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: asprintf() issue


On Fri, May 29, 2015 at 5:47 AM, Joseph Myers <joseph@codesourcery.com> wrote:
> > What are the next steps on this asprintf() issue?
>
> If a long discussion with various different views dies down, the next step
> is to post a careful analysis of that discussion to help the community in
> reaching consensus, or to help the community see what the consensus is
> that was reached.

OK here goes...

First question: should the new, specified behavior be set to NULL or
leave unmodified? The general agreement is to set to NULL: consistent
with BSD, what people are already patching glibc to do, more
developer-friendly, better at detecting bugs, etc.

Second question: what should specification/man page say? General
agreement that we should document both the new behavior, the previous
behavior, and the version in which the change occurred.

Next question: whether to just change the behavior or create a symbol
alias. If the latter, there were two options:

    - option (b): have a new symbol version with the old version
      being an alias of the new (so that new binaries that may rely
      on it being set to NULL don't run with old glibc)
    - option (c): have a new symbol version with the old version
      not changing *ptr on error

Initially most agreed creating an alias using option (c). Then Florian
said creating an alias was not worth the trouble, there was some
discussion, and then Florian came around saying "If that's the
emerging consensus, it's something with which I can agree."

So...  I think we now have consensus for option (c).

Carlos also pointed out we should add a test case that tests the
ENOMEM failure case.

Does that summary sound right?

-Archie

-- 
Archie L. Cobbs


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