musb stop is musb core call during unregister or shutting down
gadget or host musb. For graceful exit add musb_platform_exit
on musb_stop so-that it can exit the musb platform driver as well.
Tested-by: Chen-Yu Tsai <wens@csie.org> # A33-OlinuXino
Tested-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
musb have platform ops to do proper graceful exit,
so add the exit call and move musb platform exit code
instead of keeping it in driver remove.
This make proper shutdown of musb where .remove will
call disable, exit serially via musb_stop.
Tested-by: Chen-Yu Tsai <wens@csie.org> # A33-OlinuXino
Tested-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
reset0 is not available for sun4i, 5i and 7i so access
the reset0 offset from ccm via driver data for relevant
Allwinner SoC. this will eventually drop the existing
ifdef for SUN6I.
Tested-by: Chen-Yu Tsai <wens@csie.org> # A33-OlinuXino
Tested-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Allocate struct phy in private structure instead of allocating
locally and assign it to a pointer. This eventually fix miss
alignment phy which is used in another functions.
Tested-by: Chen-Yu Tsai <wens@csie.org> # A33-OlinuXino
Tested-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
When MUSB is operating in peripheral mode, probe registering
musb core using musb_register which intern return int value
for validation. so there is no scope to preserve struct musb
pointer but the same can be used in .remove musb_stop.
So fix this by return musb_register with struct musb pointer.
Cc: Igor Grinberg <grinberg@compulab.co.il>
Cc: Purna Chandra Mandal <purna.mandal@microchip.com>
Tested-by: Chen-Yu Tsai <wens@csie.org> # A33-OlinuXino
Tested-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Add an entry in usb_gadget_controller_number() for the DWC3 gadget
controller. Without it, it is not possible to bind the USB Ethernet driver.
Signed-off-by: Jean-Jacques Hiblot <jjhiblot@ti.com>
With upcoming changes that require CONFIG_BLK, this broke
USB Mass Storage on the OMAP3 boards because if CONFIG_BLK is
enabled, it assumes that DM_USB is enabled, but it wasn't yet
available on omap3 and omap4 boards.
This patch converts the OMAP2430 MUSB glue to support DM_USB and
extracts the necessary information based on the device tree.
It's based on the ti-musb driver, but there are enough significant
differences in both the architecture and device tree entires between
am33xx and OMAP3/OMAP4, that I think it makes sense to continue to
keep the separate.
Per doc/driver-model/usb-info.txt, the USB gadget stuff hasn't
migrated to DM_USB yet, so this only supports USB Host for now.
Users wanting USB Gadgets will need to disable DM_USB and leave
it the old way for now.
Signed-off-by: Adam Ford <aford173@gmail.com>
A bunch of code was encapsulated in #ifdef's whether or not
it is building or for U-Boot. Since this code is always building
for U-Boot, this patch removes the dead code.
Signed-off-by: Adam Ford <aford173@gmail.com>
As well as in K_FW_LBA_READ_10 and K_FW_LBA_ERASE_10 take device's
block size from f_rkusb->desc->blksz instead of the fixed 512 bytes.
Keep original behaviour of retry probing assigned block device on
every host request to manage late SDCard plugs.
Signed-off-by: Alberto Panizzo <alberto@amarulasolutions.com>
While downloading or uploading megabytes of data we had thousands of
printf in console like:
transfer 0x10000 bytes done
OR
Uploading 0x1000 bytes
This because transfers are chunked and there is no way on target
side to know the overall transfer size (to print one string per
overall transfer).
All these prints on serial console do slow down significantly the
transfer and does not offer a significant information to the
developer: rkdeveloptool and Rockchip original tool do use small
chunks read/writes on big transfers. This allows on workstation
to print percentage of transfer complete and as well offers to
developer the information about: transfer is running OK.
On error, either the percentage will stop or an error will be shown
on workstation console.
Signed-off-by: Alberto Panizzo <alberto@amarulasolutions.com>
This command is part of the write partition sequence performed by
rkdeveloptool: one partition is first completely erased and
than wrote.
Signed-off-by: Alberto Panizzo <alberto@amarulasolutions.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
This patch implement reading blocks form selected device with
LBA addressing.
Corresponding command on workstation is:
rkdeveloptool rl <start_blk> <blk_cnt> <file>
While we support reading more than one blocks per K_FW_LBA_READ_10
request, rkdeveloptool and original rockchip tool do perform
chunk reads limiting the maximum size per chunk far lower
than max int values.
Signed-off-by: Alberto Panizzo <alberto@amarulasolutions.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Chip Version is a string saved in BOOTROM address space Little Endian.
Ex for rk3288: 0x33323041 0x32303134 0x30383133 0x56323030
which brings: 320A20140813V200
Note that memory version do invert MSB/LSB so printing the char
buffer would show: A02341023180002V
Signed-off-by: Alberto Panizzo <alberto@amarulasolutions.com>
Two consecutive rockusb_tx_write without waiting for request complete
do results in transfer reset of first request and thus no or incomplete
data transfer. This because rockusb_tx_write do use just one USB request
to keep serialization.
So calls like:
rockusb_tx_write_str(emmc_id);
rockusb_tx_write_csw(cbw->tag, cbw->data_transfer_length, CSW_GOOD);
was succeeding only when DEBUG was defined because the time spent
printing debug info was enough for transfer to complete.
This patch fixes the issue adding a simple request complete handler
called rockusb_tx_write_csw to be set as complete handler of in_req
when sending back simple payload + CSW replies to commands.
This new handler will always send CSW_GOOD replies because in case
of error the command callback itself must send back an error CSW as
unique reply to command.
This patch fixes execution of:
$ rkdeveloptool rfi
when DEBUG is not defined.
Signed-off-by: Alberto Panizzo <alberto@amarulasolutions.com>
After the commit 6aae84769a ("gadget: f_thor: Fix memory leaks of
usb request and its buffer"), there is hang-up with ctrl-c in some
udc. It is because req of out_ep is freed before out_ep is disabled.
Fix hang-up with ctrl-c by disabling ep before free req of the ep.
Signed-off-by: Seung-Woo Kim <sw0312.kim@samsung.com>
The only platform left for the AU1x00 SoCs was the pb1x00 platform, an
apparent clone of the dbau1x00 platform. As pb1x00 had no listed
maintainer I am assuming that it is also orphaned. Remove this platform
and then remove the unused SoC support.
Cc: Daniel Schwierzeck <daniel.schwierzeck@gmail.com>
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Daniel Schwierzeck <daniel.schwierzeck@gmail.com>
On the A64 the clock for the first USB controller is actually the parent
of the clock for the second controller, so turning them off in that order
makes the system hang.
Fix this by only turning off *both* clocks when the *last* OHCI controller
is brought down. This covers the case when only one controller is used.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
When using CONFIG_OF_BOARD on rpi to use the dtb provided by the
RaspberryPi Fundation, the compatible string isn't the same, resulting
in not-functional usb from u-boot.
Signed-off-by: Oleksandr Tymoshenko <gonzo@FreeBSD.org>
Signed-off-by: Emmanuel Vadot <manu@freebsd.org>
Masking clock gate, reset register bits based on the
probed controller is proper only due to the assumption
that masking should start with 0 even thought the controller
has separate PHY or shared between OTG.
unfortunately these are fixed due to lack of separate
clock, reset drivers.
Say for example EHCI1 - EHCI3 in the datasheet (EHCI0 is for the OTG)
so we need to start reg_mask 0 - 2.
This patch calculated the mask, based on the register base
so that we can get the proper bits to set with respect to
probed controller.
We even do this masking by using PHY index specifier from dt,
but dev_read_addr_size is failing for 64-bit boards.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
For ohci, the maximam supported endpoint number is 32(in and out), and
now we have used (usb_pipeendpoint(pipe) << 1) to index the specified
endpoint descritor, usb_pipeendpoint(pipe) can reach 0xf, so we need
change the NUM_EDs from 8 to 32.
Signed-off-by: Zeng Tao <prime.zeng@hisilicon.com>
ohci-hcd casts priv_data pointer to (ohci_t *), thus it must be
the first member in private data struct.
Fixes 831cc98b1 ("usb: sunxi: Simplify ccm reg base code")
Signed-off-by: Vasily Khoruzhick <anarsoul@gmail.com>
This bug is the combination of dwc2 USB controller and lan78xx
USB ethernet controller, which is the combination in use on
the Raspberry Pi Model 3 B+.
When the host attempts to receive a packet, but a packet has not
arrived, the lan78xx controller responds by setting BIR
(Bulk-In Empty Response) to NAK. Unfortunately, this hangs
the USB controller and requires the USB controller to
be reset.
The fix proposed is to have the lan78xx controller respond
by setting BIR to ZLP.
Signed-off-by: Andrew Thomas <andrew.thomas@oracle.com>
Tested-by: Peter Robinson <pbrobinson@gmail.com>
Reviewed-by: Alexander Graf <agraf@suse.de>
struct sunxi_ccm_reg doesn't have ahb_reset0_cfg on sun4i and sun5i,
thus compilation fails with:
drivers/usb/host/ohci-sunxi.c:96:26: error: 'struct sunxi_ccm_reg' has
no member named 'ahb_reset0_cfg'
Access this reg using its offset to fix this issue.
Fixes commit 1ed9c1118 ("usb: sunxi: ehci: get rid of ifdefs")
and commit 56830cee3 ("usb: sunxi: ohci: get rid of ifdefs")
Signed-off-by: Vasily Khoruzhick <anarsoul@gmail.com>
On A64 OHCI1 clock source is OHCI0 clock, so we need to enable OHCI0
clock when OHCI1 is in use.
Fixes commit dd3228170a ("usb: sunxi: Switch to use generic-phy")
Signed-off-by: Vasily Khoruzhick <anarsoul@gmail.com>
This fixes the following compiler warning:
"warning: cast from pointer to integer of different size
[-Wpointer-to-int-cast]"
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Add WARN_ONCE definition to allow single time notification
of warnings to the user.
Taken from Linux kernel (4.17) with slight changes
(Removed __section(.data.once))
Signed-off-by: Ramon Fried <ramon.fried@gmail.com>
[trini: Drop the musb and dwc3 compat versions]
Signed-off-by: Tom Rini <trini@konsulko.com>
There is no reason to specify header with full soc name.
Symlink is setup automatically (arch -> arch-zynqmp)
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Convert USB fastboot code to use shared fastboot protocol.
Signed-off-by: Alex Kiernan <alex.kiernan@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Rename fb_mmc_flash_write/fb_mmc_erase/fb_nand_flash_write/fb_nand_erase to
fastboot_... as they form a public interface
Signed-off-by: Alex Kiernan <alex.kiernan@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Move FASTBOOT_VERSION to include/fastboot.h so when we merge the UDP code
we only have one definition.
Signed-off-by: Alex Kiernan <alex.kiernan@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Rename fb_set_reboot_flag to fastboot_set_reboot_flag so it matches
all other fastboot code in the global name space. Fix the guards around
them so that they're dependent on FASTBOOT, not just USB_FUNCTION_FASTBOOT.
Move the weak implementation of fastboot_set_reboot_flag to fb_common.c
so we can call it from non-USB fastboot code.
Signed-off-by: Alex Kiernan <alex.kiernan@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Ensure that when selecting FASTBOOT_FLASH you end up with a buildable
configuration. Prior to this you could select NAND without MTDPARTS
and end up with an image which (surprisingly) excluded NAND.
Also fix dependencies on FASTBOOT_GPT_NAME/FASTBOOT_MBR_NAME which require
you have EFI_PARTITION/DOS_PARTITION enabled.
Delete redundant FASTBOOT_FLASH_NAND_DEV from Kconfig - it was only ever
used as a guard and the value was ignored in all cases, we're using
FASTBOOT_FLASH_NAND as the guard now.
Signed-off-by: Alex Kiernan <alex.kiernan@gmail.com>
Reviewed-by: Joe Hershberger <joe.hershberger@ni.com>
Add drivers/fastboot/fb_common.c, where fastboot_okay/fail are implemented
so we can call them from a non-USB implementation.
Introduce fastboot_response which takes varargs parameters so we can
use it to generate formatted response strings. Refactor fastboot_okay/fail
to use it.
Signed-off-by: Alex Kiernan <alex.kiernan@gmail.com>
Reviewed-by: Joe Hershberger <joe.hershberger@ni.com>
Add the response string as a parameter to fastboot_okay/fail, instead
of modifying a global, to match the contract expected by the AOSP
U-Boot code.
Signed-off-by: Alex Kiernan <alex.kiernan@gmail.com>
Reviewed-by: Joe Hershberger <joe.hershberger@ni.com>
There are memory leaks of usb request and its buffer for ep0,
in_ep, and out ep. Fix memory leaks of usb request and its buffer.
Signed-off-by: Seung-Woo Kim <sw0312.kim@samsung.com>
The xhci wrapper-driver for Rockchip searches the DTS to find its
child node compatbile with 'rockchip,rk3399-usb3-phy' to retrieve the
base-address of the PHY. However, this is currently broken (and
always has been), returning NULL. However, the (wrongly) retrieved
base-address is never used.
We thus remove this code for now.
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
In xhci_set_configuration(), 'Context Entries' field in the slot
context was cleared with mask LAST_CTX_MASK, but it should have
taken the endianness into consideration.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
If a full speed hub connects to a high speed hub which supports MTT,
the MTT field of its slot context will be set to 1 when xHCI driver
setups an xHCI virtual device in xhci_setup_addressable_virt_dev().
Once usb core fetch its hub descriptor, and need to update the xHC's
internal data structures for the device, the HUB field of its slot
context will be set to 1 too, meanwhile MTT is also set before, this
will cause configure endpoint command fail. In the case, we should
clear MTT to 0 for full speed hub according to section 6.2.2.
This keeps in sync with Linux kernel commit:
096b110: usb: xhci: fix config fail of FS hub behind a HS hub with MTT
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Per xHCI spec chapter 6.2.2 table 6-7, as input, software shall
initialize the dev_state field to '0'. Though this does not seem
to cause any issue with most xHC implementations, let's do this
to conform with the spec.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Marek Vasut <marek.vasut@gmail.com>
Tested-by: Matthias Blankertz <matthias.blankertz@cetitec.com>
If a USB 3.0 hub is plugged into the root port of the xHC, the xHCI
driver will issue a 'Configure Endpoint' command to the xHC for it
to update its internal data structure for this hub device. The hub
attributes are in the slot context so we need tell xHC to update the
slot context by setting the add context flags of the input control
context to only cover the slot context.
At present the add context flags is or'ed with the slot context bit,
but it should really be accurately set to the slot context, as the
variable that holds the value of the add context flags comes from
whatever was set in the last command execution, which may contain
additional contexts that 'Configure Endpoint' command should not
touch. Some xHC implementations like x86 don't complain such, but
it was observed on Renesas RCar Gen3 platform that the RCar xHC
complains with a 'TRB error' completion codes as the response.
Reported-by: Marek Vasut <marek.vasut@gmail.com>
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Marek Vasut <marek.vasut@gmail.com>
Tested-by: Matthias Blankertz <matthias.blankertz@cetitec.com>
