Re: [PATCH 6/6] reflog-walk: stop using fake parents

From: Eric Wong <e@80x24.org>
Date: Jul 10, 2017, 09:42 UTC
Message-ID: <20170710094255.GA8170@untitled>
In-Reply-To: <20170707030224.4iubuzsz2qwdhgiw@sigill.intra.peff.net>

Jeff King <peff@peff.net> wrote:

Show 11 quoted lines

> On Fri, Jul 07, 2017 at 12:32:39AM +0000, Eric Wong wrote:
> 
> > I'm not sure why, but this is causing t1414.8 failures on 32-bit
> > x86 with the latest pu with Debian jessie (oldstable).
> > 
> > Reverting this (beafb2c62947a6d4a97b9c3baf99fe62ec8e830f) in pu
> > seems to fix the test for me.
> > 
> > +Cc: Ramsay since he also had a 32-bit environment.
> 
> Thanks, I was able to reproduce with CFLAGS=-m32.

No problem and thanks for the quick response! Latest pu is fine for me.

To notice similar errors sooner, I wonder if we should have a test target for 64-bit users to test with -m32 enabled, somehow.

Fwiw, I started having a cronjob which runs "make test-pu" on my 32-bit machine, where config.mak has this:

==> config.mak <== DEVELOPER := 1 pulog := /tmp/git-test-pu.log

cur_head = $(shell git rev-parse HEAD^0) cur_pu = $(shell git rev-parse refs/remotes/origin/pu^0)

test-pu:
	git fetch -q origin
	test x"$(cur_pu)" = x"$(cur_head)" || { \
		$(MAKE) clean && \
		git reset --hard origin/pu && \
		{ $(MAKE) -k test >$(pulog) 2>&1 || \
		 mutt -s 'pufail' e@80x24.org <$(pulog); }; }

-- 
EW