Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Date: Thu, 13 Aug 2020 11:53:58 +0300 (MSK)
From: Alexander Monakov <amonakov@...ras.ru>
To: musl@...ts.openwall.com
Subject: Re: [PATCH 1/3] setjmp: fix x86-64 longjmp argument
 adjustment



On Thu, 13 Aug 2020, Jens Gustedt wrote:

> Alexander,
> probably this has nothing to do with the issue at hand but 
> 
> on Wed, 12 Aug 2020 16:29:59 +0300 (MSK) you (Alexander Monakov
> <amonakov@...ras.ru>) wrote:
> 
> >         int v = setjmp(jb);
> 
> is a use of `setjmp` that is not conforming. `setjmp` is only allowed
> either standalone or as controlling expression either directly or
> negated.

Thanks. Yeah, this is a moot point as the test needs GCC with -O anyway, and
GCC allows setjmp in any context. But nevertheless it's easy to adjust it
by changing the offending line to 'if (setjmp(jb)) return 0;' and changing
the final 'return !v'  to 'return 1'.

Alexander

Powered by blists - more mailing lists

Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.