RISC-V/testsuite: Add missing conversion tests.

Message ID b10ba845-0699-10f3-6bf8-e6874413a25d@gmail.com
State Unresolved
Headers
Series RISC-V/testsuite: Add missing conversion tests. |

Checks

Context Check Description
snail/gcc-patch-check warning Git am fail log

Commit Message

Robin Dapp Aug. 18, 2023, 7:32 p.m. UTC
  Hi,

this patch adds some missing tests for vf[nw]cvt.

Regards
 Robin

gcc/testsuite/ChangeLog:

	* gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-run.c:
	Add tests.
	* gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-rv32gcv.c:
	Ditto.
	* gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-rv64gcv.c:
	Ditto.
	* gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-template.h:
	Ditto.
	* gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-rv32gcv.c:
	Ditto.
	* gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-rv64gcv.c:
	Ditto.
	* gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-template.h:
	Ditto.
	* gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-zvfh-run.c:
	Ditto.
	* gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-rv32gcv.c:
	Ditto.
	* gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-rv64gcv.c:
	Ditto.
	* gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-template.h:
	Ditto.
	* gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-zvfh-run.c:
	Ditto.
	* gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-run.c:
	Ditto.
	* gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-rv32gcv.c:
	Ditto.
	* gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-rv64gcv.c:
	Ditto.
	* gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-template.h:
	Ditto.
	* gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-zvfh-run.c:
	Ditto.
---
 .../rvv/autovec/conversions/vfncvt-ftoi-run.c | 96 +++++++++++++++++++
 .../autovec/conversions/vfncvt-ftoi-rv32gcv.c |  6 +-
 .../autovec/conversions/vfncvt-ftoi-rv64gcv.c |  6 +-
 .../conversions/vfncvt-ftoi-template.h        |  6 ++
 .../autovec/conversions/vfncvt-itof-rv32gcv.c |  1 +
 .../autovec/conversions/vfncvt-itof-rv64gcv.c |  4 +-
 .../conversions/vfncvt-itof-template.h        |  5 +-
 .../conversions/vfncvt-itof-zvfh-run.c        | 32 +++++++
 .../autovec/conversions/vfwcvt-ftoi-rv32gcv.c |  4 +-
 .../autovec/conversions/vfwcvt-ftoi-rv64gcv.c |  4 +-
 .../conversions/vfwcvt-ftoi-template.h        |  2 +
 .../conversions/vfwcvt-ftoi-zvfh-run.c        | 32 +++++++
 .../rvv/autovec/conversions/vfwcvt-itof-run.c | 96 +++++++++++++++++++
 .../autovec/conversions/vfwcvt-itof-rv32gcv.c |  4 +-
 .../autovec/conversions/vfwcvt-itof-rv64gcv.c |  4 +-
 .../conversions/vfwcvt-itof-template.h        | 10 +-
 .../conversions/vfwcvt-itof-zvfh-run.c        | 10 +-
 17 files changed, 302 insertions(+), 20 deletions(-)
  

Comments

juzhe.zhong@rivai.ai Aug. 18, 2023, 11:40 p.m. UTC | #1
I wonder whether this patch fix such following issues :?
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108271 
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108412 




juzhe.zhong@rivai.ai
 
From: Robin Dapp
Date: 2023-08-19 03:32
To: gcc-patches; palmer; Kito Cheng; jeffreyalaw; juzhe.zhong@rivai.ai
CC: rdapp.gcc
Subject: [PATCH] RISC-V/testsuite: Add missing conversion tests.
Hi,
 
this patch adds some missing tests for vf[nw]cvt.
 
Regards
Robin
 
gcc/testsuite/ChangeLog:
 
* gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-run.c:
Add tests.
* gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-rv32gcv.c:
Ditto.
* gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-rv64gcv.c:
Ditto.
* gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-template.h:
Ditto.
* gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-rv32gcv.c:
Ditto.
* gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-rv64gcv.c:
Ditto.
* gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-template.h:
Ditto.
* gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-zvfh-run.c:
Ditto.
* gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-rv32gcv.c:
Ditto.
* gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-rv64gcv.c:
Ditto.
* gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-template.h:
Ditto.
* gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-zvfh-run.c:
Ditto.
* gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-run.c:
Ditto.
* gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-rv32gcv.c:
Ditto.
* gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-rv64gcv.c:
Ditto.
* gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-template.h:
Ditto.
* gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-zvfh-run.c:
Ditto.
---
.../rvv/autovec/conversions/vfncvt-ftoi-run.c | 96 +++++++++++++++++++
.../autovec/conversions/vfncvt-ftoi-rv32gcv.c |  6 +-
.../autovec/conversions/vfncvt-ftoi-rv64gcv.c |  6 +-
.../conversions/vfncvt-ftoi-template.h        |  6 ++
.../autovec/conversions/vfncvt-itof-rv32gcv.c |  1 +
.../autovec/conversions/vfncvt-itof-rv64gcv.c |  4 +-
.../conversions/vfncvt-itof-template.h        |  5 +-
.../conversions/vfncvt-itof-zvfh-run.c        | 32 +++++++
.../autovec/conversions/vfwcvt-ftoi-rv32gcv.c |  4 +-
.../autovec/conversions/vfwcvt-ftoi-rv64gcv.c |  4 +-
.../conversions/vfwcvt-ftoi-template.h        |  2 +
.../conversions/vfwcvt-ftoi-zvfh-run.c        | 32 +++++++
.../rvv/autovec/conversions/vfwcvt-itof-run.c | 96 +++++++++++++++++++
.../autovec/conversions/vfwcvt-itof-rv32gcv.c |  4 +-
.../autovec/conversions/vfwcvt-itof-rv64gcv.c |  4 +-
.../conversions/vfwcvt-itof-template.h        | 10 +-
.../conversions/vfwcvt-itof-zvfh-run.c        | 10 +-
17 files changed, 302 insertions(+), 20 deletions(-)
 
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-run.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-run.c
index ce3fcfa9af8..73eda067ba3 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-run.c
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-run.c
@@ -62,6 +62,38 @@ main ()
   RUN2 (float, uint16_t, 4096)
   RUN2 (float, uint16_t, 5975)
+  RUN (float, int8_t, 3)
+  RUN (float, int8_t, 4)
+  RUN (float, int8_t, 7)
+  RUN (float, int8_t, 99)
+  RUN (float, int8_t, 119)
+  RUN (float, int8_t, 128)
+  RUN (float, int8_t, 256)
+  RUN (float, int8_t, 279)
+  RUN (float, int8_t, 555)
+  RUN (float, int8_t, 1024)
+  RUN (float, int8_t, 1389)
+  RUN (float, int8_t, 2048)
+  RUN (float, int8_t, 3989)
+  RUN (float, int8_t, 4096)
+  RUN (float, int8_t, 5975)
+
+  RUN2 (float, uint8_t, 3)
+  RUN2 (float, uint8_t, 4)
+  RUN2 (float, uint8_t, 7)
+  RUN2 (float, uint8_t, 99)
+  RUN2 (float, uint8_t, 119)
+  RUN2 (float, uint8_t, 128)
+  RUN2 (float, uint8_t, 256)
+  RUN2 (float, uint8_t, 279)
+  RUN2 (float, uint8_t, 555)
+  RUN2 (float, uint8_t, 1024)
+  RUN2 (float, uint8_t, 1389)
+  RUN2 (float, uint8_t, 2048)
+  RUN2 (float, uint8_t, 3989)
+  RUN2 (float, uint8_t, 4096)
+  RUN2 (float, uint8_t, 5975)
+
   RUN (double, int32_t, 3)
   RUN (double, int32_t, 4)
   RUN (double, int32_t, 7)
@@ -93,4 +125,68 @@ main ()
   RUN2 (double, uint32_t, 3989)
   RUN2 (double, uint32_t, 4096)
   RUN2 (double, uint32_t, 5975)
+
+  RUN (double, int16_t, 3)
+  RUN (double, int16_t, 4)
+  RUN (double, int16_t, 7)
+  RUN (double, int16_t, 99)
+  RUN (double, int16_t, 119)
+  RUN (double, int16_t, 128)
+  RUN (double, int16_t, 256)
+  RUN (double, int16_t, 279)
+  RUN (double, int16_t, 555)
+  RUN (double, int16_t, 1024)
+  RUN (double, int16_t, 1389)
+  RUN (double, int16_t, 2048)
+  RUN (double, int16_t, 3989)
+  RUN (double, int16_t, 4096)
+  RUN (double, int16_t, 5975)
+
+  RUN2 (double, uint16_t, 3)
+  RUN2 (double, uint16_t, 4)
+  RUN2 (double, uint16_t, 7)
+  RUN2 (double, uint16_t, 99)
+  RUN2 (double, uint16_t, 119)
+  RUN2 (double, uint16_t, 128)
+  RUN2 (double, uint16_t, 256)
+  RUN2 (double, uint16_t, 279)
+  RUN2 (double, uint16_t, 555)
+  RUN2 (double, uint16_t, 1024)
+  RUN2 (double, uint16_t, 1389)
+  RUN2 (double, uint16_t, 2048)
+  RUN2 (double, uint16_t, 3989)
+  RUN2 (double, uint16_t, 4096)
+  RUN2 (double, uint16_t, 5975)
+
+  RUN (double, int8_t, 3)
+  RUN (double, int8_t, 4)
+  RUN (double, int8_t, 7)
+  RUN (double, int8_t, 99)
+  RUN (double, int8_t, 119)
+  RUN (double, int8_t, 128)
+  RUN (double, int8_t, 256)
+  RUN (double, int8_t, 279)
+  RUN (double, int8_t, 555)
+  RUN (double, int8_t, 1024)
+  RUN (double, int8_t, 1389)
+  RUN (double, int8_t, 2048)
+  RUN (double, int8_t, 3989)
+  RUN (double, int8_t, 4096)
+  RUN (double, int8_t, 5975)
+
+  RUN2 (double, uint8_t, 3)
+  RUN2 (double, uint8_t, 4)
+  RUN2 (double, uint8_t, 7)
+  RUN2 (double, uint8_t, 99)
+  RUN2 (double, uint8_t, 119)
+  RUN2 (double, uint8_t, 128)
+  RUN2 (double, uint8_t, 256)
+  RUN2 (double, uint8_t, 279)
+  RUN2 (double, uint8_t, 555)
+  RUN2 (double, uint8_t, 1024)
+  RUN2 (double, uint8_t, 1389)
+  RUN2 (double, uint8_t, 2048)
+  RUN2 (double, uint8_t, 3989)
+  RUN2 (double, uint8_t, 4096)
+  RUN2 (double, uint8_t, 5975)
}
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-rv32gcv.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-rv32gcv.c
index 4bed5eb7fff..43967af1cd5 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-rv32gcv.c
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-rv32gcv.c
@@ -1,7 +1,9 @@
/* { dg-do compile } */
-/* { dg-additional-options "-std=c99 -fno-vect-cost-model -march=rv32gcv_zvfh -mabi=ilp32d --param=riscv-autovec-preference=scalable" } */
+/* { dg-additional-options "-std=c99 -fno-vect-cost-model -march=rv32gcv_zvfh -mabi=ilp32d -fno-trapping-math --param=riscv-autovec-preference=scalable" } */
#include "vfncvt-ftoi-template.h"
-/* { dg-final { scan-assembler-times {\tvfncvt\.rtz\.x\.f\.w} 3 } } */
+/* The vectorizer only performs int -> float conversions with
+   intermediate types with -fno-trapping-math.  */
+/* { dg-final { scan-assembler-times {\tvfncvt\.rtz\.x\.f\.w} 9 } } */
/* { dg-final { scan-assembler-times {\tvfncvt\.rtz\.xu\.f\.w} 3 } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-rv64gcv.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-rv64gcv.c
index 7efc3f31c26..d49370bb925 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-rv64gcv.c
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-rv64gcv.c
@@ -1,7 +1,9 @@
/* { dg-do compile } */
-/* { dg-additional-options "-std=c99 -fno-vect-cost-model -march=rv64gcv_zvfh -mabi=lp64d --param=riscv-autovec-preference=scalable" } */
+/* { dg-additional-options "-std=c99 -fno-vect-cost-model -march=rv64gcv_zvfh -mabi=lp64d -fno-trapping-math --param=riscv-autovec-preference=scalable" } */
#include "vfncvt-ftoi-template.h"
-/* { dg-final { scan-assembler-times {\tvfncvt\.rtz\.x\.f\.w} 3 } } */
+/* The vectorizer only performs int -> float conversions with
+   intermediate types with -fno-trapping-math.  */
+/* { dg-final { scan-assembler-times {\tvfncvt\.rtz\.x\.f\.w} 9 } } */
/* { dg-final { scan-assembler-times {\tvfncvt\.rtz\.xu\.f\.w} 3 } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-template.h b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-template.h
index c6efbf15eca..c334e26e60c 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-template.h
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-template.h
@@ -11,8 +11,14 @@
#define TEST_ALL()                                                             \
   TEST (double, int32_t)                                                       \
   TEST (double, uint32_t)                                                      \
+  TEST (double, int16_t)                                                       \
+  TEST (double, uint16_t)                                                      \
+  TEST (double, int8_t)                                                        \
+  TEST (double, uint8_t)                                                       \
   TEST (float, int16_t)        \
   TEST (float, uint16_t)        \
+  TEST (float, int8_t)        \
+  TEST (float, uint8_t)        \
   TEST (_Float16, int8_t)        \
   TEST (_Float16, uint8_t)        \
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-rv32gcv.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-rv32gcv.c
index dd5b95c903d..73e4644658b 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-rv32gcv.c
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-rv32gcv.c
@@ -3,5 +3,6 @@
#include "vfncvt-itof-template.h"
+/* { dg-final { scan-assembler-times {\tvfcvt\.f\.x\.v} 2 } } */
/* { dg-final { scan-assembler-times {\tvfncvt\.f\.x\.w} 2 } } */
/* { dg-final { scan-assembler-times {\tvfncvt\.f\.xu\.w} 2 } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-rv64gcv.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-rv64gcv.c
index b3bdeced81d..e9d31a70e6a 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-rv64gcv.c
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-rv64gcv.c
@@ -3,5 +3,5 @@
#include "vfncvt-itof-template.h"
-/* { dg-final { scan-assembler-times {\tvfncvt\.f\.x\.w} 2 } } */
-/* { dg-final { scan-assembler-times {\tvfncvt\.f\.xu\.w} 2 } } */
+/* { dg-final { scan-assembler-times {\tvfncvt\.f\.x\.w} 5 } } */
+/* { dg-final { scan-assembler-times {\tvfncvt\.f\.xu\.w} 1 } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-template.h b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-template.h
index b06deeb3d72..c427cd167f6 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-template.h
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-template.h
@@ -6,12 +6,15 @@
      TYPE1 *restrict a, int n)   \
   {   \
     for (int i = 0; i < n; i++)   \
-      dst[i] = (TYPE2) a[i];                                      \
+      dst[i] = (TYPE2) (a[i] & 0x7ffffffful);   \
   }
+
#define TEST_ALL()   \
   TEST (int64_t, float)   \
   TEST (uint64_t, float)   \
+  TEST (int64_t, _Float16)   \
+  TEST (uint64_t, _Float16)   \
   TEST (int32_t, _Float16)   \
   TEST (uint32_t, _Float16)   \
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-zvfh-run.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-zvfh-run.c
index b4e59c65ac7..2fb9b0c7b89 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-zvfh-run.c
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-zvfh-run.c
@@ -61,4 +61,36 @@ main ()
   RUN2 (uint32_t, _Float16, 3989)
   RUN2 (uint32_t, _Float16, 4096)
   RUN2 (uint32_t, _Float16, 5975)
+
+  RUN (int64_t, _Float16, 3)
+  RUN (int64_t, _Float16, 4)
+  RUN (int64_t, _Float16, 7)
+  RUN (int64_t, _Float16, 99)
+  RUN (int64_t, _Float16, 119)
+  RUN (int64_t, _Float16, 128)
+  RUN (int64_t, _Float16, 256)
+  RUN (int64_t, _Float16, 279)
+  RUN (int64_t, _Float16, 555)
+  RUN (int64_t, _Float16, 1024)
+  RUN (int64_t, _Float16, 1389)
+  RUN (int64_t, _Float16, 2048)
+  RUN (int64_t, _Float16, 3989)
+  RUN (int64_t, _Float16, 4096)
+  RUN (int64_t, _Float16, 5975)
+
+  RUN2 (uint64_t, _Float16, 3)
+  RUN2 (uint64_t, _Float16, 4)
+  RUN2 (uint64_t, _Float16, 7)
+  RUN2 (uint64_t, _Float16, 99)
+  RUN2 (uint64_t, _Float16, 119)
+  RUN2 (uint64_t, _Float16, 128)
+  RUN2 (uint64_t, _Float16, 256)
+  RUN2 (uint64_t, _Float16, 279)
+  RUN2 (uint64_t, _Float16, 555)
+  RUN2 (uint64_t, _Float16, 1024)
+  RUN2 (uint64_t, _Float16, 1389)
+  RUN2 (uint64_t, _Float16, 2048)
+  RUN2 (uint64_t, _Float16, 3989)
+  RUN2 (uint64_t, _Float16, 4096)
+  RUN2 (uint64_t, _Float16, 5975)
}
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-rv32gcv.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-rv32gcv.c
index ce2bea48067..0ab42af6d70 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-rv32gcv.c
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-rv32gcv.c
@@ -1,7 +1,7 @@
/* { dg-do compile } */
-/* { dg-additional-options "-std=c99 -fno-vect-cost-model -march=rv32gcv_zvfh -mabi=ilp32d --param=riscv-autovec-preference=scalable" } */
+/* { dg-additional-options "-std=c99 -fno-vect-cost-model -march=rv32gcv_zvfh -mabi=ilp32d -fno-trapping-math --param=riscv-autovec-preference=scalable" } */
#include "vfwcvt-ftoi-template.h"
-/* { dg-final { scan-assembler-times {\tvfwcvt\.rtz\.x\.f\.v} 2 } } */
+/* { dg-final { scan-assembler-times {\tvfwcvt\.rtz\.x\.f\.v} 4 } } */
/* { dg-final { scan-assembler-times {\tvfwcvt\.rtz\.xu\.f\.v} 2 } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-rv64gcv.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-rv64gcv.c
index 99aa3de4281..e1a4b631423 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-rv64gcv.c
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-rv64gcv.c
@@ -1,7 +1,7 @@
/* { dg-do compile } */
-/* { dg-additional-options "-std=c99 -fno-vect-cost-model -march=rv64gcv_zvfh -mabi=lp64d --param=riscv-autovec-preference=scalable" } */
+/* { dg-additional-options "-std=c99 -fno-vect-cost-model -march=rv64gcv_zvfh -mabi=lp64d -fno-trapping-math --param=riscv-autovec-preference=scalable" } */
#include "vfwcvt-ftoi-template.h"
-/* { dg-final { scan-assembler-times {\tvfwcvt\.rtz\.x\.f\.v} 2 } } */
+/* { dg-final { scan-assembler-times {\tvfwcvt\.rtz\.x\.f\.v} 4 } } */
/* { dg-final { scan-assembler-times {\tvfwcvt\.rtz\.xu\.f\.v} 2 } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-template.h b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-template.h
index a276351bf69..aef03859d2b 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-template.h
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-template.h
@@ -9,6 +9,8 @@
   }
#define TEST_ALL()                                                             \
+  TEST (_Float16, int64_t)                                                     \
+  TEST (_Float16, uint64_t)                                                    \
   TEST (_Float16, int32_t)                                                     \
   TEST (_Float16, uint32_t)                                                    \
   TEST (float, int64_t)        \
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-zvfh-run.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-zvfh-run.c
index 15bcd05b592..5bf4bc99037 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-zvfh-run.c
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-zvfh-run.c
@@ -61,4 +61,36 @@ main ()
   RUN2 (_Float16, uint32_t, 3989)
   RUN2 (_Float16, uint32_t, 4096)
   RUN2 (_Float16, uint32_t, 5975)
+
+  RUN (_Float16, int64_t, 3)
+  RUN (_Float16, int64_t, 4)
+  RUN (_Float16, int64_t, 7)
+  RUN (_Float16, int64_t, 99)
+  RUN (_Float16, int64_t, 119)
+  RUN (_Float16, int64_t, 128)
+  RUN (_Float16, int64_t, 256)
+  RUN (_Float16, int64_t, 279)
+  RUN (_Float16, int64_t, 555)
+  RUN (_Float16, int64_t, 1024)
+  RUN (_Float16, int64_t, 1389)
+  RUN (_Float16, int64_t, 2048)
+  RUN (_Float16, int64_t, 3989)
+  RUN (_Float16, int64_t, 4096)
+  RUN (_Float16, int64_t, 5975)
+
+  RUN2 (_Float16, uint64_t, 3)
+  RUN2 (_Float16, uint64_t, 4)
+  RUN2 (_Float16, uint64_t, 7)
+  RUN2 (_Float16, uint64_t, 99)
+  RUN2 (_Float16, uint64_t, 119)
+  RUN2 (_Float16, uint64_t, 128)
+  RUN2 (_Float16, uint64_t, 256)
+  RUN2 (_Float16, uint64_t, 279)
+  RUN2 (_Float16, uint64_t, 555)
+  RUN2 (_Float16, uint64_t, 1024)
+  RUN2 (_Float16, uint64_t, 1389)
+  RUN2 (_Float16, uint64_t, 2048)
+  RUN2 (_Float16, uint64_t, 3989)
+  RUN2 (_Float16, uint64_t, 4096)
+  RUN2 (_Float16, uint64_t, 5975)
}
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-run.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-run.c
index b9287f69558..a90faef462f 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-run.c
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-run.c
@@ -30,6 +30,70 @@
int
main ()
{
+  RUN (int8_t, float, 3)
+  RUN (int8_t, float, 4)
+  RUN (int8_t, float, 7)
+  RUN (int8_t, float, 99)
+  RUN (int8_t, float, 119)
+  RUN (int8_t, float, 128)
+  RUN (int8_t, float, 256)
+  RUN (int8_t, float, 279)
+  RUN (int8_t, float, 555)
+  RUN (int8_t, float, 1024)
+  RUN (int8_t, float, 1389)
+  RUN (int8_t, float, 2048)
+  RUN (int8_t, float, 3989)
+  RUN (int8_t, float, 4096)
+  RUN (int8_t, float, 5975)
+
+  RUN2 (uint8_t, float, 3)
+  RUN2 (uint8_t, float, 4)
+  RUN2 (uint8_t, float, 7)
+  RUN2 (uint8_t, float, 99)
+  RUN2 (uint8_t, float, 119)
+  RUN2 (uint8_t, float, 128)
+  RUN2 (uint8_t, float, 256)
+  RUN2 (uint8_t, float, 279)
+  RUN2 (uint8_t, float, 555)
+  RUN2 (uint8_t, float, 1024)
+  RUN2 (uint8_t, float, 1389)
+  RUN2 (uint8_t, float, 2048)
+  RUN2 (uint8_t, float, 3989)
+  RUN2 (uint8_t, float, 4096)
+  RUN2 (uint8_t, float, 5975)
+
+  RUN (int8_t, double, 3)
+  RUN (int8_t, double, 4)
+  RUN (int8_t, double, 7)
+  RUN (int8_t, double, 99)
+  RUN (int8_t, double, 119)
+  RUN (int8_t, double, 128)
+  RUN (int8_t, double, 256)
+  RUN (int8_t, double, 279)
+  RUN (int8_t, double, 555)
+  RUN (int8_t, double, 1024)
+  RUN (int8_t, double, 1389)
+  RUN (int8_t, double, 2048)
+  RUN (int8_t, double, 3989)
+  RUN (int8_t, double, 4096)
+  RUN (int8_t, double, 5975)
+
+  RUN2 (uint8_t, double, 3)
+  RUN2 (uint8_t, double, 4)
+  RUN2 (uint8_t, double, 7)
+  RUN2 (uint8_t, double, 99)
+  RUN2 (uint8_t, double, 119)
+  RUN2 (uint8_t, double, 128)
+  RUN2 (uint8_t, double, 256)
+  RUN2 (uint8_t, double, 279)
+  RUN2 (uint8_t, double, 555)
+  RUN2 (uint8_t, double, 1024)
+  RUN2 (uint8_t, double, 1389)
+  RUN2 (uint8_t, double, 2048)
+  RUN2 (uint8_t, double, 3989)
+  RUN2 (uint8_t, double, 4096)
+  RUN2 (uint8_t, double, 5975)
+
   RUN (int16_t, float, 3)
   RUN (int16_t, float, 4)
   RUN (int16_t, float, 7)
@@ -62,6 +126,38 @@ main ()
   RUN2 (uint16_t, float, 4096)
   RUN2 (uint16_t, float, 5975)
+  RUN (int16_t, double, 3)
+  RUN (int16_t, double, 4)
+  RUN (int16_t, double, 7)
+  RUN (int16_t, double, 99)
+  RUN (int16_t, double, 119)
+  RUN (int16_t, double, 128)
+  RUN (int16_t, double, 256)
+  RUN (int16_t, double, 279)
+  RUN (int16_t, double, 555)
+  RUN (int16_t, double, 1024)
+  RUN (int16_t, double, 1389)
+  RUN (int16_t, double, 2048)
+  RUN (int16_t, double, 3989)
+  RUN (int16_t, double, 4096)
+  RUN (int16_t, double, 5975)
+
+  RUN2 (uint16_t, double, 3)
+  RUN2 (uint16_t, double, 4)
+  RUN2 (uint16_t, double, 7)
+  RUN2 (uint16_t, double, 99)
+  RUN2 (uint16_t, double, 119)
+  RUN2 (uint16_t, double, 128)
+  RUN2 (uint16_t, double, 256)
+  RUN2 (uint16_t, double, 279)
+  RUN2 (uint16_t, double, 555)
+  RUN2 (uint16_t, double, 1024)
+  RUN2 (uint16_t, double, 1389)
+  RUN2 (uint16_t, double, 2048)
+  RUN2 (uint16_t, double, 3989)
+  RUN2 (uint16_t, double, 4096)
+  RUN2 (uint16_t, double, 5975)
+
   RUN (int32_t, double, 3)
   RUN (int32_t, double, 4)
   RUN (int32_t, double, 7)
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-rv32gcv.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-rv32gcv.c
index 898b9c17231..cf180992c5d 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-rv32gcv.c
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-rv32gcv.c
@@ -3,5 +3,7 @@
#include "vfwcvt-itof-template.h"
-/* { dg-final { scan-assembler-times {\tvfwcvt\.f\.x\.v} 3 } } */
+/* Conversions that the vectorizer does via multiple intermediate
+   types end up as signed conversions.  */
+/* { dg-final { scan-assembler-times {\tvfwcvt\.f\.x\.v} 9 } } */
/* { dg-final { scan-assembler-times {\tvfwcvt\.f\.xu\.v} 3 } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-rv64gcv.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-rv64gcv.c
index e177b63738c..b1153887bd8 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-rv64gcv.c
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-rv64gcv.c
@@ -3,5 +3,7 @@
#include "vfwcvt-itof-template.h"
-/* { dg-final { scan-assembler-times {\tvfwcvt\.f\.x\.v} 3 } } */
+/* Conversions that the vectorizer does via multiple intermediate
+   types end up as signed conversions.  */
+/* { dg-final { scan-assembler-times {\tvfwcvt\.f\.x\.v} 9 } } */
/* { dg-final { scan-assembler-times {\tvfwcvt\.f\.xu\.v} 3 } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-template.h b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-template.h
index bd6d2387063..e78647f63a0 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-template.h
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-template.h
@@ -10,11 +10,17 @@
   }
#define TEST_ALL()   \
+  TEST (int8_t, _Float16)   \
+  TEST (uint8_t, _Float16)   \
+  TEST (int8_t, float)   \
+  TEST (uint8_t, float)   \
+  TEST (int8_t, double)   \
+  TEST (uint8_t, double)   \
   TEST (int16_t, float)   \
   TEST (uint16_t, float)   \
+  TEST (int16_t, double)   \
+  TEST (uint16_t, double)   \
   TEST (int32_t, double)   \
   TEST (uint32_t, double)   \
-  TEST (int8_t, _Float16)   \
-  TEST (uint8_t, _Float16)   \
TEST_ALL ()
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-zvfh-run.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-zvfh-run.c
index f89dc46d65f..15e33200f10 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-zvfh-run.c
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-zvfh-run.c
@@ -37,9 +37,9 @@ main ()
   RUN (int8_t, _Float16, 20)
   RUN (int8_t, _Float16, 27)
-  RUN (int8_t, _Float16, 4)
-  RUN (int8_t, _Float16, 8)
-  RUN (int8_t, _Float16, 11)
-  RUN (int8_t, _Float16, 29)
-  RUN (int8_t, _Float16, 49)
+  RUN (uint8_t, _Float16, 4)
+  RUN (uint8_t, _Float16, 8)
+  RUN (uint8_t, _Float16, 11)
+  RUN (uint8_t, _Float16, 29)
+  RUN (uint8_t, _Float16, 49)
}
-- 
2.41.0
  
Jeff Law Aug. 21, 2023, 1:24 p.m. UTC | #2
On 8/18/23 13:32, Robin Dapp wrote:
> Hi,
> 
> this patch adds some missing tests for vf[nw]cvt.
> 
> Regards
>   Robin
> 
> gcc/testsuite/ChangeLog:
> 
> 	* gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-run.c:
> 	Add tests.
> 	* gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-rv32gcv.c:
> 	Ditto.
> 	* gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-rv64gcv.c:
> 	Ditto.
> 	* gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-template.h:
> 	Ditto.
> 	* gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-rv32gcv.c:
> 	Ditto.
> 	* gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-rv64gcv.c:
> 	Ditto.
> 	* gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-template.h:
> 	Ditto.
> 	* gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-zvfh-run.c:
> 	Ditto.
> 	* gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-rv32gcv.c:
> 	Ditto.
> 	* gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-rv64gcv.c:
> 	Ditto.
> 	* gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-template.h:
> 	Ditto.
> 	* gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-zvfh-run.c:
> 	Ditto.
> 	* gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-run.c:
> 	Ditto.
> 	* gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-rv32gcv.c:
> 	Ditto.
> 	* gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-rv64gcv.c:
> 	Ditto.
> 	* gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-template.h:
> 	Ditto.
> 	* gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-zvfh-run.c:
> 	Ditto.
OK.

jeff
  

Patch

diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-run.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-run.c
index ce3fcfa9af8..73eda067ba3 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-run.c
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-run.c
@@ -62,6 +62,38 @@  main ()
   RUN2 (float, uint16_t, 4096)
   RUN2 (float, uint16_t, 5975)
 
+  RUN (float, int8_t, 3)
+  RUN (float, int8_t, 4)
+  RUN (float, int8_t, 7)
+  RUN (float, int8_t, 99)
+  RUN (float, int8_t, 119)
+  RUN (float, int8_t, 128)
+  RUN (float, int8_t, 256)
+  RUN (float, int8_t, 279)
+  RUN (float, int8_t, 555)
+  RUN (float, int8_t, 1024)
+  RUN (float, int8_t, 1389)
+  RUN (float, int8_t, 2048)
+  RUN (float, int8_t, 3989)
+  RUN (float, int8_t, 4096)
+  RUN (float, int8_t, 5975)
+
+  RUN2 (float, uint8_t, 3)
+  RUN2 (float, uint8_t, 4)
+  RUN2 (float, uint8_t, 7)
+  RUN2 (float, uint8_t, 99)
+  RUN2 (float, uint8_t, 119)
+  RUN2 (float, uint8_t, 128)
+  RUN2 (float, uint8_t, 256)
+  RUN2 (float, uint8_t, 279)
+  RUN2 (float, uint8_t, 555)
+  RUN2 (float, uint8_t, 1024)
+  RUN2 (float, uint8_t, 1389)
+  RUN2 (float, uint8_t, 2048)
+  RUN2 (float, uint8_t, 3989)
+  RUN2 (float, uint8_t, 4096)
+  RUN2 (float, uint8_t, 5975)
+
   RUN (double, int32_t, 3)
   RUN (double, int32_t, 4)
   RUN (double, int32_t, 7)
@@ -93,4 +125,68 @@  main ()
   RUN2 (double, uint32_t, 3989)
   RUN2 (double, uint32_t, 4096)
   RUN2 (double, uint32_t, 5975)
+
+  RUN (double, int16_t, 3)
+  RUN (double, int16_t, 4)
+  RUN (double, int16_t, 7)
+  RUN (double, int16_t, 99)
+  RUN (double, int16_t, 119)
+  RUN (double, int16_t, 128)
+  RUN (double, int16_t, 256)
+  RUN (double, int16_t, 279)
+  RUN (double, int16_t, 555)
+  RUN (double, int16_t, 1024)
+  RUN (double, int16_t, 1389)
+  RUN (double, int16_t, 2048)
+  RUN (double, int16_t, 3989)
+  RUN (double, int16_t, 4096)
+  RUN (double, int16_t, 5975)
+
+  RUN2 (double, uint16_t, 3)
+  RUN2 (double, uint16_t, 4)
+  RUN2 (double, uint16_t, 7)
+  RUN2 (double, uint16_t, 99)
+  RUN2 (double, uint16_t, 119)
+  RUN2 (double, uint16_t, 128)
+  RUN2 (double, uint16_t, 256)
+  RUN2 (double, uint16_t, 279)
+  RUN2 (double, uint16_t, 555)
+  RUN2 (double, uint16_t, 1024)
+  RUN2 (double, uint16_t, 1389)
+  RUN2 (double, uint16_t, 2048)
+  RUN2 (double, uint16_t, 3989)
+  RUN2 (double, uint16_t, 4096)
+  RUN2 (double, uint16_t, 5975)
+
+  RUN (double, int8_t, 3)
+  RUN (double, int8_t, 4)
+  RUN (double, int8_t, 7)
+  RUN (double, int8_t, 99)
+  RUN (double, int8_t, 119)
+  RUN (double, int8_t, 128)
+  RUN (double, int8_t, 256)
+  RUN (double, int8_t, 279)
+  RUN (double, int8_t, 555)
+  RUN (double, int8_t, 1024)
+  RUN (double, int8_t, 1389)
+  RUN (double, int8_t, 2048)
+  RUN (double, int8_t, 3989)
+  RUN (double, int8_t, 4096)
+  RUN (double, int8_t, 5975)
+
+  RUN2 (double, uint8_t, 3)
+  RUN2 (double, uint8_t, 4)
+  RUN2 (double, uint8_t, 7)
+  RUN2 (double, uint8_t, 99)
+  RUN2 (double, uint8_t, 119)
+  RUN2 (double, uint8_t, 128)
+  RUN2 (double, uint8_t, 256)
+  RUN2 (double, uint8_t, 279)
+  RUN2 (double, uint8_t, 555)
+  RUN2 (double, uint8_t, 1024)
+  RUN2 (double, uint8_t, 1389)
+  RUN2 (double, uint8_t, 2048)
+  RUN2 (double, uint8_t, 3989)
+  RUN2 (double, uint8_t, 4096)
+  RUN2 (double, uint8_t, 5975)
 }
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-rv32gcv.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-rv32gcv.c
index 4bed5eb7fff..43967af1cd5 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-rv32gcv.c
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-rv32gcv.c
@@ -1,7 +1,9 @@ 
 /* { dg-do compile } */
-/* { dg-additional-options "-std=c99 -fno-vect-cost-model -march=rv32gcv_zvfh -mabi=ilp32d --param=riscv-autovec-preference=scalable" } */
+/* { dg-additional-options "-std=c99 -fno-vect-cost-model -march=rv32gcv_zvfh -mabi=ilp32d -fno-trapping-math --param=riscv-autovec-preference=scalable" } */
 
 #include "vfncvt-ftoi-template.h"
 
-/* { dg-final { scan-assembler-times {\tvfncvt\.rtz\.x\.f\.w} 3 } } */
+/* The vectorizer only performs int -> float conversions with
+   intermediate types with -fno-trapping-math.  */
+/* { dg-final { scan-assembler-times {\tvfncvt\.rtz\.x\.f\.w} 9 } } */
 /* { dg-final { scan-assembler-times {\tvfncvt\.rtz\.xu\.f\.w} 3 } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-rv64gcv.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-rv64gcv.c
index 7efc3f31c26..d49370bb925 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-rv64gcv.c
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-rv64gcv.c
@@ -1,7 +1,9 @@ 
 /* { dg-do compile } */
-/* { dg-additional-options "-std=c99 -fno-vect-cost-model -march=rv64gcv_zvfh -mabi=lp64d --param=riscv-autovec-preference=scalable" } */
+/* { dg-additional-options "-std=c99 -fno-vect-cost-model -march=rv64gcv_zvfh -mabi=lp64d -fno-trapping-math --param=riscv-autovec-preference=scalable" } */
 
 #include "vfncvt-ftoi-template.h"
 
-/* { dg-final { scan-assembler-times {\tvfncvt\.rtz\.x\.f\.w} 3 } } */
+/* The vectorizer only performs int -> float conversions with
+   intermediate types with -fno-trapping-math.  */
+/* { dg-final { scan-assembler-times {\tvfncvt\.rtz\.x\.f\.w} 9 } } */
 /* { dg-final { scan-assembler-times {\tvfncvt\.rtz\.xu\.f\.w} 3 } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-template.h b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-template.h
index c6efbf15eca..c334e26e60c 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-template.h
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-ftoi-template.h
@@ -11,8 +11,14 @@ 
 #define TEST_ALL()                                                             \
   TEST (double, int32_t)                                                       \
   TEST (double, uint32_t)                                                      \
+  TEST (double, int16_t)                                                       \
+  TEST (double, uint16_t)                                                      \
+  TEST (double, int8_t)                                                        \
+  TEST (double, uint8_t)                                                       \
   TEST (float, int16_t)							       \
   TEST (float, uint16_t)						       \
+  TEST (float, int8_t)							       \
+  TEST (float, uint8_t)							       \
   TEST (_Float16, int8_t)						       \
   TEST (_Float16, uint8_t)						       \
 
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-rv32gcv.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-rv32gcv.c
index dd5b95c903d..73e4644658b 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-rv32gcv.c
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-rv32gcv.c
@@ -3,5 +3,6 @@ 
 
 #include "vfncvt-itof-template.h"
 
+/* { dg-final { scan-assembler-times {\tvfcvt\.f\.x\.v} 2 } } */
 /* { dg-final { scan-assembler-times {\tvfncvt\.f\.x\.w} 2 } } */
 /* { dg-final { scan-assembler-times {\tvfncvt\.f\.xu\.w} 2 } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-rv64gcv.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-rv64gcv.c
index b3bdeced81d..e9d31a70e6a 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-rv64gcv.c
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-rv64gcv.c
@@ -3,5 +3,5 @@ 
 
 #include "vfncvt-itof-template.h"
 
-/* { dg-final { scan-assembler-times {\tvfncvt\.f\.x\.w} 2 } } */
-/* { dg-final { scan-assembler-times {\tvfncvt\.f\.xu\.w} 2 } } */
+/* { dg-final { scan-assembler-times {\tvfncvt\.f\.x\.w} 5 } } */
+/* { dg-final { scan-assembler-times {\tvfncvt\.f\.xu\.w} 1 } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-template.h b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-template.h
index b06deeb3d72..c427cd167f6 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-template.h
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-template.h
@@ -6,12 +6,15 @@ 
 			      TYPE1 *restrict a, int n)		  \
   {								  \
     for (int i = 0; i < n; i++)					  \
-      dst[i] = (TYPE2) a[i];                                      \
+      dst[i] = (TYPE2) (a[i] & 0x7ffffffful);			  \
   }
 
+
 #define TEST_ALL()						  \
   TEST (int64_t, float)						  \
   TEST (uint64_t, float)					  \
+  TEST (int64_t, _Float16)					  \
+  TEST (uint64_t, _Float16)					  \
   TEST (int32_t, _Float16)					  \
   TEST (uint32_t, _Float16)					  \
 
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-zvfh-run.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-zvfh-run.c
index b4e59c65ac7..2fb9b0c7b89 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-zvfh-run.c
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfncvt-itof-zvfh-run.c
@@ -61,4 +61,36 @@  main ()
   RUN2 (uint32_t, _Float16, 3989)
   RUN2 (uint32_t, _Float16, 4096)
   RUN2 (uint32_t, _Float16, 5975)
+
+  RUN (int64_t, _Float16, 3)
+  RUN (int64_t, _Float16, 4)
+  RUN (int64_t, _Float16, 7)
+  RUN (int64_t, _Float16, 99)
+  RUN (int64_t, _Float16, 119)
+  RUN (int64_t, _Float16, 128)
+  RUN (int64_t, _Float16, 256)
+  RUN (int64_t, _Float16, 279)
+  RUN (int64_t, _Float16, 555)
+  RUN (int64_t, _Float16, 1024)
+  RUN (int64_t, _Float16, 1389)
+  RUN (int64_t, _Float16, 2048)
+  RUN (int64_t, _Float16, 3989)
+  RUN (int64_t, _Float16, 4096)
+  RUN (int64_t, _Float16, 5975)
+
+  RUN2 (uint64_t, _Float16, 3)
+  RUN2 (uint64_t, _Float16, 4)
+  RUN2 (uint64_t, _Float16, 7)
+  RUN2 (uint64_t, _Float16, 99)
+  RUN2 (uint64_t, _Float16, 119)
+  RUN2 (uint64_t, _Float16, 128)
+  RUN2 (uint64_t, _Float16, 256)
+  RUN2 (uint64_t, _Float16, 279)
+  RUN2 (uint64_t, _Float16, 555)
+  RUN2 (uint64_t, _Float16, 1024)
+  RUN2 (uint64_t, _Float16, 1389)
+  RUN2 (uint64_t, _Float16, 2048)
+  RUN2 (uint64_t, _Float16, 3989)
+  RUN2 (uint64_t, _Float16, 4096)
+  RUN2 (uint64_t, _Float16, 5975)
 }
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-rv32gcv.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-rv32gcv.c
index ce2bea48067..0ab42af6d70 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-rv32gcv.c
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-rv32gcv.c
@@ -1,7 +1,7 @@ 
 /* { dg-do compile } */
-/* { dg-additional-options "-std=c99 -fno-vect-cost-model -march=rv32gcv_zvfh -mabi=ilp32d --param=riscv-autovec-preference=scalable" } */
+/* { dg-additional-options "-std=c99 -fno-vect-cost-model -march=rv32gcv_zvfh -mabi=ilp32d -fno-trapping-math --param=riscv-autovec-preference=scalable" } */
 
 #include "vfwcvt-ftoi-template.h"
 
-/* { dg-final { scan-assembler-times {\tvfwcvt\.rtz\.x\.f\.v} 2 } } */
+/* { dg-final { scan-assembler-times {\tvfwcvt\.rtz\.x\.f\.v} 4 } } */
 /* { dg-final { scan-assembler-times {\tvfwcvt\.rtz\.xu\.f\.v} 2 } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-rv64gcv.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-rv64gcv.c
index 99aa3de4281..e1a4b631423 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-rv64gcv.c
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-rv64gcv.c
@@ -1,7 +1,7 @@ 
 /* { dg-do compile } */
-/* { dg-additional-options "-std=c99 -fno-vect-cost-model -march=rv64gcv_zvfh -mabi=lp64d --param=riscv-autovec-preference=scalable" } */
+/* { dg-additional-options "-std=c99 -fno-vect-cost-model -march=rv64gcv_zvfh -mabi=lp64d -fno-trapping-math --param=riscv-autovec-preference=scalable" } */
 
 #include "vfwcvt-ftoi-template.h"
 
-/* { dg-final { scan-assembler-times {\tvfwcvt\.rtz\.x\.f\.v} 2 } } */
+/* { dg-final { scan-assembler-times {\tvfwcvt\.rtz\.x\.f\.v} 4 } } */
 /* { dg-final { scan-assembler-times {\tvfwcvt\.rtz\.xu\.f\.v} 2 } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-template.h b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-template.h
index a276351bf69..aef03859d2b 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-template.h
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-template.h
@@ -9,6 +9,8 @@ 
   }
 
 #define TEST_ALL()                                                             \
+  TEST (_Float16, int64_t)                                                     \
+  TEST (_Float16, uint64_t)                                                    \
   TEST (_Float16, int32_t)                                                     \
   TEST (_Float16, uint32_t)                                                    \
   TEST (float, int64_t)							       \
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-zvfh-run.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-zvfh-run.c
index 15bcd05b592..5bf4bc99037 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-zvfh-run.c
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-ftoi-zvfh-run.c
@@ -61,4 +61,36 @@  main ()
   RUN2 (_Float16, uint32_t, 3989)
   RUN2 (_Float16, uint32_t, 4096)
   RUN2 (_Float16, uint32_t, 5975)
+
+  RUN (_Float16, int64_t, 3)
+  RUN (_Float16, int64_t, 4)
+  RUN (_Float16, int64_t, 7)
+  RUN (_Float16, int64_t, 99)
+  RUN (_Float16, int64_t, 119)
+  RUN (_Float16, int64_t, 128)
+  RUN (_Float16, int64_t, 256)
+  RUN (_Float16, int64_t, 279)
+  RUN (_Float16, int64_t, 555)
+  RUN (_Float16, int64_t, 1024)
+  RUN (_Float16, int64_t, 1389)
+  RUN (_Float16, int64_t, 2048)
+  RUN (_Float16, int64_t, 3989)
+  RUN (_Float16, int64_t, 4096)
+  RUN (_Float16, int64_t, 5975)
+
+  RUN2 (_Float16, uint64_t, 3)
+  RUN2 (_Float16, uint64_t, 4)
+  RUN2 (_Float16, uint64_t, 7)
+  RUN2 (_Float16, uint64_t, 99)
+  RUN2 (_Float16, uint64_t, 119)
+  RUN2 (_Float16, uint64_t, 128)
+  RUN2 (_Float16, uint64_t, 256)
+  RUN2 (_Float16, uint64_t, 279)
+  RUN2 (_Float16, uint64_t, 555)
+  RUN2 (_Float16, uint64_t, 1024)
+  RUN2 (_Float16, uint64_t, 1389)
+  RUN2 (_Float16, uint64_t, 2048)
+  RUN2 (_Float16, uint64_t, 3989)
+  RUN2 (_Float16, uint64_t, 4096)
+  RUN2 (_Float16, uint64_t, 5975)
 }
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-run.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-run.c
index b9287f69558..a90faef462f 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-run.c
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-run.c
@@ -30,6 +30,70 @@ 
 int
 main ()
 {
+  RUN (int8_t, float, 3)
+  RUN (int8_t, float, 4)
+  RUN (int8_t, float, 7)
+  RUN (int8_t, float, 99)
+  RUN (int8_t, float, 119)
+  RUN (int8_t, float, 128)
+  RUN (int8_t, float, 256)
+  RUN (int8_t, float, 279)
+  RUN (int8_t, float, 555)
+  RUN (int8_t, float, 1024)
+  RUN (int8_t, float, 1389)
+  RUN (int8_t, float, 2048)
+  RUN (int8_t, float, 3989)
+  RUN (int8_t, float, 4096)
+  RUN (int8_t, float, 5975)
+
+  RUN2 (uint8_t, float, 3)
+  RUN2 (uint8_t, float, 4)
+  RUN2 (uint8_t, float, 7)
+  RUN2 (uint8_t, float, 99)
+  RUN2 (uint8_t, float, 119)
+  RUN2 (uint8_t, float, 128)
+  RUN2 (uint8_t, float, 256)
+  RUN2 (uint8_t, float, 279)
+  RUN2 (uint8_t, float, 555)
+  RUN2 (uint8_t, float, 1024)
+  RUN2 (uint8_t, float, 1389)
+  RUN2 (uint8_t, float, 2048)
+  RUN2 (uint8_t, float, 3989)
+  RUN2 (uint8_t, float, 4096)
+  RUN2 (uint8_t, float, 5975)
+
+  RUN (int8_t, double, 3)
+  RUN (int8_t, double, 4)
+  RUN (int8_t, double, 7)
+  RUN (int8_t, double, 99)
+  RUN (int8_t, double, 119)
+  RUN (int8_t, double, 128)
+  RUN (int8_t, double, 256)
+  RUN (int8_t, double, 279)
+  RUN (int8_t, double, 555)
+  RUN (int8_t, double, 1024)
+  RUN (int8_t, double, 1389)
+  RUN (int8_t, double, 2048)
+  RUN (int8_t, double, 3989)
+  RUN (int8_t, double, 4096)
+  RUN (int8_t, double, 5975)
+
+  RUN2 (uint8_t, double, 3)
+  RUN2 (uint8_t, double, 4)
+  RUN2 (uint8_t, double, 7)
+  RUN2 (uint8_t, double, 99)
+  RUN2 (uint8_t, double, 119)
+  RUN2 (uint8_t, double, 128)
+  RUN2 (uint8_t, double, 256)
+  RUN2 (uint8_t, double, 279)
+  RUN2 (uint8_t, double, 555)
+  RUN2 (uint8_t, double, 1024)
+  RUN2 (uint8_t, double, 1389)
+  RUN2 (uint8_t, double, 2048)
+  RUN2 (uint8_t, double, 3989)
+  RUN2 (uint8_t, double, 4096)
+  RUN2 (uint8_t, double, 5975)
+
   RUN (int16_t, float, 3)
   RUN (int16_t, float, 4)
   RUN (int16_t, float, 7)
@@ -62,6 +126,38 @@  main ()
   RUN2 (uint16_t, float, 4096)
   RUN2 (uint16_t, float, 5975)
 
+  RUN (int16_t, double, 3)
+  RUN (int16_t, double, 4)
+  RUN (int16_t, double, 7)
+  RUN (int16_t, double, 99)
+  RUN (int16_t, double, 119)
+  RUN (int16_t, double, 128)
+  RUN (int16_t, double, 256)
+  RUN (int16_t, double, 279)
+  RUN (int16_t, double, 555)
+  RUN (int16_t, double, 1024)
+  RUN (int16_t, double, 1389)
+  RUN (int16_t, double, 2048)
+  RUN (int16_t, double, 3989)
+  RUN (int16_t, double, 4096)
+  RUN (int16_t, double, 5975)
+
+  RUN2 (uint16_t, double, 3)
+  RUN2 (uint16_t, double, 4)
+  RUN2 (uint16_t, double, 7)
+  RUN2 (uint16_t, double, 99)
+  RUN2 (uint16_t, double, 119)
+  RUN2 (uint16_t, double, 128)
+  RUN2 (uint16_t, double, 256)
+  RUN2 (uint16_t, double, 279)
+  RUN2 (uint16_t, double, 555)
+  RUN2 (uint16_t, double, 1024)
+  RUN2 (uint16_t, double, 1389)
+  RUN2 (uint16_t, double, 2048)
+  RUN2 (uint16_t, double, 3989)
+  RUN2 (uint16_t, double, 4096)
+  RUN2 (uint16_t, double, 5975)
+
   RUN (int32_t, double, 3)
   RUN (int32_t, double, 4)
   RUN (int32_t, double, 7)
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-rv32gcv.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-rv32gcv.c
index 898b9c17231..cf180992c5d 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-rv32gcv.c
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-rv32gcv.c
@@ -3,5 +3,7 @@ 
 
 #include "vfwcvt-itof-template.h"
 
-/* { dg-final { scan-assembler-times {\tvfwcvt\.f\.x\.v} 3 } } */
+/* Conversions that the vectorizer does via multiple intermediate
+   types end up as signed conversions.  */
+/* { dg-final { scan-assembler-times {\tvfwcvt\.f\.x\.v} 9 } } */
 /* { dg-final { scan-assembler-times {\tvfwcvt\.f\.xu\.v} 3 } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-rv64gcv.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-rv64gcv.c
index e177b63738c..b1153887bd8 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-rv64gcv.c
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-rv64gcv.c
@@ -3,5 +3,7 @@ 
 
 #include "vfwcvt-itof-template.h"
 
-/* { dg-final { scan-assembler-times {\tvfwcvt\.f\.x\.v} 3 } } */
+/* Conversions that the vectorizer does via multiple intermediate
+   types end up as signed conversions.  */
+/* { dg-final { scan-assembler-times {\tvfwcvt\.f\.x\.v} 9 } } */
 /* { dg-final { scan-assembler-times {\tvfwcvt\.f\.xu\.v} 3 } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-template.h b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-template.h
index bd6d2387063..e78647f63a0 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-template.h
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-template.h
@@ -10,11 +10,17 @@ 
   }
 
 #define TEST_ALL()						  \
+  TEST (int8_t, _Float16)					  \
+  TEST (uint8_t, _Float16)					  \
+  TEST (int8_t, float)						  \
+  TEST (uint8_t, float)						  \
+  TEST (int8_t, double)						  \
+  TEST (uint8_t, double)					  \
   TEST (int16_t, float)						  \
   TEST (uint16_t, float)					  \
+  TEST (int16_t, double)					  \
+  TEST (uint16_t, double)					  \
   TEST (int32_t, double)					  \
   TEST (uint32_t, double)					  \
-  TEST (int8_t, _Float16)					  \
-  TEST (uint8_t, _Float16)					  \
 
 TEST_ALL ()
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-zvfh-run.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-zvfh-run.c
index f89dc46d65f..15e33200f10 100644
--- a/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-zvfh-run.c
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/conversions/vfwcvt-itof-zvfh-run.c
@@ -37,9 +37,9 @@  main ()
   RUN (int8_t, _Float16, 20)
   RUN (int8_t, _Float16, 27)
 
-  RUN (int8_t, _Float16, 4)
-  RUN (int8_t, _Float16, 8)
-  RUN (int8_t, _Float16, 11)
-  RUN (int8_t, _Float16, 29)
-  RUN (int8_t, _Float16, 49)
+  RUN (uint8_t, _Float16, 4)
+  RUN (uint8_t, _Float16, 8)
+  RUN (uint8_t, _Float16, 11)
+  RUN (uint8_t, _Float16, 29)
+  RUN (uint8_t, _Float16, 49)
 }