From mboxrd@z Thu Jan  1 00:00:00 1970
From: Andreas Schwab <schwab@linux-m68k.org>
Date: Wed, 24 Mar 2021 10:18:28 +0100
Subject: [PATCH v2 2/2] test: unit test for longjmp
In-Reply-To: <90e8d095-9a0a-2524-131c-49e79a47aa08@gmail.com> (Sean Anderson's
 message of "Mon, 22 Mar 2021 09:30:41 -0400")
References: <20210322110249.4387-1-xypron.glpk@gmx.de>
 <20210322110249.4387-3-xypron.glpk@gmx.de>
 <47007446-5ad9-c5b2-dc55-78400f288dc5@gmail.com>
 <90e8d095-9a0a-2524-131c-49e79a47aa08@gmail.com>
Message-ID: <875z1goqor.fsf@igel.home>
List-Id: <u-boot.lists.denx.de>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
To: u-boot@lists.denx.de

On M?r 22 2021, Sean Anderson wrote:

> int test_longjmp_ret(int i)
> {
>      jmp_buf env;
>      int ret;
>      int foo = i;
>
>      ret = setjmp(env);
>      if (ret)
>          return foo;
>      foo = 0x1000;
>      longjmp(env, i);
>      /* We should not arrive here */
>      return foo;

This is undefined.  When modifying a non-volatile auto variable between
setjmp and longjmp, there is no requirement that the value is preserved.

Andreas.

-- 
Andreas Schwab, schwab at linux-m68k.org
GPG Key fingerprint = 7578 EB47 D4E5 4D69 2510  2552 DF73 E780 A9DA AEC1
"And now for something completely different."