This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [RFC 2.0] Implementing hwcap2
- From: Roland McGrath <roland at hack dot frob dot com>
- To: "Ryan S. Arnold" <ryan dot arnold at gmail dot com>
- Cc: "Ryan S. Arnold" <rsa at us dot ibm dot com>, libc-alpha <libc-alpha at sourceware dot org>
- Date: Thu, 28 Mar 2013 12:36:26 -0700 (PDT)
- Subject: Re: [RFC 2.0] Implementing hwcap2
- References: <1364493171 dot 3473 dot 23 dot camel at localhost dot localdomain> <20130328191637 dot 163BA2C0A1 at topped-with-meat dot com> <CAAKybw-1AiQ9iv7+f2q=TD6jjoE=xt1ZwwF8EeOMcqU=2PsjAA at mail dot gmail dot com>
> I believe I identified these through warnings.. I'll verify and send a
> correctness patch if that's the case.
I can't think of a non-error warning that would be relevant to this.
Compiling happily is not a proof of correctness for this. It's likely
that something in the large web of indirect includes gets the
definition, but we only call each source file "correct" when it
includes what it uses.
> The issue that you pointed out to me in my previous patch RFC was that
> we can't presume which order AT_HWCAP and AT_HWCAP2 will be
> encountered.
Indeed.
> So I guess we have to presume that it starts out zero.
> I'll add a comment.
Or you could only |= into something that you have initialized in the function.