Re: [GIT PULL] arch/microblaze fixes for 2.6.34-rc2

Re: [GIT PULL] arch/microblaze fixes for 2.6.34-rc2

[Grant Likely]

On Thu, Mar 11, 2010 at 6:42 AM, Michal Simek <monstr-pSz03upnqPeHXe+LvDLADg@public.gmane.org> wrote:
> Hi Linus,
>
> Please pull Microblaze changes to your tree. There is support for PCI and
> new DMA interface.
>
> Thanks,
> Michal

Hi Michal.

I'm really unimpressed that you've merged the cloned PCI support code
when we talked about merging the common powerpc & microblaze bits
/before/ pushing it up to mainline.  At the very least I would have
expected a reply back from you telling me that you disagree and that
you wanted to merge anyway.  I didn't even get a reply back form you
when I asked about why the PCI bits were in your linux-next branch.

Now it needs to be merged in-place which is a lot more work.

g.

Re: [GIT PULL] arch/microblaze fixes for 2.6.34-rc2

[John Williams]

Grant,

On Sun, Mar 21, 2010 at 3:35 AM, Grant Likely <grant.likely-s3s/WqlpOiPyB63q8FvJNQ@public.gmane.org> wrote:
> On Thu, Mar 11, 2010 at 6:42 AM, Michal Simek <monstr-pSz03upnqPeHXe+LvDLADg@public.gmane.org> wrote:
>> Hi Linus,
>>
>> Please pull Microblaze changes to your tree. There is support for PCI and
>> new DMA interface.
>>
>> Thanks,
>> Michal
>
> Hi Michal.
>
> I'm really unimpressed that you've merged the cloned PCI support code
> when we talked about merging the common powerpc & microblaze bits
> /before/ pushing it up to mainline.  At the very least I would have
> expected a reply back from you telling me that you disagree and that
> you wanted to merge anyway.  I didn't even get a reply back form you
> when I asked about why the PCI bits were in your linux-next branch.
>
> Now it needs to be merged in-place which is a lot more work.

You should blame me for this one.  There was nothing deliberate about
it - I've been hassling Michal to get the DMA API upstream because
there's other work that depends on it, and with everything else going
on I plain forgot that his patchset included the PCI stuff that we
were still discussing with you.

Let me buy you a beer at ELC and you can berate me for it then!

John

Re: [GIT PULL] arch/microblaze fixes for 2.6.34-rc2

[Grant Likely]

On Sun, Mar 21, 2010 at 4:32 PM, John Williams
<john.williams-g5w7nrANp4BDPfheJLI6IQ@public.gmane.org> wrote:
> Grant,
>
> On Sun, Mar 21, 2010 at 3:35 AM, Grant Likely <grant.likely-s3s/WqlpOiPyB63q8FvJNQ@public.gmane.org> wrote:
>> On Thu, Mar 11, 2010 at 6:42 AM, Michal Simek <monstr-pSz03upnqPeHXe+LvDLADg@public.gmane.org> wrote:
>>> Hi Linus,
>>>
>>> Please pull Microblaze changes to your tree. There is support for PCI and
>>> new DMA interface.
>>>
>>> Thanks,
>>> Michal
>>
>> Hi Michal.
>>
>> I'm really unimpressed that you've merged the cloned PCI support code
>> when we talked about merging the common powerpc & microblaze bits
>> /before/ pushing it up to mainline.  At the very least I would have
>> expected a reply back from you telling me that you disagree and that
>> you wanted to merge anyway.  I didn't even get a reply back form you
>> when I asked about why the PCI bits were in your linux-next branch.
>>
>> Now it needs to be merged in-place which is a lot more work.
>
> You should blame me for this one.  There was nothing deliberate about
> it - I've been hassling Michal to get the DMA API upstream because
> there's other work that depends on it, and with everything else going
> on I plain forgot that his patchset included the PCI stuff that we
> were still discussing with you.

Blame dutifully assigned.

The next question is how to proceed.  I've not gotten to PCI code in
my rework yet, but I probably will in the next month or so.  If you or
Michal doesn't get to it first, then I'll probably move the OF
portions of the powerpc version into drivers/of and make them
conditional on !CONFIG_MICROBLAZE.  Microblaze adaptations can be
merged in at a later date and the !CONFIG_MICROBLAZE conditional
eventually dropped.

> Let me buy you a beer at ELC and you can berate me for it then!

done.

-- 
Grant Likely, B.Sc., P.Eng.
Secret Lab Technologies Ltd.