Your message dated Tue, 19 May 2009 21:55:39 +0800
with message-id <1242741339.16123.87.ca...@chianamo>
has caused the report #529429,
regarding iotop: uses wrong syscall on biarch systems
to be marked as having been forwarded to the upstream software
author(s) Guillaume Chazarain <[email protected]>
(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 [email protected]
immediately.)
--
529429: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=529429
Debian Bug Tracking System
Contact [email protected] with problems
--- Begin Message ---
On Tue, 2009-05-19 at 12:59 +0200, Piotr Engelking wrote:
> On i386 system with amd64 kernel, iotop exits without printing anything:
>
> $ iotop -b || echo $?
> 1
> $
>
> Running without -b additionally corrupts the terminal.
>
> The problem is that ioprio.find_ioprio_get_syscall_number() uses uname -m to
> find syscall numbers, which breaks on biarch systems. On i386/amd64 in
> particular, iotop calls exit_group(2) instead of ioprio_get(2) as a result.
>
> Please use something like dpkg-architecture instead of uname. Better yet,
> use a less fragile method of dispatching syscalls, if you can think of sane
> one.
Guillaume, any idea about how to fix this one? dpkg-architecture is
obviously the wrong solution since it isn't cross-distro. I guess the
right way to fix this would be to create a C-based syscall python module
that would embed some syscall numbers from #include <sys/syscall.h>?
--
bye,
pabs
http://wiki.debian.org/PaulWise
signature.asc
Description: This is a digitally signed message part
--- End Message ---
