[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: GnuTLS 2.9.12 - third and final (?) release candidate for 2.10.0
From: |
Simon Josefsson |
Subject: |
Re: GnuTLS 2.9.12 - third and final (?) release candidate for 2.10.0 |
Date: |
Mon, 21 Jun 2010 09:55:02 +0200 |
User-agent: |
Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) |
Andreas Metzler <address@hidden> writes:
> On 2010-06-18 Simon Josefsson <address@hidden> wrote:
>> Andreas Metzler <address@hidden> writes:
> [...]
>>> a workaround for http://bugs.debian.org/519006 would be nice, 2.9.x
>>> does not build on mips(el). (See e.g.
>>> http://bugs.debian.org/585476)
>
>> That is a toolchain bug on mips(el), and we don't know how to work
>> around it in GnuTLS.
>
>> I'm not sure I see why we should work around it either, the bug affects
>> so many debian packages now (coreutils, m4, unistring, libidn, ...) so
>
> Fair enough. I have worked around the issue now by building without -g
> on mips(el), which seems to have been a success:
> https://buildd.debian.org/build.php?pkg=gnutls26&arch=mips&ver=2.9.12-2
That makes the libgnutls26-dbg package useless, which it doesn't have to
be if instead self checks are disabled. But this only affects mips(el)
users so I don't care strongly...
>> the proper fix is to fix the bug in binutils.
>
> Afaiu the bug is actually in gcc-4.4 (4.5 is supposed to be ok).
The patch says the problem also affects x86_64, but I haven't seen any
evidence of this. So I wonder what makes this bug only appear on
mips(el)...
/Simon