[PATCH,rs6000] Do not check if SMS succeeds on powerpc

[PATCH,rs6000] Do not check if SMS succeeds on powerpc

[Aaron Sawdey]

These tests have become unstable and SMS either succeeds or doesn't
depending on things like changes in instruction latency. Removing
the scan-rtl-dump-times checks for powerpc*-*-*.

If bootstrap/regtest is passes, ok for trunk and backport to 11?

Thanks!
   Aaron

gcc/testsuite

	* gcc.dg/sms-1.c: Remove scan-rtl-dump-times check.
	* gcc.dg/sms-2.c: Remove scan-rtl-dump-times check.
	* gcc.dg/sms-3.c: Remove scan-rtl-dump-times check.
	* gcc.dg/sms-4.c: Remove scan-rtl-dump-times check.
	* gcc.dg/sms-6.c: Remove scan-rtl-dump-times check.
	* gcc.dg/sms-8.c: Remove scan-rtl-dump-times check.
	* gcc.dg/sms-10.c: Remove scan-rtl-dump-times check.
---
 gcc/testsuite/gcc.dg/sms-1.c  | 2 --
 gcc/testsuite/gcc.dg/sms-10.c | 3 ---
 gcc/testsuite/gcc.dg/sms-2.c  | 2 --
 gcc/testsuite/gcc.dg/sms-3.c  | 3 ---
 gcc/testsuite/gcc.dg/sms-4.c  | 3 ---
 gcc/testsuite/gcc.dg/sms-6.c  | 2 --
 gcc/testsuite/gcc.dg/sms-8.c  | 4 ----
 7 files changed, 19 deletions(-)

diff --git a/gcc/testsuite/gcc.dg/sms-1.c b/gcc/testsuite/gcc.dg/sms-1.c
index 26868c34c71..098e1aa6e45 100644
--- a/gcc/testsuite/gcc.dg/sms-1.c
+++ b/gcc/testsuite/gcc.dg/sms-1.c
@@ -40,5 +40,3 @@ main ()
   return 0;
 }
 
-/* { dg-final { scan-rtl-dump-times "SMS succeeded" 1 "sms"  { target powerpc*-*-* } } } */
-
diff --git a/gcc/testsuite/gcc.dg/sms-10.c b/gcc/testsuite/gcc.dg/sms-10.c
index d85e8e2a274..df3bba24ed0 100644
--- a/gcc/testsuite/gcc.dg/sms-10.c
+++ b/gcc/testsuite/gcc.dg/sms-10.c
@@ -113,6 +113,3 @@ main ()
 
   return 0;
 }
-
-/* { dg-final { scan-rtl-dump-times "SMS succeeded" 1 "sms" { target powerpc*-*-* } } } */
-
diff --git a/gcc/testsuite/gcc.dg/sms-2.c b/gcc/testsuite/gcc.dg/sms-2.c
index 7b96f550262..f8375f9f05d 100644
--- a/gcc/testsuite/gcc.dg/sms-2.c
+++ b/gcc/testsuite/gcc.dg/sms-2.c
@@ -31,5 +31,3 @@ fun (nb)
 	  sy = 0;
       }
 }
-
-/* { dg-final { scan-rtl-dump-times "SMS loop many exits" 1 "sms" { target powerpc*-*-* } } } */
diff --git a/gcc/testsuite/gcc.dg/sms-3.c b/gcc/testsuite/gcc.dg/sms-3.c
index 822b516af2f..5e56ecf761c 100644
--- a/gcc/testsuite/gcc.dg/sms-3.c
+++ b/gcc/testsuite/gcc.dg/sms-3.c
@@ -38,6 +38,3 @@ main ()
   foo (6, 3);
   return 0;
 }
-
-/* { dg-final { scan-rtl-dump-times "SMS succeeded" 1 "sms" { target powerpc*-*-* } } } */
-
diff --git a/gcc/testsuite/gcc.dg/sms-4.c b/gcc/testsuite/gcc.dg/sms-4.c
index f5ebb55a2f4..8416b8b9ce9 100644
--- a/gcc/testsuite/gcc.dg/sms-4.c
+++ b/gcc/testsuite/gcc.dg/sms-4.c
@@ -34,6 +34,3 @@ main ()
   foo (5, a, b, c, dst);
   return 0;
 }
-
-/* { dg-final { scan-rtl-dump-times "SMS succeeded" 1 "sms" { target powerpc*-*-* } } } */
-
diff --git a/gcc/testsuite/gcc.dg/sms-6.c b/gcc/testsuite/gcc.dg/sms-6.c
index e57e01539eb..d6fa45a2cf9 100644
--- a/gcc/testsuite/gcc.dg/sms-6.c
+++ b/gcc/testsuite/gcc.dg/sms-6.c
@@ -41,5 +41,3 @@ int main()
   
   return 0;        
 }
-
-/* { dg-final { scan-rtl-dump-times "SMS succeeded" 3 "sms" { target powerpc*-*-* } } } */
diff --git a/gcc/testsuite/gcc.dg/sms-8.c b/gcc/testsuite/gcc.dg/sms-8.c
index 7ccaa454125..dc0a3fc1f9b 100644
--- a/gcc/testsuite/gcc.dg/sms-8.c
+++ b/gcc/testsuite/gcc.dg/sms-8.c
@@ -34,7 +34,3 @@ main ()
   res = foo (3, 4);
   return 0;
 }
-
-/* { dg-final { scan-rtl-dump-times "SMS succeeded" 1 "sms" { target powerpc*-*-* } } } */
-
-
-- 
2.27.0

Re: [PATCH,rs6000] Do not check if SMS succeeds on powerpc

[Segher Boessenkool]

Hi!

On Fri, Jun 11, 2021 at 01:56:58PM -0500, Aaron Sawdey wrote:
> These tests have become unstable and SMS either succeeds or doesn't
> depending on things like changes in instruction latency. Removing
> the scan-rtl-dump-times checks for powerpc*-*-*.
> 
> If bootstrap/regtest is passes, ok for trunk and backport to 11?

Yes.  Thank you!


Segher