_15 (add, 64, float, +)
+DEF_OP_VI_15 (add, 128, float, +)
+DEF_OP_VI_15 (add, 256, float, +)
+DEF_OP_VI_15 (add, 512, float, +)
+DEF_OP_VI_15 (add, 1024, float, +)
+
+DEF_OP_VI_15 (add, 1, double, +)
+DEF_OP_VI_15 (add, 2, double, +)
+DEF_OP_VI_15 (add, 4, double, +)
+DEF_OP_VI_15 (add, 8, double, +)
+DEF_OP_VI_15 (add, 16, double, +)
+DEF_OP_VI_15 (add, 32, double, +)
+DEF_OP_VI_15 (add, 64, double, +)
+DEF_OP_VI_15 (add, 128, double, +)
+DEF_OP_VI_15 (add, 256, double, +)
+DEF_OP_VI_15 (add, 512, double, +)
+
+/* { dg-final { scan-assembler-times {vfadd\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-1.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-1.c
new file mode 100644
index 00000000000..93a9e39d00c
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-1.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VV (div, 1, _Float16, /)
+DEF_OP_VV (div, 2, _Float16, /)
+DEF_OP_VV (div, 4, _Float16, /)
+DEF_OP_VV (div, 8, _Float16, /)
+DEF_OP_VV (div, 16, _Float16, /)
+DEF_OP_VV (div, 32, _Float16, /)
+DEF_OP_VV (div, 64, _Float16, /)
+DEF_OP_VV (div, 128, _Float16, /)
+DEF_OP_VV (div, 256, _Float16, /)
+DEF_OP_VV (div, 512, _Float16, /)
+DEF_OP_VV (div, 1024, _Float16, /)
+DEF_OP_VV (div, 2048, _Float16, /)
+
+DEF_OP_VV (div, 1, float, /)
+DEF_OP_VV (div, 2, float, /)
+DEF_OP_VV (div, 4, float, /)
+DEF_OP_VV (div, 8, float, /)
+DEF_OP_VV (div, 16, float, /)
+DEF_OP_VV (div, 32, float, /)
+DEF_OP_VV (div, 64, float, /)
+DEF_OP_VV (div, 128, float, /)
+DEF_OP_VV (div, 256, float, /)
+DEF_OP_VV (div, 512, float, /)
+DEF_OP_VV (div, 1024, float, /)
+
+DEF_OP_VV (div, 1, double, /)
+DEF_OP_VV (div, 2, double, /)
+DEF_OP_VV (div, 4, double, /)
+DEF_OP_VV (div, 8, double, /)
+DEF_OP_VV (div, 16, double, /)
+DEF_OP_VV (div, 32, double, /)
+DEF_OP_VV (div, 64, double, /)
+DEF_OP_VV (div, 128, double, /)
+DEF_OP_VV (div, 256, double, /)
+DEF_OP_VV (div, 512, double, /)
+
+/* { dg-final { scan-assembler-times {vfdiv\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-2.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-2.c
new file mode 100644
index 00000000000..a5bc2a0e970
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-2.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VX (div, 1, _Float16, /)
+DEF_OP_VX (div, 2, _Float16, /)
+DEF_OP_VX (div, 4, _Float16, /)
+DEF_OP_VX (div, 8, _Float16, /)
+DEF_OP_VX (div, 16, _Float16, /)
+DEF_OP_VX (div, 32, _Float16, /)
+DEF_OP_VX (div, 64, _Float16, /)
+DEF_OP_VX (div, 128, _Float16, /)
+DEF_OP_VX (div, 256, _Float16, /)
+DEF_OP_VX (div, 512, _Float16, /)
+DEF_OP_VX (div, 1024, _Float16, /)
+DEF_OP_VX (div, 2048, _Float16, /)
+
+DEF_OP_VX (div, 1, float, /)
+DEF_OP_VX (div, 2, float, /)
+DEF_OP_VX (div, 4, float, /)
+DEF_OP_VX (div, 8, float, /)
+DEF_OP_VX (div, 16, float, /)
+DEF_OP_VX (div, 32, float, /)
+DEF_OP_VX (div, 64, float, /)
+DEF_OP_VX (div, 128, float, /)
+DEF_OP_VX (div, 256, float, /)
+DEF_OP_VX (div, 512, float, /)
+DEF_OP_VX (div, 1024, float, /)
+
+DEF_OP_VX (div, 1, double, /)
+DEF_OP_VX (div, 2, double, /)
+DEF_OP_VX (div, 4, double, /)
+DEF_OP_VX (div, 8, double, /)
+DEF_OP_VX (div, 16, double, /)
+DEF_OP_VX (div, 32, double, /)
+DEF_OP_VX (div, 64, double, /)
+DEF_OP_VX (div, 128, double, /)
+DEF_OP_VX (div, 256, double, /)
+DEF_OP_VX (div, 512, double, /)
+
+/* { dg-final { scan-assembler-times {vfdiv\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-3.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-3.c
new file mode 100644
index 00000000000..f1fb7ed2c9d
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-3.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VI_15 (div, 1, _Float16, /)
+DEF_OP_VI_15 (div, 2, _Float16, /)
+DEF_OP_VI_15 (div, 4, _Float16, /)
+DEF_OP_VI_15 (div, 8, _Float16, /)
+DEF_OP_VI_15 (div, 16, _Float16, /)
+DEF_OP_VI_15 (div, 32, _Float16, /)
+DEF_OP_VI_15 (div, 64, _Float16, /)
+DEF_OP_VI_15 (div, 128, _Float16, /)
+DEF_OP_VI_15 (div, 256, _Float16, /)
+DEF_OP_VI_15 (div, 512, _Float16, /)
+DEF_OP_VI_15 (div, 1024, _Float16, /)
+DEF_OP_VI_15 (div, 2048, _Float16, /)
+
+DEF_OP_VI_15 (div, 1, float, /)
+DEF_OP_VI_15 (div, 2, float, /)
+DEF_OP_VI_15 (div, 4, float, /)
+DEF_OP_VI_15 (div, 8, float, /)
+DEF_OP_VI_15 (div, 16, float, /)
+DEF_OP_VI_15 (div, 32, float, /)
+DEF_OP_VI_15 (div, 64, float, /)
+DEF_OP_VI_15 (div, 128, float, /)
+DEF_OP_VI_15 (div, 256, float, /)
+DEF_OP_VI_15 (div, 512, float, /)
+DEF_OP_VI_15 (div, 1024, float, /)
+
+DEF_OP_VI_15 (div, 1, double, /)
+DEF_OP_VI_15 (div, 2, double, /)
+DEF_OP_VI_15 (div, 4, double, /)
+DEF_OP_VI_15 (div, 8, double, /)
+DEF_OP_VI_15 (div, 16, double, /)
+DEF_OP_VI_15 (div, 32, double, /)
+DEF_OP_VI_15 (div, 64, double, /)
+DEF_OP_VI_15 (div, 128, double, /)
+DEF_OP_VI_15 (div, 256, double, /)
+DEF_OP_VI_15 (div, 512, double, /)
+
+/* { dg-final { scan-assembler-times {vfdiv\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-1.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-1.c
new file mode 100644
index 00000000000..3beccb275ab
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-1.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VV (mul, 1, _Float16, *)
+DEF_OP_VV (mul, 2, _Float16, *)
+DEF_OP_VV (mul, 4, _Float16, *)
+DEF_OP_VV (mul, 8, _Float16, *)
+DEF_OP_VV (mul, 16, _Float16, *)
+DEF_OP_VV (mul, 32, _Float16, *)
+DEF_OP_VV (mul, 64, _Float16, *)
+DEF_OP_VV (mul, 128, _Float16, *)
+DEF_OP_VV (mul, 256, _Float16, *)
+DEF_OP_VV (mul, 512, _Float16, *)
+DEF_OP_VV (mul, 1024, _Float16, *)
+DEF_OP_VV (mul, 2048, _Float16, *)
+
+DEF_OP_VV (mul, 1, float, *)
+DEF_OP_VV (mul, 2, float, *)
+DEF_OP_VV (mul, 4, float, *)
+DEF_OP_VV (mul, 8, float, *)
+DEF_OP_VV (mul, 16, float, *)
+DEF_OP_VV (mul, 32, float, *)
+DEF_OP_VV (mul, 64, float, *)
+DEF_OP_VV (mul, 128, float, *)
+DEF_OP_VV (mul, 256, float, *)
+DEF_OP_VV (mul, 512, float, *)
+DEF_OP_VV (mul, 1024, float, *)
+
+DEF_OP_VV (mul, 1, double, *)
+DEF_OP_VV (mul, 2, double, *)
+DEF_OP_VV (mul, 4, double, *)
+DEF_OP_VV (mul, 8, double, *)
+DEF_OP_VV (mul, 16, double, *)
+DEF_OP_VV (mul, 32, double, *)
+DEF_OP_VV (mul, 64, double, *)
+DEF_OP_VV (mul, 128, double, *)
+DEF_OP_VV (mul, 256, double, *)
+DEF_OP_VV (mul, 512, double, *)
+
+/* { dg-final { scan-assembler-times {vfmul\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-2.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-2.c
new file mode 100644
index 00000000000..b9616386177
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-2.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VX (mul, 1, _Float16, *)
+DEF_OP_VX (mul, 2, _Float16, *)
+DEF_OP_VX (mul, 4, _Float16, *)
+DEF_OP_VX (mul, 8, _Float16, *)
+DEF_OP_VX (mul, 16, _Float16, *)
+DEF_OP_VX (mul, 32, _Float16, *)
+DEF_OP_VX (mul, 64, _Float16, *)
+DEF_OP_VX (mul, 128, _Float16, *)
+DEF_OP_VX (mul, 256, _Float16, *)
+DEF_OP_VX (mul, 512, _Float16, *)
+DEF_OP_VX (mul, 1024, _Float16, *)
+DEF_OP_VX (mul, 2048, _Float16, *)
+
+DEF_OP_VX (mul, 1, float, *)
+DEF_OP_VX (mul, 2, float, *)
+DEF_OP_VX (mul, 4, float, *)
+DEF_OP_VX (mul, 8, float, *)
+DEF_OP_VX (mul, 16, float, *)
+DEF_OP_VX (mul, 32, float, *)
+DEF_OP_VX (mul, 64, float, *)
+DEF_OP_VX (mul, 128, float, *)
+DEF_OP_VX (mul, 256, float, *)
+DEF_OP_VX (mul, 512, float, *)
+DEF_OP_VX (mul, 1024, float, *)
+
+DEF_OP_VX (mul, 1, double, *)
+DEF_OP_VX (mul, 2, double, *)
+DEF_OP_VX (mul, 4, double, *)
+DEF_OP_VX (mul, 8, double, *)
+DEF_OP_VX (mul, 16, double, *)
+DEF_OP_VX (mul, 32, double, *)
+DEF_OP_VX (mul, 64, double, *)
+DEF_OP_VX (mul, 128, double, *)
+DEF_OP_VX (mul, 256, double, *)
+DEF_OP_VX (mul, 512, double, *)
+
+/* { dg-final { scan-assembler-times {vfmul\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-3.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-3.c
new file mode 100644
index 00000000000..d8e4e26bc00
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-3.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VI_15 (mul, 1, _Float16, *)
+DEF_OP_VI_15 (mul, 2, _Float16, *)
+DEF_OP_VI_15 (mul, 4, _Float16, *)
+DEF_OP_VI_15 (mul, 8, _Float16, *)
+DEF_OP_VI_15 (mul, 16, _Float16, *)
+DEF_OP_VI_15 (mul, 32, _Float16, *)
+DEF_OP_VI_15 (mul, 64, _Float16, *)
+DEF_OP_VI_15 (mul, 128, _Float16, *)
+DEF_OP_VI_15 (mul, 256, _Float16, *)
+DEF_OP_VI_15 (mul, 512, _Float16, *)
+DEF_OP_VI_15 (mul, 1024, _Float16, *)
+DEF_OP_VI_15 (mul, 2048, _Float16, *)
+
+DEF_OP_VI_15 (mul, 1, float, *)
+DEF_OP_VI_15 (mul, 2, float, *)
+DEF_OP_VI_15 (mul, 4, float, *)
+DEF_OP_VI_15 (mul, 8, float, *)
+DEF_OP_VI_15 (mul, 16, float, *)
+DEF_OP_VI_15 (mul, 32, float, *)
+DEF_OP_VI_15 (mul, 64, float, *)
+DEF_OP_VI_15 (mul, 128, float, *)
+DEF_OP_VI_15 (mul, 256, float, *)
+DEF_OP_VI_15 (mul, 512, float, *)
+DEF_OP_VI_15 (mul, 1024, float, *)
+
+DEF_OP_VI_15 (mul, 1, double, *)
+DEF_OP_VI_15 (mul, 2, double, *)
+DEF_OP_VI_15 (mul, 4, double, *)
+DEF_OP_VI_15 (mul, 8, double, *)
+DEF_OP_VI_15 (mul, 16, double, *)
+DEF_OP_VI_15 (mul, 32, double, *)
+DEF_OP_VI_15 (mul, 64, double, *)
+DEF_OP_VI_15 (mul, 128, double, *)
+DEF_OP_VI_15 (mul, 256, double, *)
+DEF_OP_VI_15 (mul, 512, double, *)
+
+/* { dg-final { scan-assembler-times {vfmul\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-1.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-1.c
new file mode 100644
index 00000000000..75fe340935c
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-1.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VV (sub, 1, _Float16, -)
+DEF_OP_VV (sub, 2, _Float16, -)
+DEF_OP_VV (sub, 4, _Float16, -)
+DEF_OP_VV (sub, 8, _Float16, -)
+DEF_OP_VV (sub, 16, _Float16, -)
+DEF_OP_VV (sub, 32, _Float16, -)
+DEF_OP_VV (sub, 64, _Float16, -)
+DEF_OP_VV (sub, 128, _Float16, -)
+DEF_OP_VV (sub, 256, _Float16, -)
+DEF_OP_VV (sub, 512, _Float16, -)
+DEF_OP_VV (sub, 1024, _Float16, -)
+DEF_OP_VV (sub, 2048, _Float16, -)
+
+DEF_OP_VV (sub, 1, float, -)
+DEF_OP_VV (sub, 2, float, -)
+DEF_OP_VV (sub, 4, float, -)
+DEF_OP_VV (sub, 8, float, -)
+DEF_OP_VV (sub, 16, float, -)
+DEF_OP_VV (sub, 32, float, -)
+DEF_OP_VV (sub, 64, float, -)
+DEF_OP_VV (sub, 128, float, -)
+DEF_OP_VV (sub, 256, float, -)
+DEF_OP_VV (sub, 512, float, -)
+DEF_OP_VV (sub, 1024, float, -)
+
+DEF_OP_VV (sub, 1, double, -)
+DEF_OP_VV (sub, 2, double, -)
+DEF_OP_VV (sub, 4, double, -)
+DEF_OP_VV (sub, 8, double, -)
+DEF_OP_VV (sub, 16, double, -)
+DEF_OP_VV (sub, 32, double, -)
+DEF_OP_VV (sub, 64, double, -)
+DEF_OP_VV (sub, 128, double, -)
+DEF_OP_VV (sub, 256, double, -)
+DEF_OP_VV (sub, 512, double, -)
+
+/* { dg-final { scan-assembler-times {vfsub\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-2.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-2.c
new file mode 100644
index 00000000000..96a6fe6df2c
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-2.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VX (sub, 1, _Float16, -)
+DEF_OP_VX (sub, 2, _Float16, -)
+DEF_OP_VX (sub, 4, _Float16, -)
+DEF_OP_VX (sub, 8, _Float16, -)
+DEF_OP_VX (sub, 16, _Float16, -)
+DEF_OP_VX (sub, 32, _Float16, -)
+DEF_OP_VX (sub, 64, _Float16, -)
+DEF_OP_VX (sub, 128, _Float16, -)
+DEF_OP_VX (sub, 256, _Float16, -)
+DEF_OP_VX (sub, 512, _Float16, -)
+DEF_OP_VX (sub, 1024, _Float16, -)
+DEF_OP_VX (sub, 2048, _Float16, -)
+
+DEF_OP_VX (sub, 1, float, -)
+DEF_OP_VX (sub, 2, float, -)
+DEF_OP_VX (sub, 4, float, -)
+DEF_OP_VX (sub, 8, float, -)
+DEF_OP_VX (sub, 16, float, -)
+DEF_OP_VX (sub, 32, float, -)
+DEF_OP_VX (sub, 64, float, -)
+DEF_OP_VX (sub, 128, float, -)
+DEF_OP_VX (sub, 256, float, -)
+DEF_OP_VX (sub, 512, float, -)
+DEF_OP_VX (sub, 1024, float, -)
+
+DEF_OP_VX (sub, 1, double, -)
+DEF_OP_VX (sub, 2, double, -)
+DEF_OP_VX (sub, 4, double, -)
+DEF_OP_VX (sub, 8, double, -)
+DEF_OP_VX (sub, 16, double, -)
+DEF_OP_VX (sub, 32, double, -)
+DEF_OP_VX (sub, 64, double, -)
+DEF_OP_VX (sub, 128, double, -)
+DEF_OP_VX (sub, 256, double, -)
+DEF_OP_VX (sub, 512, double, -)
+
+/* { dg-final { scan-assembler-times {vfsub\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-3.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-3.c
new file mode 100644
index 00000000000..0094e2cbf4b
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-3.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VI_15 (sub, 1, _Float16, -)
+DEF_OP_VI_15 (sub, 2, _Float16, -)
+DEF_OP_VI_15 (sub, 4, _Float16, -)
+DEF_OP_VI_15 (sub, 8, _Float16, -)
+DEF_OP_VI_15 (sub, 16, _Float16, -)
+DEF_OP_VI_15 (sub, 32, _Float16, -)
+DEF_OP_VI_15 (sub, 64, _Float16, -)
+DEF_OP_VI_15 (sub, 128, _Float16, -)
+DEF_OP_VI_15 (sub, 256, _Float16, -)
+DEF_OP_VI_15 (sub, 512, _Float16, -)
+DEF_OP_VI_15 (sub, 1024, _Float16, -)
+DEF_OP_VI_15 (sub, 2048, _Float16, -)
+
+DEF_OP_VI_15 (sub, 1, float, -)
+DEF_OP_VI_15 (sub, 2, float, -)
+DEF_OP_VI_15 (sub, 4, float, -)
+DEF_OP_VI_15 (sub, 8, float, -)
+DEF_OP_VI_15 (sub, 16, float, -)
+DEF_OP_VI_15 (sub, 32, float, -)
+DEF_OP_VI_15 (sub, 64, float, -)
+DEF_OP_VI_15 (sub, 128, float, -)
+DEF_OP_VI_15 (sub, 256, float, -)
+DEF_OP_VI_15 (sub, 512, float, -)
+DEF_OP_VI_15 (sub, 1024, float, -)
+
+DEF_OP_VI_15 (sub, 1, double, -)
+DEF_OP_VI_15 (sub, 2, double, -)
+DEF_OP_VI_15 (sub, 4, double, -)
+DEF_OP_VI_15 (sub, 8, double, -)
+DEF_OP_VI_15 (sub, 16, double, -)
+DEF_OP_VI_15 (sub, 32, double, -)
+DEF_OP_VI_15 (sub, 64, double, -)
+DEF_OP_VI_15 (sub, 128, double, -)
+DEF_OP_VI_15 (sub, 256, double, -)
+DEF_OP_VI_15 (sub, 512, double, -)
+
+/* { dg-final { scan-assembler-times {vfadd\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
-- 
2.34.1


^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [PATCH v1] RISC-V: Support FP ADD/SUB/MUL/DIV autovec for VLS mode
  2023-09-01  3:33 [PATCH v1] RISC-V: Support FP ADD/SUB/MUL/DIV autovec for VLS mode pan2.li
@ 2023-09-01  7:28 ` 钟居哲
  2023-09-01  8:38   ` Li, Pan2
  0 siblings, 1 reply; 3+ messages in thread
From: 钟居哲 @ 2023-09-01  7:28 UTC (permalink / raw)
  To: pan2.li, gcc-patches; +Cc: pan2.li, yanzhang.wang, kito.cheng

[-- Attachment #1: Type: text/plain, Size: 36855 bytes --]

LGTM。



juzhe.zhong@rivai.ai
 
From: pan2.li
Date: 2023-09-01 11:33
To: gcc-patches
CC: juzhe.zhong; pan2.li; yanzhang.wang; kito.cheng
Subject: [PATCH v1] RISC-V: Support FP ADD/SUB/MUL/DIV autovec for VLS mode
From: Pan Li <pan2.li@intel.com>
 
This patch would like to allow the VLS mode autovec for the
floating-point binary operation ADD/SUB/MUL/DIV.
 
Given below code example:
 
test (float *out, float *in1, float *in2)
{
  for (int i = 0; i < 128; i++)
    out[i] = in1[i] + in2[i];
}
 
Before this patch:
test:
  csrr a4,vlenb
  slli a4,a4,1
  li   a5,128
  bleu a5,a4,.L38
  mv   a5,a4
.L38:
  vsetvli  zero,a5,e32,m8,ta,ma
  vle32.v  v16,0(a1)
  vsetvli  a4,zero,e32,m8,ta,ma
  vmv.v.i  v8,0
  vsetvli  zero,a5,e32,m8,tu,ma
  vle32.v  v24,0(a2)
  vfadd.vv v8,v24,v16
  vse32.v  v8,0(a0)
  ret
 
After this patch:
test:
  li       a5,128
  vsetvli  zero,a5,e32,m1,ta,ma
  vle32.v  v1,0(a2)
  vle32.v  v2,0(a1)
  vfadd.vv v1,v1,v2
  vse32.v  v1,0(a0)
  ret
 
Please note this patch also fix the execution failure of below
vect test cases.
 
* vect-alias-check-10.c
* vect-alias-check-11.c
* vect-alias-check-12.c
* vect-alias-check-14.c
 
Signed-off-by: Pan Li <pan2.li@intel.com>
 
gcc/ChangeLog:
 
* config/riscv/autovec-vls.md (<optab><mode>3): New pattern for
vls floating-point autovec.
* config/riscv/vector-iterators.md: New iterator for
floating-point V and VLS.
* config/riscv/vector.md: Add VLS to floating-point binop.
 
gcc/testsuite/ChangeLog:
 
* gcc.target/riscv/rvv/autovec/vls/def.h:
* gcc.target/riscv/rvv/autovec/vls/floating-point-add-1.c: New test.
* gcc.target/riscv/rvv/autovec/vls/floating-point-add-2.c: New test.
* gcc.target/riscv/rvv/autovec/vls/floating-point-add-3.c: New test.
* gcc.target/riscv/rvv/autovec/vls/floating-point-div-1.c: New test.
* gcc.target/riscv/rvv/autovec/vls/floating-point-div-2.c: New test.
* gcc.target/riscv/rvv/autovec/vls/floating-point-div-3.c: New test.
* gcc.target/riscv/rvv/autovec/vls/floating-point-mul-1.c: New test.
* gcc.target/riscv/rvv/autovec/vls/floating-point-mul-2.c: New test.
* gcc.target/riscv/rvv/autovec/vls/floating-point-mul-3.c: New test.
* gcc.target/riscv/rvv/autovec/vls/floating-point-sub-1.c: New test.
* gcc.target/riscv/rvv/autovec/vls/floating-point-sub-2.c: New test.
* gcc.target/riscv/rvv/autovec/vls/floating-point-sub-3.c: New test.
---
gcc/config/riscv/autovec-vls.md               | 24 ++++++
gcc/config/riscv/vector-iterators.md          | 80 +++++++++++++++++++
gcc/config/riscv/vector.md                    | 12 +--
.../gcc.target/riscv/rvv/autovec/vls/def.h    |  8 ++
.../rvv/autovec/vls/floating-point-add-1.c    | 43 ++++++++++
.../rvv/autovec/vls/floating-point-add-2.c    | 43 ++++++++++
.../rvv/autovec/vls/floating-point-add-3.c    | 43 ++++++++++
.../rvv/autovec/vls/floating-point-div-1.c    | 43 ++++++++++
.../rvv/autovec/vls/floating-point-div-2.c    | 43 ++++++++++
.../rvv/autovec/vls/floating-point-div-3.c    | 43 ++++++++++
.../rvv/autovec/vls/floating-point-mul-1.c    | 43 ++++++++++
.../rvv/autovec/vls/floating-point-mul-2.c    | 43 ++++++++++
.../rvv/autovec/vls/floating-point-mul-3.c    | 43 ++++++++++
.../rvv/autovec/vls/floating-point-sub-1.c    | 43 ++++++++++
.../rvv/autovec/vls/floating-point-sub-2.c    | 43 ++++++++++
.../rvv/autovec/vls/floating-point-sub-3.c    | 43 ++++++++++
16 files changed, 634 insertions(+), 6 deletions(-)
create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-1.c
create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-2.c
create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-3.c
create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-1.c
create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-2.c
create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-3.c
create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-1.c
create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-2.c
create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-3.c
create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-1.c
create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-2.c
create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-3.c
 
diff --git a/gcc/config/riscv/autovec-vls.md b/gcc/config/riscv/autovec-vls.md
index 503ad691b9a..4ca640c11e2 100644
--- a/gcc/config/riscv/autovec-vls.md
+++ b/gcc/config/riscv/autovec-vls.md
@@ -208,6 +208,30 @@ (define_insn_and_split "<optab><mode>3"
[(set_attr "type" "vector")]
)
+;; -------------------------------------------------------------------------
+;; ---- [FP] Binary operations
+;; -------------------------------------------------------------------------
+;; Includes:
+;; - vfadd.vv/vfsub.vv/vfmul.vv/vfdiv.vv
+;; - vfadd.vf/vfsub.vf/vfmul.vf/vfdiv.vf
+;; -------------------------------------------------------------------------
+(define_insn_and_split "<optab><mode>3"
+  [(set (match_operand:VLSF 0 "register_operand")
+    (any_float_binop:VLSF
+     (match_operand:VLSF 1 "<binop_rhs1_predicate>")
+     (match_operand:VLSF 2 "<binop_rhs2_predicate>")))]
+  "TARGET_VECTOR && can_create_pseudo_p ()"
+  "#"
+  "&& 1"
+  [(const_int 0)]
+{
+  riscv_vector::emit_vlmax_insn (code_for_pred (<CODE>, <MODE>mode),
+ riscv_vector::BINARY_OP_FRM_DYN, operands);
+  DONE;
+}
+[(set_attr "type" "vector")]
+)
+
;; -------------------------------------------------------------------------------
;; ---- [INT] Unary operations
;; -------------------------------------------------------------------------------
diff --git a/gcc/config/riscv/vector-iterators.md b/gcc/config/riscv/vector-iterators.md
index 4023a038fe9..09fa787c0fc 100644
--- a/gcc/config/riscv/vector-iterators.md
+++ b/gcc/config/riscv/vector-iterators.md
@@ -445,6 +445,50 @@ (define_mode_iterator V_VLSI [
   (V512DI "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_64 && TARGET_MIN_VLEN >= 4096")
])
+(define_mode_iterator V_VLSF [
+  (RVVM8HF "TARGET_ZVFH") (RVVM4HF "TARGET_ZVFH") (RVVM2HF "TARGET_ZVFH")
+  (RVVM1HF "TARGET_ZVFH") (RVVMF2HF "TARGET_ZVFH")
+  (RVVMF4HF "TARGET_ZVFH && TARGET_MIN_VLEN > 32")
+  (RVVM8SF "TARGET_VECTOR_ELEN_FP_32") (RVVM4SF "TARGET_VECTOR_ELEN_FP_32") (RVVM2SF "TARGET_VECTOR_ELEN_FP_32")
+  (RVVM1SF "TARGET_VECTOR_ELEN_FP_32") (RVVMF2SF "TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN > 32")
+  (RVVM8DF "TARGET_VECTOR_ELEN_FP_64") (RVVM4DF "TARGET_VECTOR_ELEN_FP_64")
+  (RVVM2DF "TARGET_VECTOR_ELEN_FP_64") (RVVM1DF "TARGET_VECTOR_ELEN_FP_64")
+
+  (V1HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16")
+  (V2HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16")
+  (V4HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16")
+  (V8HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16")
+  (V16HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16")
+  (V32HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 64")
+  (V64HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 128")
+  (V128HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 256")
+  (V256HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 512")
+  (V512HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 1024")
+  (V1024HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 2048")
+  (V2048HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 4096")
+  (V1SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32")
+  (V2SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32")
+  (V4SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32")
+  (V8SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32")
+  (V16SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 64")
+  (V32SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 128")
+  (V64SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 256")
+  (V128SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 512")
+  (V256SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 1024")
+  (V512SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 2048")
+  (V1024SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 4096")
+  (V1DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64")
+  (V2DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64")
+  (V4DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64")
+  (V8DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 64")
+  (V16DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 128")
+  (V32DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 256")
+  (V64DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 512")
+  (V128DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 1024")
+  (V256DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 2048")
+  (V512DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 4096")
+])
+
(define_mode_iterator VF_ZVFHMIN [
   (RVVM8HF "TARGET_VECTOR_ELEN_FP_16") (RVVM4HF "TARGET_VECTOR_ELEN_FP_16") (RVVM2HF "TARGET_VECTOR_ELEN_FP_16")
   (RVVM1HF "TARGET_VECTOR_ELEN_FP_16") (RVVMF2HF "TARGET_VECTOR_ELEN_FP_16")
@@ -2494,3 +2538,39 @@ (define_mode_iterator VLSI [
   (V128DI "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_64 && TARGET_MIN_VLEN >= 1024")
   (V256DI "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_64 && TARGET_MIN_VLEN >= 2048")
   (V512DI "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_64 && TARGET_MIN_VLEN >= 4096")])
+
+(define_mode_iterator VLSF [
+  (V1HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16")
+  (V2HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16")
+  (V4HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16")
+  (V8HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16")
+  (V16HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16")
+  (V32HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 64")
+  (V64HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 128")
+  (V128HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 256")
+  (V256HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 512")
+  (V512HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 1024")
+  (V1024HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 2048")
+  (V2048HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 4096")
+  (V1SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32")
+  (V2SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32")
+  (V4SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32")
+  (V8SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32")
+  (V16SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 64")
+  (V32SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 128")
+  (V64SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 256")
+  (V128SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 512")
+  (V256SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 1024")
+  (V512SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 2048")
+  (V1024SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 4096")
+  (V1DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64")
+  (V2DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64")
+  (V4DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64")
+  (V8DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 64")
+  (V16DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 128")
+  (V32DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 256")
+  (V64DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 512")
+  (V128DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 1024")
+  (V256DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 2048")
+  (V512DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 4096")
+])
diff --git a/gcc/config/riscv/vector.md b/gcc/config/riscv/vector.md
index 451138a1b45..6fe750ca8a4 100644
--- a/gcc/config/riscv/vector.md
+++ b/gcc/config/riscv/vector.md
@@ -6026,8 +6026,8 @@ (define_insn "@pred_series<mode>"
;; -------------------------------------------------------------------------------
(define_insn "@pred_<optab><mode>"
-  [(set (match_operand:VF 0 "register_operand"           "=vd, vd, vr, vr")
- (if_then_else:VF
+  [(set (match_operand:V_VLSF 0 "register_operand"           "=vd, vd, vr, vr")
+ (if_then_else:V_VLSF
  (unspec:<VM>
    [(match_operand:<VM> 1 "vector_mask_operand" " vm, vm,Wc1,Wc1")
     (match_operand 5 "vector_length_operand"    " rK, rK, rK, rK")
@@ -6038,10 +6038,10 @@ (define_insn "@pred_<optab><mode>"
     (reg:SI VL_REGNUM)
     (reg:SI VTYPE_REGNUM)
     (reg:SI FRM_REGNUM)] UNSPEC_VPREDICATE)
-   (any_float_binop:VF
-     (match_operand:VF 3 "register_operand"       " vr, vr, vr, vr")
-     (match_operand:VF 4 "register_operand"       " vr, vr, vr, vr"))
-   (match_operand:VF 2 "vector_merge_operand"     " vu,  0, vu,  0")))]
+   (any_float_binop:V_VLSF
+     (match_operand:V_VLSF 3 "register_operand"       " vr, vr, vr, vr")
+     (match_operand:V_VLSF 4 "register_operand"       " vr, vr, vr, vr"))
+   (match_operand:V_VLSF 2 "vector_merge_operand"     " vu,  0, vu,  0")))]
   "TARGET_VECTOR"
   "vf<insn>.vv\t%0,%3,%4%p1"
   [(set_attr "type" "<float_insn_type>")
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/def.h b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/def.h
index 00a8a8d2849..476f966c427 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/def.h
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/def.h
@@ -95,6 +95,14 @@ typedef double v512df __attribute__ ((vector_size (4096)));
       a[i] = b[i] OP c[i];                                                     \
   }
+#define DEF_OP_VX(PREFIX, NUM, TYPE, OP)                                       \
+  void __attribute__ ((noinline, noclone))                                     \
+  PREFIX##_##TYPE##NUM (TYPE *restrict a, TYPE *restrict b, TYPE c)            \
+  {                                                                            \
+    for (int i = 0; i < NUM; ++i)                                              \
+      a[i] = b[i] OP c;                                                        \
+  }
+
#define DEF_OP_VI_M16(PREFIX, NUM, TYPE, OP)                                   \
   void __attribute__ ((noinline, noclone))                                     \
   PREFIX##_##TYPE##NUM (TYPE *restrict a, TYPE *restrict b, TYPE *restrict c)  \
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-1.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-1.c
new file mode 100644
index 00000000000..5c2da4df24e
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-1.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VV (add, 1, _Float16, +)
+DEF_OP_VV (add, 2, _Float16, +)
+DEF_OP_VV (add, 4, _Float16, +)
+DEF_OP_VV (add, 8, _Float16, +)
+DEF_OP_VV (add, 16, _Float16, +)
+DEF_OP_VV (add, 32, _Float16, +)
+DEF_OP_VV (add, 64, _Float16, +)
+DEF_OP_VV (add, 128, _Float16, +)
+DEF_OP_VV (add, 256, _Float16, +)
+DEF_OP_VV (add, 512, _Float16, +)
+DEF_OP_VV (add, 1024, _Float16, +)
+DEF_OP_VV (add, 2048, _Float16, +)
+
+DEF_OP_VV (add, 1, float, +)
+DEF_OP_VV (add, 2, float, +)
+DEF_OP_VV (add, 4, float, +)
+DEF_OP_VV (add, 8, float, +)
+DEF_OP_VV (add, 16, float, +)
+DEF_OP_VV (add, 32, float, +)
+DEF_OP_VV (add, 64, float, +)
+DEF_OP_VV (add, 128, float, +)
+DEF_OP_VV (add, 256, float, +)
+DEF_OP_VV (add, 512, float, +)
+DEF_OP_VV (add, 1024, float, +)
+
+DEF_OP_VV (add, 1, double, +)
+DEF_OP_VV (add, 2, double, +)
+DEF_OP_VV (add, 4, double, +)
+DEF_OP_VV (add, 8, double, +)
+DEF_OP_VV (add, 16, double, +)
+DEF_OP_VV (add, 32, double, +)
+DEF_OP_VV (add, 64, double, +)
+DEF_OP_VV (add, 128, double, +)
+DEF_OP_VV (add, 256, double, +)
+DEF_OP_VV (add, 512, double, +)
+
+/* { dg-final { scan-assembler-times {vfadd\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-2.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-2.c
new file mode 100644
index 00000000000..73a355bc085
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-2.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VX (add, 1, _Float16, +)
+DEF_OP_VX (add, 2, _Float16, +)
+DEF_OP_VX (add, 4, _Float16, +)
+DEF_OP_VX (add, 8, _Float16, +)
+DEF_OP_VX (add, 16, _Float16, +)
+DEF_OP_VX (add, 32, _Float16, +)
+DEF_OP_VX (add, 64, _Float16, +)
+DEF_OP_VX (add, 128, _Float16, +)
+DEF_OP_VX (add, 256, _Float16, +)
+DEF_OP_VX (add, 512, _Float16, +)
+DEF_OP_VX (add, 1024, _Float16, +)
+DEF_OP_VX (add, 2048, _Float16, +)
+
+DEF_OP_VX (add, 1, float, +)
+DEF_OP_VX (add, 2, float, +)
+DEF_OP_VX (add, 4, float, +)
+DEF_OP_VX (add, 8, float, +)
+DEF_OP_VX (add, 16, float, +)
+DEF_OP_VX (add, 32, float, +)
+DEF_OP_VX (add, 64, float, +)
+DEF_OP_VX (add, 128, float, +)
+DEF_OP_VX (add, 256, float, +)
+DEF_OP_VX (add, 512, float, +)
+DEF_OP_VX (add, 1024, float, +)
+
+DEF_OP_VX (add, 1, double, +)
+DEF_OP_VX (add, 2, double, +)
+DEF_OP_VX (add, 4, double, +)
+DEF_OP_VX (add, 8, double, +)
+DEF_OP_VX (add, 16, double, +)
+DEF_OP_VX (add, 32, double, +)
+DEF_OP_VX (add, 64, double, +)
+DEF_OP_VX (add, 128, double, +)
+DEF_OP_VX (add, 256, double, +)
+DEF_OP_VX (add, 512, double, +)
+
+/* { dg-final { scan-assembler-times {vfadd\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-3.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-3.c
new file mode 100644
index 00000000000..42925e5143a
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-3.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VI_15 (add, 1, _Float16, +)
+DEF_OP_VI_15 (add, 2, _Float16, +)
+DEF_OP_VI_15 (add, 4, _Float16, +)
+DEF_OP_VI_15 (add, 8, _Float16, +)
+DEF_OP_VI_15 (add, 16, _Float16, +)
+DEF_OP_VI_15 (add, 32, _Float16, +)
+DEF_OP_VI_15 (add, 64, _Float16, +)
+DEF_OP_VI_15 (add, 128, _Float16, +)
+DEF_OP_VI_15 (add, 256, _Float16, +)
+DEF_OP_VI_15 (add, 512, _Float16, +)
+DEF_OP_VI_15 (add, 1024, _Float16, +)
+DEF_OP_VI_15 (add, 2048, _Float16, +)
+
+DEF_OP_VI_15 (add, 1, float, +)
+DEF_OP_VI_15 (add, 2, float, +)
+DEF_OP_VI_15 (add, 4, float, +)
+DEF_OP_VI_15 (add, 8, float, +)
+DEF_OP_VI_15 (add, 16, float, +)
+DEF_OP_VI_15 (add, 32, float, +)
+DEF_OP_VI_15 (add, 64, float, +)
+DEF_OP_VI_15 (add, 128, float, +)
+DEF_OP_VI_15 (add, 256, float, +)
+DEF_OP_VI_15 (add, 512, float, +)
+DEF_OP_VI_15 (add, 1024, float, +)
+
+DEF_OP_VI_15 (add, 1, double, +)
+DEF_OP_VI_15 (add, 2, double, +)
+DEF_OP_VI_15 (add, 4, double, +)
+DEF_OP_VI_15 (add, 8, double, +)
+DEF_OP_VI_15 (add, 16, double, +)
+DEF_OP_VI_15 (add, 32, double, +)
+DEF_OP_VI_15 (add, 64, double, +)
+DEF_OP_VI_15 (add, 128, double, +)
+DEF_OP_VI_15 (add, 256, double, +)
+DEF_OP_VI_15 (add, 512, double, +)
+
+/* { dg-final { scan-assembler-times {vfadd\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-1.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-1.c
new file mode 100644
index 00000000000..93a9e39d00c
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-1.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VV (div, 1, _Float16, /)
+DEF_OP_VV (div, 2, _Float16, /)
+DEF_OP_VV (div, 4, _Float16, /)
+DEF_OP_VV (div, 8, _Float16, /)
+DEF_OP_VV (div, 16, _Float16, /)
+DEF_OP_VV (div, 32, _Float16, /)
+DEF_OP_VV (div, 64, _Float16, /)
+DEF_OP_VV (div, 128, _Float16, /)
+DEF_OP_VV (div, 256, _Float16, /)
+DEF_OP_VV (div, 512, _Float16, /)
+DEF_OP_VV (div, 1024, _Float16, /)
+DEF_OP_VV (div, 2048, _Float16, /)
+
+DEF_OP_VV (div, 1, float, /)
+DEF_OP_VV (div, 2, float, /)
+DEF_OP_VV (div, 4, float, /)
+DEF_OP_VV (div, 8, float, /)
+DEF_OP_VV (div, 16, float, /)
+DEF_OP_VV (div, 32, float, /)
+DEF_OP_VV (div, 64, float, /)
+DEF_OP_VV (div, 128, float, /)
+DEF_OP_VV (div, 256, float, /)
+DEF_OP_VV (div, 512, float, /)
+DEF_OP_VV (div, 1024, float, /)
+
+DEF_OP_VV (div, 1, double, /)
+DEF_OP_VV (div, 2, double, /)
+DEF_OP_VV (div, 4, double, /)
+DEF_OP_VV (div, 8, double, /)
+DEF_OP_VV (div, 16, double, /)
+DEF_OP_VV (div, 32, double, /)
+DEF_OP_VV (div, 64, double, /)
+DEF_OP_VV (div, 128, double, /)
+DEF_OP_VV (div, 256, double, /)
+DEF_OP_VV (div, 512, double, /)
+
+/* { dg-final { scan-assembler-times {vfdiv\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-2.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-2.c
new file mode 100644
index 00000000000..a5bc2a0e970
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-2.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VX (div, 1, _Float16, /)
+DEF_OP_VX (div, 2, _Float16, /)
+DEF_OP_VX (div, 4, _Float16, /)
+DEF_OP_VX (div, 8, _Float16, /)
+DEF_OP_VX (div, 16, _Float16, /)
+DEF_OP_VX (div, 32, _Float16, /)
+DEF_OP_VX (div, 64, _Float16, /)
+DEF_OP_VX (div, 128, _Float16, /)
+DEF_OP_VX (div, 256, _Float16, /)
+DEF_OP_VX (div, 512, _Float16, /)
+DEF_OP_VX (div, 1024, _Float16, /)
+DEF_OP_VX (div, 2048, _Float16, /)
+
+DEF_OP_VX (div, 1, float, /)
+DEF_OP_VX (div, 2, float, /)
+DEF_OP_VX (div, 4, float, /)
+DEF_OP_VX (div, 8, float, /)
+DEF_OP_VX (div, 16, float, /)
+DEF_OP_VX (div, 32, float, /)
+DEF_OP_VX (div, 64, float, /)
+DEF_OP_VX (div, 128, float, /)
+DEF_OP_VX (div, 256, float, /)
+DEF_OP_VX (div, 512, float, /)
+DEF_OP_VX (div, 1024, float, /)
+
+DEF_OP_VX (div, 1, double, /)
+DEF_OP_VX (div, 2, double, /)
+DEF_OP_VX (div, 4, double, /)
+DEF_OP_VX (div, 8, double, /)
+DEF_OP_VX (div, 16, double, /)
+DEF_OP_VX (div, 32, double, /)
+DEF_OP_VX (div, 64, double, /)
+DEF_OP_VX (div, 128, double, /)
+DEF_OP_VX (div, 256, double, /)
+DEF_OP_VX (div, 512, double, /)
+
+/* { dg-final { scan-assembler-times {vfdiv\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-3.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-3.c
new file mode 100644
index 00000000000..f1fb7ed2c9d
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-3.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VI_15 (div, 1, _Float16, /)
+DEF_OP_VI_15 (div, 2, _Float16, /)
+DEF_OP_VI_15 (div, 4, _Float16, /)
+DEF_OP_VI_15 (div, 8, _Float16, /)
+DEF_OP_VI_15 (div, 16, _Float16, /)
+DEF_OP_VI_15 (div, 32, _Float16, /)
+DEF_OP_VI_15 (div, 64, _Float16, /)
+DEF_OP_VI_15 (div, 128, _Float16, /)
+DEF_OP_VI_15 (div, 256, _Float16, /)
+DEF_OP_VI_15 (div, 512, _Float16, /)
+DEF_OP_VI_15 (div, 1024, _Float16, /)
+DEF_OP_VI_15 (div, 2048, _Float16, /)
+
+DEF_OP_VI_15 (div, 1, float, /)
+DEF_OP_VI_15 (div, 2, float, /)
+DEF_OP_VI_15 (div, 4, float, /)
+DEF_OP_VI_15 (div, 8, float, /)
+DEF_OP_VI_15 (div, 16, float, /)
+DEF_OP_VI_15 (div, 32, float, /)
+DEF_OP_VI_15 (div, 64, float, /)
+DEF_OP_VI_15 (div, 128, float, /)
+DEF_OP_VI_15 (div, 256, float, /)
+DEF_OP_VI_15 (div, 512, float, /)
+DEF_OP_VI_15 (div, 1024, float, /)
+
+DEF_OP_VI_15 (div, 1, double, /)
+DEF_OP_VI_15 (div, 2, double, /)
+DEF_OP_VI_15 (div, 4, double, /)
+DEF_OP_VI_15 (div, 8, double, /)
+DEF_OP_VI_15 (div, 16, double, /)
+DEF_OP_VI_15 (div, 32, double, /)
+DEF_OP_VI_15 (div, 64, double, /)
+DEF_OP_VI_15 (div, 128, double, /)
+DEF_OP_VI_15 (div, 256, double, /)
+DEF_OP_VI_15 (div, 512, double, /)
+
+/* { dg-final { scan-assembler-times {vfdiv\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-1.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-1.c
new file mode 100644
index 00000000000..3beccb275ab
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-1.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VV (mul, 1, _Float16, *)
+DEF_OP_VV (mul, 2, _Float16, *)
+DEF_OP_VV (mul, 4, _Float16, *)
+DEF_OP_VV (mul, 8, _Float16, *)
+DEF_OP_VV (mul, 16, _Float16, *)
+DEF_OP_VV (mul, 32, _Float16, *)
+DEF_OP_VV (mul, 64, _Float16, *)
+DEF_OP_VV (mul, 128, _Float16, *)
+DEF_OP_VV (mul, 256, _Float16, *)
+DEF_OP_VV (mul, 512, _Float16, *)
+DEF_OP_VV (mul, 1024, _Float16, *)
+DEF_OP_VV (mul, 2048, _Float16, *)
+
+DEF_OP_VV (mul, 1, float, *)
+DEF_OP_VV (mul, 2, float, *)
+DEF_OP_VV (mul, 4, float, *)
+DEF_OP_VV (mul, 8, float, *)
+DEF_OP_VV (mul, 16, float, *)
+DEF_OP_VV (mul, 32, float, *)
+DEF_OP_VV (mul, 64, float, *)
+DEF_OP_VV (mul, 128, float, *)
+DEF_OP_VV (mul, 256, float, *)
+DEF_OP_VV (mul, 512, float, *)
+DEF_OP_VV (mul, 1024, float, *)
+
+DEF_OP_VV (mul, 1, double, *)
+DEF_OP_VV (mul, 2, double, *)
+DEF_OP_VV (mul, 4, double, *)
+DEF_OP_VV (mul, 8, double, *)
+DEF_OP_VV (mul, 16, double, *)
+DEF_OP_VV (mul, 32, double, *)
+DEF_OP_VV (mul, 64, double, *)
+DEF_OP_VV (mul, 128, double, *)
+DEF_OP_VV (mul, 256, double, *)
+DEF_OP_VV (mul, 512, double, *)
+
+/* { dg-final { scan-assembler-times {vfmul\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-2.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-2.c
new file mode 100644
index 00000000000..b9616386177
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-2.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VX (mul, 1, _Float16, *)
+DEF_OP_VX (mul, 2, _Float16, *)
+DEF_OP_VX (mul, 4, _Float16, *)
+DEF_OP_VX (mul, 8, _Float16, *)
+DEF_OP_VX (mul, 16, _Float16, *)
+DEF_OP_VX (mul, 32, _Float16, *)
+DEF_OP_VX (mul, 64, _Float16, *)
+DEF_OP_VX (mul, 128, _Float16, *)
+DEF_OP_VX (mul, 256, _Float16, *)
+DEF_OP_VX (mul, 512, _Float16, *)
+DEF_OP_VX (mul, 1024, _Float16, *)
+DEF_OP_VX (mul, 2048, _Float16, *)
+
+DEF_OP_VX (mul, 1, float, *)
+DEF_OP_VX (mul, 2, float, *)
+DEF_OP_VX (mul, 4, float, *)
+DEF_OP_VX (mul, 8, float, *)
+DEF_OP_VX (mul, 16, float, *)
+DEF_OP_VX (mul, 32, float, *)
+DEF_OP_VX (mul, 64, float, *)
+DEF_OP_VX (mul, 128, float, *)
+DEF_OP_VX (mul, 256, float, *)
+DEF_OP_VX (mul, 512, float, *)
+DEF_OP_VX (mul, 1024, float, *)
+
+DEF_OP_VX (mul, 1, double, *)
+DEF_OP_VX (mul, 2, double, *)
+DEF_OP_VX (mul, 4, double, *)
+DEF_OP_VX (mul, 8, double, *)
+DEF_OP_VX (mul, 16, double, *)
+DEF_OP_VX (mul, 32, double, *)
+DEF_OP_VX (mul, 64, double, *)
+DEF_OP_VX (mul, 128, double, *)
+DEF_OP_VX (mul, 256, double, *)
+DEF_OP_VX (mul, 512, double, *)
+
+/* { dg-final { scan-assembler-times {vfmul\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-3.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-3.c
new file mode 100644
index 00000000000..d8e4e26bc00
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-3.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VI_15 (mul, 1, _Float16, *)
+DEF_OP_VI_15 (mul, 2, _Float16, *)
+DEF_OP_VI_15 (mul, 4, _Float16, *)
+DEF_OP_VI_15 (mul, 8, _Float16, *)
+DEF_OP_VI_15 (mul, 16, _Float16, *)
+DEF_OP_VI_15 (mul, 32, _Float16, *)
+DEF_OP_VI_15 (mul, 64, _Float16, *)
+DEF_OP_VI_15 (mul, 128, _Float16, *)
+DEF_OP_VI_15 (mul, 256, _Float16, *)
+DEF_OP_VI_15 (mul, 512, _Float16, *)
+DEF_OP_VI_15 (mul, 1024, _Float16, *)
+DEF_OP_VI_15 (mul, 2048, _Float16, *)
+
+DEF_OP_VI_15 (mul, 1, float, *)
+DEF_OP_VI_15 (mul, 2, float, *)
+DEF_OP_VI_15 (mul, 4, float, *)
+DEF_OP_VI_15 (mul, 8, float, *)
+DEF_OP_VI_15 (mul, 16, float, *)
+DEF_OP_VI_15 (mul, 32, float, *)
+DEF_OP_VI_15 (mul, 64, float, *)
+DEF_OP_VI_15 (mul, 128, float, *)
+DEF_OP_VI_15 (mul, 256, float, *)
+DEF_OP_VI_15 (mul, 512, float, *)
+DEF_OP_VI_15 (mul, 1024, float, *)
+
+DEF_OP_VI_15 (mul, 1, double, *)
+DEF_OP_VI_15 (mul, 2, double, *)
+DEF_OP_VI_15 (mul, 4, double, *)
+DEF_OP_VI_15 (mul, 8, double, *)
+DEF_OP_VI_15 (mul, 16, double, *)
+DEF_OP_VI_15 (mul, 32, double, *)
+DEF_OP_VI_15 (mul, 64, double, *)
+DEF_OP_VI_15 (mul, 128, double, *)
+DEF_OP_VI_15 (mul, 256, double, *)
+DEF_OP_VI_15 (mul, 512, double, *)
+
+/* { dg-final { scan-assembler-times {vfmul\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-1.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-1.c
new file mode 100644
index 00000000000..75fe340935c
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-1.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VV (sub, 1, _Float16, -)
+DEF_OP_VV (sub, 2, _Float16, -)
+DEF_OP_VV (sub, 4, _Float16, -)
+DEF_OP_VV (sub, 8, _Float16, -)
+DEF_OP_VV (sub, 16, _Float16, -)
+DEF_OP_VV (sub, 32, _Float16, -)
+DEF_OP_VV (sub, 64, _Float16, -)
+DEF_OP_VV (sub, 128, _Float16, -)
+DEF_OP_VV (sub, 256, _Float16, -)
+DEF_OP_VV (sub, 512, _Float16, -)
+DEF_OP_VV (sub, 1024, _Float16, -)
+DEF_OP_VV (sub, 2048, _Float16, -)
+
+DEF_OP_VV (sub, 1, float, -)
+DEF_OP_VV (sub, 2, float, -)
+DEF_OP_VV (sub, 4, float, -)
+DEF_OP_VV (sub, 8, float, -)
+DEF_OP_VV (sub, 16, float, -)
+DEF_OP_VV (sub, 32, float, -)
+DEF_OP_VV (sub, 64, float, -)
+DEF_OP_VV (sub, 128, float, -)
+DEF_OP_VV (sub, 256, float, -)
+DEF_OP_VV (sub, 512, float, -)
+DEF_OP_VV (sub, 1024, float, -)
+
+DEF_OP_VV (sub, 1, double, -)
+DEF_OP_VV (sub, 2, double, -)
+DEF_OP_VV (sub, 4, double, -)
+DEF_OP_VV (sub, 8, double, -)
+DEF_OP_VV (sub, 16, double, -)
+DEF_OP_VV (sub, 32, double, -)
+DEF_OP_VV (sub, 64, double, -)
+DEF_OP_VV (sub, 128, double, -)
+DEF_OP_VV (sub, 256, double, -)
+DEF_OP_VV (sub, 512, double, -)
+
+/* { dg-final { scan-assembler-times {vfsub\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-2.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-2.c
new file mode 100644
index 00000000000..96a6fe6df2c
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-2.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VX (sub, 1, _Float16, -)
+DEF_OP_VX (sub, 2, _Float16, -)
+DEF_OP_VX (sub, 4, _Float16, -)
+DEF_OP_VX (sub, 8, _Float16, -)
+DEF_OP_VX (sub, 16, _Float16, -)
+DEF_OP_VX (sub, 32, _Float16, -)
+DEF_OP_VX (sub, 64, _Float16, -)
+DEF_OP_VX (sub, 128, _Float16, -)
+DEF_OP_VX (sub, 256, _Float16, -)
+DEF_OP_VX (sub, 512, _Float16, -)
+DEF_OP_VX (sub, 1024, _Float16, -)
+DEF_OP_VX (sub, 2048, _Float16, -)
+
+DEF_OP_VX (sub, 1, float, -)
+DEF_OP_VX (sub, 2, float, -)
+DEF_OP_VX (sub, 4, float, -)
+DEF_OP_VX (sub, 8, float, -)
+DEF_OP_VX (sub, 16, float, -)
+DEF_OP_VX (sub, 32, float, -)
+DEF_OP_VX (sub, 64, float, -)
+DEF_OP_VX (sub, 128, float, -)
+DEF_OP_VX (sub, 256, float, -)
+DEF_OP_VX (sub, 512, float, -)
+DEF_OP_VX (sub, 1024, float, -)
+
+DEF_OP_VX (sub, 1, double, -)
+DEF_OP_VX (sub, 2, double, -)
+DEF_OP_VX (sub, 4, double, -)
+DEF_OP_VX (sub, 8, double, -)
+DEF_OP_VX (sub, 16, double, -)
+DEF_OP_VX (sub, 32, double, -)
+DEF_OP_VX (sub, 64, double, -)
+DEF_OP_VX (sub, 128, double, -)
+DEF_OP_VX (sub, 256, double, -)
+DEF_OP_VX (sub, 512, double, -)
+
+/* { dg-final { scan-assembler-times {vfsub\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-3.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-3.c
new file mode 100644
index 00000000000..0094e2cbf4b
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-3.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VI_15 (sub, 1, _Float16, -)
+DEF_OP_VI_15 (sub, 2, _Float16, -)
+DEF_OP_VI_15 (sub, 4, _Float16, -)
+DEF_OP_VI_15 (sub, 8, _Float16, -)
+DEF_OP_VI_15 (sub, 16, _Float16, -)
+DEF_OP_VI_15 (sub, 32, _Float16, -)
+DEF_OP_VI_15 (sub, 64, _Float16, -)
+DEF_OP_VI_15 (sub, 128, _Float16, -)
+DEF_OP_VI_15 (sub, 256, _Float16, -)
+DEF_OP_VI_15 (sub, 512, _Float16, -)
+DEF_OP_VI_15 (sub, 1024, _Float16, -)
+DEF_OP_VI_15 (sub, 2048, _Float16, -)
+
+DEF_OP_VI_15 (sub, 1, float, -)
+DEF_OP_VI_15 (sub, 2, float, -)
+DEF_OP_VI_15 (sub, 4, float, -)
+DEF_OP_VI_15 (sub, 8, float, -)
+DEF_OP_VI_15 (sub, 16, float, -)
+DEF_OP_VI_15 (sub, 32, float, -)
+DEF_OP_VI_15 (sub, 64, float, -)
+DEF_OP_VI_15 (sub, 128, float, -)
+DEF_OP_VI_15 (sub, 256, float, -)
+DEF_OP_VI_15 (sub, 512, float, -)
+DEF_OP_VI_15 (sub, 1024, float, -)
+
+DEF_OP_VI_15 (sub, 1, double, -)
+DEF_OP_VI_15 (sub, 2, double, -)
+DEF_OP_VI_15 (sub, 4, double, -)
+DEF_OP_VI_15 (sub, 8, double, -)
+DEF_OP_VI_15 (sub, 16, double, -)
+DEF_OP_VI_15 (sub, 32, double, -)
+DEF_OP_VI_15 (sub, 64, double, -)
+DEF_OP_VI_15 (sub, 128, double, -)
+DEF_OP_VI_15 (sub, 256, double, -)
+DEF_OP_VI_15 (sub, 512, double, -)
+
+/* { dg-final { scan-assembler-times {vfadd\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
-- 
2.34.1
 
 

^ permalink raw reply	[flat|nested] 3+ messages in thread

* RE: [PATCH v1] RISC-V: Support FP ADD/SUB/MUL/DIV autovec for VLS mode
  2023-09-01  7:28 ` 钟居哲
@ 2023-09-01  8:38   ` Li, Pan2
  0 siblings, 0 replies; 3+ messages in thread
From: Li, Pan2 @ 2023-09-01  8:38 UTC (permalink / raw)
  To: 钟居哲, gcc-patches; +Cc: Wang, Yanzhang, kito.cheng

[-- Attachment #1: Type: text/plain, Size: 37507 bytes --]

Committed, thanks Juzhe.

Pan

From: 钟居哲 <juzhe.zhong@rivai.ai>
Sent: Friday, September 1, 2023 3:28 PM
To: Li, Pan2 <pan2.li@intel.com>; gcc-patches <gcc-patches@gcc.gnu.org>
Cc: Li, Pan2 <pan2.li@intel.com>; Wang, Yanzhang <yanzhang.wang@intel.com>; kito.cheng <kito.cheng@gmail.com>
Subject: Re: [PATCH v1] RISC-V: Support FP ADD/SUB/MUL/DIV autovec for VLS mode

LGTM。

________________________________
juzhe.zhong@rivai.ai<mailto:juzhe.zhong@rivai.ai>

From: pan2.li<mailto:pan2.li@intel.com>
Date: 2023-09-01 11:33
To: gcc-patches<mailto:gcc-patches@gcc.gnu.org>
CC: juzhe.zhong<mailto:juzhe.zhong@rivai.ai>; pan2.li<mailto:pan2.li@intel.com>; yanzhang.wang<mailto:yanzhang.wang@intel.com>; kito.cheng<mailto:kito.cheng@gmail.com>
Subject: [PATCH v1] RISC-V: Support FP ADD/SUB/MUL/DIV autovec for VLS mode
From: Pan Li <pan2.li@intel.com<mailto:pan2.li@intel.com>>

This patch would like to allow the VLS mode autovec for the
floating-point binary operation ADD/SUB/MUL/DIV.

Given below code example:

test (float *out, float *in1, float *in2)
{
  for (int i = 0; i < 128; i++)
    out[i] = in1[i] + in2[i];
}

Before this patch:
test:
  csrr a4,vlenb
  slli a4,a4,1
  li   a5,128
  bleu a5,a4,.L38
  mv   a5,a4
.L38:
  vsetvli  zero,a5,e32,m8,ta,ma
  vle32.v  v16,0(a1)
  vsetvli  a4,zero,e32,m8,ta,ma
  vmv.v.i  v8,0
  vsetvli  zero,a5,e32,m8,tu,ma
  vle32.v  v24,0(a2)
  vfadd.vv v8,v24,v16
  vse32.v  v8,0(a0)
  ret

After this patch:
test:
  li       a5,128
  vsetvli  zero,a5,e32,m1,ta,ma
  vle32.v  v1,0(a2)
  vle32.v  v2,0(a1)
  vfadd.vv v1,v1,v2
  vse32.v  v1,0(a0)
  ret

Please note this patch also fix the execution failure of below
vect test cases.

* vect-alias-check-10.c
* vect-alias-check-11.c
* vect-alias-check-12.c
* vect-alias-check-14.c

Signed-off-by: Pan Li <pan2.li@intel.com<mailto:pan2.li@intel.com>>

gcc/ChangeLog:

* config/riscv/autovec-vls.md (<optab><mode>3): New pattern for
vls floating-point autovec.
* config/riscv/vector-iterators.md: New iterator for
floating-point V and VLS.
* config/riscv/vector.md: Add VLS to floating-point binop.

gcc/testsuite/ChangeLog:

* gcc.target/riscv/rvv/autovec/vls/def.h:
* gcc.target/riscv/rvv/autovec/vls/floating-point-add-1.c: New test.
* gcc.target/riscv/rvv/autovec/vls/floating-point-add-2.c: New test.
* gcc.target/riscv/rvv/autovec/vls/floating-point-add-3.c: New test.
* gcc.target/riscv/rvv/autovec/vls/floating-point-div-1.c: New test.
* gcc.target/riscv/rvv/autovec/vls/floating-point-div-2.c: New test.
* gcc.target/riscv/rvv/autovec/vls/floating-point-div-3.c: New test.
* gcc.target/riscv/rvv/autovec/vls/floating-point-mul-1.c: New test.
* gcc.target/riscv/rvv/autovec/vls/floating-point-mul-2.c: New test.
* gcc.target/riscv/rvv/autovec/vls/floating-point-mul-3.c: New test.
* gcc.target/riscv/rvv/autovec/vls/floating-point-sub-1.c: New test.
* gcc.target/riscv/rvv/autovec/vls/floating-point-sub-2.c: New test.
* gcc.target/riscv/rvv/autovec/vls/floating-point-sub-3.c: New test.
---
gcc/config/riscv/autovec-vls.md               | 24 ++++++
gcc/config/riscv/vector-iterators.md          | 80 +++++++++++++++++++
gcc/config/riscv/vector.md                    | 12 +--
.../gcc.target/riscv/rvv/autovec/vls/def.h    |  8 ++
.../rvv/autovec/vls/floating-point-add-1.c    | 43 ++++++++++
.../rvv/autovec/vls/floating-point-add-2.c    | 43 ++++++++++
.../rvv/autovec/vls/floating-point-add-3.c    | 43 ++++++++++
.../rvv/autovec/vls/floating-point-div-1.c    | 43 ++++++++++
.../rvv/autovec/vls/floating-point-div-2.c    | 43 ++++++++++
.../rvv/autovec/vls/floating-point-div-3.c    | 43 ++++++++++
.../rvv/autovec/vls/floating-point-mul-1.c    | 43 ++++++++++
.../rvv/autovec/vls/floating-point-mul-2.c    | 43 ++++++++++
.../rvv/autovec/vls/floating-point-mul-3.c    | 43 ++++++++++
.../rvv/autovec/vls/floating-point-sub-1.c    | 43 ++++++++++
.../rvv/autovec/vls/floating-point-sub-2.c    | 43 ++++++++++
.../rvv/autovec/vls/floating-point-sub-3.c    | 43 ++++++++++
16 files changed, 634 insertions(+), 6 deletions(-)
create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-1.c
create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-2.c
create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-3.c
create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-1.c
create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-2.c
create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-3.c
create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-1.c
create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-2.c
create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-3.c
create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-1.c
create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-2.c
create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-3.c

diff --git a/gcc/config/riscv/autovec-vls.md b/gcc/config/riscv/autovec-vls.md
index 503ad691b9a..4ca640c11e2 100644
--- a/gcc/config/riscv/autovec-vls.md
+++ b/gcc/config/riscv/autovec-vls.md
@@ -208,6 +208,30 @@ (define_insn_and_split "<optab><mode>3"
[(set_attr "type" "vector")]
)
+;; -------------------------------------------------------------------------
+;; ---- [FP] Binary operations
+;; -------------------------------------------------------------------------
+;; Includes:
+;; - vfadd.vv/vfsub.vv/vfmul.vv/vfdiv.vv
+;; - vfadd.vf/vfsub.vf/vfmul.vf/vfdiv.vf
+;; -------------------------------------------------------------------------
+(define_insn_and_split "<optab><mode>3"
+  [(set (match_operand:VLSF 0 "register_operand")
+    (any_float_binop:VLSF
+     (match_operand:VLSF 1 "<binop_rhs1_predicate>")
+     (match_operand:VLSF 2 "<binop_rhs2_predicate>")))]
+  "TARGET_VECTOR && can_create_pseudo_p ()"
+  "#"
+  "&& 1"
+  [(const_int 0)]
+{
+  riscv_vector::emit_vlmax_insn (code_for_pred (<CODE>, <MODE>mode),
+ riscv_vector::BINARY_OP_FRM_DYN, operands);
+  DONE;
+}
+[(set_attr "type" "vector")]
+)
+
;; -------------------------------------------------------------------------------
;; ---- [INT] Unary operations
;; -------------------------------------------------------------------------------
diff --git a/gcc/config/riscv/vector-iterators.md b/gcc/config/riscv/vector-iterators.md
index 4023a038fe9..09fa787c0fc 100644
--- a/gcc/config/riscv/vector-iterators.md
+++ b/gcc/config/riscv/vector-iterators.md
@@ -445,6 +445,50 @@ (define_mode_iterator V_VLSI [
   (V512DI "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_64 && TARGET_MIN_VLEN >= 4096")
])
+(define_mode_iterator V_VLSF [
+  (RVVM8HF "TARGET_ZVFH") (RVVM4HF "TARGET_ZVFH") (RVVM2HF "TARGET_ZVFH")
+  (RVVM1HF "TARGET_ZVFH") (RVVMF2HF "TARGET_ZVFH")
+  (RVVMF4HF "TARGET_ZVFH && TARGET_MIN_VLEN > 32")
+  (RVVM8SF "TARGET_VECTOR_ELEN_FP_32") (RVVM4SF "TARGET_VECTOR_ELEN_FP_32") (RVVM2SF "TARGET_VECTOR_ELEN_FP_32")
+  (RVVM1SF "TARGET_VECTOR_ELEN_FP_32") (RVVMF2SF "TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN > 32")
+  (RVVM8DF "TARGET_VECTOR_ELEN_FP_64") (RVVM4DF "TARGET_VECTOR_ELEN_FP_64")
+  (RVVM2DF "TARGET_VECTOR_ELEN_FP_64") (RVVM1DF "TARGET_VECTOR_ELEN_FP_64")
+
+  (V1HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16")
+  (V2HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16")
+  (V4HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16")
+  (V8HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16")
+  (V16HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16")
+  (V32HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 64")
+  (V64HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 128")
+  (V128HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 256")
+  (V256HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 512")
+  (V512HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 1024")
+  (V1024HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 2048")
+  (V2048HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 4096")
+  (V1SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32")
+  (V2SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32")
+  (V4SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32")
+  (V8SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32")
+  (V16SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 64")
+  (V32SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 128")
+  (V64SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 256")
+  (V128SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 512")
+  (V256SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 1024")
+  (V512SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 2048")
+  (V1024SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 4096")
+  (V1DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64")
+  (V2DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64")
+  (V4DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64")
+  (V8DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 64")
+  (V16DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 128")
+  (V32DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 256")
+  (V64DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 512")
+  (V128DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 1024")
+  (V256DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 2048")
+  (V512DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 4096")
+])
+
(define_mode_iterator VF_ZVFHMIN [
   (RVVM8HF "TARGET_VECTOR_ELEN_FP_16") (RVVM4HF "TARGET_VECTOR_ELEN_FP_16") (RVVM2HF "TARGET_VECTOR_ELEN_FP_16")
   (RVVM1HF "TARGET_VECTOR_ELEN_FP_16") (RVVMF2HF "TARGET_VECTOR_ELEN_FP_16")
@@ -2494,3 +2538,39 @@ (define_mode_iterator VLSI [
   (V128DI "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_64 && TARGET_MIN_VLEN >= 1024")
   (V256DI "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_64 && TARGET_MIN_VLEN >= 2048")
   (V512DI "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_64 && TARGET_MIN_VLEN >= 4096")])
+
+(define_mode_iterator VLSF [
+  (V1HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16")
+  (V2HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16")
+  (V4HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16")
+  (V8HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16")
+  (V16HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16")
+  (V32HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 64")
+  (V64HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 128")
+  (V128HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 256")
+  (V256HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 512")
+  (V512HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 1024")
+  (V1024HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 2048")
+  (V2048HF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_16 && TARGET_MIN_VLEN >= 4096")
+  (V1SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32")
+  (V2SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32")
+  (V4SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32")
+  (V8SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32")
+  (V16SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 64")
+  (V32SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 128")
+  (V64SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 256")
+  (V128SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 512")
+  (V256SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 1024")
+  (V512SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 2048")
+  (V1024SF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_32 && TARGET_MIN_VLEN >= 4096")
+  (V1DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64")
+  (V2DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64")
+  (V4DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64")
+  (V8DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 64")
+  (V16DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 128")
+  (V32DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 256")
+  (V64DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 512")
+  (V128DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 1024")
+  (V256DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 2048")
+  (V512DF "TARGET_VECTOR_VLS && TARGET_VECTOR_ELEN_FP_64 && TARGET_MIN_VLEN >= 4096")
+])
diff --git a/gcc/config/riscv/vector.md b/gcc/config/riscv/vector.md
index 451138a1b45..6fe750ca8a4 100644
--- a/gcc/config/riscv/vector.md
+++ b/gcc/config/riscv/vector.md
@@ -6026,8 +6026,8 @@ (define_insn "@pred_series<mode>"
;; -------------------------------------------------------------------------------
(define_insn "@pred_<optab><mode>"
-  [(set (match_operand:VF 0 "register_operand"           "=vd, vd, vr, vr")
- (if_then_else:VF
+  [(set (match_operand:V_VLSF 0 "register_operand"           "=vd, vd, vr, vr")
+ (if_then_else:V_VLSF
  (unspec:<VM>
    [(match_operand:<VM> 1 "vector_mask_operand" " vm, vm,Wc1,Wc1")
     (match_operand 5 "vector_length_operand"    " rK, rK, rK, rK")
@@ -6038,10 +6038,10 @@ (define_insn "@pred_<optab><mode>"
     (reg:SI VL_REGNUM)
     (reg:SI VTYPE_REGNUM)
     (reg:SI FRM_REGNUM)] UNSPEC_VPREDICATE)
-   (any_float_binop:VF
-     (match_operand:VF 3 "register_operand"       " vr, vr, vr, vr")
-     (match_operand:VF 4 "register_operand"       " vr, vr, vr, vr"))
-   (match_operand:VF 2 "vector_merge_operand"     " vu,  0, vu,  0")))]
+   (any_float_binop:V_VLSF
+     (match_operand:V_VLSF 3 "register_operand"       " vr, vr, vr, vr")
+     (match_operand:V_VLSF 4 "register_operand"       " vr, vr, vr, vr"))
+   (match_operand:V_VLSF 2 "vector_merge_operand"     " vu,  0, vu,  0")))]
   "TARGET_VECTOR"
   "vf<insn>.vv\t%0,%3,%4%p1"
   [(set_attr "type" "<float_insn_type>")
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/def.h b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/def.h
index 00a8a8d2849..476f966c427 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/def.h
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/def.h
@@ -95,6 +95,14 @@ typedef double v512df __attribute__ ((vector_size (4096)));
       a[i] = b[i] OP c[i];                                                     \
   }
+#define DEF_OP_VX(PREFIX, NUM, TYPE, OP)                                       \
+  void __attribute__ ((noinline, noclone))                                     \
+  PREFIX##_##TYPE##NUM (TYPE *restrict a, TYPE *restrict b, TYPE c)            \
+  {                                                                            \
+    for (int i = 0; i < NUM; ++i)                                              \
+      a[i] = b[i] OP c;                                                        \
+  }
+
#define DEF_OP_VI_M16(PREFIX, NUM, TYPE, OP)                                   \
   void __attribute__ ((noinline, noclone))                                     \
   PREFIX##_##TYPE##NUM (TYPE *restrict a, TYPE *restrict b, TYPE *restrict c)  \
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-1.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-1.c
new file mode 100644
index 00000000000..5c2da4df24e
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-1.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VV (add, 1, _Float16, +)
+DEF_OP_VV (add, 2, _Float16, +)
+DEF_OP_VV (add, 4, _Float16, +)
+DEF_OP_VV (add, 8, _Float16, +)
+DEF_OP_VV (add, 16, _Float16, +)
+DEF_OP_VV (add, 32, _Float16, +)
+DEF_OP_VV (add, 64, _Float16, +)
+DEF_OP_VV (add, 128, _Float16, +)
+DEF_OP_VV (add, 256, _Float16, +)
+DEF_OP_VV (add, 512, _Float16, +)
+DEF_OP_VV (add, 1024, _Float16, +)
+DEF_OP_VV (add, 2048, _Float16, +)
+
+DEF_OP_VV (add, 1, float, +)
+DEF_OP_VV (add, 2, float, +)
+DEF_OP_VV (add, 4, float, +)
+DEF_OP_VV (add, 8, float, +)
+DEF_OP_VV (add, 16, float, +)
+DEF_OP_VV (add, 32, float, +)
+DEF_OP_VV (add, 64, float, +)
+DEF_OP_VV (add, 128, float, +)
+DEF_OP_VV (add, 256, float, +)
+DEF_OP_VV (add, 512, float, +)
+DEF_OP_VV (add, 1024, float, +)
+
+DEF_OP_VV (add, 1, double, +)
+DEF_OP_VV (add, 2, double, +)
+DEF_OP_VV (add, 4, double, +)
+DEF_OP_VV (add, 8, double, +)
+DEF_OP_VV (add, 16, double, +)
+DEF_OP_VV (add, 32, double, +)
+DEF_OP_VV (add, 64, double, +)
+DEF_OP_VV (add, 128, double, +)
+DEF_OP_VV (add, 256, double, +)
+DEF_OP_VV (add, 512, double, +)
+
+/* { dg-final { scan-assembler-times {vfadd\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-2.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-2.c
new file mode 100644
index 00000000000..73a355bc085
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-2.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VX (add, 1, _Float16, +)
+DEF_OP_VX (add, 2, _Float16, +)
+DEF_OP_VX (add, 4, _Float16, +)
+DEF_OP_VX (add, 8, _Float16, +)
+DEF_OP_VX (add, 16, _Float16, +)
+DEF_OP_VX (add, 32, _Float16, +)
+DEF_OP_VX (add, 64, _Float16, +)
+DEF_OP_VX (add, 128, _Float16, +)
+DEF_OP_VX (add, 256, _Float16, +)
+DEF_OP_VX (add, 512, _Float16, +)
+DEF_OP_VX (add, 1024, _Float16, +)
+DEF_OP_VX (add, 2048, _Float16, +)
+
+DEF_OP_VX (add, 1, float, +)
+DEF_OP_VX (add, 2, float, +)
+DEF_OP_VX (add, 4, float, +)
+DEF_OP_VX (add, 8, float, +)
+DEF_OP_VX (add, 16, float, +)
+DEF_OP_VX (add, 32, float, +)
+DEF_OP_VX (add, 64, float, +)
+DEF_OP_VX (add, 128, float, +)
+DEF_OP_VX (add, 256, float, +)
+DEF_OP_VX (add, 512, float, +)
+DEF_OP_VX (add, 1024, float, +)
+
+DEF_OP_VX (add, 1, double, +)
+DEF_OP_VX (add, 2, double, +)
+DEF_OP_VX (add, 4, double, +)
+DEF_OP_VX (add, 8, double, +)
+DEF_OP_VX (add, 16, double, +)
+DEF_OP_VX (add, 32, double, +)
+DEF_OP_VX (add, 64, double, +)
+DEF_OP_VX (add, 128, double, +)
+DEF_OP_VX (add, 256, double, +)
+DEF_OP_VX (add, 512, double, +)
+
+/* { dg-final { scan-assembler-times {vfadd\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-3.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-3.c
new file mode 100644
index 00000000000..42925e5143a
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-add-3.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VI_15 (add, 1, _Float16, +)
+DEF_OP_VI_15 (add, 2, _Float16, +)
+DEF_OP_VI_15 (add, 4, _Float16, +)
+DEF_OP_VI_15 (add, 8, _Float16, +)
+DEF_OP_VI_15 (add, 16, _Float16, +)
+DEF_OP_VI_15 (add, 32, _Float16, +)
+DEF_OP_VI_15 (add, 64, _Float16, +)
+DEF_OP_VI_15 (add, 128, _Float16, +)
+DEF_OP_VI_15 (add, 256, _Float16, +)
+DEF_OP_VI_15 (add, 512, _Float16, +)
+DEF_OP_VI_15 (add, 1024, _Float16, +)
+DEF_OP_VI_15 (add, 2048, _Float16, +)
+
+DEF_OP_VI_15 (add, 1, float, +)
+DEF_OP_VI_15 (add, 2, float, +)
+DEF_OP_VI_15 (add, 4, float, +)
+DEF_OP_VI_15 (add, 8, float, +)
+DEF_OP_VI_15 (add, 16, float, +)
+DEF_OP_VI_15 (add, 32, float, +)
+DEF_OP_VI_15 (add, 64, float, +)
+DEF_OP_VI_15 (add, 128, float, +)
+DEF_OP_VI_15 (add, 256, float, +)
+DEF_OP_VI_15 (add, 512, float, +)
+DEF_OP_VI_15 (add, 1024, float, +)
+
+DEF_OP_VI_15 (add, 1, double, +)
+DEF_OP_VI_15 (add, 2, double, +)
+DEF_OP_VI_15 (add, 4, double, +)
+DEF_OP_VI_15 (add, 8, double, +)
+DEF_OP_VI_15 (add, 16, double, +)
+DEF_OP_VI_15 (add, 32, double, +)
+DEF_OP_VI_15 (add, 64, double, +)
+DEF_OP_VI_15 (add, 128, double, +)
+DEF_OP_VI_15 (add, 256, double, +)
+DEF_OP_VI_15 (add, 512, double, +)
+
+/* { dg-final { scan-assembler-times {vfadd\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-1.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-1.c
new file mode 100644
index 00000000000..93a9e39d00c
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-1.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VV (div, 1, _Float16, /)
+DEF_OP_VV (div, 2, _Float16, /)
+DEF_OP_VV (div, 4, _Float16, /)
+DEF_OP_VV (div, 8, _Float16, /)
+DEF_OP_VV (div, 16, _Float16, /)
+DEF_OP_VV (div, 32, _Float16, /)
+DEF_OP_VV (div, 64, _Float16, /)
+DEF_OP_VV (div, 128, _Float16, /)
+DEF_OP_VV (div, 256, _Float16, /)
+DEF_OP_VV (div, 512, _Float16, /)
+DEF_OP_VV (div, 1024, _Float16, /)
+DEF_OP_VV (div, 2048, _Float16, /)
+
+DEF_OP_VV (div, 1, float, /)
+DEF_OP_VV (div, 2, float, /)
+DEF_OP_VV (div, 4, float, /)
+DEF_OP_VV (div, 8, float, /)
+DEF_OP_VV (div, 16, float, /)
+DEF_OP_VV (div, 32, float, /)
+DEF_OP_VV (div, 64, float, /)
+DEF_OP_VV (div, 128, float, /)
+DEF_OP_VV (div, 256, float, /)
+DEF_OP_VV (div, 512, float, /)
+DEF_OP_VV (div, 1024, float, /)
+
+DEF_OP_VV (div, 1, double, /)
+DEF_OP_VV (div, 2, double, /)
+DEF_OP_VV (div, 4, double, /)
+DEF_OP_VV (div, 8, double, /)
+DEF_OP_VV (div, 16, double, /)
+DEF_OP_VV (div, 32, double, /)
+DEF_OP_VV (div, 64, double, /)
+DEF_OP_VV (div, 128, double, /)
+DEF_OP_VV (div, 256, double, /)
+DEF_OP_VV (div, 512, double, /)
+
+/* { dg-final { scan-assembler-times {vfdiv\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-2.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-2.c
new file mode 100644
index 00000000000..a5bc2a0e970
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-2.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VX (div, 1, _Float16, /)
+DEF_OP_VX (div, 2, _Float16, /)
+DEF_OP_VX (div, 4, _Float16, /)
+DEF_OP_VX (div, 8, _Float16, /)
+DEF_OP_VX (div, 16, _Float16, /)
+DEF_OP_VX (div, 32, _Float16, /)
+DEF_OP_VX (div, 64, _Float16, /)
+DEF_OP_VX (div, 128, _Float16, /)
+DEF_OP_VX (div, 256, _Float16, /)
+DEF_OP_VX (div, 512, _Float16, /)
+DEF_OP_VX (div, 1024, _Float16, /)
+DEF_OP_VX (div, 2048, _Float16, /)
+
+DEF_OP_VX (div, 1, float, /)
+DEF_OP_VX (div, 2, float, /)
+DEF_OP_VX (div, 4, float, /)
+DEF_OP_VX (div, 8, float, /)
+DEF_OP_VX (div, 16, float, /)
+DEF_OP_VX (div, 32, float, /)
+DEF_OP_VX (div, 64, float, /)
+DEF_OP_VX (div, 128, float, /)
+DEF_OP_VX (div, 256, float, /)
+DEF_OP_VX (div, 512, float, /)
+DEF_OP_VX (div, 1024, float, /)
+
+DEF_OP_VX (div, 1, double, /)
+DEF_OP_VX (div, 2, double, /)
+DEF_OP_VX (div, 4, double, /)
+DEF_OP_VX (div, 8, double, /)
+DEF_OP_VX (div, 16, double, /)
+DEF_OP_VX (div, 32, double, /)
+DEF_OP_VX (div, 64, double, /)
+DEF_OP_VX (div, 128, double, /)
+DEF_OP_VX (div, 256, double, /)
+DEF_OP_VX (div, 512, double, /)
+
+/* { dg-final { scan-assembler-times {vfdiv\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-3.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-3.c
new file mode 100644
index 00000000000..f1fb7ed2c9d
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-div-3.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VI_15 (div, 1, _Float16, /)
+DEF_OP_VI_15 (div, 2, _Float16, /)
+DEF_OP_VI_15 (div, 4, _Float16, /)
+DEF_OP_VI_15 (div, 8, _Float16, /)
+DEF_OP_VI_15 (div, 16, _Float16, /)
+DEF_OP_VI_15 (div, 32, _Float16, /)
+DEF_OP_VI_15 (div, 64, _Float16, /)
+DEF_OP_VI_15 (div, 128, _Float16, /)
+DEF_OP_VI_15 (div, 256, _Float16, /)
+DEF_OP_VI_15 (div, 512, _Float16, /)
+DEF_OP_VI_15 (div, 1024, _Float16, /)
+DEF_OP_VI_15 (div, 2048, _Float16, /)
+
+DEF_OP_VI_15 (div, 1, float, /)
+DEF_OP_VI_15 (div, 2, float, /)
+DEF_OP_VI_15 (div, 4, float, /)
+DEF_OP_VI_15 (div, 8, float, /)
+DEF_OP_VI_15 (div, 16, float, /)
+DEF_OP_VI_15 (div, 32, float, /)
+DEF_OP_VI_15 (div, 64, float, /)
+DEF_OP_VI_15 (div, 128, float, /)
+DEF_OP_VI_15 (div, 256, float, /)
+DEF_OP_VI_15 (div, 512, float, /)
+DEF_OP_VI_15 (div, 1024, float, /)
+
+DEF_OP_VI_15 (div, 1, double, /)
+DEF_OP_VI_15 (div, 2, double, /)
+DEF_OP_VI_15 (div, 4, double, /)
+DEF_OP_VI_15 (div, 8, double, /)
+DEF_OP_VI_15 (div, 16, double, /)
+DEF_OP_VI_15 (div, 32, double, /)
+DEF_OP_VI_15 (div, 64, double, /)
+DEF_OP_VI_15 (div, 128, double, /)
+DEF_OP_VI_15 (div, 256, double, /)
+DEF_OP_VI_15 (div, 512, double, /)
+
+/* { dg-final { scan-assembler-times {vfdiv\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-1.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-1.c
new file mode 100644
index 00000000000..3beccb275ab
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-1.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VV (mul, 1, _Float16, *)
+DEF_OP_VV (mul, 2, _Float16, *)
+DEF_OP_VV (mul, 4, _Float16, *)
+DEF_OP_VV (mul, 8, _Float16, *)
+DEF_OP_VV (mul, 16, _Float16, *)
+DEF_OP_VV (mul, 32, _Float16, *)
+DEF_OP_VV (mul, 64, _Float16, *)
+DEF_OP_VV (mul, 128, _Float16, *)
+DEF_OP_VV (mul, 256, _Float16, *)
+DEF_OP_VV (mul, 512, _Float16, *)
+DEF_OP_VV (mul, 1024, _Float16, *)
+DEF_OP_VV (mul, 2048, _Float16, *)
+
+DEF_OP_VV (mul, 1, float, *)
+DEF_OP_VV (mul, 2, float, *)
+DEF_OP_VV (mul, 4, float, *)
+DEF_OP_VV (mul, 8, float, *)
+DEF_OP_VV (mul, 16, float, *)
+DEF_OP_VV (mul, 32, float, *)
+DEF_OP_VV (mul, 64, float, *)
+DEF_OP_VV (mul, 128, float, *)
+DEF_OP_VV (mul, 256, float, *)
+DEF_OP_VV (mul, 512, float, *)
+DEF_OP_VV (mul, 1024, float, *)
+
+DEF_OP_VV (mul, 1, double, *)
+DEF_OP_VV (mul, 2, double, *)
+DEF_OP_VV (mul, 4, double, *)
+DEF_OP_VV (mul, 8, double, *)
+DEF_OP_VV (mul, 16, double, *)
+DEF_OP_VV (mul, 32, double, *)
+DEF_OP_VV (mul, 64, double, *)
+DEF_OP_VV (mul, 128, double, *)
+DEF_OP_VV (mul, 256, double, *)
+DEF_OP_VV (mul, 512, double, *)
+
+/* { dg-final { scan-assembler-times {vfmul\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-2.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-2.c
new file mode 100644
index 00000000000..b9616386177
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-2.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VX (mul, 1, _Float16, *)
+DEF_OP_VX (mul, 2, _Float16, *)
+DEF_OP_VX (mul, 4, _Float16, *)
+DEF_OP_VX (mul, 8, _Float16, *)
+DEF_OP_VX (mul, 16, _Float16, *)
+DEF_OP_VX (mul, 32, _Float16, *)
+DEF_OP_VX (mul, 64, _Float16, *)
+DEF_OP_VX (mul, 128, _Float16, *)
+DEF_OP_VX (mul, 256, _Float16, *)
+DEF_OP_VX (mul, 512, _Float16, *)
+DEF_OP_VX (mul, 1024, _Float16, *)
+DEF_OP_VX (mul, 2048, _Float16, *)
+
+DEF_OP_VX (mul, 1, float, *)
+DEF_OP_VX (mul, 2, float, *)
+DEF_OP_VX (mul, 4, float, *)
+DEF_OP_VX (mul, 8, float, *)
+DEF_OP_VX (mul, 16, float, *)
+DEF_OP_VX (mul, 32, float, *)
+DEF_OP_VX (mul, 64, float, *)
+DEF_OP_VX (mul, 128, float, *)
+DEF_OP_VX (mul, 256, float, *)
+DEF_OP_VX (mul, 512, float, *)
+DEF_OP_VX (mul, 1024, float, *)
+
+DEF_OP_VX (mul, 1, double, *)
+DEF_OP_VX (mul, 2, double, *)
+DEF_OP_VX (mul, 4, double, *)
+DEF_OP_VX (mul, 8, double, *)
+DEF_OP_VX (mul, 16, double, *)
+DEF_OP_VX (mul, 32, double, *)
+DEF_OP_VX (mul, 64, double, *)
+DEF_OP_VX (mul, 128, double, *)
+DEF_OP_VX (mul, 256, double, *)
+DEF_OP_VX (mul, 512, double, *)
+
+/* { dg-final { scan-assembler-times {vfmul\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-3.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-3.c
new file mode 100644
index 00000000000..d8e4e26bc00
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-mul-3.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VI_15 (mul, 1, _Float16, *)
+DEF_OP_VI_15 (mul, 2, _Float16, *)
+DEF_OP_VI_15 (mul, 4, _Float16, *)
+DEF_OP_VI_15 (mul, 8, _Float16, *)
+DEF_OP_VI_15 (mul, 16, _Float16, *)
+DEF_OP_VI_15 (mul, 32, _Float16, *)
+DEF_OP_VI_15 (mul, 64, _Float16, *)
+DEF_OP_VI_15 (mul, 128, _Float16, *)
+DEF_OP_VI_15 (mul, 256, _Float16, *)
+DEF_OP_VI_15 (mul, 512, _Float16, *)
+DEF_OP_VI_15 (mul, 1024, _Float16, *)
+DEF_OP_VI_15 (mul, 2048, _Float16, *)
+
+DEF_OP_VI_15 (mul, 1, float, *)
+DEF_OP_VI_15 (mul, 2, float, *)
+DEF_OP_VI_15 (mul, 4, float, *)
+DEF_OP_VI_15 (mul, 8, float, *)
+DEF_OP_VI_15 (mul, 16, float, *)
+DEF_OP_VI_15 (mul, 32, float, *)
+DEF_OP_VI_15 (mul, 64, float, *)
+DEF_OP_VI_15 (mul, 128, float, *)
+DEF_OP_VI_15 (mul, 256, float, *)
+DEF_OP_VI_15 (mul, 512, float, *)
+DEF_OP_VI_15 (mul, 1024, float, *)
+
+DEF_OP_VI_15 (mul, 1, double, *)
+DEF_OP_VI_15 (mul, 2, double, *)
+DEF_OP_VI_15 (mul, 4, double, *)
+DEF_OP_VI_15 (mul, 8, double, *)
+DEF_OP_VI_15 (mul, 16, double, *)
+DEF_OP_VI_15 (mul, 32, double, *)
+DEF_OP_VI_15 (mul, 64, double, *)
+DEF_OP_VI_15 (mul, 128, double, *)
+DEF_OP_VI_15 (mul, 256, double, *)
+DEF_OP_VI_15 (mul, 512, double, *)
+
+/* { dg-final { scan-assembler-times {vfmul\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-1.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-1.c
new file mode 100644
index 00000000000..75fe340935c
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-1.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VV (sub, 1, _Float16, -)
+DEF_OP_VV (sub, 2, _Float16, -)
+DEF_OP_VV (sub, 4, _Float16, -)
+DEF_OP_VV (sub, 8, _Float16, -)
+DEF_OP_VV (sub, 16, _Float16, -)
+DEF_OP_VV (sub, 32, _Float16, -)
+DEF_OP_VV (sub, 64, _Float16, -)
+DEF_OP_VV (sub, 128, _Float16, -)
+DEF_OP_VV (sub, 256, _Float16, -)
+DEF_OP_VV (sub, 512, _Float16, -)
+DEF_OP_VV (sub, 1024, _Float16, -)
+DEF_OP_VV (sub, 2048, _Float16, -)
+
+DEF_OP_VV (sub, 1, float, -)
+DEF_OP_VV (sub, 2, float, -)
+DEF_OP_VV (sub, 4, float, -)
+DEF_OP_VV (sub, 8, float, -)
+DEF_OP_VV (sub, 16, float, -)
+DEF_OP_VV (sub, 32, float, -)
+DEF_OP_VV (sub, 64, float, -)
+DEF_OP_VV (sub, 128, float, -)
+DEF_OP_VV (sub, 256, float, -)
+DEF_OP_VV (sub, 512, float, -)
+DEF_OP_VV (sub, 1024, float, -)
+
+DEF_OP_VV (sub, 1, double, -)
+DEF_OP_VV (sub, 2, double, -)
+DEF_OP_VV (sub, 4, double, -)
+DEF_OP_VV (sub, 8, double, -)
+DEF_OP_VV (sub, 16, double, -)
+DEF_OP_VV (sub, 32, double, -)
+DEF_OP_VV (sub, 64, double, -)
+DEF_OP_VV (sub, 128, double, -)
+DEF_OP_VV (sub, 256, double, -)
+DEF_OP_VV (sub, 512, double, -)
+
+/* { dg-final { scan-assembler-times {vfsub\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-2.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-2.c
new file mode 100644
index 00000000000..96a6fe6df2c
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-2.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VX (sub, 1, _Float16, -)
+DEF_OP_VX (sub, 2, _Float16, -)
+DEF_OP_VX (sub, 4, _Float16, -)
+DEF_OP_VX (sub, 8, _Float16, -)
+DEF_OP_VX (sub, 16, _Float16, -)
+DEF_OP_VX (sub, 32, _Float16, -)
+DEF_OP_VX (sub, 64, _Float16, -)
+DEF_OP_VX (sub, 128, _Float16, -)
+DEF_OP_VX (sub, 256, _Float16, -)
+DEF_OP_VX (sub, 512, _Float16, -)
+DEF_OP_VX (sub, 1024, _Float16, -)
+DEF_OP_VX (sub, 2048, _Float16, -)
+
+DEF_OP_VX (sub, 1, float, -)
+DEF_OP_VX (sub, 2, float, -)
+DEF_OP_VX (sub, 4, float, -)
+DEF_OP_VX (sub, 8, float, -)
+DEF_OP_VX (sub, 16, float, -)
+DEF_OP_VX (sub, 32, float, -)
+DEF_OP_VX (sub, 64, float, -)
+DEF_OP_VX (sub, 128, float, -)
+DEF_OP_VX (sub, 256, float, -)
+DEF_OP_VX (sub, 512, float, -)
+DEF_OP_VX (sub, 1024, float, -)
+
+DEF_OP_VX (sub, 1, double, -)
+DEF_OP_VX (sub, 2, double, -)
+DEF_OP_VX (sub, 4, double, -)
+DEF_OP_VX (sub, 8, double, -)
+DEF_OP_VX (sub, 16, double, -)
+DEF_OP_VX (sub, 32, double, -)
+DEF_OP_VX (sub, 64, double, -)
+DEF_OP_VX (sub, 128, double, -)
+DEF_OP_VX (sub, 256, double, -)
+DEF_OP_VX (sub, 512, double, -)
+
+/* { dg-final { scan-assembler-times {vfsub\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-3.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-3.c
new file mode 100644
index 00000000000..0094e2cbf4b
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/floating-point-sub-3.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2 --param=riscv-autovec-lmul=m8" } */
+
+#include "def.h"
+
+DEF_OP_VI_15 (sub, 1, _Float16, -)
+DEF_OP_VI_15 (sub, 2, _Float16, -)
+DEF_OP_VI_15 (sub, 4, _Float16, -)
+DEF_OP_VI_15 (sub, 8, _Float16, -)
+DEF_OP_VI_15 (sub, 16, _Float16, -)
+DEF_OP_VI_15 (sub, 32, _Float16, -)
+DEF_OP_VI_15 (sub, 64, _Float16, -)
+DEF_OP_VI_15 (sub, 128, _Float16, -)
+DEF_OP_VI_15 (sub, 256, _Float16, -)
+DEF_OP_VI_15 (sub, 512, _Float16, -)
+DEF_OP_VI_15 (sub, 1024, _Float16, -)
+DEF_OP_VI_15 (sub, 2048, _Float16, -)
+
+DEF_OP_VI_15 (sub, 1, float, -)
+DEF_OP_VI_15 (sub, 2, float, -)
+DEF_OP_VI_15 (sub, 4, float, -)
+DEF_OP_VI_15 (sub, 8, float, -)
+DEF_OP_VI_15 (sub, 16, float, -)
+DEF_OP_VI_15 (sub, 32, float, -)
+DEF_OP_VI_15 (sub, 64, float, -)
+DEF_OP_VI_15 (sub, 128, float, -)
+DEF_OP_VI_15 (sub, 256, float, -)
+DEF_OP_VI_15 (sub, 512, float, -)
+DEF_OP_VI_15 (sub, 1024, float, -)
+
+DEF_OP_VI_15 (sub, 1, double, -)
+DEF_OP_VI_15 (sub, 2, double, -)
+DEF_OP_VI_15 (sub, 4, double, -)
+DEF_OP_VI_15 (sub, 8, double, -)
+DEF_OP_VI_15 (sub, 16, double, -)
+DEF_OP_VI_15 (sub, 32, double, -)
+DEF_OP_VI_15 (sub, 64, double, -)
+DEF_OP_VI_15 (sub, 128, double, -)
+DEF_OP_VI_15 (sub, 256, double, -)
+DEF_OP_VI_15 (sub, 512, double, -)
+
+/* { dg-final { scan-assembler-times {vfadd\.vv\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 30 } } */
+/* { dg-final { scan-assembler-not {csrr} } } */
--
2.34.1



^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2023-09-01  8:41 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-09-01  3:33 [PATCH v1] RISC-V: Support FP ADD/SUB/MUL/DIV autovec for VLS mode pan2.li
2023-09-01  7:28 ` 钟居哲
2023-09-01  8:38   ` Li, Pan2

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).