[yocto] modify a kernel patch

[Bruce Ashfield]

On 12-05-08 09:03 AM, Frans Meulenbroeks wrote:
> 2012/5/8 Bruce Ashfield<bruce.ashfield at windriver.com>:
>> On 12-05-08 03:18 AM, Frans Meulenbroeks wrote:
>>>
>>> 2012/5/7 Bruce Ashfield<bruce.ashfield at windriver.com>:
>>>>
>>>> On 12-05-07 03:29 PM, Frans Meulenbroeks wrote:
>>>>>
>>>>>
>>>>> 2012/5/7 Bruce Ashfield<bruce.ashfield at windriver.com>:
>>>>>>
>>>>>>
>>>>>> On 12-05-07 10:41 AM, Frans Meulenbroeks wrote:
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> Hi all,
>>>>>>>
>>>>>>> Not sure what the best place is to mail this (the YP list or oe-core
>>>>>>> or oe-devel or ...)
>>>>>>> Apologies if this is the wrong place.
>>>>>>>
>>>>>>> The problem I'm facing is the following:
>>>>>>> I have a kernel with some additional drivers that are not upstreamed
>>>>>>> because they are very specifc.
>>>>>>> I'm moving from oe-classic to denzil and from 2.6.38 to 3.2.
>>>>>>> Now some of the patches do not apply any more.
>>>>>>> In oe-classic I would have gone to the ker>
>> Cheers
>>
>> Belen
>>
>>
>> On 08/05/2012 13:23, "Frans Meulenbroeks"<fransmeulenbroeks at gmail.com>
>> wrote:
>>
>>> 2012/5/8 Barros Pena, Belen<belen.barros.pena at intel.com>:
>>>> Hi Frans,
>>>>
>>>> Thank you so much for agreeing to talk to us. Outside working hours
>>>> should
>>>> be no problem. What time would be good for you?
>>>>
>>>>
>>>> If you are ok with it, I'd like to record our conversation: it means I
>>>> don't need to take lots of notes and I can focus on talking to you.
>>>> Using
>>>> Skype makes the recording easier: would you mind if we use that instead
>>>> of
>>>> the phone? My Skype ID is belenbarrospena.
>>>>
>>>> All the best,
>>>>
>>>> Belen
>>>>
>>>>
>>> Hi Belen,
>>>
>>> Recording is no problem. Tonight I'm a little bit low on time, but I
>>> think tomorrow evening I'm fairly flexible (this may still change
>>> though, my life is sometimes somewhat hectic).
>>> 20.30 or so will probably work for me, but if you prefer earlier or
>>> later that is also probably no problem.
>>> And recording is also no problem.
>>>
>>> Best regards, Frans
>>
>> ---------------------------------------------------------------------
>> Intel Corporation (UK) Limited
>> Registered No. 1134945 (England)
>> Registered Office: Pipers Way, Swindon SN3 1RJ
>> VAT No: 860 2173 47
>>
>> This e-mail and any attachments may contain confidential material for
>> the sole use of the intended recipient(s). Any review or distribution
>> by others is strictly prohibited. If you are not the intended
>> recipient, please contact the sender and delete all copies.
>>
>
> nel work dir, fix the issue,
>>>>>>> do a quilt refresh and copy the patch back (there is a patch that
>>>>>>> contains our driver)
>>>>>>> However, denzil uses guilt instead of quilt
>>>>>>>
>>>>>>> If I do a devshell of virtual/kernel and want to do with guilt what I
>>>>>>> used to do with quilt I run into a problem.
>>>>>>> guilt status says: Patches directory does not exist, try guilt-init
>>>>>>> and indeed there is no patches dir like there was in oe-classic
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> Do you have guilt installed locally ? You are likely using the
>>>>>> native guilt. To keep patch lists separate on a per-branch basis
>>>>>> and to be able to capture them on th meta-branch, there are some
>>>>>> patches to guilt that move the patch directory location.
>>>>>
>>>>>
>>>>>
>>>>> Guilt natively complained about a mismatch with the git version (my
>>>>> work system is still ubuntu 10.04).
>>>>> If I do a bitbake -c devshell virtual/kernel and use guilt there, I
>>>>> get the error on the patches directory that I mentioned above.
>>>>> It was my idea that I then would get the oe version in my path but I
>>>>> did not really verify that. Then again under devshell I did not get
>>>>> complaints about the git version. Will verify this tomorrow (I have no
>>>>> access to the build system from here).
>>>>>>
>>>>>>
>>>>>>>
>> Cheers
>>
>> Belen
>>
>>
>> On 08/05/2012 13:23, "Frans Meulenbroeks"<fransmeulenbroeks at gmail.com>
>> wrote:
>>
>>> 2012/5/8 Barros Pena, Belen<belen.barros.pena at intel.com>:
>>>> Hi Frans,
>>>>
>>>> Thank you so much for agreeing to talk to us. Outside working hours
>>>> should
>>>> be no problem. What time would be good for you?
>>>>
>>>>
>>>> If you are ok with it, I'd like to record our conversation: it means I
>>>> don't need to take lots of notes and I can focus on talking to you.
>>>> Using
>>>> Skype makes the recording easier: would you mind if we use that instead
>>>> of
>>>> the phone? My Skype ID is belenbarrospena.
>>>>
>>>> All the best,
>>>>
>>>> Belen
>>>>
>>>>
>>> Hi Belen,
>>>
>>> Recording is no problem. Tonight I'm a little bit low on time, but I
>>> think tomorrow evening I'm fairly flexible (this may still change
>>> though, my life is sometimes somewhat hectic).
>>> 20.30 or so will probably work for me, but if you prefer earlier or
>>> later that is also probably no problem.
>>> And recording is also no problem.
>>>
>>> Best regards, Frans
>>
>> ---------------------------------------------------------------------
>> Intel Corporation (UK) Limited
>> Registered No. 1134945 (England)
>> Registered Office: Pipers Way, Swindon SN3 1RJ
>> VAT No: 860 2173 47
>>
>> This e-mail and any attachments may contain confidential material for
>> the sole use of the intended recipient(s). Any review or distribution
>> by others is strictly prohibited. If you are not the intended
>> recipient, please contact the sender and delete all copies.
>>
>
>
>>>>>> Are you working on top of linux-yocto-3.2 ? Or some other 3.2 tree ?
>>>>>
>>>>>
>>>>>
>>>>> I downloaded the denzil tarball from yoctoproject.com. I made some
>>>>> config changed and added a few local patches (bringing things over
>>>>> from my 2.6.38 oe-classic kernel).
>>>>>
>>>>> Should I have gotten a patches dir in my tmp/work/mymachine/linux* dir
>>>>> (I have my own machine mpc8313e based)
>>>>
>>>>
>>>>
>>>> The patches are pulled underneath the kernel tree itself, so that
>>>> they can be incorporated with other git manipulations of the tree
>>>> and shared as appropriate.
>>>>
>>>> So the patches directory exists in ${S}/meta/patches/$branch_name
>>>>
>>> Ok found that dir. However....
>>> It is still not clear to me how to use this though.
>>>
>>> What I did was start a devshell for virtual/kernel.
>>> Within the devshell guilt is in the path, but still it does not work
>>> (I also tried invoking guilt with the full path).
>>>   From my devshell window:
>>>
>>>
>>> frans at frans-desktop:~/poky-denzil-7.0-build/tmp/work/syrcxx-poky-linux-uclibc/linux-yocto-3.2.11+git1+b14a08f5c7b469a5077c10942f4e1aec171faa9d_1+01e948c2bdf7f5ad9f2b30047a8d3493a1a2880a-r1.23/linux$
>>> which guilt
>>> /home/frans/poky-denzil-7.0-build/tmp/sysroots/x86_64-linux/usr/bin/guilt
>>>
>>> frans at frans-desktop:~/poky-denzil-7.0-build/tmp/work/syrcxx-poky-linux-uclibc/linux-yocto-3.2.11+git1+b14a08f5c7b469a5077c10942f4e1aec171faa9d_1+01e948c2bdf7f5ad9f2b30047a8d3493a1a2880a-r1.23/linux$
>>> /home/frans/poky-denzil-7.0-build/tmp/sysroots/x86_64-linux/usr/bin/guilt
>>> status
>>> Patches directory doesn't exist, try guilt-init
>>>
>>> This is with the standard denzil 7.0 release.
>>>
>>> I've also been to the meta/patches dir and tried guilt status in there
>>> but to no avail.
>>> (or should I have manually made a symlink or so
>>
>>
>> There's nothing extra that should have to be done, I use guilt during patch
>> carry forward and tree generation, it's simply a matter of a few
>> environment variables.
>>
>> That being said, I've largely replaced guilt with mbox manipulations,
>> so I'm a bit concerned about bit rot.
>>
>> Is there any way you can get me access to your layer and patches ?
>> A simple test run is probably the easiest way for me to see what's
>> happened.
>>
>>
>>>
>>> Not sure what I am doing wrong.
>>> Is there a small example or some doc on how to do this? or can someone
>>> give a few commands that would show how to modify an existing patch
>>> (or create a new one) with patch?
>>
>>
>> There are docs, but if guilt next, or guilt push -f aren't working, they
>> won't be of much use.
>>
>>
>>> (btw noticed that the patches are also in
>>> .git/refs/patches/...
>>> however the guilt doc says:
>>> In Guilt, all the patches are stored in .git/patches/$branch/, where
>>> $branch ...
>>
>>
>> That's an unmodified guilt, you can't commit and track the series if you
>> let guilt store information under .git, so I created a variable and
>> relocated it.
>>
>>
>>>
>>> Also tried to add a symlink from .git/patches to .git/refs/patches but
>>> that did not help either.
>>>
>>> Thanks for any suggestion!
>>
>>
>> I can easily give you the raw git commands to make this work, but the guilt
>> path is supposed to be the gentle entry to the process. We need to
>> diagnose what is happening .. and I need to see it happen in front of
>> me, so I can try a few things.
>>
>> Cheers,
>>
>> Bruce
>>
>
> Bruce,
>
> Thanks for your help.
> I'l try to make a small overlay based upon the standard denzil 7.0
> layer and send you the diffs.
> This will take a little bit of time as I want to verify here first
> that this still has the problem.

Sounds good. I'm just concerned that this didn't work right out of the
box.

Bruce

>
> Best regards, Frans