Re: Cross-compiling for powerpcspe
On Nov 01, 2010, at 09:27, David Kuehling wrote:
> "Sebastian" == Sebastian Andrzej Siewior <bigeasy@linutronix.de> writes:
>> * David Kuehling | 2010-10-29 12:18:58 [+0200]:
>>> - It seems to be impossible to find the source-packages corresponding
>>> to the binary packages in debian-ports. E.g. debian-ports contains
>>> gcc-4.4_4.4.4-4+powerpcspe4.5.dsc, however the corresponding
>>> gcc-4.4_4.4.4.orig.tar.gz is not present on any debian mirror AFAICS.
>>> Did I overlook something?
>>
>> They are there if you look into the package folder. The uploads
>> into unreleased don't update the source dist i.e. there is none.
>
> The problem seems to be that the main debian repository is already one
> version ahead, providing only gcc-4.4_4.4.5.orig.tar.gz . As Neil
> Williams pointed out, I can still get the older tgz from
> snapshot.debian.org.
Ah, yeah, we probably need to get around to updating to the newer GCC. I've been swamped on other projects myself and I don't know that Sebastian's had any free time to work on it either.
For the moment we (Sebastian and myself) are sort of stalled waiting for:
(A) A GCC developer to step up and help us get that GCC bug fixed
(B) Debian squeeze to be released and squeeze+1 to be opened up so we can get stuff merged.
(C) Additional free-time or paid-time to work on stuff :-D.
>> As Wolfang mentioned we still have atleast one gcc bug which may
>> produce wrong code under certain circumstances. Thus you should not
>> use this in any production environment. A workaround is to add
>> -fno-schedule-insns (as last paramter) another should be to disable
>> double precision floating point.
>
> Already subscribed to that bug. I take it that your powerpcspe-specific
> gcc patches do not fix that bug?
>
> There are not too much other choices for a powerpcspe gcc compiler.
> There is the Freescale supplied gcc 4.1 [1] (last updated 2007?), but I
> won't trust it more than the unofficial Debian port's GCC :)
Back when the first GCC PowerPCSPE bug was causing major operational issues for us, we (in corporate capacity) contacted FreeScale via several channels, and still had absolutely zero luck getting any support from them. We are not a terribly happy FreeScale customer right now :-(. That's one of several big reasons that our organization's next major embedded project is going back to Intel processors, albeit at higher power consumption.
Cheers,
Kyle Moffett
Reply to: