This is the mail archive of the libc-alpha@sourceware.org mailing list for the glibc 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: [PATCH 2/2] linux: Make profil_counter a compat_symbol (BZ#17726)



On 02/08/2019 12:48, Joseph Myers wrote:
> On Fri, 2 Aug 2019, Florian Weimer wrote:
> 
>> I tend to agree, but I wonder if Joseph filed this bug because it was
>> something that someone actually encountered as the result of a glibc
>> update.  In this case, it strongly points to accidental use of the
>> symbol.
> 
> I filed the bug because, when we were setting up the ABI test baselines in 
> the first place, it seemed desirable to verify them against binaries built 
> from various past glibc versions (to check as far as possible that the 
> contents of each symbol version in the test expectations were the same as 
> the contents of that version in the corresponding glibc release), and 
> doing such verification showed up that symbol as having been accidentally 
> removed.
> 
> (I did remove the _q_* symbols from powerpc soft-float; those had, in 
> glibc 2.4, been accidentally added to version GLIBC_2.2, but had never 
> been usefully usable - and made the baseline there reflect the accidental 
> removal of __fe_nomask_env which had also never been usable for that 
> configuration.)
> 

So do you think we can remove them or should we keep them? I do think
we should just remove these accidental bleeding symbols.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]