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

alignment errors on armel: what to do?

Hey debian-arm folks--

after dealing with #548815, i'm a little bit concerned about the
behavior of the kernel in the face of alignment errors on armel.

i've read http://bugs.debian.org/397616 and followed the references in
there, so i think i understand why the default is to silently fail when
alignment errors happen.

However, it seems like we should still be filing bugs against packages
which trigger alignment errors, no?  Otherwise, the authors and
maintainers of those packages (who might not have arm hardware) might
never know that their code is misbehaving on ARM.

i just turned on warnings in an NSLU2 running squeeze (a buildd for me)
and note alignment warnings from several processes:

 pdftex (reproducable with "aptitude reinstall texlive-base-bin")

 aptitude (also reproducable with "aptitude reinstall texlive-base-bin")

 apt-extracttemplates (reproducible with "apt-extracttemplates
/var/cache/apt/archives/patchutils_0.3.1-2_armel.deb" or any .deb)

 grotty (reproducible with "man gpgv")

 gpgv (reproducible during "aptitude update", haven't narrowed it down

 /usr/lib/apt/methods/http (reproducible during "aptitude update", also
haven't narrowed it further)

and this is just a quick overview from a few minutes testing.

This seems like a bad situation, particularly if these tools are
expecting "valid" behavior from whatever calls are going on.

should we be filing bugs for each of these?  Should we be doing anything


PS you can see how many userspace alignment errors have happened since
your last boot with:

  grep ^User: /proc/cpu/alignment

You can ask your kernel to log these errors with:

  echo 1 > /proc/cpu/alignment

(beware: this could cause a loop if there is an alignment error in any
process which deals with kernel output!)

Attachment: signature.asc
Description: OpenPGP digital signature

Reply to: