- 20 May, 2011 1 commit
-
-
- 19 May, 2011 29 commits
-
-
Uwe Kleine-König authored
Fixing a few "please, no space before tabs" and "empty line at end of file" warnings on the way. LAKML-Reference: 1299271882-2130-6-git-send-email-u.kleine-koenig@pengutronix.de Signed-off-by:
Uwe Kleine-König <u.kleine-koenig@pengutronix.de> Signed-off-by:
Sascha Hauer <s.hauer@pengutronix.de>
-
Gwenhael Goavec-Merou authored
APF9328 is an i.MXL based SOM (System On Module) that can be plugged on several docking/development boards. Here only basic module support is added (Ethernet, Serial, NOR Flash). Signed-off-by:
Gwenhael Goavec-Merou <gwenhael.goavec-merou@armadeus.com> Signed-off-by:
Eric Jarrige <eric.jarrige@armadeus.org> Signed-off-by:
Nicolas Colombain <nicolas.colombain@armadeus.com> Signed-off-by:
Julien Boibessot <julien.boibessot@armadeus.com> Acked-by:
-
Peter Korsgaard authored
Commit 47babe69 (mxs: dynamically allocate mmc device) added the ssp setup and mmc clocks for mx23/28, but forgot to register the mmc clocks on mx23. Signed-off-by:
Peter Korsgaard <jacmet@sunsite.dk> Acked-by:
Shawn Guo <shawn.guo@freescale.com> Signed-off-by:
-
Wolfram Sang authored
No boards should be compiled in by default. Signed-off-by:
Wolfram Sang <w.sang@pengutronix.de> Signed-off-by:
-
Uwe Kleine-König authored
LAKML-Reference: 1302464943-20721-6-git-send-email-u.kleine-koenig@pengutronix.de Signed-off-by:
-
Uwe Kleine-König authored
To be able to compile e.g. i.MX31 and i.MX51 in a single kernel image the ioremap quirk needs a runtime check. While touching this code make the comment more understandable by adding a sentence from the commit log that introduced it (eadefeff ([ARM] MX3: Use ioremap wrapper to map SoC devices nonshared)). As mach/io.h now uses cpu_is_ some header reshuffling in mach/hardware.h was necessary. (mach/mx27.h and mach/mx31.h #include <linux/io.h> which #includes <mach/io.h>. So mach/mxc.h which provides the cpu_is_ macros needs to be included before mach/mx27.h and mach/mx31.h.) LAKML-Reference: 1302464943-20721-5-git-send-email-u.kleine-koenig@pengutronix.de Signed-off-by:
-
Uwe Kleine-König authored
The two SoCs have different PHYS_OFFSETs so it's not (yet) possible to compile a single (working) kernel for these. LAKML-Reference: 1302464943-20721-4-git-send-email-u.kleine-koenig@pengutronix.de Signed-off-by:
-
Uwe Kleine-König authored
The symbols in this choice should only be used to select between the available machines that can be built into a single kernel. As these sets (will) differ e.g. depending on ARM_PATCH_PHYS_VIRT and AUTO_ZRELADDR letting them select other symbols makes the logic more complex and needs to duplicate some things. So let the machines select the corresponding symbols (indirectly via SOC_XYZ). LAKML-Reference: 1302464943-20721-2-git-send-email-u.kleine-koenig@pengutronix.de Signed-off-by:
-
Uwe Kleine-König authored
Since support for mxc91231 was introduced 2009 it only saw patches that were part of (mxc or arm) global cleanups. The only supported machine only had 4 devices (2x UART, sdhc, watchdog). Cc: Dmitriy Taychenachev <dimichxp@gmail.com> LAKML-Reference: 1302211482-17926-1-git-send-email-u.kleine-koenig@pengutronix.de Signed-off-by:
-
Sascha Hauer authored
remove usage of CONFIG_ARCH_MX1. It's mostly unused anyway, replace it with cpu_is_mx1() where necessary. Also, depend on IMX_HAVE_PLATFORM_IMX_FB instead of the architectures directly. LAKML-Reference: 20110303141244.GQ29521@pengutronix.de Acked-by:
Paul Mundt <lethal@linux-sh.org> Signed-off-by:
-
Fabio Estevam authored
There is no need for using a MX51-specific version of imx_add_gpio_keys. Remove imx51_add_gpio_keys and use imx_add_gpio_keys instead. Signed-off-by:
Fabio Estevam <fabio.estevam@freescale.com> LAKML-Reference: 1302105926-20574-1-git-send-email-fabio.estevam@freescale.com Signed-off-by:
-
Uwe Kleine-König authored
The platform id is used to determine the spi bus number, so it should better be different to the ids used for imx51-ecspi. Otherwise it's not possible to use both devices "imx51-cspi.0" and "imx51-ecspi.0". Alternative approaches are to use dynamic bus numbering as offered by the spi framework or let the machine code set the bus number. The downside of both possibilities is that the bus number isn't fixed for the same busses on different machines using i.MX51. LAKML-Reference: 1302100716-21034-1-git-send-email-u.kleine-koenig@pengutronix.de Signed-off-by:
-
Uwe Kleine-König authored
For consistency mxs has to be repeated, one for the name space and another one for the device name. LAKML-Reference: 1300308028-8922-1-git-send-email-u.kleine-koenig@pengutronix.de Signed-off-by:
-
Wolfram Sang authored
The defines for the i2c related irqs (MX23_INT_I2C_DMA and MX23_INT_I2C_ERROR) already match the reference manual. So make the base address consistent. Signed-off-by:
-
Uwe Kleine-König authored
... together with the related devices "mx3_camera" and "mx3_sdc_fb". "mx3_camera" doesn't fit the scheme of the other devices that just are allocated and registered in a single function because it needs additional care to get some dmaable memory. So currently imx31_alloc_mx3_camera duplicates most of imx_add_platform_device_dmamask, but I'm not sure it's worth to split the latter to be able to reuse more code. This gets rid of mach-mx3/devices.[ch] and so several files need to be adapted not to #include devices.h anymore. LAKML-Reference: 1299271882-2130-5-git-send-email-u.kleine-koenig@pengutronix.de Signed-off-by:
-
Uwe Kleine-König authored
It's not allowed to create an alias of system RAM for DMA. So the memory used must not be allocated using dma_alloc_coherent but has to be reserved before using memblock routines. There is no need to memzero the buffer because dma_alloc_coherent zeros the memory for us. LAKML-Reference: 1299271882-2130-4-git-send-email-u.kleine-koenig@pengutronix.de Tested-by:
Philippe Retornaz <philippe.retornaz@epfl.ch> Acked-by:
-
Uwe Kleine-König authored
There is no need to memzero the buffer because dma_alloc_coherent zeros the memory for us. This fixes: BUG: Your driver calls ioremap() on system memory. This leads <4>to architecturally unpredictable behaviour on ARMv6+, and ioremap() <4>will fail in the next kernel release. Please fix your driver. Tested-by:
Michael Grzeschik <mgr@pengutronix.de> LAKML-Reference: 1299271882-2130-3-git-send-email-u.kleine-koenig@pengutronix.de Signed-off-by:
-
Uwe Kleine-König authored
LAKML-Reference: 1299271882-2130-2-git-send-email-u.kleine-koenig@pengutronix.de Signed-off-by:
-
Fabio Estevam authored
Fix the following warning: CC arch/arm/mach-mx5/board-mx53_loco.o arch/arm/mach-mx5/board-mx53_loco.c:203: warning: initialization discards qualifiers from pointer target type Signed-off-by:
-
Fabio Estevam authored
Signed-off-by:
-
Baruch Siach authored
Signed-off-by:
Baruch Siach <baruch@tkos.co.il> LAKML-Reference: cae1d71db47204ee2654eca7391cb656ed53566b.1300095569.git.baruch@tkos.co.il Signed-off-by:
-
Richard Zhao authored
Signed-off-by:
Richard Zhao <richard.zhao@linaro.org> LAKML-Reference: 1301566583-18947-1-git-send-email-richard.zhao@freescale.com Signed-off-by:
-
Fabio Estevam authored
Current code does not set the GPIO value to zero as mentioned in the comment. Fix it by setting the initial GPIO value to zero. Signed-off-by:
-
Wolfram Sang authored
Reported-by:
Igor Trevisan <igt1972@gmail.com> Signed-off-by:
-
Uwe Kleine-König authored
Most machine files include "devices-imxXX.h" which in turn includes <mach/devices-common.h>. The latter already includes many headers that the machine files don't need to include again. These were found by: $ grep \#include arch/arm/plat-mxc/include/mach/devices-common.h > tmpfile $ git grep -l 'devices-imx' arch/arm | xargs grep -f tmpfile -F (but I kept linux/init.h, linux/kernel.h and linux/platform_device.h) LAKML-Reference: 1298912674-15153-2-git-send-email-u.kleine-koenig@pengutronix.de Signed-off-by:
-
Uwe Kleine-König authored
LAKML-Reference: 1302207841-12450-1-git-send-email-u.kleine-koenig@pengutronix.de Signed-off-by:
-
Shawn Guo authored
Signed-off-by:
-
Fabio Estevam authored
Current code inside babbage_usbhub_reset uses gpio_direction_output with initial value of the GPIO and also sets the GPIO value via gpio_set_value to the same level right after. This is not needed. By using gpio_request_one it is possible to set the direction and initial value in one shot. Signed-off-by:
-
Fabio Estevam authored
Having the silicon revision to appear on the boot log is a useful information. MX31, MX35 and MX51 already show the silicon revision on boot. Add support for displaying such information for MX53 as well. Tested on a mx53loco board, where it shows: CPU identified as i.MX53, silicon rev 2.0 Signed-off-by:
-
- 16 May, 2011 10 commits
-
-
-
Andrew Lunn authored
Signed-off-by:
Andrew Lunn <andrew@lunn.ch> Signed-off-by:
Nicolas Pitre <nico@fluxnic.net>
-
Andrew Lunn authored
Signed-off-by:
-
Andrew Lunn authored
mv78xx0 and kirkwood use identical mpp code. It should also be possible to rewrite the orion5x mpp to use this platform code. Signed-off-by:
-
Andrew Lunn authored
Signed-off-by:
-
Andrew Lunn authored
Signed-off-by:
-
Andrew Lunn authored
Signed-off-by:
-
Andrew Lunn authored
Signed-off-by:
-
Andrew Lunn authored
Signed-off-by:
-
Andrew Lunn authored
This change removes the interrupt resource. The driver does not use it. Signed-off-by:
-
