Re: PATCH to optimize two Channels methods

[Tom Tromey <tromey at redhat dot com>]

>>>>> "Per" == Per Bothner <per@bothner.com> writes:

Per> But quibbling over rules is counter-productive.  Backing out a patch
Per> without the patcher's agreement is very seldom done.

I agree with this and all the other points you make.

Per> I apologize for not responding to Richard's problem, but it's inevitable
Per> that sometimes I will drop an email.  Talking about backing out my
Per> patches because I haven't responded to one message makes me defensive
Per> and annoyed.

Please accept my apologies.

I'm extra sensitive on this particular issue because I know we've
broken Richard's target several times before.  It is an ongoing
problem.  I'm sure it frustrates Richard, and it frustrates me because
I don't have a good solution for it.  Requiring arm-newlib builds,
even for some subset of patches, seems pretty heavy.  Just living with
the occasional break-delay-fix cycle is what we've defaulted on, but I
find that unsatisfying as well.  Any ideas?

Tom