Re: [PATCH] Remove support for "rtld_" prefix on solib-svr4 probes

[Pedro Alves <palves at redhat dot com>]

On 09/25/2014 10:23 PM, Sergio Durigan Junior wrote:
> On Thursday, September 25 2014, Pedro Alves wrote:
> 
>> There's RHEL (at least, per the comment) 6.4 too, which isn't EOL'ed,
>> though.  It's reasonable to expect that people may still want to
>> build/test upstream gdb on those?
> 
> Damn, I forgot to talk about RHEL.
> 
> But yeah, RHEL-6.x is not EOL'ed, but the GDB that comes with it is from
> Red Hat as well, and contains all the necessary patches to deal with
> whatever "internal idiosyncrasy" that it may need.  OTOH, I know Gary
> has been using RHEL-6.5 to test his upstream patches, so to answer your
> question, it is still possible that people may still want to buid/test
> upstream GDB there.
> 
> My opinion is that we shouldn't need to worry about internals of each
> distro (I know we *have to do that* sometimes, but that's not an
> excuse), so I still hold the cleanup patch for approval :-).

There's the system GDB, that is usually maintained by the
distro, but then it's quite often the case that people build
and ship their own tools on top of the distro, bypassing the
system tools.

I tend to view supporting older-ish distros that people might
still be using like the proprietary OSs we "support" (in a sense).
I think that just as we'd accept a patch that makes GDB work better
on Windows 7 OOTB (e.g., to work around some debug API issue), even
though there's already Windows 8 out there, I think patches that make
GDB work better OOTB on a bit older (but still in use) distros are
fine, as long as they don't get in the way of progress and don't
impose a big maintenance burden.

IMHO, there's no harm in leaving this particular bit in
a while longer.

But I certainly won't cry over this.  I'm not personally affected.
If others are fine with yanking this out, I'll be fine with it too.

Thanks,
Pedro Alves