linux.git
7 years agoMLK-15007-3 arm64: defconfig: enable SMMUv2
Peng Fan [Mon, 26 Jun 2017 01:29:38 +0000 (09:29 +0800)]
MLK-15007-3 arm64: defconfig: enable SMMUv2

Enable SMMUv2 in defconfig.

Signed-off-by: Peng Fan <peng.fan@nxp.com>
7 years agoMLK-15007-2 arm64: dts: imx8qm: add smmu node
Peng Fan [Mon, 26 Jun 2017 01:25:02 +0000 (09:25 +0800)]
MLK-15007-2 arm64: dts: imx8qm: add smmu node

Add smmu node for i.MX8QM.

Signed-off-by: Peng Fan <peng.fan@nxp.com>
7 years agoMLK-15007-1 iommu: arm: pgtable: alloc pagetable in DMA area
Peng Fan [Wed, 28 Jun 2017 02:46:03 +0000 (10:46 +0800)]
MLK-15007-1 iommu: arm: pgtable: alloc pagetable in DMA area

Normally the iommu pagetable could be in 64bit address space,
but we have one patch to address PCIE driver, 'commit 9e03e5076269
("MLK-15064-2 ARM64: DMA: limit the dma mask to be 32bit")'

The patch restrict swiotlb and iommu dma to be in 32bit address.

So if we allocate pages in highmem, then dma_map_single will return
a 32bit address. Then, we will get "Cannot accommodate DMA
translation for IOMMU page tables", because `dma != virt_to_phys(pages)`.

So we strict the lpae iommu pgtable in DMA area to fix this issue.

Signed-off-by: Peng Fan <peng.fan@nxp.com>
7 years agoiommu/arm-smmu: Set SMTNMB_TLBEN in ACR to enable caching of bypass entries
Nipun Gupta [Fri, 4 Nov 2016 09:55:23 +0000 (15:25 +0530)]
iommu/arm-smmu: Set SMTNMB_TLBEN in ACR to enable caching of bypass entries

The SMTNMB_TLBEN in the Auxiliary Configuration Register (ACR) provides an
option to enable the updation of TLB in case of bypass transactions due to
no stream match in the stream match table. This reduces the latencies of
the subsequent transactions with the same stream-id which bypasses the SMMU.
This provides a significant performance benefit for certain networking
workloads.

With this change substantial performance improvement of ~9% is observed with
DPDK l3fwd application (http://dpdk.org/doc/guides/sample_app_ug/l3_forward.html)
on NXP's LS2088a platform.

Reviewed-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Nipun Gupta <nipun.gupta@nxp.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
(cherry picked from commit 6eb18d4a2b860ad259763c5e6d632839dcf974a1)

7 years agoMLK-15284-6: arm64: defconfig: enable UBIFS in defconfig
Han Xu [Mon, 26 Jun 2017 20:45:42 +0000 (15:45 -0500)]
MLK-15284-6: arm64: defconfig: enable UBIFS in defconfig

enable ubifs in arm64 defconfig

Signed-off-by: Han Xu <han.xu@nxp.com>
7 years agoMLK-15284-5: arm64: defconfig: add GPMI and MXS_DMA in default config
Han Xu [Fri, 23 Jun 2017 21:30:52 +0000 (16:30 -0500)]
MLK-15284-5: arm64: defconfig: add GPMI and MXS_DMA in default config

add GPMI_NAND and MXS_DMA in ARM64 default config

Signed-off-by: Han Xu <han.xu@nxp.com>
7 years agoMLK-15284-4: dma: Kconfig: add MXS_DMA dependency for ARM64
Han Xu [Fri, 23 Jun 2017 21:28:10 +0000 (16:28 -0500)]
MLK-15284-4: dma: Kconfig: add MXS_DMA dependency for ARM64

add MXS_DMA dependency for ARCH_MXC_ARM64

The patch also merge the upstreamed change that extend the dependency to
allow the mxs dma driver to be built whenever ARCH_MXS or ARCH_MXC is
selected. Refer to
https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/drivers/dma/Kconfig?h=next-20170626&id=d762e4f35601239cbebfbfd43d99876d8f220927

Signed-off-by: Han Xu <han.xu@nxp.com>
7 years agoMLK-15284-3: mtd: nand: gpmi-nand: support NAND on i.MX8QXP
Han Xu [Tue, 20 Jun 2017 21:24:42 +0000 (16:24 -0500)]
MLK-15284-3: mtd: nand: gpmi-nand: support NAND on i.MX8QXP

Enable the NAND support on i.MX8QXP

Signed-off-by: Han Xu <han.xu@nxp.com>
7 years agoMLK-15284-2: dma: mxs-dma: add i.MX8QXP support in mxs-dma
Han Xu [Fri, 2 Jun 2017 21:52:56 +0000 (16:52 -0500)]
MLK-15284-2: dma: mxs-dma: add i.MX8QXP support in mxs-dma

add one more entry for i.MX8QXP mxs-dma, also check mxs_dma in filter
function.

Signed-off-by: Han Xu <han.xu@nxp.com>
7 years agoMLK-15284-1: arm64: dts: enable the GPMI NAND module in device tree
Han Xu [Fri, 2 Jun 2017 21:47:54 +0000 (16:47 -0500)]
MLK-15284-1: arm64: dts: enable the GPMI NAND module in device tree

enable the GPMI NAND module in device tree for i.MX8QXP

Signed-off-by: Han Xu <han.xu@nxp.com>
7 years agoarm64: defconfig: Fix config option order
Leonard Crestez [Tue, 27 Jun 2017 12:31:12 +0000 (15:31 +0300)]
arm64: defconfig: Fix config option order

Continuous integration complains if savedefconfig shows differences,
even if it's just because of reordering config options.

Fixes: 42929061c62b ("MGS-3025: ARM64: dts: freescale: imx8mq: enable viv drm")
Signed-off-by: Leonard Crestez <leonard.crestez@nxp.com>
Reviewed-by: Octavian Purdila <octavian.purdila@nxp.com>
7 years agoMLK-15293 arm64: dts: imx8mq: add imx i2c support
Gao Pan [Tue, 27 Jun 2017 06:44:11 +0000 (14:44 +0800)]
MLK-15293 arm64: dts: imx8mq: add imx i2c support

add imx i2c support for imx8mq in dts files.

Signed-off-by: Gao Pan <pandy.gao@nxp.com>
7 years agoMLK-15145 ARM64: dts: Correct the system counter freqency for i.mx8mq
Bai Ping [Fri, 23 Jun 2017 02:37:17 +0000 (10:37 +0800)]
MLK-15145 ARM64: dts: Correct the system counter freqency for i.mx8mq

On i.MX8MQ, the system counter's clock frequency should be 8333333Hz.
We use 25MHz OSC, so the freqency is 25/3.

Signed-off-by: Bai Ping <ping.bai@nxp.com>
7 years agoMLK-13855-3: perf: ddr-perf: Add counter overflow handling
Tiberiu Breana [Wed, 14 Jun 2017 14:50:11 +0000 (17:50 +0300)]
MLK-13855-3: perf: ddr-perf: Add counter overflow handling

Added support for counter overflow interrupts.
When the cycles counter overflows, update all local event data,
then reset it and let it continue counting.

Signed-off-by: Tiberiu Breana <andrei-tiberiu.breana@nxp.com>
7 years agoMLK-13855-2: perf: ddr-perf: Always enable the cycles counter
Tiberiu Breana [Mon, 19 Jun 2017 08:21:52 +0000 (11:21 +0300)]
MLK-13855-2: perf: ddr-perf: Always enable the cycles counter

Always enable cycles counter 0, regardless if it is explicitly
selected by the user or not. The cycles counter generates overflow
interrupts that will be used to update other counters.

Signed-off-by: Tiberiu Breana <andrei-tiberiu.breana@nxp.com>
7 years agoMLK-13855-1: perf: ddr-perf: Clean up driver
Tiberiu Breana [Wed, 14 Jun 2017 14:45:47 +0000 (17:45 +0300)]
MLK-13855-1: perf: ddr-perf: Clean up driver

- repurpose e2c_map array to a perf_event* array
- add ddr_perf_event_enable function
- tidy up indenting

Signed-off-by: Tiberiu Breana <andrei-tiberiu.breana@nxp.com>
7 years agoMLK-14865: ARM: dts: imx6sx-sdb: Change audio PLL frequency for SSI
Daniel Baluta [Mon, 26 Jun 2017 13:49:19 +0000 (16:49 +0300)]
MLK-14865: ARM: dts: imx6sx-sdb: Change audio PLL frequency for SSI

Default frequency is 786432000 and we cannot derive an exact bitclk
for 24 bits tests.

This is similar with commit 9e3c04a3e9222a ("MLK-14781-2: ARM: dts: change
audio pll frequency for ssi master mode").

Signed-off-by: Daniel Baluta <daniel.baluta@nxp.com>
Reviewed-by: Shengjiu Wang <shengjiu.wang@nxp.com>
7 years agoMLK-15132-6 : Enable Hantro decoder on i.MX8MQ
Zhou Peng-B04994 [Tue, 27 Jun 2017 02:40:04 +0000 (10:40 +0800)]
MLK-15132-6 : Enable Hantro decoder on i.MX8MQ

Enable vpu on imx8mq-evk dts file

Signed-off-by: Zhou Peng-B04994 <eagle.zhou@nxp.com>
7 years agoMLK-15287 clk: imx: imx8mq: increase NOC clock speed
Anson Huang [Tue, 27 Jun 2017 02:03:44 +0000 (10:03 +0800)]
MLK-15287 clk: imx: imx8mq: increase NOC clock speed

NOC clock by default is running @400MHz, to achieve
best DDR access performance, increase it to 800MHz.

With CPU @1.2GHz, we can see stream copy performance
increase 24% if NOC clock is increased from 400MHz
to 800MHz.

Signed-off-by: Anson Huang <Anson.Huang@nxp.com>
7 years agoMGS-3024 dts: update i.MX8MQ device config to enable GPU
Chenyan Feng [Sat, 24 Jun 2017 08:03:44 +0000 (16:03 +0800)]
MGS-3024 dts: update i.MX8MQ device config to enable GPU

enable gpu device in imx8mq-evk board,
increase GPU memory size from 32M to 128M,
enable GPU flat mappping for full DDR range.

add IMX8MQ_CLK_GPU_AHB_DIV into gpu clock list.

Date: 26th June, 2017
Signed-of-by: Chenyan Feng <ella.feng@nxp.com>
7 years agoMGS-3023 [#imx-623] add ahb clk operation for mscale gpu
Chenyan Feng [Sat, 24 Jun 2017 07:25:40 +0000 (15:25 +0800)]
MGS-3023 [#imx-623] add ahb clk operation for mscale gpu

mscale gpu ahb clock has the different source from axi clock,
need add the separate clock operation for gpu ahb in driver.

Date: 24th June, 2017
Signed-off-by: Chenyan Feng <ella.feng@nxp.com>
7 years agoMGS-3021 [#imx-622] Disable gpu security feature for mscale
Chenyan Feng [Fri, 23 Jun 2017 09:55:39 +0000 (17:55 +0800)]
MGS-3021 [#imx-622] Disable gpu security feature for mscale

GPU security feature is only for mscale, but driver is not ready,
need disable this feature to avoid gpu kernel panic temporally,

will drop this patch when gpu security driver is ready later.

Date: 23th June, 2017
Signed-off-by: Chenyan Feng <ella.feng@nxp.com>
7 years agoMLK-15285-2 soc: imx: add HAVE_IMX_SOC for soc driver
Anson Huang [Mon, 26 Jun 2017 08:10:30 +0000 (16:10 +0800)]
MLK-15285-2 soc: imx: add HAVE_IMX_SOC for soc driver

soc driver is NOT only for i.MX8QM, add HAVE_IMX_SOC
config to improve build dependency.

Signed-off-by: Anson Huang <Anson.Huang@nxp.com>
7 years agoMLK-15285-1 soc: imx: add i.MX8MQ soc id support
Anson Huang [Mon, 26 Jun 2017 08:03:13 +0000 (16:03 +0800)]
MLK-15285-1 soc: imx: add i.MX8MQ soc id support

Add i.MX8MQ SOC ID support, users can cat soc_id
and revision via /sys/devices/soc0/:

root@imx8mq:~# cat /sys/devices/soc0/soc_id
i.MX8MQ
root@imx8mq:~# cat /sys/devices/soc0/revision
1.0

Signed-off-by: Anson Huang <Anson.Huang@nxp.com>
7 years agoMLK-15132-5 : Enable Hantro decoder on i.MX8MQ
Zhou Peng-B04994 [Mon, 26 Jun 2017 05:02:28 +0000 (13:02 +0800)]
MLK-15132-5 : Enable Hantro decoder on i.MX8MQ

Fix vpu decoder timeout issue:
Enable clock before config VPUMIX registers
Replace IMX8MQ_CLK_VPU_BUS_DIV with IMX8MQ_CLK_VPU_DEC_ROOT

Signed-off-by: Zhou Peng-B04994 <eagle.zhou@nxp.com>
7 years agoMGS-3025: ARM64: dts: freescale: imx8mq: enable viv drm
Yong Gan [Mon, 26 Jun 2017 19:03:54 +0000 (03:03 +0800)]
MGS-3025: ARM64: dts: freescale: imx8mq: enable viv drm

Change CONFIG_DRM_VIVANTE to be "m".

Date: Jun 26, 2017
Signed-off-by Meng Mingming <mingming.meng@nxp.com>

7 years agoMLK-15151: dma: imx-sdma: Fix keepings clks enabled on sdma_resume error
Leonard Crestez [Fri, 23 Jun 2017 13:05:53 +0000 (16:05 +0300)]
MLK-15151: dma: imx-sdma: Fix keepings clks enabled on sdma_resume error

This is obviously a bug but I don't know of a scenario where those errors
might happen.

Fixes: 7e84737c9c24 ("MLK-11385 dma: imx-sdma: enable clock before context restored")
Signed-off-by: Leonard Crestez <leonard.crestez@nxp.com>
7 years agoMLK-15125: arm: dts: imx6sx: Fix duplicate name in lcdif
Cristina Ciocan [Thu, 22 Jun 2017 14:46:59 +0000 (17:46 +0300)]
MLK-15125: arm: dts: imx6sx: Fix duplicate name in lcdif

lcdif2 node has a property called "display" and a subnode that is also
called "display", leading to an OF duplicate warning at boot time.

Fix this by changing the subnode's name to "display@1".

Signed-off-by: Cristina Ciocan <cristina-mihaela.ciocan@nxp.com>
7 years agoMLK-15148 ARM64: dts: freescale: imx8mq: add uart DMA chans and enable BT port
Fugang Duan [Fri, 23 Jun 2017 06:00:41 +0000 (14:00 +0800)]
MLK-15148 ARM64: dts: freescale: imx8mq: add uart DMA chans and enable BT port

Add uart DMA chans for imx8mq platform.
Enable uart3 port for Bluetooth on evk board.
Correct the earlycon name.

Signed-off-by: Fugang Duan <fugang.duan@nxp.com>
7 years agoMLK-15147 arm64: imx8mq: fix iomux header file uart pin issue
Fugang Duan [Fri, 23 Jun 2017 05:40:27 +0000 (13:40 +0800)]
MLK-15147 arm64: imx8mq: fix iomux header file uart pin issue

imx8mq iomux header file uart part select_input config are
wrong that cause most of uart pin not work.
Add DCE and DTE string to distinguish the pin is for uart
which function, and clear all select_input for output pin.

Signed-off-by: Fugang Duan <fugang.duan@nxp.com>
7 years agoMLK-15146: ASoC: fsl: add back fsl_hifi build
Shengjiu Wang [Fri, 23 Jun 2017 04:43:06 +0000 (12:43 +0800)]
MLK-15146: ASoC: fsl: add back fsl_hifi build

fsl_hifi is removed wrongly by commit a69a185aad7f ("MLK-15140-2:
ASoC: fsl: add machine driver for wm8524")

Signed-off-by: Shengjiu Wang <shengjiu.wang@freescale.com>
7 years agoMLK-15140-6: ARM64: defconfig: built-in wm8524 modules
Shengjiu Wang [Thu, 22 Jun 2017 07:41:16 +0000 (15:41 +0800)]
MLK-15140-6: ARM64: defconfig: built-in wm8524 modules

built-in wm8524 modules

Signed-off-by: Shengjiu Wang <shengjiu.wang@freescale.com>
7 years agoMLK-15140-5: ARM64: dts: enable wm8524 and sai2
Shengjiu Wang [Thu, 22 Jun 2017 07:41:09 +0000 (15:41 +0800)]
MLK-15140-5: ARM64: dts: enable wm8524 and sai2

enable wm8524 and sai2

Signed-off-by: Mihai Serban <mihai.serban@nxp.com>
Signed-off-by: Shengjiu Wang <shengjiu.wang@freescale.com>
7 years agoMLK-15140-4: clk: clk-imx8mq: Add audio ipg clock
Shengjiu Wang [Thu, 22 Jun 2017 07:40:52 +0000 (15:40 +0800)]
MLK-15140-4: clk: clk-imx8mq: Add audio ipg clock

add audio ipg clock, sai ipg clock and correct some wrong
place in clock tree.

Signed-off-by: Mihai Serban <mihai.serban@nxp.com>
Signed-off-by: Shengjiu Wang <shengjiu.wang@freescale.com>
7 years agoMLK-15140-3: ASoC: codecs: add wm8524 codec driver
Shengjiu Wang [Thu, 22 Jun 2017 07:40:29 +0000 (15:40 +0800)]
MLK-15140-3: ASoC: codecs: add wm8524 codec driver

Add wm8524 driver, there is no i2c interface for it.

Signed-off-by: Mihai Serban <mihai.serban@nxp.com>
Signed-off-by: Shengjiu Wang <shengjiu.wang@freescale.com>
7 years agoMLK-15140-2: ASoC: fsl: add machine driver for wm8524
Shengjiu Wang [Thu, 22 Jun 2017 07:39:55 +0000 (15:39 +0800)]
MLK-15140-2: ASoC: fsl: add machine driver for wm8524

This a simple machine driver for wm8524.

Signed-off-by: Shengjiu Wang <shengjiu.wang@freescale.com>
7 years agoMLK-15140-1: ASoC: fsl_sai: support latest sai module
Shengjiu Wang [Thu, 22 Jun 2017 07:39:24 +0000 (15:39 +0800)]
MLK-15140-1: ASoC: fsl_sai: support latest sai module

The version of sai is upgrate in imx8mq, which add two register
in beginning, there is VERID and PARAM. the driver need to be
update

Signed-off-by: Mihai Serban <mihai.serban@nxp.com>
Signed-off-by: Shengjiu Wang <shengjiu.wang@freescale.com>
7 years agoMLK-15144 ARM64: dts: freescale: imx8mq: add enet support
Fugang Duan [Wed, 21 Jun 2017 02:50:27 +0000 (10:50 +0800)]
MLK-15144 ARM64: dts: freescale: imx8mq: add enet support

Add enet support for imx8mq evk board.

Signed-off-by: Fugang Duan <fugang.duan@nxp.com>
7 years agoMLK-15138-2 defconfig: enable mxc ion on i.mx8
Song Bing [Thu, 22 Jun 2017 19:42:25 +0000 (11:42 -0800)]
MLK-15138-2 defconfig: enable mxc ion on i.mx8

Enable mxc ion on i.mx8.

Signed-off-by: Song Bing <bing.song@nxp.com>
7 years agoMLK-15138-1 ion: enable mxc ion on i.mx8
Song Bing [Thu, 22 Jun 2017 19:42:25 +0000 (11:42 -0800)]
MLK-15138-1 ion: enable mxc ion on i.mx8

Enable mxc ion on i.mx8.

Signed-off-by: Song Bing <bing.song@nxp.com>
7 years agoMLK-15143 ARM64: dts: freescale: imx8mq: increase CMA size
Anson Huang [Fri, 23 Jun 2017 01:49:05 +0000 (09:49 +0800)]
MLK-15143 ARM64: dts: freescale: imx8mq: increase CMA size

Increase CMA size to 640MB for i.MX8MQ.

Signed-off-by: Anson Huang <Anson.Huang@nxp.com>
7 years agoMLK-15137-02 driver: clk: Remove new_div_ack check in frac pll
Bai Ping [Thu, 22 Jun 2017 06:55:58 +0000 (14:55 +0800)]
MLK-15137-02 driver: clk: Remove new_div_ack check in frac pll

If the frac pll is powrer down or hold in reset, new_div_ack
will not be assert. Waiting for ack will failed.

Signed-off-by: Bai Ping <ping.bai@nxp.com>
7 years agoMLK-15137-01 driver: fix vpu gate clock's offset on i.mx8mq
Bai Ping [Wed, 21 Jun 2017 09:16:30 +0000 (17:16 +0800)]
MLK-15137-01 driver: fix vpu gate clock's offset on i.mx8mq

Fix vpu's gate clock register offset.

Signed-off-by: Bai Ping <ping.bai@nxp.com>
7 years agoMLK-15028: ASoC: codecs: wm8960: Remove bitclk relax condition
Daniel Baluta [Thu, 22 Jun 2017 07:18:29 +0000 (10:18 +0300)]
MLK-15028: ASoC: codecs: wm8960: Remove bitclk relax condition

Using a higher bitclk then expected doesn't always work.
Here is an example:

aplay -Dhw:0,0 -d 5 -r 48000 -f S24_LE -c 2 audio48k24b2c.wav

In this case, the required bitclk is 48000 * 24 * 2 = 2304000
but the closest bitclk that can be derived is 3072000.

Now, for format S24_LE, SAI will use slot_width = 24, but since
the clock is faster than expected, it will start to send bytes
from the next channel so the sound will be corrupted.

Thus, remove bitclk relaxation condition which was added mostly
for supporting S20_3LE format which was removed from SAI in
commit 739e6d654b5c0a ("MLK-14870: ASoC: fsl_sai: Remove support
for S20_3LE").

Signed-off-by: Daniel Baluta <daniel.baluta@nxp.com>
Reviewed-by: Shengjiu Wang <shengjiu.wang@nxp.com>
7 years agoMLK-15132-4 : Enable Hantro decoder on i.MX8MQ
Zhou Peng-B04994 [Thu, 22 Jun 2017 06:24:12 +0000 (14:24 +0800)]
MLK-15132-4 : Enable Hantro decoder on i.MX8MQ

Fix section mismatch link warning,
Removing unnecessary annotation '__init'/'__exit'

Signed-off-by: Zhou Peng-B04994 <eagle.zhou@nxp.com>
7 years agoMLK-15135-5 ARM64: configs: defconfig: enable imx-sdma driver on i.mx8
Robin Gong [Wed, 21 Jun 2017 09:29:09 +0000 (17:29 +0800)]
MLK-15135-5 ARM64: configs: defconfig: enable imx-sdma driver on i.mx8

enable sdma driver on i.mx8 defconfig.

Signed-off-by: Robin Gong <yibin.gong@nxp.com>
7 years agoMLK-15135-4: ARM: dts: fsl-imx8mq: add SDMA2 on imx8mq
Robin Gong [Wed, 21 Jun 2017 09:24:50 +0000 (17:24 +0800)]
MLK-15135-4: ARM: dts: fsl-imx8mq: add SDMA2 on imx8mq

Add Audio SDMA2 on i.mx8mscale.

Signed-off-by: Robin Gong <yibin.gong@nxp.com>
7 years agoMLK-15135-3: clk: imx8mq: add sdma clock
Robin Gong [Wed, 21 Jun 2017 09:12:08 +0000 (17:12 +0800)]
MLK-15135-3: clk: imx8mq: add sdma clock

add sdma clock and ipg clock on i.mx8mq.

Signed-off-by: Robin Gong <yibin.gong@nxp.com>
7 years agoMLK-15135-2 dma: Kconfig: add sdma on i.mx8mq
Robin Gong [Wed, 21 Jun 2017 09:10:06 +0000 (17:10 +0800)]
MLK-15135-2 dma: Kconfig: add sdma on i.mx8mq

select sdma on i.mx8mq.

Signed-off-by: Robin Gong <yibin.gong@nxp.com>
7 years agoMLK-15135-1: dma: imx-sdma: fix build warning on aarch64
Robin Gong [Wed, 21 Jun 2017 09:04:40 +0000 (17:04 +0800)]
MLK-15135-1: dma: imx-sdma: fix build warning on aarch64

fix below build warning with aarch64:
drivers/dma/imx-sdma.c: In function â€˜sdma_prep_dma_cyclic’:
drivers/dma/imx-sdma.c:1727:3: warning: format â€˜%d’ expects argument of type â€˜int’, but argument 5 has type â€˜size_t’ [-Wformat=]
   dev_dbg(sdma->dev, "entry %d: count: %d dma: %pad %s%s\n",

Signed-off-by: Robin Gong <yibin.gong@nxp.com>
7 years agoMLK-15133-02 driver: clk: Skip enable non-critical clks on imx8mq
Bai Ping [Wed, 21 Jun 2017 08:43:06 +0000 (16:43 +0800)]
MLK-15133-02 driver: clk: Skip enable non-critical clks on imx8mq

Only enable the system critical clks by default, other clks only
need to be enabled when it is used by the driver.

Signed-off-by: Bai Ping <ping.bai@nxp.com>
7 years agoMLK-15133-01 driver: clk: Enhance the frac&sccg pll code on i.mx8mq
Bai Ping [Wed, 21 Jun 2017 06:35:28 +0000 (14:35 +0800)]
MLK-15133-01 driver: clk: Enhance the frac&sccg pll code on i.mx8mq

1. Fix coding typo.

2. Add PLL lock check and fix the reload of divfi and divff for frac pll.

Signed-off-by: Bai Ping <ping.bai@nxp.com>
7 years agoMLK-15067: ASoC: fsl: imx-wm8958: Kill warning for non-gpr boards
Daniel Baluta [Tue, 20 Jun 2017 13:33:55 +0000 (16:33 +0300)]
MLK-15067: ASoC: fsl: imx-wm8958: Kill warning for non-gpr boards

Similar with 7c280619ed45b (" MLK-14663-2: ASoC: fsl: imx-wm8960: Kill
warning for non-gpr boards")

Signed-off-by: Daniel Baluta <daniel.baluta@nxp.com>
7 years agoMLK-15067: ASoC: fsl: imx-wm8958: Refactor GPR parsing
Daniel Baluta [Tue, 20 Jun 2017 10:46:53 +0000 (13:46 +0300)]
MLK-15067: ASoC: fsl: imx-wm8958: Refactor GPR parsing

This is similar with commit c79a82aec8ccf ("ASoC: fsl: imx-wm8960: Refactor
GPR parsing") and it is needed for easier adding support for non-gpr boards.

Signed-off-by: Daniel Baluta <daniel.baluta@nxp.com>
7 years agoMLK-15132-3 : Enable Hantro decoder on i.MX8MQ
Zhou Peng-B04994 [Wed, 21 Jun 2017 08:02:30 +0000 (16:02 +0800)]
MLK-15132-3 : Enable Hantro decoder on i.MX8MQ

Move hantrodec.h to uapi directory

Signed-off-by: Zhou Peng-B04994 <eagle.zhou@nxp.com>
7 years agoMLK-15132-2: Enable Hantro decoder on i.MX8MQ
Zhou Peng-B04994 [Wed, 21 Jun 2017 07:15:36 +0000 (15:15 +0800)]
MLK-15132-2: Enable Hantro decoder on i.MX8MQ

Add vpu part in fsl-imx8mq.dtsi

Signed-off-by: Zhou Peng-B04994 <eagle.zhou@nxp.com>
7 years agoMLK-15132-1: Enable Hantro decoder on i.MX8MQ
Zhou Peng-B04994 [Wed, 21 Jun 2017 06:57:33 +0000 (14:57 +0800)]
MLK-15132-1: Enable Hantro decoder on i.MX8MQ

Added hantro driver code

Signed-off-by: Zhou Peng-B04994 <eagle.zhou@nxp.com>
7 years agoMLK-15131-2 ARM64: dts: freescale: imx8mq: enable snvs rtc
Anson Huang [Wed, 21 Jun 2017 02:57:33 +0000 (10:57 +0800)]
MLK-15131-2 ARM64: dts: freescale: imx8mq: enable snvs rtc

Enable SNVS RTC by default on i.MX8MQ.

Signed-off-by: Anson Huang <Anson.Huang@nxp.com>
7 years agoMLK-15131-1 defconfig: enable SNVS RTC
Anson Huang [Wed, 21 Jun 2017 02:56:35 +0000 (10:56 +0800)]
MLK-15131-1 defconfig: enable SNVS RTC

Enable SNVS RTC for i.MX8MQ by default.

The change of CONFIG_I2C_MUX is introduced by
running savedefconfig.

Signed-off-by: Anson Huang <Anson.Huang@nxp.com>
7 years agoMLK-15128-7 clk: imx: add i.mx8mq clock driver support
Anson Huang [Tue, 20 Jun 2017 15:12:58 +0000 (23:12 +0800)]
MLK-15128-7 clk: imx: add i.mx8mq clock driver support

Add i.MX8MQ clock driver support.

Signed-off-by: Anson Huang <Anson.Huang@nxp.com>
Signed-off-by: Bai Ping <ping.bai@nxp.com>
7 years agoMLK-15128-6 soc: imx: add psci gpc support for i.mx8mq
Anson Huang [Tue, 20 Jun 2017 15:11:29 +0000 (23:11 +0800)]
MLK-15128-6 soc: imx: add psci gpc support for i.mx8mq

Add i.MX8MQ PSCI GPC virtual driver support.

Signed-off-by: Anson Huang <Anson.Huang@nxp.com>
Signed-off-by: Bai Ping <ping.bai@nxp.com>
7 years agoMLK-15128-5 ARM64: kconfig: enable i.mx8mq pinctrl driver
Anson Huang [Tue, 20 Jun 2017 15:10:41 +0000 (23:10 +0800)]
MLK-15128-5 ARM64: kconfig: enable i.mx8mq pinctrl driver

Enable i.MX8MQ pinctrl driver by default.

Signed-off-by: Anson Huang <Anson.Huang@nxp.com>
7 years agoMLK-15128-4 pinctrl: freescale: add i.mx8mq pinctrl driver support
Anson Huang [Tue, 20 Jun 2017 15:08:34 +0000 (23:08 +0800)]
MLK-15128-4 pinctrl: freescale: add i.mx8mq pinctrl driver support

Add i.MX8MQ pinctrl driver support.

Signed-off-by: Anson Huang <Anson.Huang@nxp.com>
7 years agoMLK-15128-3 pinctrl: freescale: support scu and memmap pinctrl together
Anson Huang [Tue, 20 Jun 2017 15:05:52 +0000 (23:05 +0800)]
MLK-15128-3 pinctrl: freescale: support scu and memmap pinctrl together

As i.MX8MQ is a ARM64 SoC but it does NOT use SCU pinctrl, so
need to support both SCU and MEMMAP pinctrl together for ARM64
build.

use IMX8_USE_SCU flag to distinguish SCU and MEMMAP pinctrl
type.

Signed-off-by: Anson Huang <Anson.Huang@nxp.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
7 years agoMLK-15128-2 ARM64: dts: freescale: add i.mx8mq dtsi and evk dtb
Anson Huang [Tue, 20 Jun 2017 15:03:54 +0000 (23:03 +0800)]
MLK-15128-2 ARM64: dts: freescale: add i.mx8mq dtsi and evk dtb

Add i.MX8MQ dtsi and EVK board dtb.

Signed-off-by: Anson Huang <Anson.Huang@nxp.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
Signed-off-by: Chenyan Feng <ella.feng@nxp.com>
Signed-off-by: Li Jun <jun.li@nxp.com>
Signed-off-by: Bai Ping <ping.bai@nxp.com>
7 years agoMLK-15128-1 dt-bindings: imx8mq: add clock and pinctrl head file
Anson Huang [Tue, 20 Jun 2017 15:03:02 +0000 (23:03 +0800)]
MLK-15128-1 dt-bindings: imx8mq: add clock and pinctrl head file

Add i.MX8MQ clock and pinctrl file.

Signed-off-by: Anson Huang <Anson.Huang@nxp.com>
Signed-off-by: Bai Ping <ping.bai@nxp.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
7 years agoMLK-14765: Fix DCP Aes timeout issues when used with CTS
Radu Solea [Fri, 16 Jun 2017 11:02:50 +0000 (14:02 +0300)]
MLK-14765: Fix DCP Aes timeout issues when used with CTS

The DCP driver does not obey cryptlen, when doing CTS this results in
passing to hardware input stream lengths which are not multiple of
block size. This causes the hw to misbehave. Also not honoring
cryptlen makes CTS fail. A check was introduced to prevent future
erroneous stream lengths from reaching the hardware. Code which is
splitting the input stream in internal DCP pages was changed to obey
cryptlen.

Signed-off-by: Radu Solea <radu.solea@nxp.com>
7 years agoMLK-14765: Fix DCP SHA null hashes and output length
Radu Solea [Wed, 7 Jun 2017 14:18:03 +0000 (17:18 +0300)]
MLK-14765: Fix DCP SHA null hashes and output length

On imx6sl and imx6ull DCP writes at least 32 bytes in the output
buffer instead of hash length as documented. Add intermediate buffer
to prevent write out of bounds.

When requested to produce null hashes DCP fails to produce valid
output. Add software workaround to bypass hardware and return valid output.

Signed-off-by: Radu Solea <radu.solea@nxp.com>
7 years agoMGS-2949 [#ccc] revert power management workaround for imx8
Xianzhong [Wed, 7 Jun 2017 09:27:35 +0000 (17:27 +0800)]
MGS-2949 [#ccc] revert power management workaround for imx8

SCFW crash is fixed, need drop the temporal workaround.

Date: Jun 07, 2017
Signed-off-by: Xianzhong <xianzhong.li@nxp.com>
7 years agoMGS-2970-2 [#imx-603] fix dual gpu hang with power management
Xianzhong [Sun, 18 Jun 2017 03:28:23 +0000 (11:28 +0800)]
MGS-2970-2 [#imx-603] fix dual gpu hang with power management

gpu1 hang is reproducible when run es32 cts with power mangement.

there is the gpu1 power-off/up when commit gpu0 and gpu1 early or late,
then gpu0 will break the inter-semaphore & stall from gpu1, and end up,
when gpu1 execute its command, will stuck to wait for gpu0 infinitely.

prevent the unexpected power-off bofore command commit on dual cores.

Date: Jun 18, 2017
Signed-off-by: Xianzhong <xianzhong.li@nxp.com>
7 years agoMGS-2970-1 [#imx-603] fix dual gpu hang with power management
Xianzhong [Fri, 9 Jun 2017 10:12:49 +0000 (18:12 +0800)]
MGS-2970-1 [#imx-603] fix dual gpu hang with power management

gpu mode is lost after power-off, hence gpu hang in dual mode,
this patch can configure gpu mode for each power-up properly.

Date: Jun 09, 2017
Signed-off-by: Xianzhong <xianzhong.li@nxp.com>
7 years agoMGS-2944 [#imx-290] enable GPIPE clock gating
Xianzhong [Tue, 6 Jun 2017 06:17:49 +0000 (14:17 +0800)]
MGS-2944 [#imx-290] enable GPIPE clock gating

revert the workaround which disable GPIPE clock gating.

Date: June 06, 2017
Signed-off-by: Xianzhong <xianzhong.li@nxp.com>
7 years agoMGS-2857-2 [#imx-530] cleanup spinlock debug code from gpu kernel
Xianzhong [Wed, 7 Jun 2017 09:05:47 +0000 (17:05 +0800)]
MGS-2857-2 [#imx-530] cleanup spinlock debug code from gpu kernel

not make sense to use the different codes for spinlock debug

Date: Jun 07, 2017
Signed-off-by: Xianzhong <xianzhong.li@nxp.com>
7 years agoMLK-14663-2: ASoC: fsl: imx-wm8960: Kill warning for non-gpr boards
Daniel Baluta [Thu, 15 Jun 2017 15:48:22 +0000 (18:48 +0300)]
MLK-14663-2: ASoC: fsl: imx-wm8960: Kill warning for non-gpr boards

A side effect of commit 5555277e693a7 ("MLK-13574-1: ASoC: imx-wm8960:
remove the gpr dependency") is that a warning was printed for boards
without gpr. This can be confusing.

imx7d boards do not have a gpr setting, so use imx7d-evk-wm8960
compatible string to avoid printing the warning.

Signed-off-by: Daniel Baluta <daniel.baluta@nxp.com>
7 years agoMLK-14663-1: ASoC: fsl: imx-wm8960: Refactor GPR parsing
Daniel Baluta [Thu, 15 Jun 2017 15:41:58 +0000 (18:41 +0300)]
MLK-14663-1: ASoC: fsl: imx-wm8960: Refactor GPR parsing

Refactor GPR handling into a function for easier adding support
for non-gpr boards.

Signed-off-by: Daniel Baluta <daniel.baluta@nxp.com>
7 years agoMLK-15124-06: defconfig: Add mx8 image subsystem
Sandor Yu [Tue, 20 Jun 2017 07:45:25 +0000 (15:45 +0800)]
MLK-15124-06: defconfig: Add mx8 image subsystem

Add mx8 image subsystem and v4l2 device.

Signed-off-by: Sandor Yu <Sandor.yu@nxp.com>
7 years agoMLK-15124-05: dts: Add imx8qm image subsystem property
Sandor Yu [Tue, 20 Jun 2017 07:40:31 +0000 (15:40 +0800)]
MLK-15124-05: dts: Add imx8qm image subsystem property

Add imx8qm image subsystem property.

Signed-off-by: Sandor Yu <Sandor.yu@nxp.com>
7 years agoMLK-15124-04: image ss: Add mx8 image subsystem driver
Sandor Yu [Tue, 20 Jun 2017 07:37:16 +0000 (15:37 +0800)]
MLK-15124-04: image ss: Add mx8 image subsystem driver

Add mxc media device driver.
Add mx8 isi device driver.
Add mx8 mipi csi device driver.
Add max9286 sensor driver.
mxc isi driver support CSC and scaling function.

Signed-off-by: Sandor Yu <Sandor.yu@nxp.com>
7 years agoMLK-15124-03: clk: Rename image subsystem clock name
Sandor Yu [Tue, 20 Jun 2017 07:30:25 +0000 (15:30 +0800)]
MLK-15124-03: clk: Rename image subsystem clock name

Rename image subsystem power domain name.
Rename mipi csi LIS clock name.

Signed-off-by: Sandor Yu <Sandor.yu@nxp.com>
7 years agoMLK-15124-02: clk: Add local interrupter clock
Sandor Yu [Tue, 20 Jun 2017 07:27:53 +0000 (15:27 +0800)]
MLK-15124-02: clk: Add local interrupter clock

Add mipi csi local interrupter clock

Signed-off-by: Sandor Yu <Sandor.yu@nxp.com>
7 years agoMLK-15124-01: pm: Add image subsystem power domain name
Sandor Yu [Tue, 20 Jun 2017 07:23:11 +0000 (15:23 +0800)]
MLK-15124-01: pm: Add image subsystem power domain name

Add imx8 image subsystem power domain name.

Signed-off-by: Sandor Yu <Sandor.yu@nxp.com>
7 years agoMLK-15120 ARM: imx7d: clk: select uart3 clock parent and set rate
Fugang Duan [Tue, 20 Jun 2017 06:14:03 +0000 (14:14 +0800)]
MLK-15120 ARM: imx7d: clk: select uart3 clock parent and set rate

The clock driver may enable uart clock tree when earlycon/earlyprintk
kernel param is enabled, and the clock gate specify CLK_SET_RATE_GATE,
then .of_clk_set_defaults() set the dts node assigned-rate will be failed.
So set parent and set rate in clock driver is reasonable.

Signed-off-by: Fugang Duan <fugang.duan@nxp.com>
7 years agoMLK-15104: ASoC: imx-sii902: add constraint for channels
Shengjiu Wang [Mon, 19 Jun 2017 10:36:50 +0000 (18:36 +0800)]
MLK-15104: ASoC: imx-sii902: add constraint for channels

The maximum channel supported by sii902 is 2, but machine
driver use dummy codec, and there is no constraint list
from codec, so add constraint directly in machine driver.

Signed-off-by: Shengjiu Wang <shengjiu.wang@freescale.com>
(cherry picked from commit 037051e60f3b29fb09aae6f2c97a3532482dfc2a)

7 years agoMLK-15109-4: ARM64: dts: update esai compatible string
Shengjiu Wang [Mon, 19 Jun 2017 07:47:51 +0000 (15:47 +0800)]
MLK-15109-4: ARM64: dts: update esai compatible string

update esai compatible string in imx8qm.

Signed-off-by: Shengjiu Wang <shengjiu.wang@freescale.com>
7 years agoMLK-15109-3: ASoC: fsl_esai: add constraint for dma workaround
Shengjiu Wang [Mon, 19 Jun 2017 03:28:42 +0000 (11:28 +0800)]
MLK-15109-3: ASoC: fsl_esai: add constraint for dma workaround

Use GPT dma event to instead of esai dma event can't totally
resolve/workaround the hardware issue, There is two GPT, one
GPT is to get the failing edge of dma event, then to trigger
EDMA copy data, another one is to get the raising edge, then to
clear the interrupt of both GPT, sometimes, the clear operation
may clear the failing edge event wrongly in race condition.
then the EDMA will stop.
In test result, the high sample rate and multi-channel is easy
to trigger this issue. so add constraint for them to make the
workaround stable

Signed-off-by: Shengjiu Wang <shengjiu.wang@freescale.com>
7 years agoMLK-15109-2: ASoC: fsl_esai: add constrain_period_size
Shengjiu Wang [Mon, 19 Jun 2017 03:28:42 +0000 (11:28 +0800)]
MLK-15109-2: ASoC: fsl_esai: add constrain_period_size

There is limitaion for EDMA, which can only accept the period bytes
that can be divided by maxburst with no remainder. Otherwise EDMA
will not copy the left data in the end, and it will cause noise.
so add constraint for these chips.

Signed-off-by: Shengjiu Wang <shengjiu.wang@freescale.com>
7 years agoMLK-15109-1: ASoC: fsl_esai: introduce SoC specific data
Shengjiu Wang [Mon, 19 Jun 2017 03:41:44 +0000 (11:41 +0800)]
MLK-15109-1: ASoC: fsl_esai: introduce SoC specific data

Introduce a SoC data struct which contains the differences between
the different SoCs this driver supports. This makes it easy to support
more differences without having to introduce a new switch/case each
time.

Signed-off-by: Shengjiu Wang <shengjiu.wang@freescale.com>
7 years agoMLK-15066-2: ASoC: imx-pcm-dma-v2: fix typo issue
Shengjiu Wang [Mon, 19 Jun 2017 03:28:07 +0000 (11:28 +0800)]
MLK-15066-2: ASoC: imx-pcm-dma-v2: fix typo issue

fix typo issue

Signed-off-by: Shengjiu Wang <shengjiu.wang@freescale.com>
7 years agoMLK-15066-1: ASoC: imx-pcm-dma-v2: fix noise issue with pulseaudio
Shengjiu Wang [Mon, 12 Jun 2017 08:27:14 +0000 (16:27 +0800)]
MLK-15066-1: ASoC: imx-pcm-dma-v2: fix noise issue with pulseaudio

Same as commit c55075170214 ("MLK-14582: ASoC: imx-pcm-rpmsg: fix
audio noise issue with pulseaudio"), need to add a constraint for
SNDRV_PCM_HW_PARAM_PERIODS, which make the period number integer.

Signed-off-by: Shengjiu Wang <shengjiu.wang@freescale.com>
7 years agoMLK-15075 thermal: imx: fix temp read failure on i.mx7d
Bai Ping [Mon, 19 Jun 2017 09:01:23 +0000 (17:01 +0800)]
MLK-15075 thermal: imx: fix temp read failure on i.mx7d

On i.MX7D, if the system enter LPSR mode, the tempmon module
will be power down, so the regiter's value is lost, so we need
to save the registers before suspend and restore the register after
resume back.

Signed-off-by: Bai Ping <ping.bai@nxp.com>
7 years agoMLK-15074-11 ARM64: defconfig: enable CONFIG_BLK_DEV_INITRD
Li Jun [Fri, 16 Jun 2017 16:51:18 +0000 (00:51 +0800)]
MLK-15074-11 ARM64: defconfig: enable CONFIG_BLK_DEV_INITRD

Enable CONFIG_BLK_DEV_INITRD for initramfs of MFG tool.

Signed-off-by: Li Jun <jun.li@nxp.com>
7 years agoMLK-15074-10 usb: gadget: utp: correct the set_fs condition
Li Jun [Fri, 16 Jun 2017 17:54:55 +0000 (01:54 +0800)]
MLK-15074-10 usb: gadget: utp: correct the set_fs condition

set_fs() should be done in case:
1. CONFIG_FSL_UTP is not enabled.
2. CONFIG_FSL_UTP is enabled but is_utp_device is false.

Signed-off-by: Li Jun <jun.li@nxp.com>
7 years agoMLK-15074-9 usb: gadget: utp: fix wrong fsg parameter
Li Jun [Fri, 16 Jun 2017 16:49:49 +0000 (00:49 +0800)]
MLK-15074-9 usb: gadget: utp: fix wrong fsg parameter

As common->fsg maybe have not been set correctly while enumration, so
use the correct pointer fsg for utp device check.

Signed-off-by: Li Jun <jun.li@nxp.com>
7 years agoMLK-15074-8 ARM: imx_v7_defconfig: enable CONFIG_FSL_UTP
Li Jun [Fri, 16 Jun 2017 17:40:37 +0000 (01:40 +0800)]
MLK-15074-8 ARM: imx_v7_defconfig: enable CONFIG_FSL_UTP

CONFIG_FSL_UTP is missing for MFG tool, add it.

Signed-off-by: Li Jun <jun.li@nxp.com>
7 years agoMLK-15119 PCI: imx: remove the wrong rebased codes
Richard Zhu [Mon, 19 Jun 2017 09:30:12 +0000 (17:30 +0800)]
MLK-15119 PCI: imx: remove the wrong rebased codes

Some codes are wrong rebase back into mainline in
one git pull rebase confliction.
Remove them.

Signed-off-by: Richard Zhu <hongxing.zhu@nxp.com>
7 years agoMLK-15117 ARM64: dts: imx8qxp-lpddr4-arm2: add extended dts for enet2 port
Fugang Duan [Mon, 19 Jun 2017 09:10:09 +0000 (17:10 +0800)]
MLK-15117 ARM64: dts: imx8qxp-lpddr4-arm2: add extended dts for enet2 port

Since lpddr4 arm2 baord enet2 has pin conflict with esai0,
so add extended dts for enet2 port.

Signed-off-by: Fugang Duan <fugang.duan@nxp.com>
Reviewed-by: Robin Gong <yibin.gong@nxp.com>
7 years agoMLK-15118 ARM64: dts: imx8qxp: enable lpuart1/2 and eDMA0
Fugang Duan [Mon, 19 Jun 2017 09:05:01 +0000 (17:05 +0800)]
MLK-15118 ARM64: dts: imx8qxp: enable lpuart1/2 and eDMA0

Add lpuart1, lpuart2 and eDMA0 support for imx8qxp.

Signed-off-by: Fugang Duan <fugang.duan@nxp.com>
Reviewed-by: Robin Gong <yibin.gong@nxp.com>
7 years agoMLK-15064-4 PCI: imx: enable pcie support
Richard Zhu [Tue, 13 Jun 2017 06:02:40 +0000 (14:02 +0800)]
MLK-15064-4 PCI: imx: enable pcie support

- add the cpu addr offset
  Bit[31:24]
  pciea 60 - 6f ---> 40 - 4f
  pcieb 70 - 7f ---> 80 - 8f
- internal pll is verified on imx8qxp arm2 board
- use the dma_alloc_coherent to alloc the msi region,
  because that imx8 pcie only supports up to 32bit
  msi address.

Signed-off-by: Richard Zhu <hongxing.zhu@nxp.com>
Reviewed-by: Frank Li <frank.li@nxp.com>
7 years agoMLK-15064-3 clk: imx: enable pcie support
Richard Zhu [Tue, 13 Jun 2017 05:59:39 +0000 (13:59 +0800)]
MLK-15064-3 clk: imx: enable pcie support

correct the pd definitions of the pcie clks

Signed-off-by: Richard Zhu <hongxing.zhu@nxp.com>
Reviewed-by: Frank Li <frank.li@nxp.com>
7 years agoMLK-15064-2 ARM64: DMA: limit the dma mask to be 32bit
Richard Zhu [Mon, 19 Sep 2016 08:39:44 +0000 (16:39 +0800)]
MLK-15064-2 ARM64: DMA: limit the dma mask to be 32bit

Limit the dma mask to be 32bit, because that
the imx8 doesn't have the 64bit dma capapbility
although it is 64bit soc.

Signed-off-by: Richard Zhu <hongxing.zhu@nxp.com>
Reviewed-by: Frank Li <frank.li@nxp.com>