[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#539470: marked as done (Warning: Immediates 21 and 32 will give undefined behavior.)

Your message dated Sat, 19 Sep 2009 11:54:49 +0200
with message-id <4AB4AA69.5060508@debian.org>
and subject line Re: Bug#539470: Warning: Immediates 21 and 32 will give undefined behavior.
has caused the Debian Bug report #539470,
regarding Warning: Immediates 21 and 32 will give undefined behavior.
to be marked as done.

This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.

(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact owner@bugs.debian.org
immediately.)


-- 
539470: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=539470
Debian Bug Tracking System
Contact owner@bugs.debian.org with problems
--- Begin Message --- 
Package: gcc-4.3
Tags: upstream, patch

I'm filing this BR against gcc-4.3, but according to upstream it affects
*all* versions of gcc, and in all Debian releases!

During compilations using gcc-4.3 of the upstream 2.6.31-rc4 kernel, I
twice noticed the following warning:
{standard input}: Assembler messages:
{standard input}:1345: Warning: Immediates 21 and 32 will give undefined behavior.

John David Anglin responded:
   GCC loads various constants with depwi and depdi. The deposit length was
   wrong. Helge found that a 715 failed to boot because some hand written
   assembler code had an incorrect deposit length. Thus, he believes that
   it is an issue to worry about on old machines.
   It is not a new issue. The responsible GCC code had not changed since
   1997. So, maybe it is not a problem on most PA machines.

A patch has been committed to upstream GCC trunk, 4.3 and 4.4 branches, but
not yet released:
2009-07-20  John David Anglin  <dave.anglin@nrc-cnrc.gc.ca>
        * pa.c (compute_zdepwi_operands): Limit deposit length to 32 - lsb.
        Cast "1" to unsigned HOST_WIDE_INT.
        (compute_zdepdi_operands): Limit maximum length to 64 bits.  Limit
        deposit length to the maximum length - lsb.  Extend length if
        HOST_BITS_PER_WIDE_INT is 32.

For details, see the following thread on the d-hppa list:
http://lists.debian.org/debian-hppa/2009/07/msg00093.html

For questions I suggest you contact John David Anglin or Carlos O'Donell.

Cheers,
FJP

--- End Message ---
--- Begin Message --- 
fixed in testing/unstable. I don't plan to update gcc-4.3 in stable.

On 01.08.2009 09:55, Frans Pop wrote:

Package: gcc-4.3
Tags: upstream, patch

I'm filing this BR against gcc-4.3, but according to upstream it affects
*all* versions of gcc, and in all Debian releases!

During compilations using gcc-4.3 of the upstream 2.6.31-rc4 kernel, I
twice noticed the following warning:
{standard input}: Assembler messages:
{standard input}:1345: Warning: Immediates 21 and 32 will give undefined behavior.

John David Anglin responded:
    GCC loads various constants with depwi and depdi. The deposit length was
    wrong. Helge found that a 715 failed to boot because some hand written
    assembler code had an incorrect deposit length. Thus, he believes that
    it is an issue to worry about on old machines.
    It is not a new issue. The responsible GCC code had not changed since
    1997. So, maybe it is not a problem on most PA machines.

A patch has been committed to upstream GCC trunk, 4.3 and 4.4 branches, but
not yet released:
2009-07-20  John David Anglin<dave.anglin@nrc-cnrc.gc.ca>
         * pa.c (compute_zdepwi_operands): Limit deposit length to 32 - lsb.
         Cast "1" to unsigned HOST_WIDE_INT.
         (compute_zdepdi_operands): Limit maximum length to 64 bits.  Limit
         deposit length to the maximum length - lsb.  Extend length if
         HOST_BITS_PER_WIDE_INT is 32.

For details, see the following thread on the d-hppa list:
http://lists.debian.org/debian-hppa/2009/07/msg00093.html

For questions I suggest you contact John David Anglin or Carlos O'Donell.

Cheers,
FJP

--- End Message ---
Reply to: