Ping^6: [PATCH][v3] Add dynamic linker support for $EXEC_ORIGIN.
Roland McGrath
roland@hack.frob.com
Fri Mar 21 22:31:00 GMT 2014
> Certainly. I explained the case for the feature in my initial
> description of the patch (and I'll requote that below); I'm not clear
> if you didn't see that, or if you're implying that it needs more
> elaboration and/or stronger justification. Could you clarify? I'm
> happy to elaborate if that's what you're asking for.
I saw it. My point was that this is the thing that needs consensus
(first). Sending the patch again (especially with ever-more top-posting
and > prefixing so the original message is harder and harder to actually
see in there) is not an effective way to spur discussion on the question.
Having seen it, I have not convinced myself that a) I care enough to think
about it, b) it is necessarily safe, c) the use case really justifies the
feature, etc. You need to give some justification that makes people think
it's worth thinking hard enough about the subject to become sanguine about
any particular conclusion other than the default do-nothing. I think for
most people the feature discussion that needs to happen has just gotten
lost in the sea of "please look at my patch already" that doesn't clearly
indicate what we should be discussing first.
Thanks,
Roland
More information about the Libc-alpha
mailing list