Re: Bug#845690: gcc-6: gcc creates unbootable kernel on x86-64
On 27.11.2016 19:27, Sven Joachim wrote:
> On 2016-11-27 18:34 +0100, Matthias Klose wrote:
>
>> On 27.11.2016 16:51, Sven Joachim wrote:
>>> On 2016-11-27 13:39 +0100, Matthias Klose wrote:
>>>
>>>> Control: tags -1 + help moreinfo
>>>> Control: severity -1 important
>>>>
>>>> On 27.11.2016 08:38, Sven Joachim wrote:
>>>>> Control: reassign -1 binutils 2.27.51.20161124-1
>>>>> Control: retitle -1 binutils: creates unbootable kernel on x86-64
>>>>> Control: severity -1 grave
>>>>>
>>>>> On 2016-11-26 15:13 +0100, Damien Wyart wrote:
>>>>>
>>>>>> After running further tests today, I think this is in fact *not*
>>>>>> related to gcc but to the kernel itself.
>>>>>>
>>>>>> I tested all 6.2.1-X versions as well as gcc-5 (5.4.1-3) and all the
>>>>>> kernels fail to boot (balck screen just after grub and nothing in the
>>>>>> logs).
>>>>>
>>>>> Same here, downgrading binutils to 2.27.51.20161118-2 helped. I'm
>>>>> reassigning the bug and bumping the severity, since several people have
>>>>> observed the problem.
>>>>
>>>> The original report talks about a 4.4 problem on , which afaik is superseded in
>>>> unstable by newever versions released after the GCC 6 release. This is now made
>>>> a binutils RC issue for building a kernel which is not in the archive anymore.
>>>> Please could you validate that the issue exists with the linux package in
>>>> unstable as well?
>>>
>>> I have noticed the problem with vanilla Linux 4.8.11 from kernel.org, so
>>> I suspect the Debian kernel is affected as well. There is no console
>>> output at all, the system freezes right when uncompressing the kernel.
>>>
>>> It should be noted that I haven't noticed the problem on my desktop
>>> (which has a 32-bit userland but a 64-bit kernel) where I have
>>> CONFIG_KERNEL_GZIP=y, but on my laptop which uses the default
>>> CONFIG_KERNEL_XZ=y it is reproducible.
>>
>> if it's really binutils, I prepared a package reverting the fix for PR ld/20815.
>> Would be nice if somebody could check that out:
>> https://people.debian.org/~doko/tmp/binutils_2.27.51.20161124-1.1_amd64.deb
>
> Thanks, that binutils package produces a working kernel here.
please could you check again with
https://people.debian.org/~doko/tmp/binutils_2.27.51.20161127-1.1_amd64.deb
having the suggested fix proposed at
https://sourceware.org/ml/binutils/2016-11/msg00348.html
Reply to: