[patch ping] libstdc++ testsuite cxxflags

[patch ping] libstdc++ testsuite cxxflags

[Sandra Loosemore]

It appears that this patch from last fall never got reviewed.

https://gcc.gnu.org/ml/gcc-patches/2013-10/msg02340.html

Can someone take a look?  I'll commit the patch on Cesar's behalf if 
it's approved.

-Sandra

Re: [patch ping] libstdc++ testsuite cxxflags

[Mike Stump]

On May 16, 2014, at 5:16 PM, Sandra Loosemore <sandra@codesourcery.com> wrote:
> It appears that this patch from last fall never got reviewed.

> Can someone take a look?

Tentative Ok.  Let’s let the library people have a chance to weigh in…  I’d say, let’s give them til Tuesday…  should be enough time...

Re: [patch ping] libstdc++ testsuite cxxflags

[Jonathan Wakely]

On 17 May 2014 01:16, Sandra Loosemore wrote:
> It appears that this patch from last fall never got reviewed.
>
> https://gcc.gnu.org/ml/gcc-patches/2013-10/msg02340.html
>
> Can someone take a look?  I'll commit the patch on Cesar's behalf if it's
> approved.

Libstdc++ patches need to go to the libstdc++ list, which this did, in
a separate mail that broke the threading:
https://gcc.gnu.org/ml/libstdc++/2013-10/msg00224.html
Then archives's inability to thread betweem months broke it again:
https://gcc.gnu.org/ml/libstdc++/2013-11/msg00113.html

I approved it then withdrew that approval:
https://gcc.gnu.org/ml/libstdc++/2013-11/msg00120.html
then the patch got revised:
https://gcc.gnu.org/ml/libstdc++/2013-11/msg00122.html

I'll have to refresh my memory about it.

Re: [patch ping] libstdc++ testsuite cxxflags

[Jonathan Wakely]

On 17 May 2014 10:50, Jonathan Wakely wrote:
> Then archives's inability...

Oof, not sure what my fingers were thinking there, I meant "Then the
archive's inability..."  :)

Re: [patch ping] libstdc++ testsuite cxxflags

[Jonathan Wakely]

On 17 May 2014 10:50, Jonathan Wakely wrote:
> On 17 May 2014 01:16, Sandra Loosemore wrote:
>> It appears that this patch from last fall never got reviewed.
>>
>> https://gcc.gnu.org/ml/gcc-patches/2013-10/msg02340.html
>>
>> Can someone take a look?  I'll commit the patch on Cesar's behalf if it's
>> approved.
>
> Libstdc++ patches need to go to the libstdc++ list, which this did, in
> a separate mail that broke the threading:
> https://gcc.gnu.org/ml/libstdc++/2013-10/msg00224.html
> Then archives's inability to thread betweem months broke it again:
> https://gcc.gnu.org/ml/libstdc++/2013-11/msg00113.html
>
> I approved it then withdrew that approval:
> https://gcc.gnu.org/ml/libstdc++/2013-11/msg00120.html
> then the patch got revised:
> https://gcc.gnu.org/ml/libstdc++/2013-11/msg00122.html
>
> I'll have to refresh my memory about it.

I think I'm happiest with the second version of the patch, in
https://gcc.gnu.org/ml/libstdc++/2013-11/msg00114.html

It does mean a change that might affect people using CXXFLAGS when
running the tests, so we might want to update
https://gcc.gnu.org/onlinedocs/libstdc++/manual/test.html where it says
"Or, just run the testsuites with CXXFLAGS set to -D_GLIBCXX_DEBUG or
-D_GLIBCXX_PARALLEL."

Re: [patch ping] libstdc++ testsuite cxxflags

[Sandra Loosemore]

[-- Attachment #1: Type: text/plain, Size: 1788 bytes --]

On 05/17/2014 04:07 AM, Jonathan Wakely wrote:
> On 17 May 2014 10:50, Jonathan Wakely wrote:
>> On 17 May 2014 01:16, Sandra Loosemore wrote:
>>> It appears that this patch from last fall never got reviewed.
>>>
>>> https://gcc.gnu.org/ml/gcc-patches/2013-10/msg02340.html
>>>
>>> Can someone take a look?  I'll commit the patch on Cesar's behalf if it's
>>> approved.
>>
>> Libstdc++ patches need to go to the libstdc++ list, which this did, in
>> a separate mail that broke the threading:
>> https://gcc.gnu.org/ml/libstdc++/2013-10/msg00224.html
>> Then archives's inability to thread betweem months broke it again:
>> https://gcc.gnu.org/ml/libstdc++/2013-11/msg00113.html
>>
>> I approved it then withdrew that approval:
>> https://gcc.gnu.org/ml/libstdc++/2013-11/msg00120.html
>> then the patch got revised:
>> https://gcc.gnu.org/ml/libstdc++/2013-11/msg00122.html
>>
>> I'll have to refresh my memory about it.

Whoops, I totally missed that there was already a separate thread on the 
libstdc++ mailing list only.  My bad.  :-(

> I think I'm happiest with the second version of the patch, in
> https://gcc.gnu.org/ml/libstdc++/2013-11/msg00114.html
>
> It does mean a change that might affect people using CXXFLAGS when
> running the tests, so we might want to update
> https://gcc.gnu.org/onlinedocs/libstdc++/manual/test.html where it says
> "Or, just run the testsuites with CXXFLAGS set to -D_GLIBCXX_DEBUG or
> -D_GLIBCXX_PARALLEL."

I came up with the attached patch for the wording change.  I'm having 
trouble regenerating the HTML version of the manual, though; it looks 
like I have a different version of the DocBook stylesheets around that 
are introducing lots of extraneous changes, and I'm not sure what the 
"right" version is.  :-S  Any suggestions?

-Sandra

[-- Attachment #2: doc.patch --]
[-- Type: text/x-patch, Size: 835 bytes --]

Index: libstdc++-v3/doc/xml/manual/test.xml
===================================================================
--- libstdc++-v3/doc/xml/manual/test.xml	(revision 210575)
+++ libstdc++-v3/doc/xml/manual/test.xml	(working copy)
@@ -478,9 +478,11 @@ runtest --tool libstdc++ --srcdir=/path/
     </para>
 
     <para>
-      Or, just run the testsuites with <constant>CXXFLAGS</constant>
-      set to <constant>-D_GLIBCXX_DEBUG</constant> or
-      <constant>-D_GLIBCXX_PARALLEL</constant>.
+      You can also run the testsuites by setting
+      <constant>CXXFLAGS</constant> in the environment.  In this case,
+      however, you should also make sure that <constant>CXXFLAGS</constant>
+      includes <constant>-g -O2</constant>, since some tests assume the
+      presence of these options.
     </para>
   </section>
 </section>

Re: [patch ping] libstdc++ testsuite cxxflags

[Jonathan Wakely]

On 19/05/14 14:57 -0600, Sandra Loosemore wrote:
>On 05/17/2014 04:07 AM, Jonathan Wakely wrote:
>>On 17 May 2014 10:50, Jonathan Wakely wrote:
>>>On 17 May 2014 01:16, Sandra Loosemore wrote:
>>>>It appears that this patch from last fall never got reviewed.
>>>>
>>>>https://gcc.gnu.org/ml/gcc-patches/2013-10/msg02340.html
>>>>
>>>>Can someone take a look?  I'll commit the patch on Cesar's behalf if it's
>>>>approved.
>>>
>>>Libstdc++ patches need to go to the libstdc++ list, which this did, in
>>>a separate mail that broke the threading:
>>>https://gcc.gnu.org/ml/libstdc++/2013-10/msg00224.html
>>>Then archives's inability to thread betweem months broke it again:
>>>https://gcc.gnu.org/ml/libstdc++/2013-11/msg00113.html
>>>
>>>I approved it then withdrew that approval:
>>>https://gcc.gnu.org/ml/libstdc++/2013-11/msg00120.html
>>>then the patch got revised:
>>>https://gcc.gnu.org/ml/libstdc++/2013-11/msg00122.html
>>>
>>>I'll have to refresh my memory about it.
>
>Whoops, I totally missed that there was already a separate thread on 
>the libstdc++ mailing list only.  My bad.  :-(
>
>>I think I'm happiest with the second version of the patch, in
>>https://gcc.gnu.org/ml/libstdc++/2013-11/msg00114.html
>>
>>It does mean a change that might affect people using CXXFLAGS when
>>running the tests, so we might want to update
>>https://gcc.gnu.org/onlinedocs/libstdc++/manual/test.html where it says
>>"Or, just run the testsuites with CXXFLAGS set to -D_GLIBCXX_DEBUG or
>>-D_GLIBCXX_PARALLEL."
>
>I came up with the attached patch for the wording change.  I'm having 
>trouble regenerating the HTML version of the manual, though; it looks 
>like I have a different version of the DocBook stylesheets around that 
>are introducing lots of extraneous changes, and I'm not sure what the 
>"right" version is.  :-S  Any suggestions?

You always get hundreds of changes, DocBook generates unique numeric
id attributes, which are different every run. Don't worr yabout the
docs, I can sort them out. If you and Cesar are happy with the patch
in https://gcc.gnu.org/ml/libstdc++/2013-11/msg00114.html then please
go ahead and commit that version, thanks.

Re: [patch ping] libstdc++ testsuite cxxflags

[Cesar Philippidis]

On 05/20/2014 02:11 AM, Jonathan Wakely wrote:
> On 19/05/14 14:57 -0600, Sandra Loosemore wrote:
>> On 05/17/2014 04:07 AM, Jonathan Wakely wrote:
>>> On 17 May 2014 10:50, Jonathan Wakely wrote:
>>>> On 17 May 2014 01:16, Sandra Loosemore wrote:
>>>>> It appears that this patch from last fall never got reviewed.
>>>>>
>>>>> https://gcc.gnu.org/ml/gcc-patches/2013-10/msg02340.html
>>>>>
>>>>> Can someone take a look?  I'll commit the patch on Cesar's behalf
>>>>> if it's
>>>>> approved.
>>>>
>>>> Libstdc++ patches need to go to the libstdc++ list, which this did, in
>>>> a separate mail that broke the threading:
>>>> https://gcc.gnu.org/ml/libstdc++/2013-10/msg00224.html
>>>> Then archives's inability to thread betweem months broke it again:
>>>> https://gcc.gnu.org/ml/libstdc++/2013-11/msg00113.html
>>>>
>>>> I approved it then withdrew that approval:
>>>> https://gcc.gnu.org/ml/libstdc++/2013-11/msg00120.html
>>>> then the patch got revised:
>>>> https://gcc.gnu.org/ml/libstdc++/2013-11/msg00122.html
>>>>
>>>> I'll have to refresh my memory about it.
>>
>> Whoops, I totally missed that there was already a separate thread on
>> the libstdc++ mailing list only.  My bad.  :-(
>>
>>> I think I'm happiest with the second version of the patch, in
>>> https://gcc.gnu.org/ml/libstdc++/2013-11/msg00114.html
>>>
>>> It does mean a change that might affect people using CXXFLAGS when
>>> running the tests, so we might want to update
>>> https://gcc.gnu.org/onlinedocs/libstdc++/manual/test.html where it says
>>> "Or, just run the testsuites with CXXFLAGS set to -D_GLIBCXX_DEBUG or
>>> -D_GLIBCXX_PARALLEL."
>>
>> I came up with the attached patch for the wording change.  I'm having
>> trouble regenerating the HTML version of the manual, though; it looks
>> like I have a different version of the DocBook stylesheets around that
>> are introducing lots of extraneous changes, and I'm not sure what the
>> "right" version is.  :-S  Any suggestions?
> 
> You always get hundreds of changes, DocBook generates unique numeric
> id attributes, which are different every run. Don't worr yabout the
> docs, I can sort them out. If you and Cesar are happy with the patch
> in https://gcc.gnu.org/ml/libstdc++/2013-11/msg00114.html then please
> go ahead and commit that version, thanks.

Looking back at my notes, this patch addresses the libstdc++ atomics
test failures when using a custom site.exp. Without the -O2 flag, those
tests would fail to link because of the dependency on libatomic.

I'm happy with the second patch. Sandra please commit it.

Thanks,
Cesar

Re: [patch ping] libstdc++ testsuite cxxflags

[David Edelsohn]

This patch broke the ability to run the libstdc++ testsuite on AIX.

I now see the following errors:

bad switch "-O": must be -all, -about, -indices, -inline, -expanded, -line, -lin
estop, -lineanchor, -nocase, -start, or --
    while executing
"regexp "\-O" $cxxflags"
    (procedure "libstdc++_init" line 183)
    invoked from within
"${tool}_init $test_file_name"
    invoked from within
"if [info exists tool] {
            if { [info procs "${tool}_init"] != "" } {
                ${tool}_init $test_file_name
            }
        }"
    invoked from within
"if [file exists $test_file_name] {
        set timestart [timestamp]

        if [info exists tool] {
            if { [info procs "${tool}_init"] != "" } {
                ${tool}_init..."
    (procedure "runtest" line 14)
    invoked from within
"runtest $test_name"
    ("foreach" body line 42)
    invoked from within
"foreach test_name [lsort [find ${dir} *.exp]] {
                        if { ${test_name} == "" } {
                            continue
                        }
                        # Ignore this one if asked to.
                        if { ${ignore..."
    ("foreach" body line 54)
    invoked from within
"foreach dir "${test_top_dirs}" {
                if { ${dir} != ${srcdir} } {
                    # Ignore this directory if is a directory to be
                    # ignored.
                    if {[info..."
    ("foreach" body line 121)
    invoked from within
"foreach pass $multipass {

        # multipass_name is set for `record_test' to use (see framework.exp).
        if { [lindex $pass 0] != "" } {
            set multipass_..."
    ("foreach" body line 51)
    invoked from within
"foreach current_target $target_list {
    verbose "target is $current_target"
    set current_target_name $current_target
    set tlist [split $curren..."
    (file "/gsa/yktgsa/home/e/d/edelsohn/share/dejagnu/runtest.exp" line 1625)

Re: [patch ping] libstdc++ testsuite cxxflags

[Jonathan Wakely]

On 24 May 2014 17:07, David Edelsohn wrote:
> This patch broke the ability to run the libstdc++ testsuite on AIX.
>
> I now see the following errors:
>
> bad switch "-O": must be -all, -about, -indices, -inline, -expanded, -line, -lin
> estop, -lineanchor, -nocase, -start, or --
>     while executing
> "regexp "\-O" $cxxflags"

Would this work instead?

   regexp ".*-O" $cxxflags

Re: [patch ping] libstdc++ testsuite cxxflags

[David Edelsohn]

On Sat, May 24, 2014 at 12:54 PM, Jonathan Wakely <jwakely.gcc@gmail.com> wrote:
> On 24 May 2014 17:07, David Edelsohn wrote:
>> This patch broke the ability to run the libstdc++ testsuite on AIX.
>>
>> I now see the following errors:
>>
>> bad switch "-O": must be -all, -about, -indices, -inline, -expanded, -line, -lin
>> estop, -lineanchor, -nocase, -start, or --
>>     while executing
>> "regexp "\-O" $cxxflags"
>
> Would this work instead?
>
>    regexp ".*-O" $cxxflags

That change does seem to fix the problem as seen by libstdc++ in
latest AIX testresults.

Thanks, David

Re: [patch ping] libstdc++ testsuite cxxflags

[Ramana Radhakrishnan]

On Sat, May 24, 2014 at 5:54 PM, Jonathan Wakely <jwakely.gcc@gmail.com> wrote:
> On 24 May 2014 17:07, David Edelsohn wrote:
>> This patch broke the ability to run the libstdc++ testsuite on AIX.
>>
>> I now see the following errors:
>>
>> bad switch "-O": must be -all, -about, -indices, -inline, -expanded, -line, -lin
>> estop, -lineanchor, -nocase, -start, or --
>>     while executing
>> "regexp "\-O" $cxxflags"
>
> Would this work instead?
>
>    regexp ".*-O" $cxxflags


With all the other build breakages in the past week,  I've just
started seeing the first set of testresults from an auto-tester. It
looks like on a cross test using rhe5 / x86_64  with the version of
tcl8.4 I'm using I see the same errors that David saw.

The testsuite starts running if I tried the above

regexp ".*-O" $cxxflags

Is this going to be applied - what gives ?

regards
Ramana

Re: [patch ping] libstdc++ testsuite cxxflags

[Jonathan Wakely]

[-- Attachment #1: Type: text/plain, Size: 539 bytes --]

On 30/05/14 08:53 +0100, Ramana Radhakrishnan wrote:
>With all the other build breakages in the past week,  I've just
>started seeing the first set of testresults from an auto-tester. It
>looks like on a cross test using rhe5 / x86_64  with the version of
>tcl8.4 I'm using I see the same errors that David saw.
>
>The testsuite starts running if I tried the above
>
>regexp ".*-O" $cxxflags
>
>Is this going to be applied - what gives ?

Fixed with the attached patch.

Tested x86_64-linux (but onnly with Tcl 8.5), committed to trunk.



[-- Attachment #2: p2.txt --]
[-- Type: text/plain, Size: 697 bytes --]

commit 8e72d35ed67a51df88c359249590e266c476044e
Author: Jonathan Wakely <jwakely@redhat.com>
Date:   Fri May 30 11:10:57 2014 +0100

    	* testsuite/lib/libstdc++.exp (libstdc++_init): Adjust regexp to
    	work with previous versions of Tcl.

diff --git a/libstdc++-v3/testsuite/lib/libstdc++.exp b/libstdc++-v3/testsuite/lib/libstdc++.exp
index 2b2a38b..d91bed6 100644
--- a/libstdc++-v3/testsuite/lib/libstdc++.exp
+++ b/libstdc++-v3/testsuite/lib/libstdc++.exp
@@ -283,7 +283,7 @@ proc libstdc++_init { testfile } {
     append cxxflags " "
     append cxxflags [getenv CXXFLAGS]
 
-    if ![regexp "\-O" $cxxflags] {
+    if ![regexp ".*-O" $cxxflags] {
 	append cxxflags " -g -O2"
     }