Skip to contentSkip to conversation

Final call for 3.8 arm-soc patches

DISCUSSION Arnd Bergmann 6 participants 7 messages Nov 16, 2012 – Nov 18, 2012

lore
Share & tools
lore
#0 lore raw

Hi ARM subarch maintainers,

We are currently at linux-3.7-rc5, which means the merge window is coming closer. We already have 759 changesets in arm-soc and I know of a few ones coming (tegra, clps711x, bcm), but I don't know what the status for some of the bigger ones (exynos/s5p/s3c, mvebu/orion, pxa/mmp). If you have more patches coming, please send all remaining pull requests soon, or let us know what is holding you up. I hope this way we can avoid having to reject pull requests that come too late in the cycle.

Arnd

#1 lore raw

On 11/16/2012 10:11 PM, Arnd Bergmann :

Quoted 9 lines

Hi ARM subarch maintainers,

We are currently at linux-3.7-rc5, which means the merge window is coming closer. We already have 759 changesets in arm-soc and I know of a few ones coming (tegra, clps711x, bcm), but I don't know what the status for some of the bigger ones (exynos/s5p/s3c, mvebu/orion, pxa/mmp). If you have more patches coming, please send all remaining pull requests soon, or let us know what is holding you up. I hope this way we can avoid having to reject pull requests that come too late in the cycle.

Hi Arnd,

For AT91, we have 2 or 3 branches that contain each not much more than a handful of patches. So, nothing big for this release, but I am still waiting for some information to finish them.
The beginning of next week should be ok for finishing the work and send you pull requests.

Best regards,

Signature · 1 line
Nicolas Ferre
#2 lore raw
In reply to Arnd Bergmann
Quoted 9 lines · Arnd Bergmann

Hi ARM subarch maintainers,

We are currently at linux-3.7-rc5, which means the merge window is coming closer. We already have 759 changesets in arm-soc and I know of a few ones coming (tegra, clps711x, bcm), but I don't know what the status for some of the bigger ones (exynos/s5p/s3c, mvebu/orion, pxa/mmp). If you have more patches coming, please send all remaining pull requests soon, or let us know what is holding you up. I hope this way we can avoid having to reject pull requests that come too late in the cycle.

Thanks for the status update, this helps the subarch maintainers too to stop piling up patches so at least I would like to see it every merge cycle.

For omaps, we're pretty much done except for the following multiplatform related changes that I'd like to get out of the way:

1. Move omap iommu/iovmm code to drivers, this is a series of six patches still waiting for an ack from Joerg.

2. Get rid of #include <plat/omap-serial.h>, need to coordinate with Russell's omap-serial.c changes that are now in linux next. This should be a trivial patch after Russell's changes, just need to update my patch to move it to include/linux/platform_data.

3. Move dma-omap.h out of the way for multiplatform. The options here are to add a nasty hack to arch/arm/Makefile to include plat again for omaps, or just move it to the already existing new header we have in include/linux/omap-dma.h. I'm thinking the second option makes more sense until the custom DMA functions are removed.

So that's about eight patches left to before mach-omap2 is multiplatform clean. These patches don't cause any functional changes and I should have those ready to pull by Wednesday next week if that's not too late.

Regards,

Tony

#3 lore raw
Quoted 3 lines · Tony Lindgren

So that's about eight patches left to before mach-omap2 is multiplatform clean. These patches don't cause any functional changes and I should have those ready to pull by Wednesday next week if that's not too late.

Ok, very good. Don't worry about the timing, you have sent a lot of patches early enough that I wouldn't complain about a bunch of others coming after we're closed for everybody else. Same thing for LinusW and Simon who have been great at sending ux500 and shmobile patches early in the cycle.

By contrast, people who haven't sent a single patch yet will have a much harder time arguing why we should take a large number of patches at this point. Of course, bug fixes are completely exempt and can always come.

Arnd

#4 lore raw
In reply to Arnd Bergmann

Arnd,

Quoted 7 lines · Arnd Bergmann

We are currently at linux-3.7-rc5, which means the merge window is coming closer. We already have 759 changesets in arm-soc and I know of a few ones coming (tegra, clps711x, bcm), but I don't know what the status for some of the bigger ones (exynos/s5p/s3c, mvebu/orion, pxa/mmp). If you have more patches coming, please send all remaining pull requests soon, or let us know what is holding you up. I hope this way we can avoid having to reject pull requests that come too late in the cycle.

We still have quite a lot of things to send from the Marvell side, at least:

* A new network driver for Armada 370/XP, which has been ACKed by David Miller

* Clock framework support for Kirkwood, Dove, Armada 370 and Armada XP.

* Support for the Globalscale Mirabox and PlatHome OpenBlocks platforms.

* Rework of the XOR Marvell driver and addition of a Device Tree binding to it.

* Conversion of Kirkwood boards to the pinctrl framework.

* Conversion of Kirkwood boards to the usage of regulators.

* L2 cache support for Armada 370/XP.

Even though quite a bit of this code as been around for some time, Jason Cooper hasn't had the time until now to send the pull requests. I hope Jason will manage to do the corresponding pull requests soon, or I'll have to do them myself because we really don't want to miss the 3.8 merge window for those changes. I'll try to see with Jason either later today or tomorrow what are his plans.

Best regards,

Thomas

Signature · 4 lines
Thomas Petazzoni, Free Electrons
Kernel, drivers, real-time and embedded Linux
development, consulting, training and support.
http://free-electrons.com
#5 lore raw
Quoted 30 lines · Thomas Petazzoni

Arnd,

Quoted 7 lines · Arnd Bergmann

We are currently at linux-3.7-rc5, which means the merge window is coming closer. We already have 759 changesets in arm-soc and I know of a few ones coming (tegra, clps711x, bcm), but I don't know what the status for some of the bigger ones (exynos/s5p/s3c, mvebu/orion, pxa/mmp). If you have more patches coming, please send all remaining pull requests soon, or let us know what is holding you up. I hope this way we can avoid having to reject pull requests that come too late in the cycle.

We still have quite a lot of things to send from the Marvell side, at least:

* A new network driver for Armada 370/XP, which has been ACKed by David Miller

* Clock framework support for Kirkwood, Dove, Armada 370 and Armada XP.

* Support for the Globalscale Mirabox and PlatHome OpenBlocks platforms.

* Rework of the XOR Marvell driver and addition of a Device Tree binding to it.

* Conversion of Kirkwood boards to the pinctrl framework.

* Conversion of Kirkwood boards to the usage of regulators.

Quoted 8 lines

* L2 cache support for Armada 370/XP.

Even though quite a bit of this code as been around for some time, Jason Cooper hasn't had the time until now to send the pull requests. I hope Jason will manage to do the corresponding pull requests soon, or I'll have to do them myself because we really don't want to miss the 3.8 merge window for those changes. I'll try to see with Jason either later today or tomorrow what are his plans.

Yes, a lot of the hold up has been waiting for driver maintainer Acks, cache-l2x0 (rmk's part) to land, etc. My time has also been short lately. I'll pull this together over the next few days.

thx,

Jason.

#6 lore raw
In reply to Arnd Bergmann
Quoted 3 lines · Arnd Bergmann

We are currently at linux-3.7-rc5, which means the merge window is coming closer.

Thanks for your gentle reminder.

Quoted 3 lines

coming (tegra, clps711x, bcm), but I don't know what the status for some of the bigger ones (exynos/s5p/s3c, mvebu/orion, pxa/mmp). If you have more patches coming, please send all remaining pull requests soon, or

Yeah, there are patches for samsung stuff in my tree for 3.8 and I had a plan to send it to arm-soc in the beginning of this week ;-) I will send soon, probably by tomorrow.

Quoted 1 line

let us know what is holding you up.

I have new exynos5440(Quad Cortex-A15) patches and I'm holding it up due to dependency of Samsung common clock stuff is under reviewing in mailing list. I will apply exynos5440 stuff without Samsung common clock so that it can be sent to upstream for v3.8.

Thanks.

Best regards,
Kgene.
--
Kukjin Kim <[email protected]>, Senior Engineer, SW Solution Development Team, Samsung Electronics Co., Ltd.