Simon Horman [Mon, 16 Mar 2015 21:50:29 +0000 (06:50 +0900)]
Merge tag 'renesas-da9063-da9210-quirk-for-v4.1' into next
Renesas ARM Based SoC da9063/da9210 Regulator Quirk for v4.1
The r8a7790/lager and r8a7791/koelsch development boards have da9063 and
da9210 regulators. Both regulators have their interrupt request lines
tied to the same interrupt pin (IRQ2) on the SoC.
After cold boot or da9063-induced restart, both the da9063 and da9210
seem to assert their interrupt request lines. Hence as soon as one
driver requests this irq, it gets stuck in an interrupt storm, as it
only manages to deassert its own interrupt request line, and the other
driver hasn't installed an interrupt handler yet.
To handle this, install a quirk that masks the interrupts in both the
da9063 and da9210. This quirk has to run after the i2c master driver
has been initialized, but before the i2c slave drivers are initialized.
As it depends on i2c, select I2C if one of the affected platforms is
enabled in the kernel config.
Simon Horman [Wed, 11 Mar 2015 23:51:40 +0000 (08:51 +0900)]
Merge tag 'renesas-sh73a0-multiplatform-for-v4.1' into requests/enqueued
Renesas ARM Based SoC sh73a0 Multiplatform Updates for v4.1
* Add multiplatform support to sh73a0 and its kzm9g board
* Use Bus State Controller to enable ethernet for multiplatform sh73a0/kzm9g
* Add PM domain support to multiplatform sh73a0
ARM: shmobile: R-Car Gen2: CONFIG_COMMON_CLK is always set
Since commit e042681894b62d60 ("ARM: shmobile: r8a7790: Remove legacy
code"), all R-Car Gen2 SoCs are supported by multiplatform kernels only.
As CONFIG_COMMON_CLK is always set for multiplatform kernels, we can
remove related #ifdefs in code specific to R-Car Gen2 SoCs.
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be> Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
The r8a7790/lager and r8a7791/koelsch development boards have da9063 and
da9210 regulators. Both regulators have their interrupt request lines
tied to the same interrupt pin (IRQ2) on the SoC.
After cold boot or da9063-induced restart, both the da9063 and da9210
seem to assert their interrupt request lines. Hence as soon as one
driver requests this irq, it gets stuck in an interrupt storm, as it
only manages to deassert its own interrupt request line, and the other
driver hasn't installed an interrupt handler yet.
To handle this, install a quirk that masks the interrupts in both the
da9063 and da9210. This quirk has to run after the i2c master driver
has been initialized, but before the i2c slave drivers are initialized.
As it depends on i2c, select I2C if one of the affected platforms is
enabled in the kernel config.
On koelsch, the following happens:
- Cold boot or reboot using the da9063 restart handler:
rcar_gen2_regulator_quirk: IRQ2 is not asserted, not installing quirk
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be> Tested-by: Wolfram Sang <wsa+renesas@sang-engineering.com> Reviewed-by: Mark Brown <broonie@kernel.org> Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
ARM: shmobile: ape6evm dts: Enable pull-up for GPIO switches
The switches on r8a73a4/ape6evm do not have pull-up registers. The
schematics say: "Need to use APE6 internal PullUp", hence enable pull-up
using pinctrl.
Without this, the switches don't really work, as the GPIO inputs are
more likely to pick up ghost signals through capacitive coupling than
actual keypresses.
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be> Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
Simon Horman [Thu, 5 Mar 2015 00:41:29 +0000 (09:41 +0900)]
Merge tag 'renesas-dt-for-v4.1' into requests/merged
Renesas ARM Based SoC DT Updates for v4.1
* Increase hardware coverage of DT for various SoCs
* Add PFC information for emev2 SoC
* Remap entire APMU region for r8a7791 and r8a7790 SoCs
* Declare the full 512 MiB of RAM for kzm9g board
* Add selectable sources to DIV6 clocks to sh73a0 SoC
* Add missing INTCA0 clock for irqpin module on sh73a0 SoC
* Set control-parent for all irqpin node on sh73a0 SoC
ARM: shmobile: r8a73a4 dtsi: Add PM domain support
Add a device node for the System Controller, with subnodes that
represent the hardware power area hierarchy.
Hook up all devices to their respective PM domains.
Add a minimal device node for the Coresight-ETM hardware block, and
hook it up to the D4 PM domain, so the R-Mobile System Controller
driver can keep the domain powered, until the new Coresight code
handles runtime PM.
The System Controller is also used by the R-Mobile Reset driver, which
can now restart the system.
Add a device node for the System Controller, with subnodes that
represent the hardware power area hierarchy.
Hook up all devices to their respective PM domains.
Note that unlike on R-Mobile A1 (r8a7740), PM domain D4 can be powered
down without ill effects on s2ram behavior, just like on SH-Mobile AP4
(sh7372). Hence we can postpone adding a (minimal) device node for the
Coresight-ETM hardware block.
The System Controller is also used by the R-Mobile Reset driver, which
can now restart the system.
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be> Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
"ARM: shmobile: kzm9g-reference: Remove board C code and DT file"
removed the last user of sh73a0_add_standard_devices_dt(). Hence remove
this function.
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be> Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
Add a node for the Private Timer and Watchdog, as found in the Cortex-A9
MPCore.
Without this, there's no clocksource available during early kernel
initialization, before cmt1 is initialized, leading to a lock-up if
CONFIG_CPU_IDLE=y.
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be> Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
Simon Horman [Mon, 23 Feb 2015 22:10:05 +0000 (07:10 +0900)]
ARM: shmobile: kzm9g-reference: Remove board C code and DT file
Now that the sh73a0 generic multiplatform case has the same feature set
as the kzm9g DT reference board code, we get rid of the latter.
DT reference code in the future shall make use of the sh73a0
multiplatform support code with the generic SoC machine vector.
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be> Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
ARM: shmobile: kzm9g dts: Move Ethernet node to BSC
Move the Ethernet node from the root of the device tree to the BSC node,
as it's connected to the Bus State Controller. This allows the system
to know the right position of the Ethernet node in the clock and PM
domain hierarchy, and manage the clock and PM domain appropriately.
Also rename the node's name from "lan9220" to "ethernet", to conform to
ePAPR generic name recomendations.
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be> Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
Magnus Damm [Fri, 9 Jan 2015 13:08:51 +0000 (14:08 +0100)]
ARM: shmobile: kzm9g dts: Sync with kzm9g-reference dts
Sync the two DTS for the KZM9G board. The target is the file
"sh73a0-kzm9g.dts" and it is made identical to the DT reference
case with the exception of the compatbile string. In the future
the DT reference file will go away.
Signed-off-by: Magnus Damm <damm+renesas@opensource.se>
[geert: Update for recent changes to sh73a0-kzm9g-reference.dts] Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be> Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
Magnus Damm [Fri, 9 Jan 2015 13:08:50 +0000 (14:08 +0100)]
ARM: shmobile: sh73a0: Add Multiplatform support
Enable sh73a0 Multiplatform support for the generic sh73a0
machine vector. No board support is enabled, and the board
code for KZM9G DT Reference is left by itself.
Signed-off-by: Magnus Damm <damm+renesas@opensource.se> Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be> Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
Add a generic sh73a0 machine setup callback for the upcoming
sh73a0 multiplatform case. Cache needs to be configured, and
legacy clocks must be omitted.
Signed-off-by: Magnus Damm <damm+renesas@opensource.se> Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be> Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
drivers: sh: Disable PM runtime for multi-platform r8a7740 with genpd
If the default PM domain using PM_CLK is used for PM runtime, the real PM
domain(s) cannot be registered from DT later.
Hence do not enable it when running a multi-platform kernel with genpd
support on an r8a7740. The R-Mobile PM domain driver will take care of
PM runtime management of the module clocks.
The default PM domain is still needed for:
- platforms without genpd support,
- the legacy (non-DT) case, where genpd may take over later, except
for the C5 "always on" PM domain.
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be> Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
Magnus Damm [Wed, 21 Jan 2015 05:01:37 +0000 (14:01 +0900)]
ARM: shmobile: No R-Car Gen2 CMA reservation when HIGHMEM=n
Allow R-Car Gen2 platforms to boot with CMA enabled
and HIGHMEM disabled. This patch adds code to check
if the R-Car Gen2 specific memory reservation window
is included in the kernel memory range or not. When
HIGHMEM is disabled the R-Car Gen2 reservation area is
outside the kernel memory range and in such case the
memory reservation is simply skipped over.
Without this patch the kernel boot hangs when CMA is
enabled and HIGHMEM is disabled on the r8a7791 Koelsch
hardware platform:
WARNING: CPU: 1 PID: 1 at mm/cma.c:113 cma_init_reserved_areas+0x88/0x1d4()
...
WARNING: CPU: 1 PID: 1 at mm/cma.c:121 cma_init_reserved_areas+0xf8/0x1d4()
...
Unable to handle kernel NULL pointer dereference at virtual address 00000160
pgd = c0003000
[00000160] *pgd=80000040004003, *pmd=00000000
Internal error: Oops: 206 [#1] SMP ARM
Modules linked in:
CPU: 1 PID: 1 Comm: swapper/0 Tainted: G W 3.19.0-rc4-koelsch-01450-g7f9b6075ce12c3ea-dirty #735
Hardware name: Generic R8A7791 (Flattened Device Tree)
task: edc553c0 ti: edc56000 task.ti: edc56000
PC is at set_pfnblock_flags_mask+0x54/0xa0
LR is at 0x440
In the current shmobile_defconfig HIGHMEM is enabled
while CMA is disabled, so to trigger this the kernel
configuration for both CMA and HIGHMEM needs to be
adjusted.
Signed-off-by: Magnus Damm <damm+renesas@opensource.se> Tested-by: Geert Uytterhoeven <geert+renesas@glider.be> Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
By default only the non-boot CPUs can be hotplugged if the
smp_operations structure doesn't have the cpu_disable function
pointer set. r8a7779_cpu_disable() implements the same logic,
only non-boot CPUs can be hotplugged, so it's redundant to
override the default behavior.
Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Acked-by: Magnus Damm <damm+renesas@opensource.se> Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
Magnus Damm [Tue, 20 Jan 2015 11:58:10 +0000 (20:58 +0900)]
ARM: shmobile: Remove mach/clkdev.h
Move over ARCH_SHMOBILE_LEGACY to use the default legacy ARM
implementations of __clk_get() and __clk_put() in <asm/clkdev.h>
by deselecting HAVE_MACH_CLKDEV. This has the nice side effect
that <mach/clkdev.h> is no longer used and can be removed.
Signed-off-by: Magnus Damm <damm+renesas@opensource.se> Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com> Acked-by: Geert Uytterhoeven <geert+renesas@glider.be> Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
Magnus Damm [Mon, 26 Jan 2015 06:18:08 +0000 (15:18 +0900)]
ARM: shmobile: mackerel: Remove Legacy C board code
Remove legacy C code for the sh7372 Mackerel board. There is no
DT multiplatform implementation available for the sh7372 SoC so
simply phase out the board and SoC code support.
Signed-off-by: Magnus Damm <damm+renesas@opensource.se> Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com> Acked-by: Geert Uytterhoeven <geert+renesas@glider.be> Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
Magnus Damm [Mon, 26 Jan 2015 06:17:58 +0000 (15:17 +0900)]
ARM: shmobile: mackerel: Remove ZBOOT code
Remove ZBOOT support code for Mackerel. With this patch in place
it is no longer possible to boot a self-contained kernel directly
from the reset vector. ZBOOT is still supported on kzm9g.
Signed-off-by: Magnus Damm <damm+renesas@opensource.se> Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com> Acked-by: Geert Uytterhoeven <geert+renesas@glider.be> Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
ARM: shmobile: ape6evm dts: Drop console= bootargs parameter
Since ("ARM: shmobile: r8a73a4: ape6evm: Remove legacy platform"),
ape6evm is restricted to booting from DT, so chosen/stdout-path is
always used, and we can drop the "console=" parameter from
chosen/bootargs.
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be> Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
Ulrich Hecht [Tue, 17 Feb 2015 15:04:44 +0000 (16:04 +0100)]
ARM: shmobile: ape6evm-reference: Remove board C code and DT file
Now that the r8a73a4 generic multiplatform case has the same features as the
APE6EVM DT reference board code, we get rid of the latter. DT reference
code in the future shall make use of the r8a73a4 multiplatform support code
with the generic SoC machine vector.
ARM: shmobile: r8a73a4: Move pfc node to work around probe ordering bug
Currently the pin function controller (which is also a GPIO controller)
is instantiated before the interrupt controllers due to the order in the
DTS. At that time, the irq domains for the interrupt controllers
referenced by its interrupts-extended property cannot be found yet:
irq: no irq domain found for /interrupt-controller@e61c0000 !
Nevertheless, the core OF probing code ignores this failure, besides a
debug message that's not normally printed:
not all legacy IRQ resources mapped for pfc
and continues initialization of the device. Then, the sh-pfc driver
cannot find any IRQ resources, and thinks no interrupts are available,
causing gpio-keys to fail later:
gpio-keys keyboard: Unable to claim irq 0; error -22
gpio-keys: probe of keyboard failed with error -22
Move the pin function controller node after the interrupt controller
nodes it references to work around the bug in the core OF probing code.
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be> Signed-off-by: Simon Horman <horms+renesas@verge.net.au>