Bin Meng [Wed, 19 Jul 2017 13:51:21 +0000 (21:51 +0800)]
usb: xhci: Enable TT to support LS/FS devices behind a HS hub
So far LS/FS devices directly attached to xHC root port can be
successfully enumerated by xHCI driver, but if they are connected
behind a hub, the enumeration process fails to address the device.
It turns out xHCI driver still misses a part that in the device's
input slot context, all Transaction Translator (TT) related fields
are not programmed. The xHCI spec defines how to enable TT.
Now LS/FS devices like USB keyboard/mouse can be enumerated behind
a high speed hub.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org>
There is no way to know whether the attached device is a hub or
not in advance before the device's descriptor is fetched. But
once we know it's a high speed hub, per the xHCI spec, we need
to tell xHC it's a hub device by initializing hub-related fields
in the input slot context.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Bin Meng [Wed, 19 Jul 2017 13:51:17 +0000 (21:51 +0800)]
dm: usb: Add a new USB controller operation 'update_hub_device'
For USB host controllers like xHC, its internal representation of
hub needs to be updated after the hub descriptor is fetched. This
adds a new op that does this.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Bin Meng [Wed, 19 Jul 2017 13:51:16 +0000 (21:51 +0800)]
usb: hub: Parse and save TT details from device descriptor
A high speed hub has a special responsibility to handle full speed/
low speed devices connected on downstream ports. In this case, the
hub must isolate the high speed signaling environment from the full
speed/low speed signaling environment with the help of Transaction
Translator (TT). TT details are provided by hub descriptors and we
parse and save it to hub uclass_priv for later use.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Bin Meng [Wed, 19 Jul 2017 13:51:15 +0000 (21:51 +0800)]
usb: xhci: Program 'route string' in the input slot context
xHCI spec says: the values of the 'route string' field shall be
initialized by the first 'Address Device' command issued to a
device slot, and shall not be modified by any other command.
So far U-Boot does not program this field, and it does not prevent
SS device directly attached to root port, or HS device behind an HS
hub, from working, due to the fact that 'route string' is used by
the xHC to target SS packets. But in order to enumerate devices
behind an SS hub, this field must be programmed.
With this commit and along with previous commits, now SS & HS devices
attached to a USB 3.0 hub can be enumerated by U-Boot.
As usual, this new feature is only available when DM is on.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org>
For future extension, change xhci_setup_addressable_virt_dev()
signature to accept a pointer to 'struct usb_device', instead
of its members slot_id & speed, as the struct already contains
these two plus some other useful information of the device.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Bin Meng [Wed, 19 Jul 2017 13:51:13 +0000 (21:51 +0800)]
usb: hub: Support 'set hub depth' request for USB 3.0 hubs
USB 3.0 hub uses a hub depth value multiplied by four as an offset
into the 'route string' to locate the bits it uses to determine the
downstream port number. We shall set the hub depth value of a USB
3.0 hub after it is configured.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Bin Meng [Wed, 19 Jul 2017 13:51:12 +0000 (21:51 +0800)]
usb: hub: Translate USB 3.0 hub port status into old version
USB 3.0 hub port status field has different bit positions from 2.0
hubs. Since U-Boot only understands the old version, translate the
new one into the old one.
Since we are going to add USB 3.0 hub support, this feature is only
available with driver model USB.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Bin Meng [Wed, 19 Jul 2017 13:51:11 +0000 (21:51 +0800)]
usb: hub: Add a new API to test if a hub device is root hub
Sometimes we need know if a given hub device is root hub or not.
Add a new API to test this. This removes the xHCI driver's own
version is_root_hub() and change to use the new API.
While we are here, remove the unused/commented out get_usb_device()
in the xHCI driver too.
Bin Meng [Wed, 19 Jul 2017 13:51:09 +0000 (21:51 +0800)]
usb: hub: Use 'struct usb_hub_device' as hub device's uclass_priv
Use USB hub device's dev->uclass_priv to point to 'usb_hub_device'
so that with driver model usb_hub_reset() and usb_hub_allocate()
are no longer needed.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Bin Meng [Wed, 19 Jul 2017 13:50:10 +0000 (21:50 +0800)]
x86: minnowmax: Enable USB xHCI support
BayTrail SoC supports both EHCI and xHCI controllers. However only
one host controller (either EHCI or xHCI) can be used. To enable
HSIC and SS ports, xHCI must be used. This turns on xHCI support on
Intel MinnowMax board.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org> Reviewed-by: Stefan Roese <sr@denx.de> Tested-by: Stefan Roese <sr@denx.de>
Bin Meng [Wed, 19 Jul 2017 13:50:09 +0000 (21:50 +0800)]
x86: minnowmax: Add a environment variable for USB power-on delay
Occasionally it was observed that on Intel MinnowMax board, with a
USB 2.0 device connected to the bottom port, when doing 'usb start'
on the xHCI controller:
scanning bus 0 for devices... cannot reset port 3!?
But neither of the two USB ports is routed to xHCI root port 3.
Adding some debug information shows that xHCI port 3 PORTSC register
mysteriously reports both CCS = 1 and CSC = 1.
This is not seen every time. After increasing the timeout to wait
for power to become stable, the issue is gone. So this indicates
current default USB power-on delay (20ms) might be at a critical
region where power is stable/unstable. U-Boot provides a mechanism
to have a environment variable to override the default one. Add
one for MinnowMax.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org> Reviewed-by: Stefan Roese <sr@denx.de> Tested-by: Stefan Roese <sr@denx.de>
Bin Meng [Wed, 19 Jul 2017 13:50:08 +0000 (21:50 +0800)]
usb: xhci: Convert CONFIG_USB_XHCI_PCI to Kconfig
Add CONFIG_USB_XHCI_PCI as a Kconfig option.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org> Reviewed-by: Stefan Roese <sr@denx.de> Tested-by: Stefan Roese <sr@denx.de>
Bin Meng [Wed, 19 Jul 2017 13:50:07 +0000 (21:50 +0800)]
usb: cmd: Print actual packet size for super speed devices
USB 3.0 defines bMaxPacketSize0 field in the device descriptor as
the exponent of 2, so let's print the calculated actual size.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org> Reviewed-by: Stefan Roese <sr@denx.de> Tested-by: Stefan Roese <sr@denx.de>
Bin Meng [Wed, 19 Jul 2017 13:50:06 +0000 (21:50 +0800)]
configs: Remove CONFIG_SYS_USB_EHCI_MAX_ROOT_PORTS in all boards
Now that EHCD does not use CONFIG_SYS_USB_EHCI_MAX_ROOT_PORTS,
remove it in all boards' config files.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org> Reviewed-by: Stefan Roese <sr@denx.de> Tested-by: Stefan Roese <sr@denx.de>
Bin Meng [Wed, 19 Jul 2017 13:50:05 +0000 (21:50 +0800)]
usb: ehci: Get rid of CONFIG_SYS_USB_EHCI_MAX_ROOT_PORTS
EHC reports supported maximum number of ports in the HCSPARAMS
register, so it's unnecessary to use a hardcoded config option
CONFIG_SYS_USB_EHCI_MAX_ROOT_PORTS.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org> Reviewed-by: Stefan Roese <sr@denx.de> Tested-by: Stefan Roese <sr@denx.de>
Bin Meng [Wed, 19 Jul 2017 13:50:04 +0000 (21:50 +0800)]
configs: Remove CONFIG_SYS_USB_XHCI_MAX_ROOT_PORTS in all boards
Now that xHCD does not use CONFIG_SYS_USB_XHCI_MAX_ROOT_PORTS,
remove it in all boards' config files.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org> Reviewed-by: Stefan Roese <sr@denx.de> Tested-by: Stefan Roese <sr@denx.de>
Bin Meng [Wed, 19 Jul 2017 13:50:03 +0000 (21:50 +0800)]
usb: xhci: Get rid of CONFIG_SYS_USB_XHCI_MAX_ROOT_PORTS
xHC reports supported maximum number of ports in the HCSPARAMS1
register, so it's unnecessary to use a hardcoded config option
CONFIG_SYS_USB_XHCI_MAX_ROOT_PORTS.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org> Reviewed-by: Stefan Roese <sr@denx.de> Tested-by: Stefan Roese <sr@denx.de>
Bin Meng [Wed, 19 Jul 2017 13:50:02 +0000 (21:50 +0800)]
usb: xhci: Change MAX_HC_PORTS to 255
HCSPARAMS1:MaxPorts field specifies the maximum port number value,
and its valid values are in the range of 1 to 255.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org> Reviewed-by: Stefan Roese <sr@denx.de> Tested-by: Stefan Roese <sr@denx.de>
Bin Meng [Wed, 19 Jul 2017 13:50:01 +0000 (21:50 +0800)]
usb: hub: Add 3.0 hub port status mask of 2.0 hub
USB 3.0 hub port status has different bit position regarding to
port power, port speed, etc. But others are the same as 2.0 hubs.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org> Reviewed-by: Stefan Roese <sr@denx.de> Tested-by: Stefan Roese <sr@denx.de>
Bin Meng [Wed, 19 Jul 2017 13:50:00 +0000 (21:50 +0800)]
usb: hub: Change USB hub descriptor to match USB 3.0 hubs
USB 3.0 hubs have a slightly different hub descriptor than USB 2.0
hubs, with a fixed (rather than variable length) size. Change the
host controller drivers that access those last two fields
(DeviceRemovable and PortPowerCtrlMask) to use the union.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org> Reviewed-by: Stefan Roese <sr@denx.de> Tested-by: Stefan Roese <sr@denx.de>
Bin Meng [Wed, 19 Jul 2017 13:49:59 +0000 (21:49 +0800)]
usb: hub: Revise wLength for 'get port status' request
For accuracy, we should use 'sizeof(struct usb_port_status)' as the
wLength for 'get port status' request, although it happens to be
equal to 'sizeof(struct usb_hub_status)'.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org> Reviewed-by: Stefan Roese <sr@denx.de> Tested-by: Stefan Roese <sr@denx.de>
Bin Meng [Wed, 19 Jul 2017 13:49:58 +0000 (21:49 +0800)]
usb: hub: Send correct wValue to get hub descriptor of a USB 3.0 hub
Testing a USB 3.0 hub by connecting it to the xHCI port on Intel
MinnowMax, when issuing 'get hub descriptor' to the hub, xHCI
reports a transfer event TRB with a completion code 6 which means
'Stall Error'.
In fact super speed USB hub descriptor type is 0x2a, not 0x29.
Sending correct SETUP packet to the hub makes it not stall anymore.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org> Reviewed-by: Stefan Roese <sr@denx.de> Tested-by: Stefan Roese <sr@denx.de>
Bin Meng [Wed, 19 Jul 2017 13:49:57 +0000 (21:49 +0800)]
usb: hub: Update handling connect status/change in usb_scan_port()
It was observed that on Intel MinnowMax board, when xHCI is enabled
in the BayTrail SoC, with a USB 3.0 device connected to the bottom
USB 3.0 port (mapped to xHCI root port #7), its PORTSC register is
always 0x201203 (CCS = 1, CSC = 0). The root cause of such behavior
is unknown yet. Connect status change bit is set on the same port
with a USB 2.0 device (mapped to xHCI port #1, which is a different
port on the root hub).
With current logic in usb_scan_port(), the enumeration process will
abort if it does not detect a connect status change on a hub port.
However since a device connection status is correctly reported, the
enumeration process can still continue.
With this change, USB device connected to the bottom blue port on
MinnowMax board can be enumerated under either SS or HS mode.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Stefan Roese <sr@denx.de> Tested-by: Stefan Roese <sr@denx.de> Tested-by: Dinh Nguyen <dinguyen@kernel.org>
Bin Meng [Wed, 19 Jul 2017 13:49:56 +0000 (21:49 +0800)]
usb: xhci: Add input slot context in xhci_set_configuration()
A valid input slot context for a 'configure endpoint' command requires
the 'Context Entries' field to be initialized to the index of the last
valid endpoint context that is defined by the target configuration. We
set up the 'Context Entries' field, but we forget to include the input
slot context in the input control context 'Add Context flags' bitmap.
So xHC will simply ignore input slot context and continue using its own
which contains old information of the device.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org> Reviewed-by: Stefan Roese <sr@denx.de> Tested-by: Stefan Roese <sr@denx.de>
Bin Meng [Wed, 19 Jul 2017 13:49:55 +0000 (21:49 +0800)]
usb: xhci: Initialize scratchpad buffer array and scratchpad buffers
The scratchpad buffer array is used to define the locations of
statically allocated memory pages that are available for the
private use of the xHC. The xHCI spec explicitly mentions that
system software shall allocate the scratchpad buffers before
placing the xHC in to Run mode (Run/Stop (R/S) = ‘1’), however
U-Boot is missing this part.
This causes xHC on Intel platform does not respond the very first
'enable slot' command that is given to xHC and the 'enable slot'
command completion event TRB is never generated and xHC seems to
hang forever.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org> Reviewed-by: Stefan Roese <sr@denx.de> Tested-by: Stefan Roese <sr@denx.de>
In xhci_queue_command(), when the command is not 'reset endpoint',
'stop endpoint' or 'set TR dequeue pointer', endpoint ID should not
be encoded in the TRB.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org> Reviewed-by: Stefan Roese <sr@denx.de> Tested-by: Stefan Roese <sr@denx.de>
Bin Meng [Wed, 19 Jul 2017 13:49:53 +0000 (21:49 +0800)]
usb: xhci: Remove incorrect comments for struct xhci_container_ctx
There is no member called 'dma' in struct xhci_container_ctx. Remove
the comments that mentions it.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org> Reviewed-by: Stefan Roese <sr@denx.de> Tested-by: Stefan Roese <sr@denx.de>
usb: host: ehci-generic: add error path and .remove callback
Use an array to save enabled clocks reference and deasserted resets
in order to respectively disabled and asserted them in case of error
during probe() or during driver removal.
Signed-off-by: Patrice Chotard <patrice.chotard@st.com> Reviewed-by: Simon Glass <sjg@chromium.org>
This function is usefull to get phandle number contained
in a property list.
For example, this allows to allocate the right amount
of memory to keep clock's reference contained into the
"clocks" property.
To implement it, either of_count_phandle_with_args() or
fdtdec_parse_phandle_with_args() are used respectively
for live tree and flat tree.
By passing index = -1, these 2 functions returns the
number of phandle contained into the property list.
Add also the dev_count_phandle_with_args() based on
ofnode_count_phandle_with_args()
Signed-off-by: Patrice Chotard <patrice.chotard@st.com> Reviewed-by: Simon Glass <sjg@chromium.org>
A few lines (defines and declarations) had been duplicated when the
puma-rk3399 board was initially merged. This removes the duplicates
and changes the style to use local constants instead of pasted
literals.
Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
[fixed up commit-message & converted to use 'const u32':] Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Andy Yan [Mon, 24 Jul 2017 09:52:24 +0000 (17:52 +0800)]
rockchip: add u-boot specific dts for rk3036 sdk
Add this dts to enable debug uart releated devices
before relocation.
Signed-off-by: Andy Yan <andy.yan@rock-chips.com> Reviewed-by: Simon Glass <sjg@chromium.org> Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Andy Yan [Mon, 24 Jul 2017 09:52:01 +0000 (17:52 +0800)]
rockchip: use puts instead of printf when back to bootrom
printf will increase the code size more than 1kb, but platform
like rk3036 has no enough space for it.
Signed-off-by: Andy Yan <andy.yan@rock-chips.com> Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Andy Yan [Mon, 24 Jul 2017 09:51:30 +0000 (17:51 +0800)]
rockchip: enable SPL_LIBGENERIC for rk3036 based boards
function board_init_f_init_reserve will call memset, which
is implemented in lib, and enabled by CONFIG_SPL_LIBGENERIC_SUPPORT
in spl stage.
To reduce the code size, also enable SPL_TINY_MEMSET.
As rk3036 will return to bootrom immediately after dram
initialization, there is no need to run DM, so disable
SPL_DM_SERIAL.
Signed-off-by: Andy Yan <andy.yan@rock-chips.com> Reviewed-by: Simon Glass <sjg@chromium.org> Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Andy Yan [Mon, 24 Jul 2017 09:51:07 +0000 (17:51 +0800)]
rockchip: disable SPL_ARCH_MEMCPY/MEMSET for rk3036
RK3036 has no enough sapce use ARCH_MEMCPY/MEMSET in spl stage
Signed-off-by: Andy Yan <andy.yan@rock-chips.com> Reviewed-by: Simon Glass <sjg@chromium.org> Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Andy Yan [Mon, 24 Jul 2017 09:50:46 +0000 (17:50 +0800)]
rockchip: set malloc pool size to 0 before relocation in spl state on rk3036 based board
RK3036 only has 4kb sram, the spl code will use
3.4 ~ 3.5 kb, the last 0.5kb are used for SP and
GD, so there is no space for malloc. Also, the spl
will directly return to bootrom after dram initialized,
they never need the space for malloc.
Signed-off-by: Andy Yan <andy.yan@rock-chips.com> Reviewed-by: Simon Glass <sjg@chromium.org> Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Andy Yan [Mon, 24 Jul 2017 09:49:59 +0000 (17:49 +0800)]
sandbox: use CONFIG_VAL(SYS_MALLOC_F_LEN) to distinguish malloc pool size before relocation
SPL and normal u-boot stage use different malloc pool size
configuration before relocation, so use CONFIG_VAL(SYS_MALLOC_F_LEN)
to fit different boot stage.
Signed-off-by: Andy Yan <andyshrk@gmail.com>
Changes in v3:
- use CONFIG_VAL(), which suggested by Simon
Changes in v2: None
arch/sandbox/cpu/start.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-) Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Andy Yan [Mon, 24 Jul 2017 09:49:01 +0000 (17:49 +0800)]
microblaze: spl: configure SYS_MALLOC_F_LEN independently for SPL and full U-Boot
Some platforms have very limited SRAM to run SPL code, so there may
not be the same amount space for a malloc pool before relocation in
the SPL stage as the normal U-Boot stage.
Make SPL and (the full) U-Boot stage use independent SYS_MALLOC_F_LEN,
so the size of pre-relocation malloc pool can be configured memory
space independently.
Signed-off-by: Andy Yan <andy.yan@rock-chips.com> Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
[fixed up commit-message:] Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Andy Yan [Mon, 24 Jul 2017 09:47:27 +0000 (17:47 +0800)]
powerpc: spl: configure SYS_MALLOC_F_LEN independently for SPL and full U-Boot
Some platforms have very limited SRAM to run SPL code, so there may
not be the same amount space for a malloc pool before relocation in
the SPL stage as the normal U-Boot stage.
Make SPL and (the full) U-Boot stage use independent SYS_MALLOC_F_LEN,
so the size of pre-relocation malloc pool can be configured memory
space independently.
Signed-off-by: Andy Yan <andy.yan@rock-chips.com> Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
[fixed up commit-message:] Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Andy Yan [Mon, 24 Jul 2017 09:45:27 +0000 (17:45 +0800)]
mips: spl: configure SYS_MALLOC_F_LEN independently for SPL and full U-Boot
Some platforms have very limited SRAM to run SPL code, so there may
not be the same amount space for a malloc pool before relocation in
the SPL stage as the normal U-Boot stage.
Make SPL and (the full) U-Boot stage use independent SYS_MALLOC_F_LEN,
so the size of pre-relocation malloc pool can be configured memory
space independently.
Signed-off-by: Andy Yan <andy.yan@rock-chips.com> Acked-by: Daniel Schwierzeck <daniel.schwierzeck@gmail.com> Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
[fixed up commit-message:] Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Andy Yan [Mon, 24 Jul 2017 09:43:34 +0000 (17:43 +0800)]
spl: make SPL and normal u-boot stage use independent SYS_MALLOC_F_LEN
Some platforms have very limited SRAM to run SPL code, so there may
not be the same amount space for a malloc pool before relocation in
the SPL stage as the normal U-Boot stage.
Make SPL and (the full) U-Boot stage use independent SYS_MALLOC_F_LEN,
so the size of pre-relocation malloc pool can be configured memory
space independently.
Signed-off-by: Andy Yan <andy.yan@rock-chips.com> Reviewed-by: Tom Rini <trini@konsulko.com> Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
[fixed up commit-message:] Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Leo Wen [Tue, 25 Jul 2017 12:47:47 +0000 (20:47 +0800)]
rockchip: firefly: Add "usb start" to auto-start USB device
Add "preboot=usb start" to ROCKCHIP_DEVICE_SETTINGS,you don't
need to input "usb start" in command line of u-boot console,it
can auto-start the USB device,after that usb keyboard can work.
Signed-off-by: Leo Wen <leo.wen@rock-chips.com> Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Leo Wen [Tue, 25 Jul 2017 12:47:46 +0000 (20:47 +0800)]
rockchip: firefly: Set the environment variable 'usbkbd' to the stdin
Add the 'usbkbd' environment variable to the 'stdin', the contents of
the keyboard input can be auto-displayed on the serial terminal,so
you don't need to manually set the environment variable 'stdin'.
Signed-off-by: Leo Wen <leo.wen@rock-chips.com> Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Leo Wen [Tue, 25 Jul 2017 12:47:45 +0000 (20:47 +0800)]
rockchip: firefly: Add some macros to enable the usb keyboard
Add four macros of CONFIG_USB_KEYBOARD,CONFIG_DM_KEYBOARD,etc in the
firefly-rk3288_defconfig,can support usb keyboard device when these four
macros are enabled.
Signed-off-by: Leo Wen <leo.wen@rock-chips.com> Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Philipp Tomsich [Wed, 26 Jul 2017 10:29:01 +0000 (12:29 +0200)]
rockchip: rk3399: enable SPL_SERIAL_SUPPORT and SPL_DRIVERS_MISC_SUPPORT via Kconfig
SPL_SERIAL_SUPPORT and SPL_DRIVERS_MISC_SUPPORT were previously
enabled through rk3399_common.h. This change implies these options
through Kconfig.
These need to always be active for the RK3399, as follows:
- SPL_SERIAL_SUPPORT is needed to pass the SPL build
- SPL_DRIVERS_MISC_SUPPORT is needed to pass the SPL build
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
rockchip: rk3288: Add support for drive-strength in PINCTRL
Currently, drive-strenght to 12ma are described and supposed to be used
on RK3288. However, the pinctrl driver for this SoC only handles muxing
and pull up/pull down via PU/PD control registers. So complex IPs like
GMAC are working in normal ethernet 100mbps, but not at 1gbps typically.
This commit adds support for handling drive-strength of 12ma, when it's
defined in the DT.
Signed-off-by: Romain Perier <romain.perier@collabora.com> Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Kever Yang [Tue, 25 Jul 2017 04:00:05 +0000 (12:00 +0800)]
rockchip: use UUID for root partitions
We use to use /dev/mmcbl0p7 as root partition, and pass it
to kernel by cmdline, but the mmc number in kernel in not
fixed, we need to change the bootargs to adapt it from time
to time.
We can use the UUID to fix it, the ID is from:
https://www.freedesktop.org/wiki/Specifications/DiscoverablePartitionsSpec/
ARM 32bit: 69dad710-2ce4-4e3c-b16c-21a1d49abed3
ARM 64bit: b921b045-1df0-41c3-af44-4c6f280d3fae
Signed-off-by: Kever Yang <kever.yang@rock-chips.com> Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Philipp Tomsich [Wed, 19 Jul 2017 12:32:23 +0000 (14:32 +0200)]
rockchip: dts: rk3399-puma: put EFI partition entries at 2MB
When creating a EFI/GUID partition map for the RK3399-Q7 through
U-Boot, the partition entries should be places at a 1MB offset from
the start of the device to give us space for the environment (at 16KB
on SD/MMC devices), the SPL stage (at 32KB on SD/MMC devices) and the
image payload (at 256KB on SD/MMC devices).
This change sets this up through the u-boot,efi-partition-entries-offset
/config property in the RK3399-Q7 DTSI.
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Philipp Tomsich [Wed, 19 Jul 2017 12:32:22 +0000 (14:32 +0200)]
rockchip: dts: rk3399-puma: put environment (in MMC/SD configurations) before SPL
As our SPL stage can grow quite large (80KB+ are not unusual) on the
RK3399-Q7, the default setting for the environment location (in
include/configs/rockchip-common.h) can overlap our SPL.
This change finally makes use of the 'u-boot,mmc-env-offset' DTS
property to override the environment location and put it at 16KB into
the device, which is right before the SPL (located at 32KB).
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Philipp Tomsich [Tue, 11 Jul 2017 18:49:44 +0000 (20:49 +0200)]
rockchip: clk: rk3399: remove unused fields from priv-structures
This removes the unused 'rate' field from both rk3399_pmuclk_priv and
rk3399_clk_priv. I didn't bother to check where this came from (i.e.
what the historical context of these was), but only verified that
these are indeed unused across all code-paths.
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Philipp Tomsich [Tue, 11 Jul 2017 18:49:43 +0000 (20:49 +0200)]
rockchip: clk: rk3368: remove unused fields from rk3368_clk_priv
The rk3368_clk_priv has two unused fields: rate, has_bwadj. This
removes them as there's no need for either (i.e. has_bwadj is always
true for the RK3368, according to its TRM).
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Philipp Tomsich [Tue, 11 Jul 2017 18:59:45 +0000 (20:59 +0200)]
rockchip: clk: rk3368: use correct (i.e. 'rk3368_clk_priv') structure for auto-alloc
The clk driver for the RK3368 picked the wrong data structure's size
for its auto-alloc size: the size was calculated on the structure
representing the CRU hardware block instead of the priv structure.
As the CRU's register file is much larger than the driver's priv,
this did not cause any pain (except wasting memory).
Fix this by using the correct data structure's size.
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Philipp Tomsich [Tue, 11 Jul 2017 19:04:45 +0000 (21:04 +0200)]
rockchip: spl: make boot0 hook TPL safe
When building for a TPL/SPL setup (e.g. on the RK3368), we need the
TPL stage to have the extra space for for the 'Rockchip SPL name'
(i.e. 'RK33' word). Yet, the SPL will start execution at its first
word (i.e. the first word in the SPL binary needs to be a valid
instruction). To make things a bit more involved, CONFIG_SPL_BUILD
is defined both for the SPL and the TPL stage.
To avoid having to explicitly test for the first stage (TPL, if and
only if TPL and SPL are built, SPL otherwise), this commit modifies
the sequence to repeat the 'b reset' (instead of reserving 4 bytes
of undefined space) at the start of the boot0 hook: if overwritten
(and execution starts at the second word), the first instruction is
still a 'b reset'... if not overwritten, we start on a 'b reset' as
well.
This solution wouldn't even require the check whether we are in the
SPL/TPL build (i.e. CONFIG_SPL_BUILD), but we leave this check in for
documentation purposes.
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Kever Yang [Wed, 19 Jul 2017 11:54:22 +0000 (19:54 +0800)]
power: pwm_regulator: remove redundant code
The regulator_enable() should be called from upper layer like
regulators_enable_boot_on(), remove it from pwm regulator driver.
Signed-off-by: Kever Yang <kever.yang@rock-chips.com> Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
[fixed up typo in commit message:] Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Kever Yang [Wed, 19 Jul 2017 11:54:21 +0000 (19:54 +0800)]
power: pwm_regulator: fix the pwm_set_config parameter order
The rkpwm reg order has fixed by below patch: e3ef41d rockchip: pwm: fix the register layout for the PWM controller
We need to correct the parameter order for pwm_set_config() to make
the pwm regulator works correctly.
Signed-off-by: Kever Yang <kever.yang@rock-chips.com> Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Kever Yang [Wed, 19 Jul 2017 11:54:20 +0000 (19:54 +0800)]
rockchip: dts: correct vdd_log setting for firefly-rk3399
Add regulator-init-microvolt for driver to init the regulator,
and the min output value is not 800000mV for the PWM2 io domain has
changed to VCC3V0 instead of VCC1V8 in rockchip evb, we need to
correct it with the value measured when PWM2 output HIGH.
Signed-off-by: Kever Yang <kever.yang@rock-chips.com> Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Kever Yang [Tue, 18 Jul 2017 13:55:08 +0000 (21:55 +0800)]
rockchip: dts: firefly using ddr3 1600
According to my test, some of firefly-rk3399 hang after dram init
when using ddr3-1333 config, while using ddr3-1600 config works
for all the board I have test.
Signed-off-by: Kever Yang <kever.yang@rock-chips.com> Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
The Identification Page (32 byte) is an additional page which can be written
and (later) permanently locked in Read-only mode.
phyCORE-RK3288 SoMs are using this page to describe the module variant.
This page also contains a MAC.
Our boards can be equipped with a different amount of EEPROMs. To make
this more transparent let's add an alias for the eeprom which stores the
module variant.
Signed-off-by: Wadim Egorov <w.egorov@phytec.de> Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
rockchip: rk3288: Revert MAC_TXCLK in pinctrl for GMAC
This reverts TXCLK toggling that was accidently dropped while reworking
commit 2454b719fb87 ("rockchip: rk3288: Add pinctrl support for the gmac
ethernet interface"). So the TX clock is enabled and we can use
GMAC_ROCKCHIP in 1Gbps when basic PINCTRL support is enabled
(!PINTRL_FULL).
Fixes: 2454b719fb87 ("rockchip: rk3288: Add pinctrl support for the...") Signed-off-by: Romain Perier <romain.perier@collabora.com> Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Philipp Tomsich [Fri, 14 Jul 2017 16:09:25 +0000 (18:09 +0200)]
rockchip: timer: make register sizes explicit
We are about to reuse the rockchip timer (header file) for 64bit ARMv8
chips, so it seems a good time to make the register sizes explicit by
changing from 'unsigned int' to 'u32'.
Reorders the header-includes in rk_timer.c to ensure that 'u32' is
definded before it is used by 'asm/arch/timer.h'.
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Tom Rini [Tue, 18 Jul 2017 18:14:19 +0000 (14:14 -0400)]
FIT: List kernel_noload in the list of types
In the source_file_format.txt file we talk about how to construct a
valid FIT image. While it already says to look at the source for the
full list, add kernel_noload to the explicit list of types. This is
arguably the most important type to use as most often we are including a
kernel that will run from wherever it is loaded into memory and execute.
This for example, allows you to create a single FIT image for Linux that
can be used on both OMAP and i.MX devices as the kernel will not need to
be moved in memory.
STM32F7 and H7 shared the same SDRAM control block.
On STM32H7 few control bits has been added.
The current driver need some minor adaptation as FMC block
enable/disable for H7.
Retrieve RAM base address from DT instead of using STM32_SDRAM_FMC
For STM32F7, FMC block base address is 0xA0000000, but SDRAM
registers are located at offset 0x140 inside FMC block.
Update the stm32_fmc_regs fields with all FMC registers
to map SDRAM registers at the right address.
This allow to remove include/dm/platform_data/serial_stm32x7.h
which was included in the past by stm32x7 driver and by
stm32f746-disco.c board file.
Since patch 42bf5e7c27 "serial: stm32f7: add device tree support"
this file is no more needed in board file.
Add clock framework .get_rate callback.
This step will allow to convert all drivers which was using
proprietary clock_get() to use clock framework .get_rate().