Hi Lukasz, On Wed, Apr 11, 2012 at 03:37:40PM +0200, Lukasz Majewski wrote: > > On Fri, Mar 23, 2012 at 12:48:47PM +0100, Lukasz Majewski wrote: > > > From: Joonyoung Shim <jy0922.shim@xxxxxxxxxxx> > > > > > > The s3c-hsotg driver controls S3C64XX specific registers directly > > > but this driver can be to EXYNOS also. This removes arch specific > > > parts from driver using platform data callbacks. > > > > > > Signed-off-by: Joonyoung Shim <jy0922.shim@xxxxxxxxxxx> > > > Signed-off-by: Kyungmin Park <kyungmin.park@xxxxxxxxxxx> > > > --- > > > drivers/usb/gadget/s3c-hsotg.c | 94 > > > ++++++---------------------------------- 1 files changed, 14 > > > insertions(+), 80 deletions(-) > > > > > > diff --git a/drivers/usb/gadget/s3c-hsotg.c > > > b/drivers/usb/gadget/s3c-hsotg.c index 105b206..0e8e2f4 100644 > > > --- a/drivers/usb/gadget/s3c-hsotg.c > > > +++ b/drivers/usb/gadget/s3c-hsotg.c > > > @@ -35,9 +35,9 @@ > > > > > > #include <plat/regs-usb-hsotg-phy.h> > > > #include <plat/regs-usb-hsotg.h> > > > -#include <mach/regs-sys.h> > > > #include <plat/udc-hs.h> > > > #include <plat/cpu.h> > > > +#include <plat/usb-phy.h> > > > > please put platform_data under <linux/platform_data/> > > But this code is phy related and used by other Samsung's devices (e.g. > exynos4, host controllers, etc.). > > Following files are using it: > > ./arch/arm/mach-exynos/setup-usb-phy.c:20:#include <plat/usb-phy.h> > ./arch/arm/mach-exynos/dev-ohci.c:21:#include <plat/usb-phy.h> > ./arch/arm/mach-s5pv210/setup-usb-phy.c:19:#include <plat/usb-phy.h> > ./arch/arm/mach-s5pv210/setup-usb-phy.c~:20:#include <plat/usb-phy.h> > ./arch/arm/mach-s3c64xx/setup-usb-phy.c:21:#include <plat/usb-phy.h> > ./arch/arm/plat-samsung/devs.c:62:#include <plat/usb-phy.h> > ./drivers/usb/gadget/s3c-hsotg.c.orig:41:#include <plat/usb-phy.h> > ./drivers/usb/gadget/s3c-hsotg.c:42:#include <plat/usb-phy.h> > ./drivers/usb/host/ehci-s5p.c:18:#include <plat/usb-phy.h> > ./drivers/usb/host/ohci-exynos.c:17:#include <plat/usb-phy.h> The idea is just so drivers don't depend on ARCH-specific includes. I really want to make sure all of those drivers compile on all archs so we can better use linux-next. > > > #define DMA_ADDR_INVALID (~((dma_addr_t)0)) > > > > you should move to the generic map/unmap routines and get rid of this > > define. > > > > The DMA mode is still not complete (if it was ever), so I'd like to > leave it as it is. Moreover this driver uses PIO instead of DMA, so > for this fixes patch set it would be better to leave the DMA alone. > > Will you prepare the USB for-next branch, so I could resend the > s3c-hsotg patches? You can use either my gadget branch or my master branch. I pushed it a couple of hours ago, so it should have already synced up on kernel.org -- balbi
Attachment:
signature.asc
Description: Digital signature
