This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: new abi file for x86_64-linux-gnu

Jakub Jelinek <jakub@redhat.com> writes:

> On Sat, Jun 14, 2003 at 05:37:15PM +0200, Andreas Jaeger wrote:
>> 
>> The abi file for x86_64-linux-gnu was not changed since the 3.2
>> release.  Here's a patch created with make new-abi-baseline on the 3.3
>> branch.  I'd like to commit this to both 3.3 and mainline.  
>> 
>> Ok to commit?
>> Andreas
>> 
>> 2003-06-14  Andreas Jaeger  <aj@suse.de>
>> 
>> 	* libstdc++-v3/config/abi/x86_64-linux-gnu/baseline_symbols.txt:
>> 	Regenerate.
>
>> @@ -1531,16 +1547,12 @@ FUNC:_ZNSt6localeC2Ev@@GLIBCPP_3.2
>>  FUNC:_ZNSt6localeD1Ev@@GLIBCPP_3.2
>>  FUNC:_ZNSt6localeD2Ev@@GLIBCPP_3.2
>>  FUNC:_ZNSt6localeaSERKS_@@GLIBCPP_3.2
>> -FUNC:_ZNSt7codecvtIcc11__mbstate_tEC1EP15__locale_structm@@GLIBCPP_3.2
>>  FUNC:_ZNSt7codecvtIcc11__mbstate_tEC1Em@@GLIBCPP_3.2
>> -FUNC:_ZNSt7codecvtIcc11__mbstate_tEC2EP15__locale_structm@@GLIBCPP_3.2
>>  FUNC:_ZNSt7codecvtIcc11__mbstate_tEC2Em@@GLIBCPP_3.2
>>  FUNC:_ZNSt7codecvtIcc11__mbstate_tED0Ev@@GLIBCPP_3.2
>>  FUNC:_ZNSt7codecvtIcc11__mbstate_tED1Ev@@GLIBCPP_3.2
>>  FUNC:_ZNSt7codecvtIcc11__mbstate_tED2Ev@@GLIBCPP_3.2
>> -FUNC:_ZNSt7codecvtIwc11__mbstate_tEC1EP15__locale_structm@@GLIBCPP_3.2
>>  FUNC:_ZNSt7codecvtIwc11__mbstate_tEC1Em@@GLIBCPP_3.2
>> -FUNC:_ZNSt7codecvtIwc11__mbstate_tEC2EP15__locale_structm@@GLIBCPP_3.2
>>  FUNC:_ZNSt7codecvtIwc11__mbstate_tEC2Em@@GLIBCPP_3.2
>>  FUNC:_ZNSt7codecvtIwc11__mbstate_tED0Ev@@GLIBCPP_3.2
>>  FUNC:_ZNSt7codecvtIwc11__mbstate_tED1Ev@@GLIBCPP_3.2
>
> This doesn't look right. I think we should never throw away exported symbols
> if binary compatibility is to be ensured (which I thought it
> is between 3.2 and 3.3). So it would be good to investigate why these symbols
> went away.

The initial commit of the ABI was done when the ABI was still in flux
and apparently we forgot to update this.

I just check gcc 3.2.1 and it does not contain these symbols, so I
think we're fine.

We should really run this regularly to catch problems (I'll do now as
part of my builds),

Andreas
-- 
 Andreas Jaeger, SuSE Linux AG, aj@suse.de, http://www.suse.de/~aj
  GPG fingerprint = 93A3 365E CE47 B889 DF7F  FED1 389A 563C C272 A126
Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]