Simon Glass [Sat, 15 Apr 2017 19:11:31 +0000 (13:11 -0600)]
rockchip: Print a message when returning to the bootrom
At present if the return to bootrom fails (e.g. because you are not using
the Rockchip's bootrom's pointer table in MMC) then the board prints
SPL message and hangs. Print a message first if we can, to help in
understanding what happened when it hangs.
Tom Rini [Fri, 14 Apr 2017 23:47:50 +0000 (19:47 -0400)]
buildman: Allow 'gnueabihf' toolchains for ARM
Many toolchains for ARM use the 'gnueabihf' suffix rather than just
'gnueabi', so allow these to be used, but with a lower priority than
'gnueabi' ones.
Cc: Simon Glass <sjg@chromium.org> Signed-off-by: Tom Rini <trini@konsulko.com> Reviewed-by: Simon Glass <sjg@chromium.org>
sysreset: psci: support system reset in a generic way with PSCI
If the system is running PSCI firmware, the System Reset function
(func ID: 0x80000009) is supposed to be handled by PSCI, that is,
the SoC/board specific reset implementation should be moved to PSCI.
U-Boot should call the PSCI service according to the arm-smccc
manner.
The arm-smccc is supported on ARMv7 or later. Especially, ARMv8
generation SoCs are likely to run ARM Trusted Firmware BL31. In
this case, U-Boot is a non-secure world boot loader, so it should
not be able to reset the system directly.
Tyler Hall [Wed, 12 Apr 2017 20:29:17 +0000 (16:29 -0400)]
cramfs: basic symlink support
Handle symlinks to files in the current directory. Other cases could be
handled with additional code, but this is a start.
Add explicit errors for absolute paths and links found in the middle of
a path (directories). Other cases like '..' or '.' will result with the
file not being found as when those path components are explicitly
provided.
Add a helper to decompress a null-terminated link name which is shared
with cramfs_list_inode.
Tyler Hall [Wed, 12 Apr 2017 20:29:15 +0000 (16:29 -0400)]
cmd: cramfs: use map_sysmem for sandbox support
As with most other commands, this needs to factor in the sysmem offset
in the sandbox or it will try to dereference the simulated physical
address directly.
Simon Glass [Sun, 19 Mar 2017 18:59:21 +0000 (12:59 -0600)]
x86: Display the SPL banner only once
At present on a cold reboot we must reset the CPU to get it to full speed.
With 64-bit U-Boot this happens in SPL. At present we print the banner
before doing this, the end result being that we print the banner twice.
Print the banner a little later (after the CPU is ready) to avoid this.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com> Tested-by: Bin Meng <bmeng.cn@gmail.com>
Shengzhou Liu [Mon, 10 Apr 2017 08:00:08 +0000 (16:00 +0800)]
powerpc/board/t1024rdb: enable board-level reset when issuing reset command
As board-specific reset logic, it needs to issue reset signal
via CPLD when issuing 'reset' command in u-boot, this patch
solves the issue of reset command not working on T1024RDB.
Signed-off-by: Shengzhou Liu <Shengzhou.Liu@nxp.com> Reviewed-by: York Sun <york.sun@nxp.com>
Ruchika Gupta [Thu, 2 Mar 2017 08:42:41 +0000 (14:12 +0530)]
powerpc: e6500: Lock/unlock 1 cache instead of L1 as init_ram
For E6500 cores, L2 cache has been used as init_ram. L1 cache is a
write through cache on E6500.If lines are not locked in both L1 and
L2 caches, crashes are observed during secure boot. This patch locks/
unlocks both L1 and L2 cache to prevent the crash.
Signed-off-by: Ruchika Gupta <ruchika.gupta@nxp.com> Reviewed-by: York Sun <york.sun@nxp.com>
- Add SD secure boot target for ls1046ardb.
- Change the u-boot size defined by a macro for copying the main
U-Boot by SPL to also include the u-boot Secure Boot header size
as header is appended to u-boot image. So header will also be
copied from SD to DDR.
- CONFIG_MAX_SPL_SIZE is limited to 90KB. SPL is copied to OCRAM
(128K) where 32K are reserved for use by boot ROM and 6K for the
header.
- Reduce the size of CAAM driver for SPL Blobification functions
and descriptors, that are not required at the time of SPL are
disabled. Further error code conversion to strings is disabled
for SPL build.
Signed-off-by: Vinitha Pillai <vinitha.pillai@nxp.com> Signed-off-by: Sumit Garg <sumit.garg@nxp.com> Signed-off-by: Ruchika Gupta <ruchika.gupta@nxp.com> Reviewed-by: York Sun <york.sun@nxp.com>
- Change the u-boot size defined by a macro for copying the main
U-Boot by SPL to also include the u-boot Secure Boot header size as
header is appended to u-boot image. So header will also be copied
from SD to DDR.
- MACRO for CONFIG_BOOTSCRIPT_COPY_RAM is enabled to copy Bootscript
from NAND to DDR. Offsets for Bootscript on NAND and DDR have been
also defined.
Signed-off-by: Vinitha Pillai <vinitha.pillai@nxp.com> Signed-off-by: Sumit Garg <sumit.garg@nxp.com> Signed-off-by: Ruchika Gupta <ruchika.gupta@nxp.com> Reviewed-by: York Sun <york.sun@nxp.com>
- Add SD secure boot target for ls1043ardb.
- Implement FSL_LSCH2 specific spl_board_init() to setup CAAM stream
ID and corresponding stream ID in SMMU.
- Change the u-boot size defined by a macro for copying the main
U-Boot by SPL to also include the u-boot Secure Boot header size as
header is appended to u-boot image. So header will also be copied
from SD to DDR.
- CONFIG_MAX_SPL_SIZE is limited to 90KB. SPL is copied to OCRAM
(128K) where 32K are reserved for use by boot ROM and 6K for secure
boto header.
- Error messages during SPL boot are limited to error code numbers
instead of strings to reduce the size of SPL image.
Signed-off-by: Vinitha Pillai-B57223 <vinitha.pillai@nxp.com> Signed-off-by: Sumit Garg <sumit.garg@nxp.com> Signed-off-by: Ruchika Gupta <ruchika.gupta@nxp.com> Reviewed-by: York Sun <york.sun@nxp.com>
Sumit Garg [Thu, 30 Mar 2017 04:23:13 +0000 (09:53 +0530)]
armv8: ls1046ardb: SPL size reduction
Using changes in this patch we were able to reduce approx 4k
size of u-boot-spl.bin image. Following is breif description of
changes to reduce SPL size:
1. Changes in board/freescale/ls1046ardb/Makefile to remove
compilation of eth.c and cpld.c in case of SPL build.
2. Changes in board/freescale/ls1046ardb/ls1046ardb.c to keep
only ddr_init and board_early_init_f funcations in case of SPL
build.
3. Changes in ls1046a_common.h & ls1046ardb.h to remove driver
specific macros due to which static data was being compiled in
case of SPL build.
4. Disable MMC driver from bieng compiled in case of SPL NAND
build and NAND driver from bieng compiled in case of SPL MMC build.
Signed-off-by: Vinitha Pillai <vinitha.pillai@nxp.com> Signed-off-by: Sumit Garg <sumit.garg@nxp.com> Reviewed-by: York Sun <york.sun@nxp.com>
Sumit Garg [Thu, 30 Mar 2017 04:22:38 +0000 (09:52 +0530)]
armv8: ls1043ardb: SPL size reduction
Using changes in this patch we were able to reduce approx 10k
size of u-boot-spl.bin image. Following is breif description of
changes to reduce SPL size:
1. Changes in board/freescale/ls1043ardb/Makefile to remove
compilation of eth.c and cpld.c in case of SPL build.
2. Changes in board/freescale/ls1043ardb/ls1043ardb.c to keep
only ddr_init and board_early_init_f funcations in case of SPL
build.
3. Changes in ls1043a_common.h & ls1043ardb.h to remove driver
specific macros due to which static data was being compiled in
case of SPL build.
4. Disable MMC driver from bieng compiled in case of SPL NAND
build and NAND driver from bieng compiled in case of SPL MMC build.
5. Remove I2C driver support from SPL in case of LS1043ARDB.
Signed-off-by: Vinitha Pillai <vinitha.pillai@nxp.com> Signed-off-by: Sumit Garg <sumit.garg@nxp.com> Reviewed-by: York Sun <york.sun@nxp.com>
Depending on DDR configuration, gcc-6.x will show up unused-const-
variable messages. Use __maybe_unused specifier for all dynamic_odt
variable definitions to remove these warnings.
Memory footprint will not increase as gcc will optimize out unused
constants.
Signed-off-by: Thomas Schaefer <thomas.schaefer@kontron.com> Signed-off-by: York Sun <york.sun@nxp.com>
Philipp Tomsich [Sat, 1 Apr 2017 10:59:25 +0000 (12:59 +0200)]
rockchip: rk3399: spl: add UART0 support for SPL
The RK3399-Q7 ("Puma") SoM exposes UART0 as the Qseven UART (i.e. the
serial line available via standardised pins on the edge connector and
available on a RS232 connector).
To support boards (such as the RK3399-Q7) that require UART0 as a
debug console, we match CONFIG_DEBUG_UART_BASE and add the appropriate
iomux setup to the rk3399 SPL code.
As we are already touching this code, we also move the board-specific
UART setup (i.e. iomux setup) into board_debug_uart_init(). This will
be called from the debug UART init when CONFIG_DEBUG_UART_BOARD_INIT
is set.
As the RK3399 needs to use its board_debug_uart_init() function, we
have Kconfig enable it by default for RK3399 builds.
With everything set up to define CONFIG_BAUDRATE via defconfig and
with to have the SPL debug UART either on UART0 or UART2, the configs
for the RK3399 EVB are then update (the change for the RK3399-Q7 is
left for later to not cause issues on applying the change).
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Reviewed-by: Kever Yang <kever.yang@rock-chips.com> Acked-by: Simon Glass <sjg@chromium.org>
when enable PMIC rk808,the system will halt at very
early stage,log is shown as bellow.
INFO: plat_rockchip_pmu_init(1211): pd status 3e
INFO: BL31: Initializing runtime services
INFO: BL31: Preparing for EL3 exit to normal world
INFO: Entry point address = 0x200000
INFO: SPSR = 0x3c9
time 44561b, 0 (<<----Just stop here)
It's caused by the absence of "{ }" in syscon_rk3399.c
,which will lead to memory overflow like below.According
to Sysmap file ,we can find the function buck_get_value
of rk808 is just follow the compatible struct,the pointer
"of_match" point to "buck_get_value",but it is not a
struct and don't have member of compatible, In this case,
system crash. So,on the face, it looks like that rk808 is
guilty.but he is really innocent.
while (of_match->compatible) { <<----------
if (!strcmp(of_match->compatible, compat)) {
*of_idp = of_match;
return 0;
}
of_match++;
}
Signed-off-by: Eric Gao <eric.gao@rock-chips.com> Reviewed-by: Kever Yang <kever.yang@rock-chips.com> Tested-by: Kever Yang <kever.yang@rock-chips.com>
Klaus Goger [Fri, 7 Apr 2017 17:13:38 +0000 (19:13 +0200)]
rockchip: ARM64: split RK3399-Q7 board off the RK3399-EVB board
The RK3399-Q7 SoM is a Qseven-compatible (70mm x 70mm, MXM-230
connector) system-on-module from Theobroma Systems, featuring the
Rockchip RK3399.
It provides the following feature set:
* up to 4GB DDR3
* on-module SPI-NOR flash
* on-module eMMC (with 8-bit interace)
* SD card (on a baseboad) via edge connector
* Gigabit Ethernet w/ on-module Micrel KSZ9031 GbE PHY
* HDMI/eDP/MIPI displays
* 2x MIPI-CSI
* USB
- 1x USB 3.0 dual-role (direct connection)
- 2x USB 3.0 host + 1x USB 2.0 (on-module USB 3.0 hub)
* on-module STM32 Cortex-M0 companion controller, implementing:
- low-power RTC functionality (ISL1208 emulation)
- fan controller (AMC6821 emulation)
- USB<->CAN bridge controller
Note that we use a multi-payload FIT image for booting and have
Cortex-M0 payload in a separate subimage: we thus rely on the FIT
image loader to put it into the SRAM region that ATF expects it in.
Signed-off-by: Klaus Goger <klaus.goger@theobroma-systems.com> Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Fixed build warning on puma-rk3399: Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Simon Glass <sjg@chromium.org>
rockchip: sysreset: rk3188: Make sure remap is off on warm-resets
The warm-reset of rk3188 socs keeps the remap setting as it was, so if
it was enabled, the cpu would start from address 0x0 of the sram instead
of address 0x0 of the bootrom, thus making the reset hang.
Therefore make sure the remap is disabled before attempting a warm reset.
Cold reset is not affected by this at all.
Signed-off-by: Heiko Stuebner <heiko@sntech.de> Acked-by: Simon Glass <sjg@chromium.org>
Most Rockchip socs have the ability to either map the bootrom or a sram
area to the starting address of the cpu by flipping a bit in the GRF.
Newer socs leave this untouched and mapped to the bootrom but the legacy
loaders on rk3188 and before enabled the remap functionality and the
current smp implementation in the Linux kernel also requires it to be
enabled, to bring up secondary cpus.
So to keep smp working in the kernel, mimic the behaviour of the legacy
bootloaders and enable the remap functionality.
Signed-off-by: Heiko Stuebner <heiko@sntech.de> Acked-by: Simon Glass <sjg@chromium.org>
Eddie Cai [Sat, 1 Apr 2017 06:49:54 +0000 (14:49 +0800)]
rockchip: Add USB to the default boot targets
Now that most rockchip SoC based board have usb host support, enable
USB boot targets by default.
Signed-off-by: Eddie Cai <eddie.cai.linux@gmail.com> Acked-by: Simon Glass <sjg@chromium.org>
Fixed build errors when CONFIG_CMD_USB not defined: Signed-off-by: Simon Glass <sjg@chromium.org>
Philipp Tomsich [Wed, 29 Mar 2017 19:20:28 +0000 (21:20 +0200)]
rockchip: spl: rk3399: disable DDR security regions for SPL
The RK3399 hangs during DMA of the Designware MMC controller, when
performing DMA-based transactions in SPL due to the DDR security settings
left behind by the BootROM (i.e. accesses to the first MB of DRAM are
restricted... however, the DMA is likely to target this first MB, as it
transfers from/to the stack).
System security is not affected, as the final security configuration is
performed by the ATF, which is executed after the SPL stage.
With this fix in place, we can now drop 'fifo-mode' in the DTS for the
RK3399-Q7 (Puma).
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Reviewed-by: Kever Yang <kever.yang@rock-chips.com> Acked-by: Simon Glass <sjg@chromium.org>
Wenyou Yang [Fri, 14 Apr 2017 00:51:46 +0000 (08:51 +0800)]
board: sama5d3_xplained: Clean up code
Due to the introduction of the pinctrl and clk driver, and using
device tree files, remove the unneeded hardcoded pin configuration
and clock enabling code from the board file.
Signed-off-by: Wenyou Yang <wenyou.yang@atmel.com>
Wenyou Yang [Fri, 14 Apr 2017 00:51:45 +0000 (08:51 +0800)]
board: sama5d3_xplained: Update to support DM/DT
Update the configuration files to support the device tree and driver
model, so do SPL. The device clock and pins configuration are handled
by the clock and the pinctrl drivers respectively.
Signed-off-by: Wenyou Yang <wenyou.yang@atmel.com>
Fix build error with sama5d3_xplained_mmc: Signed-off-by: Simon Glass <sjg@chromium.org>
Wenyou Yang [Fri, 14 Apr 2017 00:51:43 +0000 (08:51 +0800)]
board: sama5d3xek: Clean up code
Due to the introduction of the pinctrl and clk driver, and using
device tree files, remove the unneeded hardcoded pin configuration
and clock enabling code from the board file.
Signed-off-by: Wenyou Yang <wenyou.yang@atmel.com>
Remove CONFIG_PHY_MICREL as per previous patch: Signed-off-by: Simon Glass <sjg@chromium.org>
Wenyou Yang [Fri, 14 Apr 2017 00:51:42 +0000 (08:51 +0800)]
board: sama5d3xek: Update to support DM/DT
Update the configuration files to support the device tree and
driver model, so do SPL. The device clock and pins configuration
are handled by the clock and the pinctrl drivers respectively.
Signed-off-by: Wenyou Yang <wenyou.yang@atmel.com>
Add back CONFIG_PHY_MICREL to prevent a build error: Signed-off-by: Simon Glass <sjg@chromium.org>
Simon Glass [Mon, 10 Apr 2017 17:34:54 +0000 (11:34 -0600)]
dm: led: Adjust the LED uclass
At present this is very simple, supporting only on and off. We want to
also support toggling and blinking. As a first step, change the name of
the main method and use an enum to indicate the state.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Ziping Chen <techping.chan@gmail.com>
With DM_SCSI enabled, blk_create_devicef() is called with blkz = 0, leading
to a divide-by-0 exception.
scsi_detect_dev() can be used to get the required parameters (block size
and number of blocks) from the drive before calling blk_create_devicef().
Signed-off-by: Jean-Jacques Hiblot <jjhiblot@ti.com> Reviewed-by: Tom Rini <trini@konsulko.com> Reviewed-by: Simon Glass <sjg@chromium.org>
scsi: make the LUN a parameter of scsi_detect_dev()
This is a cosmetic change. target and LUN have kind of the same role in
this function. One of them was passed as a parameter and the other was
embedded in a structure. For consistency, pass both of them as parameters.
Signed-off-by: Jean-Jacques Hiblot <jjhiblot@ti.com> Reviewed-by: Tom Rini <trini@konsulko.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Tom Rini [Fri, 14 Apr 2017 14:06:28 +0000 (10:06 -0400)]
buildman: Translate more strings to latin-1
When writing out some of our results we may now have UTF-8 characters
in there as well. Translate these to latin-1 and ignore any errors (as
this is for diagnostic and given the githash anything else can be
reconstructed by the user.
Reviewed-by: Simon Glass <sjg@chromium.org> Signed-off-by: Tom Rini <trini@konsulko.com>
which caused part_init to be called when it was not previously.
Without this patch, the following happens when a USB sd card reader is used.
=> usb start
starting USB...
USB0: Port not available.
USB1: USB EHCI 1.00
scanning bus 1 for devices... 3 USB Device(s) found
scanning usb for storage devices... Device NOT ready
Request Sense returned 02 3A 00
### ERROR ### Please RESET the board ###
Eddie Cai [Thu, 6 Apr 2017 03:37:04 +0000 (11:37 +0800)]
usb: dwc2: invalidate the dcache before starting the DMA
We should invalidate the dcache before starting the DMA. In case there are
any dirty lines from the DMA buffer in the cache, subsequent cache-line
replacements may corrupt the buffer in memory while the DMA is still going on.
Cache-line replacement can happen if the CPU tries to bring some other memory
locations into the cache while the DMA is going on.
Signed-off-by: Eddie Cai <eddie.cai.linux@gmail.com> Reviewed-by: Stefan Brüns <stefan.bruens@rwth-aachen.de>
Philipp Tomsich [Thu, 6 Apr 2017 14:58:53 +0000 (16:58 +0200)]
usb: dwc3: gadget: make cache-maintenance on event buffers more robust
Merely using dma_alloc_coherent does not ensure that there is no stale
data left in the caches for the allocated DMA buffer (i.e. that the
affected cacheline may still be dirty).
The original code was doing the following (on AArch64, which
translates a 'flush' into a 'clean + invalidate'):
# during initialisation:
1. allocate buffers via memalign
=> buffers may still be modified (cached, dirty)
# during interrupt processing
2. clean + invalidate buffers
=> may commit stale data from a modified cacheline
3. read from buffers
This could lead to garbage info being written to buffers before
reading them during even-processing.
To make the event processing more robust, we use the following sequence
for the cache-maintenance:
# during initialisation:
1. allocate buffers via memalign
2. clean + invalidate buffers
(we only need the 'invalidate' part, but dwc3_flush_cache()
always performs a 'clean + invalidate')
# during interrupt processing
3. read the buffers
(we know these lines are not cached, due to the previous
invalidation and no other code touching them in-between)
4. clean + invalidate buffers
=> writes back any modification we may have made during event
processing and ensures that the lines are not in the cache
the next time we enter interrupt processing
Note that with the original sequence, we observe reproducible
(depending on the cache state: i.e. running dhcp/usb start before will
upset caches to get us around this) issues in the event processing (a
fatal synchronous abort in dwc3_gadget_uboot_handle_interrupt on the
first time interrupt handling is invoked) when running USB mass
storage emulation on our RK3399-Q7 with data-caches on.
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Philipp Tomsich [Thu, 6 Apr 2017 14:58:52 +0000 (16:58 +0200)]
usb: dwc3: ensure consistent types for dwc3_flush_cache
The dwc3_flush_cache() call was declared and used inconsistently:
* The declaration assumed 'int' for addresses (a potential issue
when running in a LP64 memory model).
* The invocation cast the address to 'long'.
This change ensures that both the declaration and usage of this
function consistently uses 'uintptr_t' for correct behaviour even
when the allocated buffers (to be flushed) reside outside of the
lower 32bits of memory.
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Felipe Balbi [Wed, 22 Feb 2017 15:12:40 +0000 (17:12 +0200)]
usb: gadget: g_dnl: only set iSerialNumber if we have a serial#
We don't want to claim that we support a serial number string and
later return nothing. Because of that, if g_dnl_serial is an empty
string, let's skip setting iSerialNumber to a valid number.
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
projects / u-boot.git / log