u-boot.git
5 years agoMLK-18243 DTS: imx8mm_evk: Add DTS files for iMX8MM LPDDR4 and DDR4 EVK boards
Ye Li [Fri, 19 Apr 2019 03:26:12 +0000 (20:26 -0700)]
MLK-18243 DTS: imx8mm_evk: Add DTS files for iMX8MM LPDDR4 and DDR4 EVK boards

Add two DTS files, one for iMX8MM LPDDR4 EVK (fsl-imx8mm-evk.dts)
and another for iMX8MM DDR4 EVK.
Two boards share same base board, the major difference is the DDR4 board
installs NAND chip and removes flexspi and eMMC.

Signed-off-by: Ye Li <ye.li@nxp.com>
5 years agoMLK-21845-4 imx8m: Add imximage for flexspi boot image
Ye Li [Fri, 19 Apr 2019 03:24:42 +0000 (20:24 -0700)]
MLK-21845-4 imx8m: Add imximage for flexspi boot image

Since flexspi has different IVT offset with SD/MMC, add a new imximage
for it.

Signed-off-by: Ye Li <ye.li@nxp.com>
5 years agoMLK-21836-4 spi: fspi: Add compatible string for iMX8MM
Ye Li [Fri, 19 Apr 2019 02:52:23 +0000 (19:52 -0700)]
MLK-21836-4 spi: fspi: Add compatible string for iMX8MM

iMX8MM DTS uses new compatible string "fsl,imx8mm-flexspi" in flexspi
node. Add it into the flexspi driver.

Signed-off-by: Ye Li <ye.li@nxp.com>
5 years agoMLK-18590 pmic: bd71837: Change to use new fdt API
Ye Li [Wed, 13 Jun 2018 02:14:29 +0000 (19:14 -0700)]
MLK-18590 pmic: bd71837: Change to use new fdt API

Changed to use dev_read_subnode to get the ofnode type of the bd71837 device node.
Because the pmic_bind_children is changed to use ofnode.

Signed-off-by: Ye Li <ye.li@nxp.com>
(cherry picked from commit acdc5c297a9630a464896ba507f33f4d4e9820c4)

5 years agoMLK-14422 imx7d: wdog: Overwrite the reset_cpu to turn off internal reset signal
Ye Li [Tue, 7 Feb 2017 14:48:01 +0000 (22:48 +0800)]
MLK-14422 imx7d: wdog: Overwrite the reset_cpu to turn off internal reset signal

Set wdog WCR register SRS bit to turn off internal reset signal WDOG_RESET_B_DEB
for mx7d. So that the warm reset is disabled.
The WDA is cleared to output WDOG_B immediately to reset the board.

Signed-off-by: Ye Li <ye.li@nxp.com>
(cherry picked from commit 6d09863801695d975060fbc240147a9067dbafbf)
(cherry picked from commit cde504077e1ce94d9ad34ece0f3e882972404297)

5 years agoMLK-19726-2 arm: Don't remove all devices when power domain driver is enabled
Ye Li [Wed, 26 Sep 2018 09:15:00 +0000 (02:15 -0700)]
MLK-19726-2 arm: Don't remove all devices when power domain driver is enabled

Because we power off all devices in board_quiesce_devices which is prior then
executing dm_remove_devices_flags. So any access to HW in dm_remove_devices_flags
will cause problem.
However, some drivers like ethernet which implements the pre_remove callback is always
called without any flags check, and this finally accesses FEC controller.

Since we don't need to remove all devices in u-boot before starting kernel, disable
this feature when power domain is enabled.

Signed-off-by: Ye Li <ye.li@nxp.com>
(cherry picked from commit 741f2ea182bf293d8270bdc4a217a96db22c414c)

5 years agoMLK-21019 TMU: Check the TEMP range for iMX8MM
Ye Li [Thu, 28 Feb 2019 07:50:39 +0000 (23:50 -0800)]
MLK-21019 TMU: Check the TEMP range for iMX8MM

On iMX8MM, the V flag in TRISTR register only reflect the state of SNSR
value, not the calibrated TEMP value. So checking this flag is not
reliable. Per IC suggestion, change to read the TEMP/AVG_TEMP directly
and check whether it in valid range 10-125C.

Signed-off-by: Ye Li <ye.li@nxp.com>
(cherry picked from commit fe41c5fa6af88cce7f5a9723c82d6ad4e61357ce)

5 years agoMLK-20966 TMU: Fix for temperature out of range
Ye Li [Fri, 22 Feb 2019 05:12:57 +0000 (21:12 -0800)]
MLK-20966 TMU: Fix for temperature out of range

When the temperature is out of sensor's range, the Valid bit won't be
set in TRITSR register. So the polling loop won't go out.

Change the codes to retry 10 times with 100ms interval for the Valid bit.
If the timeout, we give a warning for the invalid data.

Signed-off-by: Ye Li <ye.li@nxp.com>
Reviewed-by: Bai Ping <ping.bai@nxp.com>
(cherry picked from commit 7ea2168e06d4f77a872f51a167ee1ed6bf2b0632)

5 years agoMLK-20784-2 imx8mm: Load fuse for TMU TCALIV and TASR
Ye Li [Mon, 21 Jan 2019 08:33:38 +0000 (00:33 -0800)]
MLK-20784-2 imx8mm: Load fuse for TMU TCALIV and TASR

On iMX8MM, the default value of TMU registers TCALIV and TASR need
be loaded from fuse. HW won't do this, it expect SW loads them before
using TMU.

Signed-off-by: Ye Li <ye.li@nxp.com>
Reviewed-by: Bai Ping <ping.bai@nxp.com>
(cherry picked from commit 9dada8a697b1c103fdb28f528f168da7ecc20849)

5 years agoMLK-20784-1 TMU: Add a interface for TMU arch level initialization
Ye Li [Mon, 21 Jan 2019 08:47:04 +0000 (00:47 -0800)]
MLK-20784-1 TMU: Add a interface for TMU arch level initialization

Since imx8mm TMU needs to load some registers from fuse, this is arch
dependent operation and may vary on different platforms. So add
a interface for arch level initialization.

Signed-off-by: Ye Li <ye.li@nxp.com>
(cherry picked from commit 1a66350d1baeef355d51da609c505aebb233cefb)

5 years agoMLK-18341-1 thermal: nxp_tmu: Add support for thermal sensor on iMX8MM
Ye Li [Fri, 18 May 2018 02:21:46 +0000 (19:21 -0700)]
MLK-18341-1 thermal: nxp_tmu: Add support for thermal sensor on iMX8MM

The analog sensors on iMX8MM are new, used for 14LPP process. So the
Temperature Sensor Monitoring Unit (TMU) has some change accordingly.
We use version 2 in TMU driver to represent the new TMU, so the one driver
can service for both MQ and MM.

Signed-off-by: Ye Li <ye.li@nxp.com>
(cherry picked from commit 86fab918e607221249acc5c9a8f8eca01c17ec37)
(cherry picked from commit f68b8d4e0b256d376fd4ea8f1f4be0334327eca1)

5 years agoMLK-18322 HAB: Add suport for iMX8MM
Ye Li [Thu, 18 Apr 2019 07:20:40 +0000 (00:20 -0700)]
MLK-18322 HAB: Add suport for iMX8MM

The imx8mm has changed the address of rvt_hab, use new address for imx8mm.

The authentication procedure is same as imx8mq. In u-boot, the authentication
uses SIP call to trap ATF to run HAB authenticate.
Need to work with ATF commit:
(commit 7a4d6f90e999ed413d520310cc199901b52b7a04)

Users need to add CONFIG_SECURE_BOOT=y to defconfig to enable the feature.

Signed-off-by: Ye Li <ye.li@nxp.com>
Acked-by: Peng Fan <peng.fan@nxp.com>
5 years agoMLK-17997: video: imx: hdp: Add HDMI RX firmware loading
Sandor Yu [Mon, 9 Apr 2018 03:29:47 +0000 (11:29 +0800)]
MLK-17997: video: imx: hdp: Add HDMI RX firmware loading

Add iMX8QM HDMI RX firmware loading.

Signed-off-by: Sandor Yu <Sandor.yu@nxp.com>
(cherry picked from commit 969d93660e1b34765627ff9f6f8dcab08dcd1250)
(cherry picked from commit 43bc31a2700f4078ffc5f55ff3fea0ed554f4506)

5 years agoMLK-18316 imx8: Fix USBOTG disconnect when XRDC blocking is enabled
Ye Li [Tue, 15 May 2018 09:03:19 +0000 (02:03 -0700)]
MLK-18316 imx8: Fix USBOTG disconnect when XRDC blocking is enabled

We clean USBOTG register USBCMD if it is used in serial download mode.
When XRDC blocking is enabled, we can't write this register directly,
must enable the OTG power, otherwise the kernel will get SError
exception in mfgtool.

Signed-off-by: Ye Li <ye.li@nxp.com>
(cherry picked from commit cf2143dc97b2a8f21b828c7386c59ee965d981f2)
(cherry picked from commit 55dd2c8e17eca27b3d393be9fd84a153ce7bf37b)

5 years agoMLK-18290-4 usb: ci_udc: Add function to remove usb device
Ye Li [Fri, 11 May 2018 08:32:39 +0000 (01:32 -0700)]
MLK-18290-4 usb: ci_udc: Add function to remove usb device

When unregister gadget driver in ci_udc, the usb device is not
removed or stop. This causes next "usb start" fails to work.

Add a new interface "usb_remove_ehci_gadget" in usb-uclass to
remove the usb device for DM driver. Using "usb_lowlevel_stop" for
non-DM driver.

Signed-off-by: Ye Li <ye.li@nxp.com>
(cherry picked from commit c73d137e0a4a613a49f6f63ec63332508afb88c0)
(cherry picked from commit c2f83b46c21e738fa8176a8f83661bf3603067f8)

5 years agoMLK-18290-3 imx8mm: Add SOC level support for OTG USB
Ye Li [Thu, 18 Apr 2019 07:09:24 +0000 (00:09 -0700)]
MLK-18290-3 imx8mm: Add SOC level support for OTG USB

Enable the OTG power on for EHCI OTG controller on iMX8MM

Signed-off-by: Ye Li <ye.li@nxp.com>
5 years agoMLK-18290-2 ehci-mx6: Update driver to support i.MX8MM
Ye Li [Fri, 11 May 2018 08:26:30 +0000 (01:26 -0700)]
MLK-18290-2 ehci-mx6: Update driver to support i.MX8MM

Since the i.MX8MM reuses the otg controllers on i.MX7D. We can use
CONFIG_USB_EHCI_MX7 for them.

Due the TCPC and load switch are used on Typec circuit. Add the
board_usb_init and board_usb_cleanup to ehci-mx6 DM driver. So
we can implement the TCPC settings in these board functions.

Signed-off-by: Ye Li <ye.li@nxp.com>
(cherry picked from commit 67699e88aed8bd36e919f54f9555ae15595faaf7)
(cherry picked from commit 13c19c6214ebb69706bb6710ac3ff1db5c2de185)

5 years agoMLK-18243-17: arm: mach-imx: fix sd to mmc1 and emmc to mmc2 for i.MX8MM
Ye Li [Wed, 9 May 2018 06:07:27 +0000 (14:07 +0800)]
MLK-18243-17: arm: mach-imx: fix sd to mmc1 and emmc to mmc2 for i.MX8MM

Since the SD is usdhc2 and eMMC is usdhc3, this cause mapping problem
for spl_boot_device. So far hard coded them to correct MMC index, so
that SD and eMMC boot can work.

Signed-off-by: Ye Li <ye.li@nxp.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
(cherry picked from commit 3e4123fb3d6945a466563f4505a56bb0ea6f4ca0)
(cherry picked from commit 6c6206ad1a5a3c70118a3ecf526b0ff00088adf7)

5 years agoMLK-18243-16 arm: dts: add imx8mm dtsi and binding files
Peng Fan [Wed, 9 May 2018 05:59:54 +0000 (13:59 +0800)]
MLK-18243-16 arm: dts: add imx8mm dtsi and binding files

Sync dts from 4.14.98 Linux kernel commit
e88899128d81ea8b82dfd7d294572f21c388e568
("MLK-21424 can: flexcan: fix normal CAN can't receive
 remote frame after setting fd mode").

Signed-off-by: Peng Fan <peng.fan@nxp.com>
(cherry picked from commit 0e7a5484878338d0dff871b6d21092a0479f07b4)
(cherry picked from commit 54a97f31802ea568e04285e0f18689811eecedcb)
Signed-off-by: Ye Li <ye.li@nxp.com>
5 years agoMLK-18243-15 spi: fspi: Fix 3bytes access
Ye Li [Thu, 18 Apr 2019 06:33:38 +0000 (23:33 -0700)]
MLK-18243-15 spi: fspi: Fix 3bytes access

The address size is 3 bytes when using BAR mode, but the driver
supposes the 4bytes mode is enabled and uses 4 bytes address. This
cause address issue.

Signed-off-by: Ye Li <ye.li@nxp.com>
5 years agoMLK-18243-14 spi: fspi: init unused LUT to 0
Peng Fan [Wed, 9 May 2018 05:52:20 +0000 (13:52 +0800)]
MLK-18243-14 spi: fspi: init unused LUT to 0

If not initialize unused LUT to 0, the value is random which might
cause qspi command failure.

On i.MX8QM/QXP, it works ok because ROM inittialize them to 0, but on
i.MX8MM, ROM not initialize them, so let's do it here.

Signed-off-by: Peng Fan <peng.fan@nxp.com>
(cherry picked from commit c4bd49c7b249073415f052fb28cd5a4ad374a318)
(cherry picked from commit bce752cfe43b3378dbd84b64d46f1d1818e0300f)

5 years agoMLK-18243-12 arm: imx8m: add cpu revision for i.MX8MM
Peng Fan [Wed, 9 May 2018 05:48:59 +0000 (13:48 +0800)]
MLK-18243-12 arm: imx8m: add cpu revision for i.MX8MM

Add cpu revision for i.MX8MM
Add helper function

Signed-off-by: Peng Fan <peng.fan@nxp.com>
(cherry picked from commit 5fdfc7d73157a5eb9254b43f65edd1bb5f13fd16)
(cherry picked from commit e33f0d97423c29e130ddecbd8d7567ca8bf1826d)

5 years agoMLK-18243-11 arm: imx8m: get cpu chip id for i.MX8MM
Ye Li [Wed, 9 May 2018 05:37:22 +0000 (13:37 +0800)]
MLK-18243-11 arm: imx8m: get cpu chip id for i.MX8MM

Using the cpu chip id in DIGPROG registers 0x824110

Signed-off-by: Ye Li <ye.li@nxp.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
(cherry picked from commit 330051b4fbcdda201b67a94e72345ec47bd85a96)
(cherry picked from commit dcc51cc1fe69b0dcdc6a6e4f76743f927fe49b47)

5 years agoMLK-18243-10 arm: imx8m: soc: enable SCTR clock before timer init
Peng Fan [Wed, 9 May 2018 07:50:07 +0000 (15:50 +0800)]
MLK-18243-10 arm: imx8m: soc: enable SCTR clock before timer init

To i.MX8MM SCTR clock is disabled by ROM, so before timer init
need to enable it.
To i.MX8MQ, it does not hurt the clock is enabled again.

Signed-off-by: Peng Fan <peng.fan@nxp.com>
(cherry picked from commit 2347541e9e79f833716fe88f7de860b5704b5e48)
(cherry picked from commit f5cfd600cfa58d4a768e94c29aaa890fd39a2d63)

5 years agoMLK-20051-2 imx8mq_arm2: Split DDR to two banks for iMX8MQ DDR4 ARM2
Ye Li [Thu, 1 Nov 2018 01:37:45 +0000 (18:37 -0700)]
MLK-20051-2 imx8mq_arm2: Split DDR to two banks for iMX8MQ DDR4 ARM2

Because the iMX8MQ DDR4 ARM2 has 4GB DDR. To fit new MMU settings, we split
it to two banks. The first bank has 3GB DDR, reach to 4GB memory map.
The second bank has 1GB DDR, is beyond 4GB memory map.

Notice: there is no OPTEE for ARM2 board. The trust zone setting in OPTEE
for iMX8MQ EVK is not match with DDR size on ARM2 board. So ARM2 Only can
work without OPTEE.

Signed-off-by: Ye Li <ye.li@nxp.com>
Reviewed-by: Peng Fan <peng.fan@nxp.com>
5 years agoMLK-20051-1 imx8m: Fix MMU table issue for OPTEE memory
Ye Li [Thu, 18 Apr 2019 06:00:01 +0000 (23:00 -0700)]
MLK-20051-1 imx8m: Fix MMU table issue for OPTEE memory

When running with OPTEE, the MMU table in u-boot does not remove the OPTEE
memory from its settings. So ARM speculative prefetch in u-boot may access
that OPTEE memory. Due to trust zone is enabled by OPTEE and that memory
is set to secure access, then the speculative prefetch will fail and cause
various memory issue in u-boot.
The fail address register and int_status register in trustzone has logged
that speculative access from u-boot.

Signed-off-by: Peng Fan <peng.fan@nxp.com>
Signed-off-by: Ye Li <ye.li@nxp.com>
5 years agoMLK-18243-5 arm: mach-imx: iomux: add i.MX8MM PE property
Ye Li [Wed, 9 May 2018 03:21:46 +0000 (11:21 +0800)]
MLK-18243-5 arm: mach-imx: iomux: add i.MX8MM PE property

i.MX8MM has a PE property, it does not have LVTTL

Signed-off-by: Ye Li <ye.li@nxp.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
(cherry picked from commit 96169c1f2de1ac90a244166ab5bb6c874cdfd6bd)
(cherry picked from commit 6ccb18feb68644292ed0df02588e775ffa942872)

5 years agoMLK-18243-4 arm: imx8m: add pin header for i.MX8MM
Peng Fan [Wed, 9 May 2018 03:16:50 +0000 (11:16 +0800)]
MLK-18243-4 arm: imx8m: add pin header for i.MX8MM

Add pin header file for i.MX8MM

To IMX8MM_PAD_NAND_WE_B_USDHC3_CLK, IOMUX_CONFIG_SION needs to be
selected.

Signed-off-by: Peng Fan <peng.fan@nxp.com>
(cherry picked from commit e5d3d27f79df1c3a2b8261929e666e6bfdb69abc)
(cherry picked from commit 1127c7f41b967cffd041905254adb426bde0c48a)

5 years agoMLK-21845-3 imx8m: Append HDMI Firmware only for imx8mq
Ye Li [Mon, 22 Apr 2019 08:32:21 +0000 (01:32 -0700)]
MLK-21845-3 imx8m: Append HDMI Firmware only for imx8mq

Since only i.MX8MQ needs HDMI firmware packing into boot image. Change
the imximage.cfg with CONFIG_IMX8MQ used.

Signed-off-by: Ye Li <ye.li@nxp.com>
5 years agoMLK-21845-2 imx8mm: add clk support for i.MX8MM
Ye Li [Thu, 18 Apr 2019 05:46:02 +0000 (22:46 -0700)]
MLK-21845-2 imx8mm: add clk support for i.MX8MM

Introduce clk implementation for i.MX8MM, including pll configuration,
pll decoding, ccm configuration.

Signed-off-by: Peng Fan <peng.fan@nxp.com>
Signed-off-by: Ye Li <ye.li@nxp.com>
5 years agoMLK-21845-1 imx8mq: restructure clock
Ye Li [Thu, 18 Apr 2019 03:37:43 +0000 (20:37 -0700)]
MLK-21845-1 imx8mq: restructure clock

i.MX8MQ and i.MX8MM use different analog pll design, but they
share same ccm design. Add clock_imx8mq.h for i.MX8MQ and
keep common part in clock.h

Signed-off-by: Peng Fan <peng.fan@nxp.com>
Signed-off-by: Ye Li <ye.li@nxp.com>
5 years agoMLK-18243-2 arm: imx8m: add regs definition for i.MX8MM
Ye Li [Thu, 18 Apr 2019 02:33:15 +0000 (19:33 -0700)]
MLK-18243-2 arm: imx8m: add regs definition for i.MX8MM

Move the original imx-regs.h to imx-regs-imx8mq.h
Introduce a new file dedicated for i.MX8MM
Use CONFIG_IMX8MQ/M to choose the header in imx-regs.h

Signed-off-by: Peng Fan <peng.fan@nxp.com>
Signed-off-by: Ye Li <ye.li@nxp.com>
5 years agoMLK-18243-1 arm: Kconfig: Add i.MX8MM Kconfig entry
Peng Fan [Wed, 9 May 2018 02:55:02 +0000 (10:55 +0800)]
MLK-18243-1 arm: Kconfig: Add i.MX8MM Kconfig entry

i.MX8MM is one variant of i.MX8M, so add IMX8MM entry and
change the previous i.MX8MQ boards to select IMX8MQ.

Signed-off-by: Peng Fan <peng.fan@nxp.com>
(cherry picked from commit f20e63c5e16fe84ece2686c7c16e2ba2f2862378)
(cherry picked from commit b4cd0b4267ec25d6d929df9f278cc248e6ecb133)

5 years agoMLK-18387 pmic: Add pmic driver for BD71837
Ye Li [Wed, 23 May 2018 05:43:00 +0000 (22:43 -0700)]
MLK-18387 pmic: Add pmic driver for BD71837

The iMX8MM EVK board uses BD71837MWV pmic. Add its driver to u-boot.

Signed-off-by: Ye Li <ye.li@nxp.com>
(cherry picked from commit e9a3bec2e95a4b2b4641223c8ee4ebd8da76d7f9)

5 years agoMLK-18384 imx8mq: Add mfgtool boot check
Ye Li [Wed, 23 May 2018 02:00:19 +0000 (19:00 -0700)]
MLK-18384 imx8mq: Add mfgtool boot check

Need to define the is_boot_from_usb, so that u-boot can check whether
it is booted for mfgtool.

Signed-off-by: Ye Li <ye.li@nxp.com>
(cherry picked from commit f3d2fba8183ac9b9274348f9bbe4c85da266984d)

5 years agoMLK-18044-5: crypto: caam: Add secure memory vid 3 support
Aymen Sghaier [Thu, 3 May 2018 12:19:57 +0000 (14:19 +0200)]
MLK-18044-5: crypto: caam: Add secure memory vid 3 support

  In i.MX8M platforms the secure memory block has a newer version
 than those used in i.MX6/7 platforms, this patch update the driver
 to use the correct registers offsets.

Signed-off-by: Aymen Sghaier <aymen.sghaier@nxp.com>
(cherry picked from commit 0daa9c142a1809b473d101f6db9700574ce3d522)
(cherry picked from commit 7184ec3307c4347a7a7765ba72415580cd65c341)

5 years agoMLK-18044-4: crypto: caam: Fix pointer size to 32bit for i.MX8M
Aymen Sghaier [Thu, 3 May 2018 10:17:53 +0000 (12:17 +0200)]
MLK-18044-4: crypto: caam: Fix pointer size to 32bit for i.MX8M

  The CAAM block used in i.MX8M is 32 bits address size but when the flag
 PHYS_64BIT is enabled for armv8, the CAAM driver will try to use a
 wrong pointer size.
  This patch fixes this issue.

Signed-off-by: Aymen Sghaier <aymen.sghaier@nxp.com>
(cherry picked from commit 9e1ef25c121240cf3f6a142058e617dd7df0030b)
(cherry picked from commit 1733916155ca319be754c45861130df327d8a57d)

5 years agoMLK-18044-3: crypto: Add blob command support for i.MX8M platforms
Aymen Sghaier [Wed, 2 May 2018 08:34:27 +0000 (10:34 +0200)]
MLK-18044-3: crypto: Add blob command support for i.MX8M platforms

 This patch enable blob command for mScale platforms.

Signed-off-by: Aymen Sghaier <aymen.sghaier@nxp.com>
(cherry picked from commit 895669394f6aae633abf6ea3f327d6093562edde)
(cherry picked from commit bac8ed98778c93ef43ce9093efa3b9999d650576)

5 years agoMLK-18044-2: crypto: caam: Fix build warnings pointer casting
Aymen Sghaier [Tue, 1 May 2018 16:42:21 +0000 (18:42 +0200)]
MLK-18044-2: crypto: caam: Fix build warnings pointer casting

  Enabling CAAM driver for i.MX8M platforms, a 64 bits architecture,
 lead to casting warnings: from/to pointer to/from integer with
 different size. This patch fix these warnings

Signed-off-by: Aymen Sghaier <aymen.sghaier@nxp.com>
(cherry picked from commit d02fbc7d2957f4788ced017ccf17fd35ab968121)
(cherry picked from commit 79e90af14af32e86fd7efd3c84e09a0bd572ab5b)

5 years agoMLK-18044-1: crypto: caam: Add CAAM support to i.MX8M platforms
Aymen Sghaier [Tue, 1 May 2018 09:37:22 +0000 (11:37 +0200)]
MLK-18044-1: crypto: caam: Add CAAM support to i.MX8M platforms

  This patch enable CAAM support for i.MX8M platforms.

Signed-off-by: Aymen Sghaier <aymen.sghaier@nxp.com>
(cherry picked from commit 1fc92e6e34b06bdee81240ce06326aca1d9c02d8)
(cherry picked from commit b0f889b77b367b69aa0778b1d03a2ec30fdee243)

5 years agoMLK-18043 4/4: imx7ulp: Enable support for cmd blob
Franck LENORMAND [Wed, 2 May 2018 10:01:56 +0000 (12:01 +0200)]
MLK-18043 4/4: imx7ulp: Enable support for cmd blob

Signed-off-by: Franck LENORMAND <franck.lenormand@nxp.com>
(cherry picked from commit 03cb71cd22ff6ab1e2005a4bcb183c65f49b6e4e)
(cherry picked from commit 5d797db35da7fa3f7971df8811b6a1e800786ced)

5 years agoMLK-18043 2/4: caam: enable support for iMX7ULP
Franck LENORMAND [Wed, 2 May 2018 10:16:36 +0000 (12:16 +0200)]
MLK-18043 2/4: caam: enable support for iMX7ULP

Signed-off-by: Franck LENORMAND <franck.lenormand@nxp.com>
(cherry picked from commit 5f429339ab5a5f0d1f00a2880ce05cf479f11ba5)
(cherry picked from commit c26db25a130ef1f0abd43a3133877f74431eb217)

5 years agoMLK-18043 1/4: caam: change JR running loop
Franck LENORMAND [Wed, 2 May 2018 10:16:25 +0000 (12:16 +0200)]
MLK-18043 1/4: caam: change JR running loop

Signed-off-by: Franck LENORMAND <franck.lenormand@nxp.com>
(cherry picked from commit 6503c1ed0264fc61945c2b3b5948b793f104fc32)
(cherry picked from commit d1d10af5a15473377fefbca56c1dac293fb06f3b)

5 years agoMLK-18172-2: imx8: Introduce xenlinux_addr for dom0 kernel
Leonard Crestez [Fri, 27 Apr 2018 11:49:01 +0000 (14:49 +0300)]
MLK-18172-2: imx8: Introduce xenlinux_addr for dom0 kernel

Avoid overloading variable meaning and overlap with hdp_addr.

Signed-off-by: Leonard Crestez <leonard.crestez@nxp.com>
(cherry picked from commit 9e4b3e80b4b2f40bb35a2c8f1b5ede9868788a34)
(cherry picked from commit 7dea0df52f0100f2b363b79d793f271fac5b6fab)

5 years agoMLK-18172-1: imx8qm_mek: Load hdmi fw before dom0 kernel for xenboot
Leonard Crestez [Thu, 26 Apr 2018 19:17:45 +0000 (22:17 +0300)]
MLK-18172-1: imx8qm_mek: Load hdmi fw before dom0 kernel for xenboot

Xen boot writes ${filesize} to /chosen/module but if HDMI fw is present
then the size of hdmitxfw.bin is written instead of the size of dom0
kernel.

Signed-off-by: Leonard Crestez <leonard.crestez@nxp.com>
(cherry picked from commit 8823d52896aedf740899c59db3da73f8478c40ed)
(cherry picked from commit 0efebadd915a999030c2462be9c714db3ae51895)

5 years agoMA-11789 Flash system.img failed due to erasing timeout
guoyin.chen [Sun, 22 Apr 2018 10:41:20 +0000 (18:41 +0800)]
MA-11789 Flash system.img failed due to erasing timeout

Flash system partition with fastboot will earse the partition firstly
The 1.2s timeout will fail on some SD Card.
Enlarge it to 5s to make it works for most of sdcard

Change-Id: I285df411c7a07025251fd19f4c8e8b549bee2421
Signed-off-by: guoyin.chen <guoyin.chen@nxp.com>
(cherry picked from commit 642d77fb6d6412095faa6584eeef7bb0132cae57)
(cherry picked from commit 49ae851ccd00a027d122338d7b39402135041ad8)

5 years agoMLK-18129-2 imx8qxp: Add secure boot environment
Ye Li [Wed, 25 Apr 2018 09:45:32 +0000 (02:45 -0700)]
MLK-18129-2 imx8qxp: Add secure boot environment

Add the secure boot relevant environment variables to ARM2 and MEK.
When CONFIG_AHAB_BOOT is enabled, we will switch to boot in this
new way:
1. Load the OS container to address 0x98000000
2. Using auth_cntr to authenticate the OS container. It will load the
   kernel and FDT to destination address.
3. Using booti to boot kernel.

Signed-off-by: Ye Li <ye.li@nxp.com>
Acked-by: Peng Fan <peng.fan@nxp.com>
(cherry picked from commit 545c8fe112244f20377e97921389c86eacc8551f)
(cherry picked from commit cba5acebbe6be0736f229ef719b1a1e5835e1a5f)

5 years agoMLK-18129-1 imx8: ahab: Add secure boot support
Ye Li [Wed, 25 Apr 2018 09:33:58 +0000 (02:33 -0700)]
MLK-18129-1 imx8: ahab: Add secure boot support

Add function and new command "auth_cntr" for secure boot support.
When booting with life cycle set to OEM closed, we need to use
this function to authenticate the OS container and load kernel & FDT
from OS container to their destination.

Users can set CONFIG_AHAB_BOOT=y to enable the feature. It is not
set at default.

Signed-off-by: Ye Li <ye.li@nxp.com>
(cherry picked from commit 3b0496c267a4c09ff5b790ccf8e4f658af460c6c)
(cherry picked from commit 1c83e14fb8e8ceec99f2c42ee1442596d8ffe517)

5 years agoMLK-18168 imx8qm_arm2: Add xen boot commands
Leonard Crestez [Wed, 29 Nov 2017 12:56:54 +0000 (14:56 +0200)]
MLK-18168 imx8qm_arm2: Add xen boot commands

These commands try to be as generic as possible and avoid duplicating
argument values used for regular boot.

The fdt_file parameter still needs to be set to a .dtb for dom0. Booting
xen with stock dtb would require more hacks, for example lpuart
interrupt-parent = <&wu> needs to be removed.

Signed-off-by: Leonard Crestez <leonard.crestez@nxp.com>
(cherry picked from commit 4c671ef548e2b7f352be184215306766ebc1cc45)
(cherry picked from commit d705425c698a71fb070a04b327248bac00681111)

5 years agoMLK-17998 configs: imx8qxp_mek: add xenmmcboot and xennetboot
Peng Fan [Wed, 11 Apr 2018 07:53:18 +0000 (15:53 +0800)]
MLK-17998 configs: imx8qxp_mek: add xenmmcboot and xennetboot

Add xenmmcboot and xennetboot command.

Signed-off-by: Peng Fan <peng.fan@nxp.com>
(cherry picked from commit 13f791db9a2bd0fe442115e77c91e8a56a4cf217)
(cherry picked from commit f71b5a989d0292e17f8a5ad526ea8faa3d28ea60)

5 years agoMLK-17842 imx8: introduce partition cmd
Peng Fan [Tue, 20 Mar 2018 03:17:53 +0000 (11:17 +0800)]
MLK-17842 imx8: introduce partition cmd

Add partition API cmd support.
And parse dtb support, see
arch/arm64/boot/dts/freescale/fsl-imx8qm-mek-dom0.dts for example.

Signed-off-by: Peng Fan <peng.fan@nxp.com>
(cherry picked from commit f61119240922ad13fd573f16c02361d7228fbd67)
(cherry picked from commit ef488a85953398838c5242aed66c917547f7f918)

5 years agoMLK-21831-5 sci_api: Update SCFW API for partition
Ye Li [Wed, 17 Apr 2019 09:26:12 +0000 (02:26 -0700)]
MLK-21831-5 sci_api: Update SCFW API for partition

Add relevants SCFW APIs used in imx8 partition implementation

Signed-off-by: Ye Li <ye.li@nxp.com>
5 years agoMLK-21831-4 sci_api: Update SCFW API for ahab
Ye Li [Wed, 17 Apr 2019 08:53:36 +0000 (01:53 -0700)]
MLK-21831-4 sci_api: Update SCFW API for ahab

Add relevants SCFW APIs used in AHAB implementation

Signed-off-by: Ye Li <ye.li@nxp.com>
5 years agoMLK-17786: imx8qm: add xen boot script
Peng Fan [Tue, 13 Mar 2018 01:46:39 +0000 (09:46 +0800)]
MLK-17786: imx8qm: add xen boot script

Add XEN boot script to ease booting.

Signed-off-by: Peng Fan <peng.fan@nxp.com>
(cherry picked from commit 2fd813fcfcb82f28f353fbdc727765d6e848bfd7)
(cherry picked from commit 28caf1e050765a040b1290c210b40563b30cca3c)

5 years agoMLK-17143 imx8qxp: mek: add xen env
Peng Fan [Sun, 10 Dec 2017 02:01:30 +0000 (10:01 +0800)]
MLK-17143 imx8qxp: mek: add xen env

Add xen env, "run xenboot" to boot xen.

Signed-off-by: Peng Fan <peng.fan@nxp.com>
(cherry picked from commit d5f756f0de780d6e8db2f43f8388083f9454dcac)
(cherry picked from commit 9c9f854b7a5952f66423145ea18c77a9dfb0a99c)

5 years agoMLK-17919: imx: Avoid hardcoded Job Ring Max size
Breno Lima [Fri, 23 Mar 2018 20:25:15 +0000 (17:25 -0300)]
MLK-17919: imx: Avoid hardcoded Job Ring Max size

Prior instantiating RNG we have to ensure if the CAAM job rings are
available. Avoid hardcoded job ring max size and use the definition at
fsl_sec.h

Signed-off-by: Breno Lima <breno.lima@nxp.com>
Reviewed-by: Ye Li <ye.li@nxp.com>
(cherry picked from commit b81379d3d2228760cc061a56c3c610b4e9fa9783)
(cherry picked from commit f2e2f48f64264f06ed0a20e23903852e0a0fd8f9)

5 years agoMLK-17918: imx: Ensure CAAM clock is enabled prior getting out_jr_size
Breno Lima [Fri, 23 Mar 2018 20:21:38 +0000 (17:21 -0300)]
MLK-17918: imx: Ensure CAAM clock is enabled prior getting out_jr_size

Prior calling sec_in32() we have to ensure CAAM clock is enabled, the
function sec_in32() is reading CAAM registers and if CAAM clock is disabled
the system will hang.

Signed-off-by: Breno Lima <breno.lima@nxp.com>
Reviewed-by: Ye Li <ye.li@nxp.com>
(cherry picked from commit c4da7540e7548f36c6ad90292db7057b828a7956)
(cherry picked from commit e7832c25579cedafe53b1973550664f654d38894)

5 years agoMLK-17917: imx: Avoid hardcoded output ring size register offset (ORSR)
Breno Lima [Fri, 23 Mar 2018 20:17:04 +0000 (17:17 -0300)]
MLK-17917: imx: Avoid hardcoded output ring size register offset (ORSR)

The CAAM output ring size register offset is currently defined in fsl_sec.h
as FSL_CAAM_ORSR_JRa_OFFSET, use this definition to avoid hardcoded value in
i.MX common code.

Signed-off-by: Breno Lima <breno.lima@nxp.com>
Reviewed-by: Ye Li <ye.li@nxp.com>
(cherry picked from commit 9d381708210270a1cf9bf25924c8932c436967e8)
(cherry picked from commit 42db685ace092bedc569cd2c846b51b393630759)

5 years agoMLK-17915: cmd: blob: Instantiate RNG before running CMD_BLOB
Breno Lima [Fri, 23 Mar 2018 20:05:47 +0000 (17:05 -0300)]
MLK-17915: cmd: blob: Instantiate RNG before running CMD_BLOB

U-Boot can instantiate CAAM RNG if needed by crypto operations.
Call sec_init() prior running a blob operation to ensure
RNG is correctly instantiated.

Make sure CAAM clock is enabled and check if a job ring is
available for that operation.

Signed-off-by: Breno Lima <breno.lima@nxp.com>
Reviewed-by: Ye Li <ye.li@nxp.com>
(cherry picked from commit b2b745515d06f05ea288524e26cf2389d6924ff2)
(cherry picked from commit fe33f7296336b7364fb24ad21b0be32eb5810262)

5 years agoMLK-17914: cmd: blob: Add SECURE_BOOT and CAAM supported SoCs as dependency
Breno Lima [Fri, 23 Mar 2018 19:25:33 +0000 (16:25 -0300)]
MLK-17914: cmd: blob: Add SECURE_BOOT and CAAM supported SoCs as dependency

In order to build CMD_BLOB on i.MX CAAM supported devices it's
necessary to select SECURE_BOOT. Add SECURE_BOOT and CAAM supported
SoCs as dependency.

Signed-off-by: Breno Lima <breno.lima@nxp.com>
Reviewed-by: Ye Li <ye.li@nxp.com>
(cherry picked from commit 4afb519038d190687cac6d410e9ad65d0655f1c0)
(cherry picked from commit acda6159eefbefaee15b19568fd1c4665c047432)

5 years agoMLK-18169 imx: imx7 Support for Manufacturing Protection
Breno Lima [Tue, 14 Nov 2017 22:10:45 +0000 (20:10 -0200)]
MLK-18169 imx: imx7 Support for Manufacturing Protection

This code was originally developed by Raul Cardenas <raul.casas@nxp.com>
and modified to be applied in U-Boot imx_v2017.03.

More information about the initial submission can be seen
in the link below:
https://lists.denx.de/pipermail/u-boot/2016-February/245273.html

i.MX7D has an a protection feature for Manufacturing process.
This feature uses asymmetric encryption to sign and verify
authenticated software handled between parties. This command
enables the use of such feature.

The private key is unique and generated once per device.
And it is stored in secure memory and only accessible by CAAM.
Therefore, the public key generation and signature functions
are the only functions available for the user.

The manufacturing-protection authentication process can be used to
authenticate the chip to the OEM's server.

Command usage:

Print the public key for the device.
- mfgprot pubk

Generates Signature over given data.
- mfgprot sign <data_address> <data_size>

Signed-off-by: Raul Ulises Cardenas <raul.casas@nxp.com>
Signed-off-by: Breno Lima <breno.lima@nxp.com>
Reviewed-by: Fabio Estevam <fabio.estevam@nxp.com>
Reviewed-by: Ye Li <ye.li@nxp.com>
(cherry picked from commit db2dbf622d3c711b2fbd85e6814992e023479dad)

5 years agoMLK-16753-1 imx8m: add optee node according to runtime status
Peng Fan [Thu, 2 Nov 2017 05:47:24 +0000 (13:47 +0800)]
MLK-16753-1 imx8m: add optee node according to runtime status

If TEE is enabled according to rom_pointer[0,1] passed to BL33
from ATF, uboot need to add the optee node in dts to let
Kernel could probe the TEE driver.

Signed-off-by: Peng Fan <peng.fan@nxp.com>
(cherry picked from commit 4341d35cc99a6226f2671e15154d3148ebf49c47)

5 years agoMLK-17404-7 config: Adding support for i.MX8M HDMI splash
Oliver Brown [Mon, 29 Jan 2018 20:25:58 +0000 (14:25 -0600)]
MLK-17404-7 config: Adding support for i.MX8M HDMI splash

Adding support for i.MX8M HDMI splash screen

Signed-off-by: Oliver Brown <oliver.brown@nxp.com>
(cherry picked from commit d8aac64be7d83847cff8a2235e1be677964a1a9d)
(cherry picked from commit a84dedb3516d14f947dccf6a5e177b00d0509017)

5 years agoMLK-17404-6 video: Adding i.MX8 HDMI support to configuration
Oliver Brown [Mon, 29 Jan 2018 20:23:56 +0000 (14:23 -0600)]
MLK-17404-6 video: Adding i.MX8 HDMI support to configuration

Adding i.MX8 HDMI support to build configuration.

Signed-off-by: Oliver Brown <oliver.brown@nxp.com>
(cherry picked from commit 2edbbe51df0817876b943760a272640e3c9bff70)
(cherry picked from commit 0857e58bc7eaea5da6cdde14cdd8f72293a83273)

5 years agoMLK-17404-5 board: freescale: Adding support for i.MX8MQ EVK display
Ye Li [Fri, 20 Apr 2018 06:37:28 +0000 (23:37 -0700)]
MLK-17404-5 board: freescale: Adding support for i.MX8MQ EVK display

Adding basic display support for splash screen.

Signed-off-by: Oliver Brown <oliver.brown@nxp.com>
Signed-off-by: Ye Li <ye.li@nxp.com>
(cherry picked from commit 31058332c273d181390560ae2b88209b628f5842)
(cherry picked from commit 367e8bdaf391dc83455ef2e7e28d6dbf00dbbe2c)

5 years agoMLK-17404-4 arch: arm: Adding support for i.MX8M display
Ye Li [Fri, 20 Apr 2018 06:28:13 +0000 (23:28 -0700)]
MLK-17404-4 arch: arm: Adding support for i.MX8M display

Adding Display support for splash screen.

Signed-off-by: Oliver Brown <oliver.brown@nxp.com>
Signed-off-by: Ye Li <ye.li@nxp.com>
(cherry picked from commit 8733549494dfa18d9317d76a8fabad8b41e6dcf6)
(cherry picked from commit 18d86b0683b57da3b778cc15ee214bd92ee35ac2)

5 years agoMLK-17404-3 arch: arm: Adding support for i.MX8M HDMI
Oliver Brown [Mon, 29 Jan 2018 20:11:06 +0000 (14:11 -0600)]
MLK-17404-3 arch: arm: Adding support for i.MX8M HDMI

Adding basic HDMI support for i.MX8M for splash screen.

Signed-off-by: Oliver Brown <oliver.brown@nxp.com>
(cherry picked from commit 390a15a523c5aefb99008a5f67d6c8f9ec7ef63c)
(cherry picked from commit d0569911819015931bc26686aa854a62e04fc530)

5 years agoMLK-17404-2 include: Adding header for i.MX8 HDMI support
Oliver Brown [Mon, 29 Jan 2018 20:08:44 +0000 (14:08 -0600)]
MLK-17404-2 include: Adding header for i.MX8 HDMI support

Adding header file for i.MX8 HDMI support

Signed-off-by: Oliver Brown <oliver.brown@nxp.com>
(cherry picked from commit 42f1298bdeac2e37d30ce4c1265696b3c46fb107)
(cherry picked from commit 80a5d47915da7bcb87fa40b80051d687a0f756b4)

5 years agoMLK-17404-1 video: imx: hdp: Adding support for HDMI splash screen
Oliver Brown [Mon, 29 Jan 2018 20:05:05 +0000 (14:05 -0600)]
MLK-17404-1 video: imx: hdp: Adding support for HDMI splash screen

Adding HDMI support for splash screen.

Signed-off-by: Oliver Brown <oliver.brown@nxp.com>
(cherry picked from commit a95018c0d3b0ac8f0893b9408ae598324de4a530)
(cherry picked from commit 485bccc47619fdaf967f893dd672ccc9d375a8af)

5 years agoMLK-17205-3 video: imx: hdp: Adding configs for HDP firmware loading
Oliver Brown [Wed, 13 Dec 2017 23:14:58 +0000 (17:14 -0600)]
MLK-17205-3 video: imx: hdp: Adding configs for HDP firmware loading

Added default environment for hdp loading.
Added hdp loading as default for iMX8QM ARM2 and MEK boards.

Signed-off-by: Oliver Brown <oliver.brown@nxp.com>
(cherry picked from commit 432e5b4347841095c3b5a8a0d106f35deadd006e)
(cherry picked from commit cb78a32fa8eb8c37932be003ebe4fa1f8c46c1d5)

5 years agoMLK-17205-2 video: imx: hdp: Adding HDP firmware loading to the build
Oliver Brown [Wed, 13 Dec 2017 23:12:45 +0000 (17:12 -0600)]
MLK-17205-2 video: imx: hdp: Adding HDP firmware loading to the build

Adding the HDP firmware loading to the build.

Signed-off-by: Oliver Brown <oliver.brown@nxp.com>
(cherry picked from commit d4702582552aa1921fce58329ae40c24481fba3a)

5 years agoMLK-17205-1 video: imx: hdp: Adding support for HDP firmware loading
Oliver Brown [Wed, 13 Dec 2017 23:09:14 +0000 (17:09 -0600)]
MLK-17205-1 video: imx: hdp: Adding support for HDP firmware loading

This adds a command to load the HDP firmware and supporting libraries.

Signed-off-by: Oliver Brown <oliver.brown@nxp.com>
(cherry picked from commit 94c9c7022c5b6cffb74bb2e34336cffc2790d0d8)

5 years agoMLK-18161-19 imx8qxp_arm2: Enable HS400 Enhanced mode and SDR104 for ARM2 board
Ye Li [Wed, 17 Apr 2019 06:05:05 +0000 (23:05 -0700)]
MLK-18161-19 imx8qxp_arm2: Enable HS400 Enhanced mode and SDR104 for ARM2 board

There is eMMC5.1 chip on imx8qxp ARM2 board, enable the HS400 Enhanced
mode(ES) support and SD3.0 SDR104 support.

Signed-off-by: Ye Li <ye.li@nxp.com>
5 years agoMLK-16034-03: support NAND on i.MX8QXP ARM2 board
Ye Li [Wed, 18 Apr 2018 10:34:31 +0000 (03:34 -0700)]
MLK-16034-03: support NAND on i.MX8QXP ARM2 board

NAND module is pin conflict with SD/eMMC on i.MX8QXP ARM2 board,
add new config to disable SD/eMMC when booting from NAND.

Signed-off-by: Han Xu <han.xu@nxp.com>
Signed-off-by: Ye Li <ye.li@nxp.com>
(cherry picked from commit 0a498b1f591fb00a0187650446e0de81dd3453ad)

5 years agoMLK-21844-2 MXS_NAND: Add clock support for iMX8
Ye Li [Wed, 17 Apr 2019 02:59:56 +0000 (19:59 -0700)]
MLK-21844-2 MXS_NAND: Add clock support for iMX8

Since iMX8 has enabled clock uclass, we can parse the clocks from DTB
and enable them in GPMI driver.

Signed-off-by: Ye Li <ye.li@nxp.com>
5 years agoMLK-21844-1 DTS: fsl-imx8dx: Add APBH-DMA clock to GPMI node
Ye Li [Wed, 17 Apr 2019 02:56:12 +0000 (19:56 -0700)]
MLK-21844-1 DTS: fsl-imx8dx: Add APBH-DMA clock to GPMI node

Since the apbh-dma driver does not support DM, we add its clock to
GPMI node. So when GPMI is probed, its driver can help to enable
the clock.

Signed-off-by: Ye Li <ye.li@nxp.com>
5 years agoMLK-16034-02: enable GPMI NAND driver for i.MX8
Han Xu [Wed, 19 Jul 2017 16:43:08 +0000 (11:43 -0500)]
MLK-16034-02: enable GPMI NAND driver for i.MX8

enable the GPMI NAND driver for i.MX8, the major changes

- register defination for i.mx8
- Makefile change for misc.c
- DMA structure must be 32bit address

Signed-off-by: Han Xu <han.xu@nxp.com>
(cherry picked from commit 474c4270108551647c7064a23abdc2e11d7f37ab)
(cherry picked from commit 029cce25cce94c30dd0305bb9b17ba7f939ee1af)

5 years agoMLK-16034-01: change the maximum oob size
Han Xu [Wed, 19 Jul 2017 16:40:59 +0000 (11:40 -0500)]
MLK-16034-01: change the maximum oob size

change the maximum oob size to 1872 for the MT29F128G08CBCCB raw NAND
chip on i.MX8QXP

Signed-off-by: Han Xu <han.xu@nxp.com>
(cherry picked from commit 58dee6ac965000f93693c36c93fb3d453293c43f)
(cherry picked from commit d297f33f4719502aa415dc7c7002c437a6af6c28)

5 years agoMLK-17153-2 imx8mq_arm2: Resolve malloc memory problem for SPL NAND boot
Ye Li [Tue, 12 Dec 2017 06:02:19 +0000 (00:02 -0600)]
MLK-17153-2 imx8mq_arm2: Resolve malloc memory problem for SPL NAND boot

When using SPL NAND boot, the required malloc memory is larger than SD/eMMC boot.
Since we have used out OCRAM (for ATF) and OCRAM_S, there is no enough memory could be
allocated.

The solution is moving the malloc memory pool to DDR. The malloc pool is initialized in
board_init_r, so we moved the VDD_DRAM adjustment and DDR init to board_init_f. Then the DDR
can setup before memory pool initialization. Because the i2c and PMIC driver needs to malloc
data, this change has to enable malloc_f pool by setting CONFIG_MALLOC_F_ADDR to previous
malloc address on OCRAM_S.

Signed-off-by: Ye Li <ye.li@nxp.com>
Reviewed-by: Peng Fan <peng.fan@nxp.com>
(cherry picked from commit ab1895fb5e38d216ff86de292a44776099d159ab)
(cherry picked from commit de084efbc91de971f10c745bbc94f960cdb813a9)

5 years agoMLK-17153-1 nand: Update SPL MXS NAND mini driver
Ye Li [Tue, 16 Apr 2019 09:47:18 +0000 (02:47 -0700)]
MLK-17153-1 nand: Update SPL MXS NAND mini driver

Update the mini driver to add support for getting ecc info from ONFI and
support read image data from page unaligned NAND address.

Signed-off-by: Ye Li <ye.li@nxp.com>
5 years agoMLK-17109-4 imx8mq_arm2: Add i.MX8MQ DDR3L and DDR4 ARM2 boards support
Ye Li [Wed, 15 Nov 2017 06:58:18 +0000 (00:58 -0600)]
MLK-17109-4 imx8mq_arm2: Add i.MX8MQ DDR3L and DDR4 ARM2 boards support

Add board codes, configurations, DTS and DDR initialization codes for the
DDR3L and DDR4 ARM2 boards.

Supported modules
    - DDR3L ARM2: Two RANK DDR3L, QSPI B, eMMC/SD, RMII ENET, UART.
    - DDR4  ARM2: Two RANK DDR4, SD, NAND, RGMII ENET, UART.
      NAND read/write/erase is ok in u-boot, NAND SPL boot will be tested later
      when tool is ready.

Signed-off-by: Ye Li <ye.li@nxp.com>
Acked-by: Peng Fan <peng.fan@nxp.com>
(cherry picked from commit 3c3a9e90faa2c3bbb1bcd9adac5915cfd6f2f204)

5 years agoMLK-17109-3 mxs_nand: Add support for i.MX8M
Ye Li [Thu, 16 Nov 2017 04:58:05 +0000 (22:58 -0600)]
MLK-17109-3 mxs_nand: Add support for i.MX8M

Update the gpmi/apbh_dma/bch drivers and relevant registers for i.MX8M.

Signed-off-by: Ye Li <ye.li@nxp.com>
(cherry picked from commit 6cb839cabb42b81e37214e00448fc5dac89fd1f1)
(cherry picked from commit 468509f86a2d040398aa6b019bb6644bfb0ef11c)

5 years agoMLK-12693-2 nand: mxs: correct bitflip for erased NAND page
Peng Fan [Sat, 7 May 2016 08:58:24 +0000 (16:58 +0800)]
MLK-12693-2 nand: mxs: correct bitflip for erased NAND page

This patch is a porting of
http://git.freescale.com/git/cgit.cgi/imx/linux-2.6-imx.git/commit/?h=imx_4.1.15_1.0.0_ga&id=e4dacc44d22e9474ec456cb330df525cd805ea38
"
i.MX6QP and i.MX7D BCH module integrated a new feature to detect the
bitflip number for erased NAND page. So for these two platform, set the
erase threshold to gf/2 and if bitflip detected, GPMI driver will
correct the data to all 0xFF.

Also updated the imx6qp dts file to ditinguish the GPMI module for i.MX6Q
with the one for i.MX6QP.
"

In this patch, i.MX6UL is added and threshold changed to use ecc_strength.

Signed-off-by: Peng Fan <peng.fan@nxp.com>
(cherry picked from commit 489929be0221bb7d4c46bb5bc6083650b78f73e0)
Signed-off-by: Ye Li <ye.li@nxp.com>
(cherry picked from commit 37d7f9614aa357f270312d7ceaab0f7006dc5aea)
(cherry picked from commit 5f50a850dd42d28b6105ee7e1b4b1822e7ba569b)

5 years agoMLK-12693-1 nand: mxs: fix the bitflips for erased page when uncorrectable error
Peng Fan [Fri, 6 May 2016 04:13:38 +0000 (12:13 +0800)]
MLK-12693-1 nand: mxs: fix the bitflips for erased page when uncorrectable error

This patch is porting from linux:
http://git.freescale.com/git/cgit.cgi/imx/linux-2.6-imx.git/commit/?h=imx_4.1.15_1.0.0_ga&id=3d42fcece496224fde59f9343763fb2dfc5b0768

"
We may meet the bitflips in reading an erased page(contains all 0xFF),
this may causes the UBIFS corrupt, please see the log from Elie:

-----------------------------------------------------------------
[    3.831323] UBI warning: ubi_io_read: error -74 (ECC error) while reading 16384 bytes from PEB 443:245760, read only 16384 bytes, retry
[    3.845026] UBI warning: ubi_io_read: error -74 (ECC error) while reading 16384 bytes from PEB 443:245760, read only 16384 bytes, retry
[    3.858710] UBI warning: ubi_io_read: error -74 (ECC error) while reading 16384 bytes from PEB 443:245760, read only 16384 bytes, retry
[    3.872408] UBI error: ubi_io_read: error -74 (ECC error) while reading 16384 bytes from PEB 443:245760, read 16384 bytes
...
[    4.011529] UBIFS error (pid 36): ubifs_recover_leb: corrupt empty space LEB 27:237568, corruption starts at 9815
[    4.021897] UBIFS error (pid 36): ubifs_scanned_corruption: corruption at LEB 27:247383
[    4.030000] UBIFS error (pid 36): ubifs_scanned_corruption: first 6569 bytes from LEB 27:247383
-----------------------------------------------------------------

This patch does a check for the uncorrectable failure in the following steps:

   [0] set the threshold.
       The threshold is set based on the truth:
       "A single 0 bit will lead to gf_len(13 or 14) bits 0 after the BCH
        do the ECC."

        For the sake of safe, we will set the threshold with half the gf_len, and
        do not make it bigger the ECC strength.

   [1] count the bitflips of the current ECC chunk, assume it is N.

   [2] if the (N <= threshold) is true, we continue to read out the page with
       ECC disabled. and we count the bitflips again, assume it is N2.
       (We read out the whole page, not just a chunk, this makes the check
        more strictly, and make the code more simple.)

   [3] if the (N2 <= threshold) is true again, we can regard this is a erased
       page. This is because a real erased page is full of 0xFF(maybe also has
       several bitflips), while a page contains the 0xFF data will definitely
       has many bitflips in the ECC parity areas.

   [4] if the [3] fails, we can regard this is a page filled with the '0xFF'
       data.
"

Signed-off-by: Peng Fan <peng.fan@nxp.com>
(cherry picked from commit ceb324a2914487aa517a6c70a06a20b5e3438fda)
(cherry picked from commit 026751697e41c7376414a8716cf0ea4bf998b85f)
(cherry picked from commit 93b481f07b8cb59c733f420bebea77ac484f9036)

5 years agoMLK-12601: mtd: gpmi: provide the option to use legacy bch geometry
Ye Li [Tue, 16 Apr 2019 04:06:04 +0000 (21:06 -0700)]
MLK-12601: mtd: gpmi: provide the option to use legacy bch geometry

Provide an option in DT to use legacy bch geometry, which compatible
with the 3.10 kernel bch setting. To enable the feature, adding
"fsl,legacy-bch-geometry" under gpmi-nand node.

NOTICE: The feature must be enabled/disabled in both u-boot and kernel.

Signed-off-by: Han Xu <han.xu@nxp.com>
Signed-off-by: Ye Li <ye.li@nxp.com>
5 years agoMLK-12214 NAND:apbh_dma: Fix logically dead code issue
Ye.Li [Wed, 13 Jan 2016 02:06:59 +0000 (10:06 +0800)]
MLK-12214 NAND:apbh_dma: Fix logically dead code issue

The list_first_entry always assumes the list is not empty, it won't return NULL pointer when
the list is empty. So the "if (pdesc == NULL)" becomes a dead code. Fix the issue by calling
the list_empty before the list_first_entry.

(Coverity CID 29934)

Signed-off-by: Ye.Li <ye.li@nxp.com>
(cherry picked from commit ff3923f294cc2e15f436d7520b4042736b1b48a6)
(cherry picked from commit 64c6a7b5d621080b8bd948c061a4f223a8c2d886)
(cherry picked from commit b9846931a2196fa2bf29a9cb0304d11e96c2c40a)

5 years agoMLK-11719-4: mtd: gpmi: change the BCH layout setting for large oob NAND
Ye Li [Tue, 16 Apr 2019 03:50:53 +0000 (20:50 -0700)]
MLK-11719-4: mtd: gpmi: change the BCH layout setting for large oob NAND

The code change updated the NAND driver BCH ECC layout algorithm to
support large oob size NAND chips(oob > 1024 bytes) and proposed a new
way to set ECC layout.

Current implementation requires each chunk size larger than oob size so
the bad block marker (BBM) can be guaranteed located in data chunk. The
ECC layout always using the unbalanced layout(Ecc for both meta and
Data0 chunk), but for the NAND chips with oob larger than 1k, the driver
cannot support because BCH doesn’t support GF 15 for 2K chunk.

The change keeps the data chunk no larger than 1k and adjust the ECC
strength or ECC layout to locate the BBM in data chunk. General idea for
large oob NAND chips is

1.Try all ECC strength from the minimum value required by NAND spec to
the maximum one that works, any ECC makes the BBM locate in data chunk
can be chosen.

2.If none of them works, using separate ECC for meta, which will add one
extra ecc with the same ECC strength as other data chunks. This extra
ECC can guarantee BBM located in data chunk, of course, we need to check
if oob can afford it.

Previous code has two methods for ECC layout setting, the
legacy_calc_ecc_layout and calc_ecc_layout_by_info, the difference
between these two methods is, legacy_calc_ecc_layout set the chunk size
larger chan oob size and then set the maximum ECC strength that oob can
afford. While the calc_ecc_layout_by_info set chunk size and ECC
strength according to NAND spec. It has been proved that the first
method cannot provide safe ECC strength for some modern NAND chips, so
in current code,

1. Driver read NAND parameters first and then chose the proper ECC
layout setting method.

2. If the oob is large or NAND required data chunk larger than oob size,
chose calc_ecc_for_large_oob, otherwise use calc_ecc_layout_by_info

3. legacy_calc_ecc_layout only used for some NAND chips does not contains
necessary information. So this is only a backup plan, it is NOT
recommended to use these NAND chips.

Signed-off-by: Han Xu <b45815@freescale.com>
Signed-off-by: Ye Li <ye.li@nxp.com>
5 years agoMLK-19945 mx6dq: hab: Fix chip version in hab.h code
Breno Lima [Mon, 15 Oct 2018 17:11:26 +0000 (14:11 -0300)]
MLK-19945 mx6dq: hab: Fix chip version in hab.h code

Since commit 8891410c729b ("MLK-19848 mx6dq: Fix chip version issue for
rev1.3") it's not possible to call the HAB API functions on i.MX6DQ
SoC Rev 1.3:

Authenticate image from DDR location 0x12000000...
undefined instruction
pc : [<412c00dc>]          lr : [<8ff560bc>]
reloc pc : [<c8b6d0dc>]    lr : [<178030bc>]
sp : 8ef444a8  ip : 126e8068     fp : 8ff59aa8
r10: 8ffd51e4  r9 : 8ef50eb0     r8 : 006e8000
r7 : 00000000  r6 : 126ea01f     r5 : 0000002b  r4 : 126e8000
r3 : 412c00dd  r2 : 00000001     r1 : 00000001  r0 : 00000063
Flags: nzCv  IRQs off  FIQs off  Mode SVC_32
Resetting CPU ...

resetting ...

The hab.h code is defining the HAB API base address according to the
old SoC revision number, thus failing when calling the HAB API
authenticate_image() function.

Fix this issue by using mx6dq rev 1.3 instead of mx6dq rev 1.5.

Signed-off-by: Breno Lima <breno.lima@nxp.com>
Reviewed-by: Ye Li <ye.li@nxp.com>
(cherry picked from commit adc4d93e6e1726f111e86ca74fb76215215dbc86)

5 years agoMLK-19876 imx: hab: Enable hab.c to authenticate additional images in open configuration
Breno Lima [Tue, 9 Oct 2018 02:25:20 +0000 (23:25 -0300)]
MLK-19876 imx: hab: Enable hab.c to authenticate additional images in open configuration

Currently it's not possible to authenticate additional boot images in HAB
open configuration.

The hab.c code is checking if the SEC_CONFIG[1] fuse is programmed prior
to calling the hab_authenticate_image() API function. Users cannot check
if their additional boot images has been correctly signed prior to closing
their device.

Enable hab.c to authenticate additional boot images in open mode so HAB
events can be retrieved through get_hab_status() function.

Signed-off-by: Breno Lima <breno.lima@nxp.com>
Reviewed-by: Ye Li <ye.li@nxp.com>
(cherry picked from commit 15d3e93bd16b5375f4dbf525ce61f0912ece4c0b)

5 years agoMLK-19263-4 hab: Change calling to ROM API failsafe
Ye Li [Mon, 20 Aug 2018 07:00:33 +0000 (00:00 -0700)]
MLK-19263-4 hab: Change calling to ROM API failsafe

Modify to use hab_rvt_failsafe function for failsafe ROM API, not
directly call its ROM address. This function will wrap the sip call for iMX8M
platforms.

Signed-off-by: Ye Li <ye.li@nxp.com>
(cherry picked from commit d6b9992db082920b55c08d33df1cb03942939e7f)

5 years agoMLK-17935: imx: HAB: Validate IVT before authenticating image
Utkarsh Gupta [Tue, 3 Apr 2018 00:41:16 +0000 (19:41 -0500)]
MLK-17935: imx: HAB: Validate IVT before authenticating image

Calling csf_is_valid() with an un-signed image may lead to data abort
as the CSF pointer could be pointing to a garbage address when accessed
in HAB_HDR_LEN(*(const struct hab_hdr *)(ulong)ivt_initial->csf).

Authenticate image from DDR location 0x80800000...
Check CSF for Write Data command before authenticating image
data abort
pc : [<fff5494c>]          lr : [<fff54910>]
reloc pc : [<8780294c>]    lr : [<87802910>]
sp : fdf45dc8  ip : 00000214     fp : 00000000
r10: fffb6170  r9 : fdf4fec0     r8 : 00722020
r7 : 80f20000  r6 : 80800000     r5 : 80800000  r4 : 00720000
r3 : 17a5aca3  r2 : 00000000     r1 : 80f2201f  r0 : 00000019
Flags: NzcV  IRQs off  FIQs off  Mode SVC_32
Resetting CPU ...

resetting ...

To avoid such errors during authentication process, validate IVT structure
by calling validate_ivt function which checks the following values in an IVT:

IVT_HEADER = 0x4X2000D1
ENTRY != 0x0
RES1 = 0x0
DCD = 0x0       /* Recommended */
SELF != 0x0     /* Absoulute address of IVT */
CSF != 0x0
RES2 = 0x0

This commit also checks if Image's start address is 4 byte aligned.

commit "0088d127 MLK-14945 HAB: Check if IVT valid before authenticating image"
removed as this patch addresses the issue.

Signed-off-by: Utkarsh Gupta <utkarsh.gupta@nxp.com>
(cherry picked from commit dabffd1b04df3b0393ef6a9a35b5fd816edd8c63)
Signed-off-by: Ye Li <ye.li@nxp.com>
(cherry picked from commit b4bda0ae3c5c5be0524148c1bcdfea908da859d6)

5 years agoMLK-17585 imx8/imx8m: Disable TRUST_OF_CHAIN for secure boot
Ye Li [Wed, 11 Apr 2018 14:09:06 +0000 (07:09 -0700)]
MLK-17585 imx8/imx8m: Disable TRUST_OF_CHAIN for secure boot

In the board common directory, the CONFIG_TRUST_OF_CHAIN is automatically
enabled once the CONFIG_SECURE_BOOT is set. Since we don't support this
trust chain on all i.MX8 platforms, modify the Kconfig to disable it.

Signed-off-by: Ye Li <ye.li@nxp.com>
(cherry picked from commit 0eea75587e40d461cdddd215bc582526b1529bf6)

5 years agoMLK-18159-12 imx: cmd_dek: Enable DEK only for chips supporting CAAM
Ye Li [Wed, 11 Apr 2018 14:06:42 +0000 (07:06 -0700)]
MLK-18159-12 imx: cmd_dek: Enable DEK only for chips supporting CAAM

Since cmd_dek is using CAAM JR, so enable the CMD_DEK only when
HAS_CAAM is set

Signed-off-by: Ye Li <ye.li@nxp.com>
(cherry picked from commit 8a552c8d98df953c13a0144972b8c16ed7684e92)

5 years agoMLK-17044-3 booti: Add kernel image authentication for secure boot
Ye Li [Thu, 30 Nov 2017 07:32:39 +0000 (01:32 -0600)]
MLK-17044-3 booti: Add kernel image authentication for secure boot

When secure boot is enabled, add authenticate_image in booti to authenticate
kernel image.

Signed-off-by: Ye Li <ye.li@nxp.com>
(cherry picked from commit f29a143cdb8c74566113737e9be7e1bcd8c625f4)
(cherry picked from commit 1e33f493a55dad7e016f948b932000ec295c6df4)

5 years agoMLK-17044-2 HAB: Add SIP call for ARM64 platform
Ye Li [Thu, 30 Nov 2017 07:29:54 +0000 (01:29 -0600)]
MLK-17044-2 HAB: Add SIP call for ARM64 platform

When current EL is not EL3, the direct calling to HAB will fail because
CAAM/SNVS can't initialize at non-secure mode. In this case, we use
SIP call to run the HAB in ATF.

Signed-off-by: Ye Li <ye.li@nxp.com>
(cherry picked from commit bca4e251b6083476e987d92c428136a1c81bcb0b)
(cherry picked from commit e31a75cc78ad2d5e6ab5d58851058ab57de1566d)

5 years agoMLK-16787-2 iMX8M: Add support to enable CONFIG_SECURE_BOOT
Ye Li [Thu, 9 Nov 2017 03:43:56 +0000 (21:43 -0600)]
MLK-16787-2 iMX8M: Add support to enable CONFIG_SECURE_BOOT

Add some SOC level codes and build configurations to use HAB lib for
CONFIG_SECURE_BOOT, like adding the SEC_CONFIG fuse, enable fuse driver,
CAAM clock function, and add CAAM secure RAM to MMU table.

The FSL_CAAM is temporally not enabled for iMX8M when CONFIG_SECURE_BOOT is set,
because we don't need the CAAM driver for SPL.

Signed-off-by: Ye Li <ye.li@nxp.com>
Reviewed-by: Peng Fan <peng.fan@nxp.com>
(cherry picked from commit a0cba5678b10827bc4b5e4fb2e40945a5c332baa)
(cherry picked from commit 330a3108e3806fc2b9fa500f84ebfb4feb821c6d)

5 years agoMLK-16787-1 HAB: Update hab codes to support ARM64 and i.MX8M
Ye Li [Thu, 9 Nov 2017 03:25:44 +0000 (21:25 -0600)]
MLK-16787-1 HAB: Update hab codes to support ARM64 and i.MX8M

There are some changes to support ARM64 i.MX8M platform in this patches:
1. The hab_rvt base and function vectors are different as i.MX6/7

2. Need to bypass an workaround for i.MX6 to fix problem in MMU.

3. The x18 register needed save & restore before calling any HAB API. According
   to ARM procedure call spec, the x18 is caller saved when it is used as
   temporary register. So calling HAB API may scratch this register, and
   cause crash once accessing the gd pointer.

   On ARMv7, the r9 is callee saved when it is used as variable register. So
   no need to save & restore it.

4. Add SEC_CONFIG fuse for iMX8M

Signed-off-by: Ye Li <ye.li@nxp.com>
Reviewed-by: Peng Fan <peng.fan@nxp.com>
(cherry picked from commit 5f6ef97a9d13361895180df7014f4e4fb48a5875)
(cherry picked from commit 465407632f436cb55db1d261ee0adb7458220045)

5 years agoMLK-12500-1 HAB: Add kernel image authentication in image loading
Ye Li [Tue, 8 Mar 2016 14:03:47 +0000 (22:03 +0800)]
MLK-12500-1 HAB: Add kernel image authentication in image loading

To support the trust boot chain, we integrate the authentication
into the kernel image loading process. The kernel image will be verified
at its load address. So when signing the kernel image, we need to
use this load address which may change on different platforms.

Signed-off-by: Ye Li <ye.li@nxp.com>
(cherry picked from commit 3c118b8d6bbe1a25ca8c8bafeb528309f16fc73d)
(cherry picked from commit fd9a9759ed9b3a9fc26b18aff00880382213b1ca)
(cherry picked from commit 98d4faefdb83579d4a5a170e06af5efb64ad2b3c)

5 years agoMLK-18158 imx: hab: Add function to authenticate kernel image
Ye Li [Wed, 11 Apr 2018 10:30:19 +0000 (03:30 -0700)]
MLK-18158 imx: hab: Add function to authenticate kernel image

When loading kernel image, the image size is parsed from header, so it
does not include the CSF and IVT.

Add back the authenticate_image function to wrap the imx_hab_authenticate_image
with calculating IVT offset and full image size.

Signed-off-by: Ye Li <ye.li@nxp.com>
(cherry picked from commit 86bf433c3d6940ce71749e97a37efbff5ac7a79f)

5 years agoMLK-21842-4 imx8mq_evk: Add QSPI NOR support
Ye Li [Wed, 17 Apr 2019 05:29:30 +0000 (22:29 -0700)]
MLK-21842-4 imx8mq_evk: Add QSPI NOR support

Add QSPI NOR relavant configurations and QSPI clock init to board
codes

Signed-off-by: Ye Li <ye.li@nxp.com>