Re: Potential Mono removal from 32-bit big endian PowerPC in next upload
On 08/09/15 16:08, Lennart Sorensen wrote:
> On Mon, Sep 07, 2015 at 03:42:03PM +0100, Jo Shields wrote:
>> I can get things in under a "trivial" exception. One patch is in, the
>> other is stuck in "please rewrite this" hell @
>> https://github.com/mono/referencesource/pull/16 - I'll try to get to it
>> when I have time, but I have a backlog of other tasks
> Objection sounds fair.
>
>> The bigger issue is, sadly as feared, the Debian PowerPC builder boxes
>> don't like it. It works on a Mac Mini I was given access to, but it's
>> not sustainable for me to build/sign all PowerPC builds manually.
>>
>> https://buildd.debian.org/status/fetch.php?pkg=mono&arch=powerpc&ver=4.0.4.1%2Bdfsg-1&stamp=1441629526&file=log
>>
>> Any bright ideas from anyone else in the peanut gallery? It seems dumb
>> as hell to be to need to pull the package when it *does* work - just not
>> on the right computer for it to be considered "fine".
> Well so far I can't make it fail on my debian powerpc box.
>
> Which kernel is currently running on the machine where it fails?
3.16.0-4-powerpc64
> I see the build machines are power8, while I am using power7. One would
> hope that doesn't matter. Almost makes me wonder if there is a power8
> specific bug in libc or something that is causing the segfault.
> That would suck. I certainly don't have access to any power8 machines.
partch.debian.org, the porterbox where it fails:
cpu : POWER7 (architected), altivec supported
model : IBM,8231-E2B
Reply to: