From: Jeff King <peff@peff.net>
Date: Mon, 27 Jun 2016 14:36:48 GMT
Subject: Re: [PATCH 1/2] xread: retry after poll on EAGAIN/EWOULDBLOCK
Message-ID: <20160627143648.GA2618@sigill.intra.peff.net>
In-Reply-To: <xmqqoa6mdbu3.fsf@gitster.mtv.corp.google.com>

On Mon, Jun 27, 2016 at 06:02:12AM -0700, Junio C Hamano wrote:

> Jeff King <peff@peff.net> writes:
> 
> > I also wondered how we managed to miss such an obvious point in review
> > of the original patch. Sadly, we _did_ notice it[1] but it looks like we
> > never fixed the problem. That is even more disturbing.
> 
> Yes indeed.
> 
> I try to pay attention to "this is broken because..."  comments in
> discussions to make a note in my copy of "What's cooking" report for
> a problematic topic, as that is where I work from when merging
> topics down, but apparently that procedure failed work in this case.
> There needs a stronger mechanism to stop a known-buggy patch from
> going thru, but I am not sure offhand what that should be.

I was the one who saw the bug. I could have followed the series more
closely to make sure my concern was addressed. Or possibly pointed out
the bug more prominently than an in a "PS" as part of the discussion.

I think part of the problem was that this particular series was
large-ish and involved a lot of re-rolls, and I got sick of looking at
it. I dunno.

It's also true that our error rate will never be 0%. So some bugs will
always slip through, some review comments will be forgotten, etc. Eric
did find and fix the bug just now, so the "many eyes" theory did work
here eventually.

-Peff