Bug#428105: buffer broken on amd64: buffer (writer): write of data failed: Bad address

2008-05-10 Thread Tim Connors
On Fri, 9 May 2008, Martin Buck wrote:

 On Wed, May 07, 2008 at 05:10:15PM +0200, Martin Buck wrote:
  Pointers to these buffers are also stored inside the shared memory segment,
  so it looks like they somehow get corrupted. I just don't understand how
  and why this happens only on amd64 under unknown circumstances.

 I think I know what's wrong and I wonder how this code could ever work and
 fail only on amd64 under some circumstances...

 Please give the attached binary a try. It should fix the problem. If so,
 I'll prepare a new release in the next few days.

Yes - it does seem to work.

Ta.

-- 
TimC
Modus Ponens in action:
- Nothing is better than world peace.
- A turkey sandwich is better than nothing.
  ==  Ergo, a turkey sandwich is better than world peace.  --unknown



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Bug#428105: buffer broken on amd64: buffer (writer): write of data failed: Bad address

2008-05-09 Thread Martin Buck
On Wed, May 07, 2008 at 05:10:15PM +0200, Martin Buck wrote:
 Pointers to these buffers are also stored inside the shared memory segment,
 so it looks like they somehow get corrupted. I just don't understand how
 and why this happens only on amd64 under unknown circumstances.

I think I know what's wrong and I wonder how this code could ever work and
fail only on amd64 under some circumstances...

Please give the attached binary a try. It should fix the problem. If so,
I'll prepare a new release in the next few days.

Martin


buffer.gz
Description: Binary data


Bug#428105: buffer broken on amd64: buffer (writer): write of data failed: Bad address

2007-06-09 Thread Martin Buck
tags 428105 unreproducible
thanks

 On amd64, however, it seems 100% broken:

Works fine for me:

$ dpkg-architecture 
DEB_BUILD_ARCH=amd64
DEB_BUILD_ARCH_OS=linux
DEB_BUILD_ARCH_CPU=amd64
DEB_BUILD_GNU_CPU=x86_64
DEB_BUILD_GNU_SYSTEM=linux-gnu
DEB_BUILD_GNU_TYPE=x86_64-linux-gnu
DEB_HOST_ARCH=amd64
DEB_HOST_ARCH_OS=linux
DEB_HOST_ARCH_CPU=amd64
DEB_HOST_GNU_CPU=x86_64
DEB_HOST_GNU_SYSTEM=linux-gnu
DEB_HOST_GNU_TYPE=x86_64-linux-gnu
$ echo foo | ./buffer 
foo
$

Could you please run one of your test cases under strace -f and send me
the output? And please also include your kernel configuration since this
*could* have to do with how SysV shared memory is configured on your
machine.

Thanks,
Martin


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Bug#428105: buffer broken on amd64: buffer (writer): write of data failed: Bad address

2007-06-08 Thread Wolfgang Weisselberg
Package: buffer
Version: 1.19-9
Severity: important

Buffer works on x86.


On amd64, however, it seems 100% broken:

$ /dev/zero buffer
buffer (writer): write of data failed: Bad address
bytes to write=10240, bytes written=-1, total written  0K

$ /dev/zero buffer  /tmp/out
buffer (writer): write of data failed: Bad address
bytes to write=10240, bytes written=-1, total written  0K

$ find -type f | buffer
buffer (writer): write of data failed: Bad address
bytes to write=10240, bytes written=-1, total written  0K

$ ls | buffer
buffer (writer): write of data failed: Bad address
bytes to write=10240, bytes written=-1, total written  0K

$ echo -n  123 | buffer
buffer (writer): write of data failed: Bad address
bytes to write=3, bytes written=-1, total written  0K


-- System Information:
Debian Release: lenny/sid
  APT prefers unstable
  APT policy: (500, 'unstable'), (500, 'testing'), (500, 'stable')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.20.4 (SMP w/2 CPU cores; PREEMPT)
Locale: LANG=de_DE.UTF-8, LC_CTYPE=de_DE.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/bash

Versions of packages buffer depends on:
ii  libc6 2.5-10 GNU C Library: Shared libraries

buffer recommends no packages.

-- no debconf information


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]