At Mon, 15 Jun 2009 12:22:08 +1000, Benjamin Herrenschmidt wrote:
On Wed, 2009-06-10 at 16:55 +0200, Takashi Iwai wrote:
At Wed, 10 Jun 2009 16:39:01 +0200, Geert Uytterhoeven wrote:
Signed-off-by: Geert Uytterhoeven Geert.Uytterhoeven@sonycom.com Cc: alsa-devel@alsa-project.org Cc: Takashi Iwai tiwai@suse.de
Thanks, applied these three patches (26,27,28) to sound git tree.
Are you sure they don't depend on the previous patches ? We're already having fallover from the SCSI tree merging some of them while I haven't merge the previous ones yet...
These three patches are just trivial changes: coding-style fixes, shuffle of function lines, and addition of __devinit/__devexit notations. So they should be fairly safe and independent (thus I applied them). At least, the test cross-build was OK.
Geert, there are several "procedural" issues with your series here: In addition to the whole thing being submitted too late that is :-)
You don't have the same CC list on all patches. That is fine except that you also don't provide merge instructions after the signature, such as "this patch depends on XX from this series" etc...
A bunch of them do have such dependencies. In fact, there's even a patch that changes something in the block layer (it's not a big deal, just adding an accessor, but I still -need- an Ack from Jens to merge it) and so I have to block all patches that depend on this one. But then, we have indirect dependencies where some patch don't apply because they touch the same file that was patched by the patch who had the dependency etc...
I ended up stopping mid-way through the series. I'll push some of the patches to my -next branch tonight, and the rest once I get the Ack from Jens.
I recommend all maintainers (scsi, alsa) just drop these patches for now and instead give me Ack's so I can merge them via the powerpc tree along with their respective dependencies. James, feel free to do a revert if you don't want to rebase, that will break bisect on ps3 for a little while, but that's the price Geert will have to pay for his mistakes :-)
The alsa part also already reached to the upstream, so we have to give revert patches if needed. But, in the case of sound bits, I think they can remain there as they are independent changes. But, if you think it's better to revert for the whole maintainability, I'll revert in the next pull request that'll be sent soon later.
thanks,
Takashi