[poky] GCC 4.6.0 on ARM?

[Khem Raj]

On Wed, May 11, 2011 at 4:00 AM, Gary Thomas <gary at mlbassoc.com> wrote:
> On 05/10/2011 09:28 PM, Kamble, Nitin A wrote:
>>
>>
>>> -----Original Message-----
>>> From: poky-bounces at yoctoproject.org [mailto:poky-
>>> bounces at yoctoproject.org] On Behalf Of Gary Thomas
>>> Sent: Monday, May 02, 2011 5:57 AM
>>> To: Khem Raj
>>> Cc: Poky Project; Patches and discussions about the oe-core layer
>>> Subject: Re: [poky] GCC 4.6.0 on ARM?
>>>
>>> On 04/30/2011 04:10 PM, Khem Raj wrote:
>>>>
>>>> On Thu, Apr 28, 2011 at 11:23 AM, Khem Raj<raj.khem at gmail.com>
>>>
>>> wrote:
>>>>>
>>>>> On Thu, Apr 28, 2011 at 10:09 AM, Gary Thomas<gary at mlbassoc.com>
>>>
>>> wrote:
>>>>>>
>>>>>> I got my OMAP target to build with GCC-4.6.0 using the latest
>>>>>> Poky tree after a fresh rebuild.  Sadly, the Linux kernel
>>>
>>> (2.6.37.3)
>>>>>>
>>>>>> does not run correctly when built with this compiler.  It
>>>>>> fails when enumerating the USB bus :-(
>>>>>>
>>>>>> Are there known issues with the Linux kernel on ARM and GCC-4.6?
>>>>>>
>>>>>
>>>>> There has been fixes on top of 4.6.0 the one we have right now is
>>>
>>> vanilla
>>>>>
>>>>> 4.6.0 release. I plan to add second set of patches but you have to
>>>
>>> dig a bit
>>>>>
>>>>> more to identify the problem
>>>>>
>>>>
>>>> I have some updates for gcc 4.6.0
>>>> they are staged in
>>>>
>>>> http://git.openembedded.org/cgit.cgi/openembedded-core-
>>>
>>> contrib/log/?h=kraj/gcc-4.6
>>>>
>>>> (top 3 patches)
>>>>
>>>> Please test them out and if you are lucky
>>>> your problem might be fixed too.
>>>>
>>>> All, please try these patches and provide any feedback
>>>
>>> I was able to build my kernel (ARM OMAP/3530 based) using this, but
>>> the USB still doesn't work, so no improvement, sorry.
>>
>> Gary,
>>  Is the USB breakage is due to gcc 4.6.0 ?
>
> Yes, that's the only change.  Building the kernel from the same
> source repository with 4.5.2 (using patches from you and Khem Raj)
> works perfectly.
>

Unless we dig down and narrow the issue its hard to say if its a gcc
issue or a latent problem that
gcc 4.6 brings out.