- 23 Feb, 2010 1 commit
-
-
Ben Dooks authored
Move common headers from plat-s3c's include/plat directory into plat-samsung. No need to fix any files, these are still included via <plat/> Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
- 22 Feb, 2010 2 commits
-
-
Ben Dooks authored
Merge branch 'next-samsung-s3c64xx-platdeletion' into next-samsung Conflicts: arch/arm/mach-s3c64xx/Makefile (fixed)
-
Ben Dooks authored
Now we've move the support out of plat-s3c64xx for everything, eliminate the platform directory arch/arm/plat-s3c64xx and remove it from the ARM build configuration. Note, PLAT_S3C64XX is kept around for the moment until the drivers that depend on it can be updated, so it is moved to the mach-s3c64xx Kconfig. Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
- 21 Feb, 2010 1 commit
-
-
Ben Dooks authored
dma-plat.h is the last file left in plat-s3c64xx, but to remove it we must also change the use of dma-plat.h by the core code and the s3c24xx implementation. Rename the s3c24xx dma-plat.h in the common plat-samsung directory as it may be used for other ports. Move the specific dma bits into the mach-s3c64xx directory and update the build as needed. Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
- 20 Feb, 2010 9 commits
-
-
Ben Dooks authored
Move the pm-core.h to the machine include directory in the process of eliminating the plat-s3c64xx. Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
Ben Dooks authored
Remove the Kconfig and PLAT_S3C64XX defines for the previous S3C64XX directory structure now that the code is moved into mach-s3c64xx. Note, we cannot currently remove plat-s3c64xx directory as we have a pair of include files used within plat-s3c and plat-samsung that need to find a new home. Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
Ben Dooks authored
Turn the init sequence of s3c24xx_register_baseclocks(xtal); s3c64xx_register_clocks(); s3c6400_register_clocks(S3C6410_CLKDIV0_ARM_MASK); into a single call as this is now contained within one file. Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
Ben Dooks authored
Merge plat-s3c64xx/s3c6400-clock.c mach-s3c64xx/clock.c placing all the clock code into one place. Note, no effort is made in this patch to squash the init functions together. Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
Ben Dooks authored
Since this file is small, and is compiled for both systems in this architecture merge it into the cpu support file and remove the original instead of moving it. Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
Ben Dooks authored
Move the core S3C64XX support to mach-s3c64xx as it is unlikely to be used outside of this directory. Also move the SoC header files in with it. This includes the clock, cpu, cpufreq, dma, gpiolib and pll support. Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
Ben Dooks authored
Move the S3C64XX specific device and setup files into mach-s3c64xx as they are unlikely to be used outside of this code. Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
Ben Dooks authored
Move IRQ support to mach-s3c64xx as it is unlikely to be re-used outside this machine. Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
Ben Dooks authored
Move the register and GPIO definition files from plat-s3c64xx into the machine include direcotry as they are unlikely to be reused outside mach-s3c64xx. This move includes removing the empty <mach/regs-clock.h> and replacing it with the <plat/regs-clock.h> implementation. Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
- 03 Feb, 2010 9 commits
-
-
Ben Dooks authored
Merge branch 'next-samsung-s3c244x-merge' into next-samsung
-
Ben Dooks authored
Merge branch 'next-samsung-devupdates2' into next-samsung
-
Maurus Cuelenaere authored
The S3C64XX SoC series support 12bit ADC data, enable this and mask the data accordingly. Signed-off-by: Maurus Cuelenaere <mcuelenaere@gmail.com> Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
Ben Dooks authored
Making the code depend on CONFIG_SND_S3C24XX_SOC means that if the ASoC code is build modularily, the boards will fail to link due to the device code also being built as a module. As per Mark Brown's suggestions, just compile this always as it is expected many machines will have audio. Also move the obj-y line out of the device setup area into a new area just for devices. Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
Ben Dooks authored
As part of the development process, it is hoped PLAT_S3C is either removed from all of the PLAT_SAMSUNG derived platforms or removed entirely. It is also better to use PLAT_SAMSUNG as this is the current base of all Samsung devices. Change the two places that use PLAT_S3C to use PLAT_SAMSUNG Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
Ben Dooks authored
Add SERIAL_SAMSUNG_UARTS_4 to mop up the number of cases currently using four Samsung compatible UARTS. Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
Michel Pollet authored
The S3C2412 has a fractional baud rate register, this patch adds the corresponding flag to the UART definition to allow generation of more precise baud rates for the various clock combinations. Signed-off-by: Michel Pollet <buserror@gmail.com> Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
Jassi Brar authored
The SPI controller platform devices should be compiled in independent of the driver support, otherwise we might end up with dev-spi built as kernel module. Change this to make every machine select if it has some SPI device and wants to build device definitions. Signed-off-by: Jassi Brar <jassi.brar@samsung.com> Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
Ben Dooks authored
Add the required DMA masks to the hs-otg device definition to allow DMA to work with it. Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
- 01 Feb, 2010 7 commits
-
-
Ben Dooks authored
Merge these two headers into one, these two SoCs are so similar. Note, correct fault in mach-smdk2443.h including the wrong header. Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
Ben Dooks authored
Remove the old common S3C2440 and S3C2442 SoC support from plat-s3c24xx into mach-s3c2440 now this directory is serving both SoCs. Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
Ben Dooks authored
Remove the old mach-s3c2442 directory now all is merged into mach-s3c2440 and remove it from the arch/arm build process. Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
Ben Dooks authored
Move the remaining S3C2442 code into mach-s3c2440 as there is only one file currently in there and these two SoCs are very similar. Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
Ben Dooks authored
Fix the entries for SMDK2440 should have allowed for S3C2440 or S3C2440 SoC selection but this depended on ARCH_S3C2440 which has not been around for a while. Remove the dependency to allow this to be selected. Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
Ben Dooks authored
Merge s3c2442.c and clock.c as the s3c242.c does not contain much and the clock parts are always built for s3c2442 anyway. Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
Ben Dooks authored
Move the pair of machines in arch/arm/mach-s3c2442 into the mach-s3c2440 directory as the S3C2440 and S3C2442 are very close. Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
- 29 Jan, 2010 1 commit
-
-
Atul Dahiya authored
Currently, S5P_TIMER_IRQ is based at the end of VICs. This patch changes the S5P_TIMER_IRQ base from end of VICs to 11 in ISA IRQ space. No of VICs varies between SOCs. This causes an exception on S5P6442. Signed-off-by: Atul Dahiya <atul.dahiya@samsung.com> Signed-off-by: Kukjin Kim <kgene.kim@samsung.com> Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
- 27 Jan, 2010 2 commits
-
-
Kukjin Kim authored
1. Moved common memory map definitions for S5P such as S5P_VA_XXX into plat-s5p/include/mach/map-s5p.h from mach-s5p6440/include/mach. 2. Removed unnecessary definitions in the map.h and irq.c 3. Removed the unnecessary support for unaligned UART address 4. Renamed S5P_VA_VICx definitions as VA_VICx 5. Moved the definitons of VIC_BASE to plat-s5p/include/plat/irqs.h Signed-off-by: Kukjin Kim <kgene.kim@samsung.com> Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
Kukjin Kim authored
The s5p6440-clock.c and s5p6440-init.c code in the plat-s5p is specific to s5p6440 SoC based systems. So these files are being moved to the machine directory of s5p6440. And these files are renamed in the machine directory. Signed-off-by: Kukjin Kim <kgene.kim@samsung.com> Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
- 26 Jan, 2010 8 commits
-
-
Ben Dooks authored
Merge branch 'next-smdk6410-defconfig' into next-samsung
-
Ben Dooks authored
Add an error print to s3c24xx_register_clocks to provide more useful information when failing to register the clock. I belive this was originally left out due to the possibility of a problem with low-level debugging code. However, if the low-level debug code is not functional by now there will be a whole other set of problems being presented to the system. Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
Ben Dooks authored
Add some kerneldoc documentation to the s3c24xx_register_clock and the s3c24xx_register_clocks() call. Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
Ben Dooks authored
Squash the SDHCI setup for both the S3C6400 and S3C6410 into one file and make the S3C6410 case use the S3C6400 code. Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
Ben Dooks authored
The iotable in arch/arm/mach-s3c64xx/s3c6410.c is currently empty and therefore can be removed from the build. Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
Ben Dooks authored
As per discussions with Russell King on linux-arm-kernel, it appears that both mach-s3c6400 and mach-s3c6410 are so close together that they should simply be merged into mach-s3c64xx. Note, this patch does not eliminate any of the bits that are still common, it is simply a move of the two directories together, any further common code will be eliminated or moved in further patches. Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
Ben Dooks authored
Functions should be marked __init, not __initdata. Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-
Maurus Cuelenaere authored
ARM: SAMSUNG: Fix null pointer dereference in ADC driver Use struct adc_device instead of relying on client (which could be NULL) when checking for SoC type. Signed-off-by: Maurus Cuelenaere <mcuelenaere@gmail.com> Signed-off-by: Ben Dooks <ben-linux@fluff.org>
-