https://gcc.gnu.org/g:9e0b94dba0759d9fb992674e3a57f7672864fc86
commit r15-7653-g9e0b94dba0759d9fb992674e3a57f7672864fc86
Author: Richard Sandiford <richard.sandif...@arm.com>
Date: Thu Feb 20 17:10:13 2025 +0000
testsuite: Fix sve/var_stride_*.c failures
gcc.target/aarch64/sve/var_stride_2.c started failing after
r15-268-g9dbff9c05520, but the change was an improvement:
@@ -36,13 +36,11 @@
b.any .L9
ret
.L17:
- ubfiz x5, x3, 10, 16
- ubfiz x4, x2, 10, 16
- add x5, x1, x5
- add x4, x0, x4
- cmp x0, x5
- ccmp x1, x4, 2, ls
uxtw x4, w2
+ add x6, x1, x3, lsl 10
+ cmp x0, x6
+ add x5, x0, x4, lsl 10
+ ccmp x1, x5, 2, ls
ccmp w2, 0, 4, hi
beq .L3
cntb x7
This patch therefore changes the test to expect the new output
for var_stride_2.c.
The changes for var_stride_4.c were a wash, with both versions
having 18(!) arithmetic instructions before the alias check branch.
Both versions sign-extend the n and m arguments as part of this
sequence; the question is whether they do it first or later.
This patch therefore changes the test to accept either the old
or the new code for var_stride_4.c.
gcc/testsuite/
* gcc.target/aarch64/sve/var_stride_2.c: Expect ADD+LSL.
* gcc.target/aarch64/sve/var_stride_4.c: Accept LSL or SBFIZ.
Diff:
---
gcc/testsuite/gcc.target/aarch64/sve/var_stride_2.c | 3 +--
gcc/testsuite/gcc.target/aarch64/sve/var_stride_4.c | 3 +--
2 files changed, 2 insertions(+), 4 deletions(-)
diff --git a/gcc/testsuite/gcc.target/aarch64/sve/var_stride_2.c
b/gcc/testsuite/gcc.target/aarch64/sve/var_stride_2.c
index b8afea70207f..33b9f0f197e4 100644
--- a/gcc/testsuite/gcc.target/aarch64/sve/var_stride_2.c
+++ b/gcc/testsuite/gcc.target/aarch64/sve/var_stride_2.c
@@ -16,8 +16,7 @@ f (TYPE *x, TYPE *y, unsigned short n, unsigned short m)
/* { dg-final { scan-assembler {\tldr\tw[0-9]+} } } */
/* { dg-final { scan-assembler {\tstr\tw[0-9]+} } } */
/* Should multiply by (257-1)*4 rather than (VF-1)*4 or (VF-2)*4. */
-/* { dg-final { scan-assembler-times {\tubfiz\tx[0-9]+, x2, 10, 16\n} 1 } } */
-/* { dg-final { scan-assembler-times {\tubfiz\tx[0-9]+, x3, 10, 16\n} 1 } } */
+/* { dg-final { scan-assembler-times {\tadd\tx[0-9]+, x[0-9]+, x[0-9]+, lsl
10\n} 2 } } */
/* { dg-final { scan-assembler-not {\tcmp\tx[0-9]+, 0} } } */
/* { dg-final { scan-assembler-not {\tcmp\tw[0-9]+, 0} } } */
/* { dg-final { scan-assembler-not {\tcsel\tx[0-9]+} } } */
diff --git a/gcc/testsuite/gcc.target/aarch64/sve/var_stride_4.c
b/gcc/testsuite/gcc.target/aarch64/sve/var_stride_4.c
index d2e74f9d4175..71b826a4c1bb 100644
--- a/gcc/testsuite/gcc.target/aarch64/sve/var_stride_4.c
+++ b/gcc/testsuite/gcc.target/aarch64/sve/var_stride_4.c
@@ -16,8 +16,7 @@ f (TYPE *x, TYPE *y, int n, int m)
/* { dg-final { scan-assembler {\tldr\tw[0-9]+} } } */
/* { dg-final { scan-assembler {\tstr\tw[0-9]+} } } */
/* Should multiply by (257-1)*4 rather than (VF-1)*4. */
-/* { dg-final { scan-assembler-times {\tsbfiz\tx[0-9]+, x2, 10, 32\n} 1 } } */
-/* { dg-final { scan-assembler-times {\tsbfiz\tx[0-9]+, x3, 10, 32\n} 1 } } */
+/* { dg-final { scan-assembler-times {\t(?:lsl\tx[0-9]+, x[0-9]+,
10|sbfiz\tx[0-9]+, x[0-9]+, 10, 32)\n} 2 } } */
/* { dg-final { scan-assembler {\tcmp\tw2, 0} } } */
/* { dg-final { scan-assembler {\tcmp\tw3, 0} } } */
/* { dg-final { scan-assembler-times {\tcsel\tx[0-9]+} 4 } } */
