Kindly ping.
Pan
-----Original Message-----
From: Li, Pan2 <pan2...@intel.com>
Sent: Wednesday, August 2, 2023 3:59 PM
To: gcc-patches@gcc.gnu.org
Cc: juzhe.zh...@rivai.ai; kito.ch...@sifive.com; Li, Pan2 <pan2...@intel.com>;
Wang, Yanzhang <yanzhang.w...@intel.com>
Subject: [PATCH v1] RISC-V: Enhance the test case for RVV vfsub/vfrsub rounding
From: Pan Li <pan2...@intel.com>
This patch would like to enhance the vfsub/vfrsub rounding API test for
below 2 purposes.
* The non-rm API has no frm related insn generated.
* The rm API has the frm backup/restore/set insn generated.
Signed-off-by: Pan Li <pan2...@intel.com>
gcc/testsuite/ChangeLog:
* gcc.target/riscv/rvv/base/float-point-single-rsub.c: Enhance
cases.
* gcc.target/riscv/rvv/base/float-point-single-sub.c: Ditto.
---
.../riscv/rvv/base/float-point-single-rsub.c | 16 +++++++++++++++-
.../riscv/rvv/base/float-point-single-sub.c | 16 +++++++++++++++-
2 files changed, 30 insertions(+), 2 deletions(-)
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/float-point-single-rsub.c
b/gcc/testsuite/gcc.target/riscv/rvv/base/float-point-single-rsub.c
index 1d770adc32c..86c56b7c6cb 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/base/float-point-single-rsub.c
+++ b/gcc/testsuite/gcc.target/riscv/rvv/base/float-point-single-rsub.c
@@ -16,4 +16,18 @@ test_vfrsub_vf_f32m1_rm_m (vbool32_t mask, vfloat32m1_t op1,
float32_t op2,
return __riscv_vfrsub_vf_f32m1_rm_m (mask, op1, op2, 3, vl);
}
-/* { dg-final { scan-assembler-times
{vfrsub\.v[vf]\s+v[0-9]+,\s*v[0-9]+,\s*[fav]+[0-9]+} 2 } } */
+vfloat32m1_t
+test_vfrsub_vf_f32m1 (vfloat32m1_t op1, float32_t op2, size_t vl) {
+ return __riscv_vfrsub_vf_f32m1 (op1, op2, vl);
+}
+
+vfloat32m1_t
+test_vfrsub_vf_f32m1_m (vbool32_t mask, vfloat32m1_t op1, float32_t op2,
+ size_t vl) {
+ return __riscv_vfrsub_vf_f32m1_m (mask, op1, op2, vl);
+}
+
+/* { dg-final { scan-assembler-times
{vfrsub\.v[vf]\s+v[0-9]+,\s*v[0-9]+,\s*[fav]+[0-9]+} 4 } } */
+/* { dg-final { scan-assembler-times {frrm\s+[axs][0-9]+} 2 } } */
+/* { dg-final { scan-assembler-times {fsrm\s+[axs][0-9]+} 2 } } */
+/* { dg-final { scan-assembler-times {fsrmi\s+[01234]} 2 } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/float-point-single-sub.c
b/gcc/testsuite/gcc.target/riscv/rvv/base/float-point-single-sub.c
index 34ed03a31d9..8075dced0b9 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/base/float-point-single-sub.c
+++ b/gcc/testsuite/gcc.target/riscv/rvv/base/float-point-single-sub.c
@@ -27,4 +27,18 @@ test_vfsub_vf_f32m1_rm_m (vbool32_t mask, vfloat32m1_t op1,
float32_t op2,
return __riscv_vfsub_vf_f32m1_rm_m (mask, op1, op2, 3, vl);
}
-/* { dg-final { scan-assembler-times
{vfsub\.v[vf]\s+v[0-9]+,\s*v[0-9]+,\s*[fav]+[0-9]+} 4 } } */
+vfloat32m1_t
+test_riscv_vfsub_vv_f32m1 (vfloat32m1_t op1, vfloat32m1_t op2, size_t vl) {
+ return __riscv_vfsub_vv_f32m1 (op1, op2, vl);
+}
+
+vfloat32m1_t
+test_vfsub_vv_f32m1_m (vbool32_t mask, vfloat32m1_t op1, vfloat32m1_t op2,
+ size_t vl) {
+ return __riscv_vfsub_vv_f32m1_m (mask, op1, op2, vl);
+}
+
+/* { dg-final { scan-assembler-times
{vfsub\.v[vf]\s+v[0-9]+,\s*v[0-9]+,\s*[fav]+[0-9]+} 6 } } */
+/* { dg-final { scan-assembler-times {frrm\s+[axs][0-9]+} 4 } } */
+/* { dg-final { scan-assembler-times {fsrm\s+[axs][0-9]+} 4 } } */
+/* { dg-final { scan-assembler-times {fsrmi\s+[01234]} 4 } } */
--
2.34.1
