[PATCH 2/16] SPI: Refactor spi-orion to use SPI framework queue.
From: andrew@lunn.ch (Andrew Lunn)
Date: 2012-07-22 09:53:28
On Sun, Jul 22, 2012 at 09:22:09AM +0200, Linus Walleij wrote:
On Fri, Jul 20, 2012 at 7:35 PM, Andrew Lunn [off-list ref] wrote:quoted
Replace the deprecated master->transfer with transfer_one_message() and allow the SPI subsystem handle all the queuing of messages. Signed-off-by: Andrew Lunn <andrew@lunn.ch> Acked-by: Linus Walleij <redacted> Acked-by: Sebastian Hesselbarth <redacted>Send these with Mark Brown on the To: line right now, I think he's busy collecting SPI patches for the next merge window.
Hi Linus, Mark and now Arnd. I was planning sending all these via arm-soc. This patch is part of a bigger collection of Orion changes. The rest are contributions from others which have already been ack-by, etc, on the lists. Somewhere near the beginning of this series is: 4385fbf SPI: Refactor spi-orion to use SPI framework queue. d9d8293 spi-orion: remove uneeded spi_info ee13b49 spi-orion: add device tree binding 76f0b84 ARM: kirkwood: use devicetree for SPI on dreamplug 05f504f ARM: kirkwood: use devicetree for orion-spi You can see the chain of dependencies here. So at minimum i would need to send the first three via Mark. I then need to cross my fingers and hope he really picks them up, because without them, we are going to have a number of boards which fail to find their root file system on an SPI flash. Now, maybe i've done all this wrong, since i'm a newbie to collecting patches together for submission upstream.... You can see the complete set of patches here: git://github.com/lunn/linux.git v3.5-rc7-for-next-v5 There is one exception and that is the very first patch: I2C: MV64XXX: Add Device Tree support which Wolfram Sang insists goes via him. I see this less critical. If for some reason it does not make it in, all it means is temperature sensors and RTCs are missing, but at least the boards still boot. Andrew