[PATCH] testsuite: info-shared.exp: Expect leading dot on ppc64's symbols

[PATCH] testsuite: info-shared.exp: Expect leading dot on ppc64's symbols

[Edjunior Barbosa Machado]

Hi,

With this patch, the testcase permits the leading `.' found on ppc64's symbols.
It fixes 10 false unexpected failures on ppc64. Ok?

Thanks,
Edjunior.

gdb/testsuite/
2014-01-22  Edjunior Barbosa Machado  <emachado@linux.vnet.ibm.com>

	* gdb.base/info-shared.exp: Expect leading `.' on ppc64's symbols.

---
 gdb/testsuite/gdb.base/info-shared.exp | 20 ++++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

diff --git a/gdb/testsuite/gdb.base/info-shared.exp b/gdb/testsuite/gdb.base/info-shared.exp
index b7be174..a7cbe1d 100644
--- a/gdb/testsuite/gdb.base/info-shared.exp
+++ b/gdb/testsuite/gdb.base/info-shared.exp
@@ -89,18 +89,18 @@ gdb_breakpoint "foo" allow-pending
 gdb_breakpoint "bar" allow-pending
 
 # Run to the first stop and check that only the first library is loaded.
-gdb_continue_to_breakpoint "library load #1" "stop .*"
+gdb_continue_to_breakpoint "library load #1" "\\.?stop .*"
 check_info_shared "info sharedlibrary #2" 1 0
 
 # Run to the second stop and check that both libraries are loaded.
-gdb_continue_to_breakpoint "library load #2" "stop .*"
+gdb_continue_to_breakpoint "library load #2" "\\.?stop .*"
 check_info_shared "info sharedlibrary #3" 1 1
 
 # Check that the next stop is in foo.
-gdb_continue_to_breakpoint "library function #1" "foo .*"
+gdb_continue_to_breakpoint "library function #1" "\\.?foo .*"
 
 # Check that the next stop is in bar.
-gdb_continue_to_breakpoint "library function #2" "bar .*"
+gdb_continue_to_breakpoint "library function #2" "\\.?bar .*"
 
 # Restart the inferior and make sure there are no breakpoint reset
 # errors.  These can happen with the probes-based runtime linker
@@ -124,23 +124,23 @@ gdb_test_multiple "" $test {
 check_info_shared "info sharedlibrary #4" 0 0
 
 # Run to the first stop and check that only the first library is loaded.
-gdb_continue_to_breakpoint "library load #3" "stop .*"
+gdb_continue_to_breakpoint "library load #3" "\\.?stop .*"
 check_info_shared "info sharedlibrary #5" 1 0
 
 # Run to the second stop and check that both libraries are loaded.
-gdb_continue_to_breakpoint "library load #4" "stop .*"
+gdb_continue_to_breakpoint "library load #4" "\\.?stop .*"
 check_info_shared "info sharedlibrary #6" 1 1
 
 # Check that the next stop is in foo.
-gdb_continue_to_breakpoint "library function #3" "foo .*"
+gdb_continue_to_breakpoint "library function #3" "\\.?foo .*"
 
 # Check that the next stop is in bar.
-gdb_continue_to_breakpoint "library function #4" "bar .*"
+gdb_continue_to_breakpoint "library function #4" "\\.?bar .*"
 
 # Run to the next stop and check that the first library has been unloaded.
-gdb_continue_to_breakpoint "library unload #1" "stop .*"
+gdb_continue_to_breakpoint "library unload #1" "\\.?stop .*"
 check_info_shared "info sharedlibrary #7" 0 1
 
 # Run to the last stop and check that both libraries are gone.
-gdb_continue_to_breakpoint "library unload #2" "stop .*"
+gdb_continue_to_breakpoint "library unload #2" "\\.?stop .*"
 check_info_shared "info sharedlibrary #8" 0 0
-- 
1.8.1.4

Re: [PATCH] testsuite: info-shared.exp: Expect leading dot on ppc64's symbols

[Sergio Durigan Junior]

On Wednesday, January 22 2014, Edjunior Barbosa Machado wrote:

> Hi,
>
> With this patch, the testcase permits the leading `.' found on ppc64's symbols.
> It fixes 10 false unexpected failures on ppc64. Ok?

I believe this also fixes the regressions I found on PPC64 + RHEL 6.5
for the 7.7 branch.  I will test the patch and get back with the results.

Thanks,

-- 
Sergio

Re: [PATCH] testsuite: info-shared.exp: Expect leading dot on ppc64's symbols

[Sergio Durigan Junior]

On Wednesday, January 22 2014, I wrote:

> I believe this also fixes the regressions I found on PPC64 + RHEL 6.5
> for the 7.7 branch.  I will test the patch and get back with the results.

Confirmed, it fixes the failures indeed.

IMO this patch could/should also be applied to the 7.7 branch.

Thanks,

-- 
Sergio

Re: [PATCH] testsuite: info-shared.exp: Expect leading.on ppc64's symbols

[Ulrich Weigand]

Edjunior Barbosa Machado wrote:

> gdb/testsuite/
> 2014-01-22  Edjunior Barbosa Machado  <emachado@linux.vnet.ibm.com>
> 
> 	* gdb.base/info-shared.exp: Expect leading `.' on ppc64's symbols.

This is OK.

Thanks,
Ulrich

-- 
  Dr. Ulrich Weigand
  GNU/Linux compilers and toolchain
  Ulrich.Weigand@de.ibm.com

Re: [PATCH] testsuite: info-shared.exp: Expect leading.on ppc64's symbols

[Joel Brobecker]

> Edjunior Barbosa Machado wrote:
> 
> > gdb/testsuite/
> > 2014-01-22  Edjunior Barbosa Machado  <emachado@linux.vnet.ibm.com>
> > 
> > 	* gdb.base/info-shared.exp: Expect leading `.' on ppc64's symbols.
> 
> This is OK.

I think this patch might be one of the patches that Sergio mentioned
in: https://www.sourceware.org/ml/gdb-patches/2014-01/msg00990.html

If that's the case, they definitely seem OK for the branch as well.

-- 
Joel

Re: [PATCH] testsuite: info-shared.exp: Expect leading.on ppc64's symbols

[Edjunior Barbosa Machado]

On 01/28/2014 02:36 PM, Joel Brobecker wrote:
>> Edjunior Barbosa Machado wrote:
>>
>>> gdb/testsuite/
>>> 2014-01-22  Edjunior Barbosa Machado  <emachado@linux.vnet.ibm.com>
>>>
>>> 	* gdb.base/info-shared.exp: Expect leading `.' on ppc64's symbols.
>>
>> This is OK.
> 
> I think this patch might be one of the patches that Sergio mentioned
> in: https://www.sourceware.org/ml/gdb-patches/2014-01/msg00990.html
> 
> If that's the case, they definitely seem OK for the branch as well.

Thank you folks, I've just pushed the changes to both master and gdb-7.7
branches:

<https://sourceware.org/ml/gdb-cvs/2014-01/msg00131.html>

<https://sourceware.org/ml/gdb-cvs/2014-01/msg00132.html>

--
Edjunior

Re: [PATCH] testsuite: info-shared.exp: Expect leading.on ppc64's symbols

[Sergio Durigan Junior]

On Tuesday, January 28 2014, Joel Brobecker wrote:

> I think this patch might be one of the patches that Sergio mentioned
> in: https://www.sourceware.org/ml/gdb-patches/2014-01/msg00990.html
>
> If that's the case, they definitely seem OK for the branch as well.

Yes, that's it Joel.  Thank you both!

-- 
Sergio