18 Apr, 2012
22 commits
-
These functions allow the driver core to automatically clean up any
allocations made by drivers, thus leading to simplified drivers.Signed-off-by: Stephen Warren
Signed-off-by: Linus Walleij -
Signed-off-by: Dong Aisheng
Signed-off-by: Linus Walleij -
When we compile pinctrl layer for platforms without CONFIG_PINCONF, we get
following compilation errors:drivers/built-in.o: In function `pinctrl_show':
linux-2.6/drivers/pinctrl/core.c:1116: undefined
reference to `pinconf_show_setting'
drivers/built-in.o: In function `pinctrl_maps_show':
linux-2.6/drivers/pinctrl/core.c:1071: undefined
reference to `pinconf_show_map'
drivers/built-in.o: In function `pinctrl_init_device_debugfs':
linux-2.6/drivers/pinctrl/core.c:1224: undefined
reference to `pinconf_init_device_debugfs'
make[1]: *** [.tmp_vmlinux1] Error 1This patch fixes this.
Signed-off-by: Viresh Kumar
Signed-off-by: Linus Walleij -
Add a pinconf op so that pin controller drivers can decode their pin
config settings for debugfs.Signed-off-by: Stephen Warren
Signed-off-by: Linus Walleij -
Only provide prototypes for pin{mux,conf}.c debugfs-related functions
when both CONFIG_PIN* /and/ CONFIG_DEBUG_FS are enabled, otherwise
provide static inlines.Signed-off-by: Stephen Warren
Signed-off-by: Linus Walleij -
With the finalization of the external driver API and the device
tree support, this subsystem is now mature and can be promoted to
non-experimental status.Acked-by: Stephen Warren
Signed-off-by: Linus Walleij -
Implement pinctrl_ops dt_node_to_map() and dt_free_map(). These allow
complete specification of the desired pinmux configuration using device
tree.Signed-off-by: Stephen Warren
Acked-by: Dong Aisheng
Signed-off-by: Linus Walleij -
Define a new binding for the Tegra pin controller, which is capable of
defining all aspects of desired pin multiplexing and pin configuration.
This is all based on the new common pinctrl bindings.Add Tegra30 binding based on Tegra20 binding.
Add some basic stuff that was missing before:
* How many and what reg property entries must be provided.
* An example.Signed-off-by: Stephen Warren
Acked-by: Dong Aisheng
Signed-off-by: Linus Walleij -
This places the file in the new location for all pin controller bindings.
Also, rename the file using the full compatible value for easier
avoidance of conflicts between multiple bindings.Signed-off-by: Stephen Warren
Signed-off-by: Linus Walleij -
The core pin controller bindings define:
* The fact that pin controllers expose pin configurations as nodes in
device tree.
* That the bindings for those pin configuration nodes is defined by the
individual pin controller drivers.
* A standardized set of properties for client devices to define numbered
or named pin configuration states, each referring to some number of the
afore-mentioned pin configuration nodes.
* That the bindings for the client devices determines the set of numbered
or named states that must exist.Signed-off-by: Stephen Warren
Acked-by: Shawn Guo
Acked-by: Tony Lindgren
Acked-by: Linus Walleij
Acked-by: Simon Glass
Acked-by: Dong Aisheng
Signed-off-by: Linus Walleij -
This patch adds macros of_property_for_each_u32() and
of_property_for_each_string(), which iterate over an array of values
within a device-tree property. Usage is for example:struct property *prop;
const __be32 *p;
u32 u;
of_property_for_each_u32(np, "propname", prop, p, u)
printk("U32 value: %x\n", u);struct property *prop;
const char *s;
of_property_for_each_string(np, "propname", prop, s)
printk("String value: %s\n", s);Based on work by Rob Herring
Cc: Grant Likely
Signed-off-by: Stephen Warren
Acked-by: Rob Herring
Signed-off-by: Linus Walleij -
If drivers try to obtain pinctrl handles for a pin controller that
has not yet registered to the subsystem, we need to be able to
back out and retry with deferred probing. So let's return
-EPROBE_DEFER whenever this location fails. Also downgrade the
errors to info, maybe we will even set them to debug once the
deferred probing is commonplace.Cc: Arnd Bergmann
Reviewed-by: Mark Brown
Acked-by: Stephen Warren
Signed-off-by: Linus Walleij -
This patch can avoid kernel oops in case the mux or config
function is not supported by driver.Acked-by: Stephen Warren
Signed-off-by: Dong Aisheng
Signed-off-by: Linus Walleij -
Do not use get_functions_count before checking.
Acked-by: Stephen Warren
Signed-off-by: Dong Aisheng
Signed-off-by: Linus Walleij -
Most of the SoC drivers implement list_groups() and list_functions()
routines for pinctrl and pinmux. These routines continue returning
zero until the selector argument is greater than total count of
available groups or functions.This patch replaces these list_*() routines with get_*_count()
routines, which returns the number of available selection for SoC
driver. pinctrl layer will use this value to check the range it can
choose.This patch fixes all user drivers for this change. There are other
routines in user drivers, which have checks to check validity of
selector passed to them. It is also no more required and hence
removed.Documentation updated as well.
Acked-by: Stephen Warren
Signed-off-by: Viresh Kumar
[Folded in fix and fixed a minor merge artifact manually]
Signed-off-by: Linus Walleij -
As long as there is no other non-const variable marked __initdata in the
same compilation unit it doesn't hurt. If there were one however
compilation would fail witherror: $variablename causes a section type conflict
because a section containing const variables is marked read only and so
cannot contain non-const variables.Signed-off-by: Uwe Kleine-König
Cc: Randy Dunlap
Signed-off-by: Linus Walleij -
Missed one group from the documentation when proofreading.
Signed-off-by: Viresh Kumar
Signed-off-by: Linus Walleij -
pinctrl/devicetree.c won't compile when !CONFIG_PINCTRL, since the
pinctrl headers don't declare some types when !PINCTRL. Make sure
pinctrl/Makefile only attempts to compile devicetree.c when OF &&
PINCTRL.Acked-by: Dong Aisheng
Signed-off-by: Stephen Warren
Signed-off-by: Linus Walleij -
During pinctrl_get(), if the client device has a device tree node, look
for the common pinctrl properties there. If found, parse the referenced
device tree nodes, with the help of the pinctrl drivers, and generate
mapping table entries from them.During pinctrl_put(), free any results of device tree parsing.
Acked-by: Dong Aisheng
Signed-off-by: Stephen Warren
Signed-off-by: Linus Walleij -
Most code assumes that the pinctrl ops are present. Validate this when
registering a pinctrl driver. Remove the one place in the code that
was checking whether one of these non-optional ops was present.Signed-off-by: Stephen Warren
Signed-off-by: Linus Walleij -
Macros in call ARRAY_SIZE(), the definition of
which eventually calls BUILD_BUG_ON_ZERO(), which is defined in
. Include that so that every .c file using the pinctrl macros
doesn't have to do that itself.Signed-off-by: Stephen Warren
Signed-off-by: Linus Walleij -
The pinctrl_register_mappings is defined in core.c, so change the dependent
macro from CONFIG_MUX to CONFIG_PINCTRL.The compile error message is:
drivers/pinctrl/core.c:886: error: redefinition of 'pinctrl_register_mappings'
include/linux/pinctrl/machine.h:160: note: previous definition of 'pinctrl_register_mappings' was here
make[2]: *** [drivers/pinctrl/core.o] Error 1
make[1]: *** [drivers/pinctrl] Error 2
make: *** [drivers] Error 2Acked-by: Stephen Warren
Signed-off-by: Dong Aisheng
Signed-off-by: Linus Walleij
16 Apr, 2012
11 commits
-
Pull ARM fixes from Russell King:
"Nothing too disasterous, the biggest thing being the removal of the
regulator support for vcore in the AMBA driver; only one SoC was using
this and it got broken during the last merge window, which then
started causing problems for other people. Mutual agreement was
reached for it to be removed."* 'fixes' of git://git.linaro.org/people/rmk/linux-arm:
ARM: 7386/1: jump_label: fixup for rename to static_key
ARM: 7384/1: ThumbEE: Disable userspace TEEHBR access for !CONFIG_ARM_THUMBEE
ARM: 7382/1: mm: truncate memory banks to fit in 4GB space for classic MMU
ARM: 7359/2: smp_twd: Only wait for reprogramming on active cpus
ARM: 7383/1: nommu: populate vectors page from paging_init
ARM: 7381/1: nommu: fix typo in mm/Kconfig
ARM: 7380/1: DT: do not add a zero-sized memory property
ARM: 7379/1: DT: fix atags_to_fdt() second call site
ARM: 7366/3: amba: Remove AMBA level regulator support
ARM: 7377/1: vic: re-read status register before dispatching each IRQ handler
ARM: 7368/1: fault.c: correct how the tsk->[maj|min]_flt gets incremented -
The 'max' range needs to be unsigned, since the size of the user address
space is bigger than 2GB.We know that 'count' is positive in 'long' (that is checked in the
caller), so we will truncate 'max' down to something that fits in a
signed long, but before we actually do that, that comparison needs to be
done in unsigned.Bug introduced in commit 92ae03f2ef99 ("x86: merge 32/64-bit versions of
'strncpy_from_user()' and speed it up"). On x86-64 you can't trigger
this, since the user address space is much smaller than 63 bits, and on
x86-32 it works in practice, since you would seldom hit the strncpy
limits anyway.I had actually tested the corner-cases, I had only tested them on
x86-64. Besides, I had only worried about the case of a pointer *close*
to the end of the address space, rather than really far away from it ;)This also changes the "we hit the user-specified maximum" to return
'res', for the trivial reason that gcc seems to generate better code
that way. 'res' and 'count' are the same in that case, so it really
doesn't matter which one we return.Signed-off-by: Linus Torvalds
-
c5905afb0 ("static keys: Introduce 'struct static_key'...") renamed
struct jump_label_key to struct static_key. Fixup ARM for this to
eliminate these build warnings:include/linux/jump_label.h:113:2:
warning: passing argument 1 of 'arch_static_branch' from incompatible pointer type
include/asm/jump_label.h:17:82:
note: expected 'struct jump_label_key *' but argument is of type 'struct static_key *'Signed-off-by: Rabin Vincent
Signed-off-by: Russell King -
Currently when ThumbEE is not enabled (!CONFIG_ARM_THUMBEE) the ThumbEE
register states are not saved/restored at context switch. The default state
of the ThumbEE Ctrl register (TEECR) allows userspace accesses to the
ThumbEE Base Handler register (TEEHBR). This can cause unexpected behaviour
when people use ThumbEE on !CONFIG_ARM_THUMBEE kernels, as well as allowing
covert communication - eg between userspace tasks running inside chroot
jails.This patch sets up TEECR in order to prevent user-space access to TEEHBR
when !CONFIG_ARM_THUMBEE. In this case, tasks are sent SIGILL if they try to
access TEEHBR.Cc: stable@vger.kernel.org
Reviewed-by: Will Deacon
Signed-off-by: Jonathan Austin
Signed-off-by: Russell King -
If a bank of memory spanning the 4GB boundary is added on a !CONFIG_LPAE
kernel then we will hang early during boot since the memory bank will
have wrapped around to zero.This patch truncates memory banks for !LPAE configurations when the end
address is not representable in 32 bits.Acked-by: Nicolas Pitre
Signed-off-by: Will Deacon
Signed-off-by: Russell King -
During booting of cpu1, there is a short window where cpu1
is online, but not active where cpu1 is occupied by waiting
to become active. If cpu0 then decides to schedule something
on cpu1 and wait for it to complete, before cpu0 has set
cpu1 active, we have a deadlock.Typically it's this CPU frequency transition that happens at
this time, so let's just not wait for it to happen, it will
happen whenever the CPU eventually comes online instead.Cc: Peter Zijlstra
Cc: stable@kernel.org
Signed-off-by: Jonas Aaberg
Reviewed-by: Rickard Andersson
Signed-off-by: Linus Walleij
Signed-off-by: Russell King -
Commit 26f41062f28d ("PCI: check for pci bar restore completion and
retry") attempted to address problems with PCI BAR restoration on
systems where FLR had not been completed before pci_restore_state() was
called, but it did that in an utterly wrong way.First off, instead of retrying the writes for the BAR registers only, it
did that for all of the PCI config space of the device, including the
status register (whose value after the write quite obviously need not be
the same as the written one). Second, it added arbitrary delay to
pci_restore_state() even for systems where the PCI config space
restoration was successful at first attempt. Finally, the mdelay(10) it
added to every iteration of the writing loop was way too much of a delay
for any reasonable device.All of this actually caused resume failures for some devices on Mikko's
system.To fix the regression, make pci_restore_state() only retry the writes
for BAR registers and only wait if the first read from the register
doesn't return the written value. Additionaly, make it wait for 1 ms,
instead of 10 ms, after every failing attempt to write into config
space.Reported-by: Mikko Vinni
Signed-off-by: Rafael J. Wysocki
Signed-off-by: Linus Torvalds -
Pull "ARM: a few more SoC fixes for 3.4-rc" from Olof Johansson:
- A handful of warning and build fixes for Qualcomm MSM
- Build/warning and bug fixes for Samsung Exynos
- A fix from Rob Herring that removes misplaced interrupt-parent
properties from a few device trees
- A fix to OMAP dealing with cpufreq build errors, removing some of the
offending code since it was redundant anyway* tag 'fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc:
ARM: OMAP: clock: cleanup CPUfreq leftovers, fix build errors
ARM: dts: remove blank interrupt-parent properties
ARM: EXYNOS: Fix Kconfig dependencies for device tree enabled machine files
ARM: EXYNOS: Remove broken config values for touchscren for NURI board
ARM: EXYNOS: set fix xusbxti clock for NURI and Universal210 boards
ARM: EXYNOS: fix regulator name for NURI board
ARM: SAMSUNG: make SAMSUNG_PM_DEBUG select DEBUG_LL
ARM: msm: Fix section mismatches in proc_comm.c
video: msm: Fix section mismatches in mddi.c
arm: msm: trout: fix compile failure
arm: msm: halibut: remove unneeded fixup
ARM: EXYNOS: Add PDMA and MDMA physical base address defines
ARM: S5PV210: Fix compiler warning in dma.c file
ARM: EXYNOS: Fix compile error in exynos5250-cpufreq.c
ARM: EXYNOS: Add missing definition for IRQ_I2S0
ARM: S5PV210: fix unused LDO supply field from wm8994_pdata -
Pull another round of sound fixes from Takashi Iwai:
"A few regression fixes for Realtek HD-audio codecs, mainly specific to
some laptop models."* tag 'sound-3.4' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
ALSA: hda/realtek - Fix mem leak (and rid us of trailing whitespace).
ALSA: hda/realtek - Add quirk for Mac Pro 5,1 machines
ALSA: hda/realtek - Add a fixup entry for Acer Aspire 8940G
ALSA: hda/realtek - Fix GPIO1 setup for Acer Aspire 4930 & co
ALSA: hda/realtek - Add a few ALC882 model strings back -
Commit 18a4d0a22ed6 ("[SCSI] Handle disk devices which can not process
medium access commands") introduced a bug in which we would attempt to
dereference the scsi driver even when the device had no ULD attached.Ensure that a driver is registered and make the driver accessor function
more resilient to errors during device discovery.Reported-by: Elric Fu
Reported-by: Bart Van Assche
Signed-off-by: Martin K. Petersen
Signed-off-by: Linus Torvalds
15 Apr, 2012
4 commits
-
…/git/kgene/linux-samsung into fixes
* 'v3.4-samsung-fixes-2' of git://git.kernel.org/pub/scm/linux/kernel/git/kgene/linux-samsung:
ARM: EXYNOS: Fix Kconfig dependencies for device tree enabled machine files
ARM: EXYNOS: Remove broken config values for touchscren for NURI board
ARM: EXYNOS: set fix xusbxti clock for NURI and Universal210 boards
ARM: EXYNOS: fix regulator name for NURI board
ARM: SAMSUNG: make SAMSUNG_PM_DEBUG select DEBUG_LL
ARM: EXYNOS: Add PDMA and MDMA physical base address defines
ARM: S5PV210: Fix compiler warning in dma.c file
ARM: EXYNOS: Fix compile error in exynos5250-cpufreq.c
ARM: EXYNOS: Add missing definition for IRQ_I2S0
ARM: S5PV210: fix unused LDO supply field from wm8994_pdata -
Now that we have OPP layer, and OMAP CPUfreq driver is using it, we no
longer need/use the clock framework code for filling up CPUfreq
tables. Remove it.Removing this code also eliminates build errors when CPU_FREQ_TABLE
support is not enabled.Thanks to Russell King for pointing out the parts I missed under
plat-omap in the original version and also pointing out the build
errors when CPUFREQ_TABLE support was not enabled.Cc: Russell King
Signed-off-by: Kevin Hilman
Acked-by: Paul Walmsley
Acked-by: Tony Lindgren
Signed-off-by: Olof Johansson -
These were incorrectly introduced and can cause problems for of_irq_init.
The correct way to define a root controller is no interrupt-parent set at
all or the interrupt-parent is set to the root controller itself when
inherited from a parent node.Signed-off-by: Rob Herring
Tested-by: Lee Jones
Signed-off-by: Olof Johansson -
From David Brown:
"Here are some fixes for msm that fix problems caused by the latest
ARM code. The ones from Daniel remove unneeded fixups that now
cause compilation failures. Mine fix section mismatches, that were
incompletely fixed earlier."* 'msm-fix' of git://git.kernel.org/pub/scm/linux/kernel/git/davidb/linux-msm:
ARM: msm: Fix section mismatches in proc_comm.c
video: msm: Fix section mismatches in mddi.c
arm: msm: trout: fix compile failure
arm: msm: halibut: remove unneeded fixup
14 Apr, 2012
3 commits
-
Add config dependency for Exynos4 and Exynos5 device tree enabled machine
files on config options ARCH_EXYNOS4 and ARCH_EXYNOS5 respectively.
Enabling machine support without proper ARCH support enabled is incorrect.Signed-off-by: Sachin Kamat
Signed-off-by: Kukjin Kim -
Pull perf tool fixes from Ingo Molnar.
* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf kvm: Finding struct machine fails for PERF_RECORD_MMAP
perf annotate: Validate addr in symbol__inc_addr_samples
perf hists browser: Fix NULL deref in hists browsing code
perf hists: Catch and handle out-of-date hist entry maps.
perf annotate: Fix hist decay
perf top: Add intel_idle to the skip list -
GCC's NULL is actually __null, which allows detecting some questionable
NULL usage and warn about it. Moreover each platform/compiler should
have its own stddef.h anyway (which is different from linux/stddef.h).So there's no good reason to leak kernel's NULL to userspace and
override what the compiler provides.Signed-off-by: Luboš Luňák
Acked-by: Arnd Bergmann
Signed-off-by: Linus Torvalds
