This is the mail archive of the
binutils@sourceware.org
mailing list for the binutils project.
Re: PATCH: PR binutils/12283: bfd/doc doesn't support parallel build
- From: "H.J. Lu" <hjl dot tools at gmail dot com>
- To: Ralf Wildenhues <Ralf dot Wildenhues at gmx dot de>
- Cc: binutils at sourceware dot org
- Date: Thu, 2 Dec 2010 10:16:07 -0800
- Subject: Re: PATCH: PR binutils/12283: bfd/doc doesn't support parallel build
- References: <20101202170032.GA19376@intel.com> <20101202180829.GA16143@gmx.de>
On Thu, Dec 2, 2010 at 10:08 AM, Ralf Wildenhues <Ralf.Wildenhues@gmx.de> wrote:
> Hello H.J.,
>
> * H.J. Lu wrote on Thu, Dec 02, 2010 at 06:00:32PM CET:
>> Any objection to this patch?
>
> The rule to update $(MKDOC):
>
> $(MKDOC): $(srcdir)/chew.c
> ? ? ? ?$(CC_FOR_BUILD) -o chew.$$$$ $(CFLAGS_FOR_BUILD) $(LDFLAGS_FOR_BUILD) \
> ? ? ? ? ?$(H_CFLAGS) $(AM_CPPFLAGS) $(srcdir)/chew.c; \
> ? ? ? ?$(SHELL) $(srcdir)/../../move-if-change chew.$$$$ $(MKDOC)
>
> should be atomic (the ';' should be '&&' though). ?So, I would be
Failure is very random. There is a race condition. "make -j8" on a 16core
machine can start making $(MKDOC) at the same time. How can move-if-change
be 100% atomic?
> interested to see the last 100 or so lines of a failure happening,
> because I can't spot one nor reproduce it. ?Also, configure flags.
>
--
H.J.