Apt alignment trap.

[Andrea Gasparini]

Hi,
I'm pretty new to porting debian on arm, so please be kind with me.... ;)

Jokes apart, I've the following issue: I'm seeing a lot of alignment trap
on apt-get, even on a simple "apt-get update".

Should I file a bug against apt (but it seems very strange to me that no
one encountered this yet), or did I miss something ?

The kernel was compiled with CodeSourcery toolchain, and my processor is a
Ti Davinci Dm355, don't know if it matters...

# file `which apt-get`
/usr/bin/apt-get: ELF 32-bit LSB executable, ARM, version 1 (SYSV),
dynamically linked (uses shared libs), for GNU/Linux 2.6.18, stripped

# cat /proc/cpuinfo
Processor : ARM926EJ-S rev 5 (v5l)

Thanks, bye.
--
-gaspa-
-----------------------------------------------
-------- https://launchpad.net/~gaspa ---------
------ HomePage: iogaspa.altervista.org -------
-Il lunedi'dell'arrampicatore: www.lunedi.org -


--
To UNSUBSCRIBE, email to debian-arm-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

[martinwguy]

On 10/8/09, Andrea Gasparini <gaspa@...> wrote:
> I'm seeing a lot of alignment trap
>  on apt-get, even on a simple "apt-get update".
>
>  Should I file a bug against apt (but it seems very strange to me that no
>  one encountered this yet), or did I miss something ?

I guess you're using squeeze or sid release. Yes, something is causing
apt and many other programs to get alignment traps.
The bug is actually in glibc bugs.debian.org/548842

If you'd like to help debug this, you can
echo 5 > /proc/cpu/alignment
and run apt-get under gdb - it will be killed with a Bus Error at the bad code.

    M


--
To UNSUBSCRIBE, email to debian-arm-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

[John Reiser-2]

> If you'd like to help debug this, you can
> echo 5>  /proc/cpu/alignment
> and run apt-get under gdb - it will be killed with a Bus Error at the bad code.

>   *reloc_addr += value

Some shared library has been built with an initialized pointer, where the storage
for the pointer itself is not aligned on a 4-byte boundary.  The problem is not
in glibc(ld-linux); the problem lies in some shared library that the app requires.

Debug this via
     setenv LD_DEBUG reloc
     ./my_app args...
or perhaps [in bash]:
     LD_DEBUG=reloc,files  ./my_app args...

Set LD_DEBUG=help to get info on other options for debugging the processing
that ld-linux does.

--


--
To UNSUBSCRIBE, email to debian-arm-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

[martinwguy]

reassign 548842 gcc-4.3 4.3.4-2
thanks

On 10/9/09, John Reiser <vendor@...> wrote:

>  Some shared library has been built with an initialized pointer, where the
> storage
>  for the pointer itself is not aligned on a 4-byte boundary.  The problem is
> not
>  in glibc(ld-linux); the problem lies in some shared library that the app
> requires.
>
>  Debug this via
>     setenv LD_DEBUG reloc
>     ./my_app args...

Thanks! It seems to affect any C++ program on armel, including hello.cc

martin@n2100:~$ LD_DEBUG=reloc ./a.out
      6836:
      6836: relocation processing: /lib/libc.so.6 (lazy)
      6836:
      6836: relocation processing: /lib/libgcc_s.so.1 (lazy)
      6836:
      6836: relocation processing: /lib/libm.so.6 (lazy)
      6836:
      6836: relocation processing: /usr/lib/libstdc++.so.6 (lazy)
Bus error

and libstdc++.so.6 seems to be provided by gcc-4.3, so I'm reassigning
the bug...


--
To UNSUBSCRIBE, email to debian-arm-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...