Re: Poky on "standard" PC

[Saul Wold <saul.wold@intel.com>]

On 01/25/2011 08:45 AM, Gary Thomas wrote:
> On 01/25/2011 09:40 AM, Bruce Ashfield wrote:
>> On 11-01-25 11:37 AM, Gary Thomas wrote:
>>> On 01/25/2011 06:45 AM, Bruce Ashfield wrote:
>>>> On Tue, Jan 25, 2011 at 7:08 AM, Gary Thomas<gary@mlbassoc.com> wrote:
>>>>> On 01/24/2011 03:54 PM, Darren Hart wrote:
>>>>>>
>>>>>> On 01/24/2011 04:39 AM, Richard Purdie wrote:
>>>>>>>
>>>>>>> On Mon, 2011-01-24 at 05:05 -0700, Gary Thomas wrote:
>>>>>>>>
>>>>>>>> I wanted to try Poky on a spare laptop, so I went looking
>>>>>>>> for an appropriate configuration to build.
>>>>>>>>
>>>>>>>> From 'README.hardware' (last updated 2010-12-16), I read:
>>>>>>>> Poky Boot CD (bootcdx86)
>>>>>>>> ========================
>>>>>>>>
>>>>>>>> The Poky boot CD iso images are designed as a demonstration of the
>>>>>>>> Poky
>>>>>>>> environment and to show the versatile image formats Poky can
>>>>>>>> generate.
>>>>>>>> It will
>>>>>>>> run on Pentium2 or greater PC style computers. The iso image can be
>>>>>>>> burnt to CD and then booted from.
>>>>>>>>
>>>>>>>> When I tried it (it's also listed in the default local.conf),
>>>>>>>> I got this error:
>>>>>>>> Please set TARGET_ARCH directly, or choose a MACHINE or DISTRO
>>>>>>>> that does
>>>>>>>> so.
>>>>>>>> Please set a valid MACHINE in your local.conf
>>>>>>>>
>>>>>>>> There seems to be no machine named 'bootcd*' anywhere :-(
>>>>>>>>
>>>>>>>> What should I be using?
>>>>>>>
>>>>>>> bootcdx86 was superceeded by the -live images. I'd suggest using the
>>>>>>> qemux86 or atom-pc machines and building a -live image, dd that to
>>>>>>> a usb
>>>>>>> stick and boot that on the laptop. An iso format image is also
>>>>>>> generated
>>>>>>> that can be placed onto a CD.
>>>>>>
>>>>>> It will have to be atom-pc as qemu doesn't provide live images
>>>>>> (last I
>>>>>> checked, a few weeks back).
>>>>>
>>>>> Sadly, this doesn't build :-( Failed in compile_perf, log attached.
>>>>
>>>> Hmm. This one was worked around a while ago via:
>>>>
>>>> commit d38ad3271829b2d282c973e69edd265ef52174c8
>>>> Author: Tom Zanussi<tom.zanussi@intel.com>
>>>> Date: Fri Jan 14 16:33:05 2011 -0600
>>>>
>>>> linux-yocto: hard-code NO_LIBPERL/NO_LIBPYTHON
>>>>
>>>> ExtUtils::Embed ccopts is getting the host's -I/usr/local/include and
>>>> using it to compile perf, which results in a compilation error that
>>>> started appearing just recently.
>>>>
>>>> This turns the code that makes use of ExtUtils::Embed off and simply
>>>> hard-codes NO_LIBPERL.
>>>>
>>>> It does the same for LIBPYTHON while we're at it, since it probably
>>>> suffers from a similar underlying problem and just by chance hasn't
>>>> broken anything yet.
>>>>
>>>> This will be re-enabled after I familiarize myself with the perf
>>>> recipe and am able to create a proper fix.
>>>>
>>>> Signed-off-by: Tom Zanussi<tom.zanussi@intel.com>
>>>>
>>>> :100644 100644 21023da... 7833f65... M
>>>> meta/recipes-kernel/linux/linux-yocto_git.bb
>>>>
>>>> But the issue hadn't been showing up on the -stable .. until now. We
>>>> can cherry
>>>> pick this patch and commit to at least temporarily work around it.
>>>>
>>>> I would have expected the nightly builds to show this one, since not
>>>> all
>>>> supported targets have moved to 2.6.37.
>>>
>>> I just updated to the latest master and started from scratch.
>>> It's still happening :-(
>>>
>>> OE Build Configuration:
>>> BB_VERSION = "1.11.0"
>>> METADATA_BRANCH = "master"
>>> METADATA_REVISION = "beb9166b3adf83c5e864afcb56cf9ea24f55a7c3"
>>> TARGET_ARCH = "i586"
>>> TARGET_OS = "linux"
>>> MACHINE = "atom-pc"
>>> DISTRO = "poky"
>>> DISTRO_VERSION = "0.9+snapshot-20110125"
>>> TARGET_FPU = ""
>>>
>>> | CC util/scripting-engines/trace-event-perl.o
>>> | cc1: warnings being treated as errors
>>> | cc1: error: include location "/usr/local/include" is unsafe for
>>> cross-compilation
>>> | make: *** [util/scripting-engines/trace-event-perl.o] Error 1
>>> | make: Leaving directory
>>> `/home/local/pc_poky/tmp/work/atom-pc-poky-linux/linux-yocto-stable-2.6.34+git1+e1f85a470934a0cf6abde5d95533e74501822c6b_0+35521a5a70316785a67aca1de1d39a7b84c49ccf-r1/linux/tools/perf'
>>>
>>>
>>>
>>> | FATAL: oe_runmake failed
>>> | Function 'do_compile_perf' failed (see
>>> /home/local/pc_poky/tmp/work/atom-pc-poky-linux/linux-yocto-stable-2.6.34+git1+e1f85a470934a0cf6abde5d95533e74501822c6b_0+35521a5a70316785a67aca1de1d39a7b84c49ccf-r1/temp/log.do_compile_perf.8191
>>>
>>>
>>> for further information)
>>> | ERROR: Function 'do_compile_perf' failed (see
>>> /home/local/pc_poky/tmp/work/atom-pc-poky-linux/linux-yocto-stable-2.6.34+git1+e1f85a470934a0cf6abde5d95533e74501822c6b_0+35521a5a70316785a67aca1de1d39a7b84c49ccf-r1/temp/log.do_compile_perf.8191
>>>
>>>
>>> for further information)
>>> NOTE: package
>>> linux-yocto-stable-2.6.34+git1+e1f85a470934a0cf6abde5d95533e74501822c6b_0+35521a5a70316785a67aca1de1d39a7b84c49ccf-r1:
>>>
>>> task do_compile_perf: Failed
>>>
>>> Maybe it has something to do with my build host (Fedora-13, x86)?
>>
>> In this case the host is fine. That commit was on the
>> 2.6.37 kernel, and I used your report as a trigger and
>> backported it to 2.6.34. I have a pull request out now,
>> which should fixup the perf compilation issues once
>> merged.
>
> Thanks, I'll try it with your patch.
>
This has been patched in master now also.

Sau!