There's no need to store an array of QTD pointers in the controller.
Since the calculation is so simple, just have ci_get_qtd() perform it
at run-time, rather than pre-calculating everything.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
2 QTDs are allocated for each EP. The current allocation scheme aligns
the first QTD in each pair, but simply adds the struct size to calculate
the second QTD's address. This will result in a non-cache-aligned
addresss IF the system's ARCH_DMA_MINALIGN is not 32 bytes (i.e. the
size of struct ept_queue_item).
Similarly, the original ilist_ent_sz calculation aligned the value to
ARCH_DMA_MINALIGN but didn't take the USB HW's 32-byte alignment
requirement into account. This doesn't cause a practical issue unless
ARCH_DMA_MINALIGN < 32 (which I suspect is quite unlikely), but we may
as well fix the code to be explicit, so it's obviously completely
correct.
The new value of ILIST_ENT_SZ takes all alignment requirements into
account, so we can simplify ci_{flush,invalidate}_qtd() by simply using
that macro rather than calling roundup().
Similarly, the calculation of controller.items[i] can be simplified,
since each QTD is evenly spaced at its individual alignment requirement,
rather than each pair being aligned, and entries within the pair being
spaced apart only by structure size.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
This will allow functions other than ci_udc_probe() to make use of the
constants in a future change.
This in turn requires converting the const int variables to #defines,
since the initialization of one global const int can't depend on the
value of another const int; the compiler thinks it's non-constant if
that dependency exists.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Fix ci_ep_submit_next_request()'s ZLP transmission code to explicitly
call ci_get_qtd() to find the address of the other QTD to use. This
will allow us to correctly align each QTD individually in the future,
which may involve leaving a gap between the QTDs.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
ci_udc_probe() initializes a pair of QHs and QTDs for each EP. After
each pair has been initialized, the pair is cache-flushed. The
conversion from QH/QTD index [0..2*NUM_END_POINTS) to EP index
[0..NUM_ENDPOINTS] is incorrect; it simply subtracts 1 (which yields
the QH/QTD index of the first entry in the pair) rather than dividing
by two (which scales the range). Fix this.
On my system, this avoids cache debug prints due to requests to flush
unaligned ranges. This is caused because the flush calls happen before
the items[] array entries are initialized for all but EP0.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
cb_getvar tries to prevent overflowing the response buffer
by using strncat. But strncat takes the number of data bytes
copied as a limit not the total buffer length so it can still
overflow. Pass the correct value instead.
cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
cc: Rob Herring <robh@kernel.org>
Signed-off-by: Jeroen Hofstee <jeroen@myspectrum.nl>
since ALLOC_CACHE_ALIGN_BUFFER defines a pointer and not a
buffer, the memset with sizeof(rqt) likely does something else
then intended. Since there is a memcpy directly after it with
the full size, drop the memset completely.
Cc: Lukasz Majewski <l.majewski@samsung.com>
Cc: Marek Vasut <marex@denx.de>
Signed-off-by: Jeroen Hofstee <jeroen@myspectrum.nl>
Acked-by: Lukasz Majewski <l.majewski@samsung.com>
Since the struct fsg_common is calloced, reset it completely
with zero's when reused. While at it, make checkpatch happy.
cc: Lukasz Majewski <l.majewski@samsung.com>
cc: Piotr Wilczek <p.wilczek@samsung.com>
cc: Kyungmin Park <kyungmin.park@samsung.com>
cc: Marek Vasut <marek.vasut@gmail.com>
Signed-off-by: Jeroen Hofstee <jeroen@myspectrum.nl>
Acked-by: Marek Vasut <marex@denx.de>
Acked-by: Lukasz Majewski <l.majewski@samsung.com>
ci_udc.c's usb_gadget_unregister_driver() doesn't call driver->unbind()
unlike other USB gadget drivers. Fix it to do this.
Without this, when ether.c's CDC Ethernet device is torn down,
eth_unbind() is never called, so dev->gadget is never set to NULL.
For some reason, usb_eth_halt() is called both at the end of the first
use of the Ethernet device, and prior to any subsequent use. Since
dev->gadget is never cleared, all calls to usb_eth_halt() attempt to
stop, disconnect, and clean up the device, resulting in double cleanup,
which hangs U-Boot on my Tegra device at least.
ci_udc allocates its own singleton EP0 request object, and cleans it up
during usb_gadget_unregister_driver(). This appears necessary when using
the USB gadget framework in U-Boot, since that does not allocate/free
the EP0 request. However, the CDC Ethernet driver *does* allocate and
free its own EP0 requests. Consequently, we must protect
ci_ep_free_request() against double-freeing the request.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Sometimes, a zero-length packet is required at the end of an IN
transaction so that the host knows the device is done sending data.
Enhance ci_udc to send a zlp when necessary. See the comments for
more details.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
usb_gadget_unregister_driver() is called to tear down the USB device mode
stack. Fix the driver to stop the USB HW (which causes any attached host
to notice the disappearance of the device), and free all allocations
(which obviously prevents memory leaks).
Signed-off-by: Stephen Warren <swarren@nvidia.com>
ci_ep_alloc_request() avoids allocating multiple request objects for ep0
by keeping a record of the first req allocated for ep0, and always
returning that instead of allocating a new req. However, if this req is
ever freed, the record of the previous allocation is not cleared, so
ci_ep_alloc_request() will keep returning this stale pointer. Fix
ci_ep_free_request() to clear the record of the previous allocation.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
ci_pullup()'s !is_on path contains a cut/paste copy of udc_disconnect().
Remove the duplication by simply calling udc_disconnect() instead.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
handle_setup() currently assumes that the response to a Setup transaction
will be an OUT transaction, and any subsequent packet (if any) will be an
IN transaction. This appears to be valid in many cases; both USB
enumeration and Mass Storage work OK with this restriction. However, DFU
uses ep0 to transfer data in both directions. This renders the assumption
invalid; when sending data from device to host, the Data Stage is an IN
transaction, and the Status Stage is an OUT transaction. Enhance
handle_setup() to deduce the correct direction for the USB transactions
based on Setup transaction data.
ep0's request object only needs to be automatically re-queued when the
Data Stage completes, in order to implement the Status Stage. Once the
Status Stage transaction is complete, there is no need to re-queue the
USB request, so don't do that.
Don't sent USB request completion callbacks for Status Stage transactions.
These were queued by ci_udc itself, and only serve to confuse the USB
function code. For example, f_dfu attempts to interpret the 0-length data
buffers for Status Stage transactions as DFU packets. These buffers
contain stale data from the previous transaction. This causes f_dfu to
complain about a sequence number mismatch.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Allocate ep0's USB request object when the UDC driver is probed. This
solves a couple of issues in the current code:
a) A request object always exists for ep0. Prior to this patch, if setup
transactions arrived in an unexpected order, handle_setup() would need
to reply to a setup transaction before any ep0 usb_req was created.
This issue was introduced in commit 2813006fec "usb: ci_udc: allow
multiple buffer allocs per ep."
b) handle_ep_complete no longer /has/ to queue the ep0 request again
after every single request completion. This is currently required, since
handle_setup() assumes it can find some request object in ep0's request
queue. This patch doesn't actually stop handle_ep_complete() from always
requeueing the request, but the next patch will.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
ci_udc currently points ep->desc at separate descriptors for IN and OUT.
These descriptors only differ in the ep address IN/OUT field. Modify the
code to use a single descriptor, and change that descriptor's ep address
to indicate IN/OUT as required. This removes some data duplication.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
The flipping of ep0 between IN and OUT relies on ci_ep_queue() consuming
the current IN/OUT setting immediately. If this is deferred to a later
point when the req is pulled out of ci_req->queue, then the IN/OUT
setting may have been changed since the req was queued, and state will
get out of sync. This condition doesn't occur today, but could if bugs
were introduced later, and this error-check will save a lot of debugging
time.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Since dfu_flush() can write raw data, dfu_write() with zero size
can be removed from download_tail() in thor gadget.
Signed-off-by: Przemyslaw Marczak <p.marczak@samsung.com>
Cc: Lukasz Majewski <l.majewski@samsung.com>
Cc: Heiko Schocher <hs@denx.de>
Cc: Marek Vasut <marex@denx.de>
ci_udc only allocates a single QTD structure per EP. All data needs to be
extracted from the DTD prior to calling ci_ep_submit_next_request(), since
that fills the QTD with next transaction's parameters. Fix
handle_ep_complete() to extract the transaction (remaining) length before
kicking off the next transaction.
In practice, this only causes writes to UMS devices to fail for me. I may
have tested the final versions of my previous ci_udc patch only with
reads. More recently, I had patches applied locally that allocated a QTD
per USB request rather than per USB EP, although since that doesn't give
any performance benefit, I'm dropping those.
Fixes: 2813006fec ("usb: ci_udc: allow multiple buffer allocs per ep")
Signed-off-by: Stephen Warren <swarren@nvidia.com>
add a possibility to add a medium specific polltimeout
function. So it is possible to define different
poll timeouts.
Used on nand medium, for setting the DFU_MANIFEST_POLL_TIMEOUT
only on nand ubi partitions, which is currently the only
usecase.
Change-Id: If1db5f49b32d93fefa7481e8dfe5b7ccc0e65af4
Signed-off-by: Heiko Schocher <hs@denx.de>
Cc: Kyungmin Park <kyungmin.park@samsung.com>
Cc: Marek Vasut <marex@denx.de>
Cc: Pantelis Antoniou <panto@antoniou-consulting.com>
Reviewed-by: Marek Vasut <marex@denx.de>
Acked-by: Lukasz Majewski <l.majewski@samsung.com>
This patch contains an implementation of the fastboot protocol on the
device side and documentation. This is based on USB download gadget
infrastructure. The fastboot function implements the getvar, reboot,
download and reboot commands. What is missing is the flash handling i.e.
writting the image to media.
v3 (Rob Herring):
This is based on http://patchwork.ozlabs.org/patch/126798/ with the
following changes:
- Rebase to current mainline and updates for current gadget API
- Use SPDX identifiers for licenses
- Traced the history and added missing copyright to cmd_fastboot.c
- Use load_addr/load_size for transfer buffer
- Allow vendor strings to be optional
- Set vendor/product ID from config defines
- Allow Ctrl-C to exit fastboot mode
v4:
- Major re-write to use the USB download gadget. Consolidated function
code to a single file.
- Moved globals into single struct.
- Use puts and putc as appropriate.
- Added CONFIG_USB_FASTBOOT_BUF_ADDR and CONFIG_USB_FASTBOOT_BUF_SIZE to
set the fastboot transfer buffer.
v5:
- Add CONFIG option documentation to README
- Rebase using new downloader registration
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Rob Herring <robh@kernel.org>
Now that the ci_udc driver supports allocating multiple requests per
endpoint, we can revert the special-case added by a022c1e13c "usb:
ums: use only 1 buffer for CI_UDC".
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Modify ci_ep_alloc_request() to return a dynamically allocated request
object, rather than a singleton that's part of the endpoint. This
requires moving various state from the endpoint structure to the request
structure, since we need one copy per request.
The "fast bounce buffer" b_fast is removed by this change rather than
moved to the request object. Instead, we enhance the bounce buffer logic
in ci_bounce()/ci_debounce() to keep the bounce buffer around between
request submissions. This avoids the need to allocate an arbitrarily-
sized bounce buffer up-front, yet avoids incurring the allocation
overhead each time a request is submitted.
A future enhancement would be to actually submit multiple requests to HW
at once. The Linux driver shows that this is possible. That might improve
throughput (depending on the USB protocol in use), since USB could be
performing a transfer to one HW buffer in parallel with whatever SW
actions U-Boot performs on another buffer. However, I have not made this
change as part of this patch, in order to keep SW changes related to
buffer management separate from any change in the way the HW is
programmed.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
g_dnl_register() currently first attempts to register a composite
driver by name, and then saves the driver name once it's registered.
Internally to the registration code, g_dnl_do_config() is called and
attempts to compare the composite device's name with the list of known
device names. This fails since the composite device's name has not yet
been stored. This means that the first time "ums 0 0" is run, it fails,
but subsequent attempts succeed.
Re-order the name-saving and registration code to solve this.
Fixes: e5b834e07f51 ("USB: gadget: added a saner gadget downloader registration API")
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Preprocessor definitions and hardcoded implementation selection in
g_dnl core were replaced by a linker list made of (usb_function_name,
bind_callback) pairs.
Signed-off-by: Mateusz Zalega <m.zalega@samsung.com>
Acked-by: Lukasz Majewski <l.majewski@samsung.com>
Acked-by: Marek Vasut <marex@denx.de>
Former usb_cable_connected() patch broke compilation of boards which do
not support this feature.
I've renamed usb_cable_connected() to g_dnl_usb_cable_connected() and added
its default implementation to gadget downloader driver code. There's
only one driver of this kind and it's unlikely there'll be another, so
there's no point in keeping it in /common.
Previously this function was declared in usb.h. I've moved it, since
it's more appropriate to keep it in g_dnl.h - usb.h seems to be intended
for USB host implementation.
Existing code, confronted with default -EOPNOTSUPP return value,
continues as if the cable was connected.
CONFIG_USB_CABLE_CHECK was removed.
Change-Id: Ib9198621adee2811b391c64512f14646cefd0369
Signed-off-by: Mateusz Zalega <m.zalega@samsung.com>
Acked-by: Marek Vasut <marex@denx.de>
Acked-by: Lukasz Majewski <l.majewski@samsung.com>
Allow ci_udc.o to be built when using the new(?) USB gadget framework,
as enabled by CONFIG_USB_GADGET.
Note that this duplicates the Makefile entry for ci_udc.o, since it's
also included inside #ifdef CONFIG_USB_ETHER. I'm not sure what that
define means; perhaps an old style of Ethernet-specific USB gadget
implementation?
I wonder if the line that this patch adds shouldn't be outside all of
the ifdefs, so it stands on its own, similar to how e.g. epautoconf.o
is shared between the two?
Signed-off-by: Stephen Warren <swarren@nvidia.com>
ci_udc.c allocates only a single buffer for each endpoint, which
ci_ep_alloc_request() returns as a hard-coded value rather than
dynamically allocating. Consequently, storage_common.c must limit
itself to using a single buffer at a time. Add a special case
to the definition of FSG_NUM_BUFFERS for this.
Another option would be to fix ci_ep_alloc_request() to dynamically
allocate the buffers like some/all(?) other device mode drivers do.
However, I don't think that ci_ep_queue() supports queueing up
multiple buffers either yet, and I'm not familiar enough with the
controller yet to implement that. As such, any attempt to use multiple
buffers simply results in data corruption and other errors.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Tegra's USB controller appears to be a variant of the ChipIdea
controller; perhaps derived from it, or simply a different version of
the IP core to what U-Boot supports today.
In this variant, at least the following difference are present:
- Some registers are moved about.
- Setup transaction completion is reported in a separate 'epsetupstat'
register, rather than in 'epstat' (which still exists, perhaps for
other transaction types).
- USB connection speed is reported in a separate 'hostpc1_devlc'
register, rather than 'portsc'.
- The registers used by ci_udc.c begin at offset 0x130 from the USB
register base, rather than offset 0x140. However, this is handled
by the associated EHCI controller driver, since the register address
is stored in controller.ctrl->hcor.
Introduce define CONFIG_CI_UDC_HAS_HOSTPC to indicate which variant of
the controller should be supported. The "HAS_HOSTPC" part of this name
mirrors the similar "has_hostpc" field used by the Linux EHCI controller
core to represent the presence/absence of the hostpc1_devlc register.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
usb_gadget_register_driver() currently unconditionally programs PORTSC
to select a ULPI PHY. This is incorrect on at least the Tegra boards I
am testing with, which use a UTMI PHY for the OTG ports. Make the PHY
selection code conditional upon the specific EHCI controller that is in
use.
Ideally, I believe that the PHY initialization code should be part of
ehci_hcd_init() in the relevant EHCI controller driver, or some board-
specific function that ehci_hcd_init() calls.
For MX6, I'm not sure this PHY initialization code is correct even before
this patch, since ehci-mx6's ehci_hcd_init() already configures PORTSC to
a board-specific value, and it seems likely that the code in ci_udc.c is
incorrectly undoing this. Perhaps this is not an issue if the PHY
selection register bits aren't implemented on this instance of the MX6
USB controller?
ehci-mxs.c doens't appear to touch PORTSC, so this code is likely still
required there.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
At least drivers/usb/gadget/storage_common.c expects that ep->req.actual
contain the number of bytes actually transferred. (At least in practice,
I observed it failing to work correctly unless this was the case).
However, ci_udc.c modifies ep->req.length instead. I assume that .length
is supposed to represent the allocated buffer size, whereas .actual is
supposed to represent the actual number of bytes transferred. In the OUT
transaction case, this may happen simply because the host sends a smaller
packet than the max possible size, which is quite legal. In the IN case,
transferring fewer bytes than requested could presumably happen as an
error.
Modify handle_ep_complete() to write to .actual rather than modifying
.length.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
ci_ep_queue() currently only fills in the page0/page1 fields in the
queue item. If the buffer is larger than 4KiB (unaligned) or 8KiB
(page-aligned), then this prevents the HW from knowing where to write
the balance of the data.
Fix this by initializing all 5 pageN pointers, which allows up to
16KiB (potentially non-page-aligned) buffers.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Allow a NULL table to be passed to usb_gadget_get_string for cases
when a string table may not be populated.
Signed-off-by: Rob Herring <robh@kernel.org>
Reviewed-by: Tom Rini <trini@ti.com>
Acked-by: Marek Vasut <marex@denx.de>
Acked-by: Lukasz Majewski <l.majewski@samsung.com>
Since dfu read/write operations needs to be flushed manually,
writing to filesystem on MMC by thor was broken. MMC raw write
actually is working fine because current dfu_flush() function
writes filesystem only. This commit adds dfu_flush() to f_thor
and now filesystem write is working.
This change was tested on Trats2 board.
Signed-off-by: Przemyslaw Marczak <p.marczak@samsung.com>
Cc: Lukasz Majewski <l.majewski@samsung.com>
Cc: Marek Vasut <marex@denx.de>
Cc: Heiko Schocher <hs@denx.de>
Cc: Tom Rini <trini@ti.com>
In thor's download_tail() function, dfu_get_entity() is called
before each dfu_write() call and the returned entity pointers
are the same. So dfu_get_entity() can be called just once and
this patch changes this.
Signed-off-by: Przemyslaw Marczak <p.marczak@samsung.com>
Cc: Lukasz Majewski <l.majewski@samsung.com>
Cc: Marek Vasut <marex@denx.de>
Cc: Heiko Schocher <hs@denx.de>
Cc: Tom Rini <trini@ti.com>
on nand flash using ubi, after the download of the new image into
the flash, the "rest" of the nand sectors get erased while flushing
the medium. With current u-boot version dfu-util may show:
Starting download: [##################################################] finished!
state(7) = dfuMANIFEST, status(0) = No error condition is present
unable to read DFU status
as get_status is not answered while erasing sectors, if erasing
needs some time.
So do the following changes to prevent this:
- introduce dfuManifest state
According to dfu specification
( http://www.usb.org/developers/devclass_docs/usbdfu10.pdf ) section 7:
"the device enters the dfuMANIFEST-SYNC state and awaits the solicitation
of the status report by the host. Upon receipt of the anticipated
DFU_GETSTATUS, the device enters the dfuMANIFEST state, where it
completes its reprogramming operations."
- when stepping into dfuManifest state, sending a PollTimeout
DFU_MANIFEST_POLL_TIMEOUT in ms, to the host, so the host
(dfu-util) waits the PollTimeout before sending a get_status again.
Signed-off-by: Heiko Schocher <hs@denx.de>
Cc: Lukasz Majewski <l.majewski@samsung.com>
Cc: Kyungmin Park <kyungmin.park@samsung.com>
Cc: Marek Vasut <marex@denx.de>
Cc: Pantelis Antoniou <panto@antoniou-consulting.com>
Copied from Linux sources "include/linux/sizes.h" commit
413541dd66d51f791a0b169d9b9014e4f56be13c
Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
Cc: Vineet Gupta <vgupta@synopsys.com>
Cc: Tom Rini <trini@ti.com>
Cc: Stefan Roese <sr@denx.de>
Cc: Albert Aribaud <albert.u.boot@aribaud.net>
Acked-by: Tom Rini <trini@ti.com>
Acked-by: Stefan Roese <sr@denx.de>
[trini: Add bcm Kona platforms to the patch]
Signed-off-by: Tom Rini <trini@ti.com>
The mv_udc is not marvell-specific anymore. The mv_udc is used to drive
generic ChipIdea CI13xxx series OTG cores, so rename the driver to ci_udc
instead.
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Eric Nelson <eric.nelson@boundarydevices.com>
Cc: Stefano Babic <sbabic@denx.de>
Apparently debug memset (with a 0x55 value) has been overlooked in the
f_thor code.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Cc: Marek Vasut <marex@denx.de>
Now it is possible to allocate static request - which receives data from
the host (OUT transaction) to the size of THOR packet.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Cc: Marek Vasut <marex@denx.de>
The Samsung's UDC driver is not anymore copying data from USB requests to
aligned internal buffers. Now it works directly in data allocated in the
upper layers like UMS, DFU, THOR.
This change is possible since those gadgets now must take care to allocate
buffers aligned to cache line (CONFIG_SYS_CACHELINE_SIZE).
This can be achieved by using DEFINE_CACHE_ALIGN_BUFFER() or
ALLOC_CACHE_ALIGN_BUFFER() macros. Those take care to allocate buffer
aligned to cache line in both starting address and its size.
Sometimes it is enough to just use memalign() with size being a
multiplication of cache line size.
Test condition
- test HW + measurement: Trats - Exynos4210 rev.1
- test HW Trats2 - Exynos4412 rev.1
400 MiB compressed rootfs image download with `thor 0 mmc 0`
Measurement:
Transmission speed: 27.04 MiB/s
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Cc: Marek Vasut <marex@denx.de>
This patch removed obscure restriction on the HW setting of DMA transfers.
Before this change each transaction sent up to 512 bytes (with packet count
equal to 1) for non EP0 transfer.
Now it is possible to setup DMA transaction up to DMA_BUFFER_SIZE.
Test condition
- test HW + measurement: Trats - Exynos4210 rev.1
- test HW Trats2 - Exynos4412 rev.1
400 MiB compressed rootfs image download with `thor 0 mmc 0`
Measurement:
Transmission speed: 20.74 MiB/s
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Cc: Marek Vasut <marex@denx.de>
A set of cache operations (both invalidation and flush) were redundant
in the S3C HS OTG Samsung driver:
1. s3c_udc_ep0_zlp - to transmit EP0's ZLP packets one don't need to flush
the cache (since it is the zero length transmission)
2. s3c_udc_pre_setup and s3c_ep0_complete_out - cache invalidation is not
needed when the buffer for OUT EP0 transmission is setup, since no data
has yet arrived.
Cache cleanups presented above don't contribute much to transmission speed
up, hence shall be regarded as cosmetic changes.
3. setdma_rx - here the s3c UDC driver's internal buffers were invalidated.
This call is not needed anymore since we reuse the buffers passed from
gadgets. This is a key contribution to transmission speed improvement.
Test condition
- test HW + measurement: Trats - Exynos4210 rev.1
- test HW Trats2 - Exynos4412 rev.1
400 MiB compressed rootfs image download with `thor 0 mmc 0`
Measurements:
Base values (without improvement):
Transmission speed: 9.51 MiB/s
After the change:
Transmission speed: 10.15 MiB/s
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Cc: Marek Vasut <marex@denx.de>
Calls to malloc() have been replaced by memalign. It now provides proper
buffer alignment.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Cc: Marek Vasut <marex@denx.de>
The fifo size of ep0 is 64 bytes, and if the packet size grater than
64 bytes, the driver would have to fill up the fifo multiple times,
and before filling up the fifo, the driver should make sure the fifo
is empty by checking fifo empty indication.
However there is a hardware bug that the fifo empty indication is
somehow a bit earlier than fifo reset. So if I don't add an extra
delay here, the data might be corrupted. (i.e., 1 byte missing)
After a couple of tests, it truns out that 1 usec is good enough.
This workaround should be applied to all hardware revisions.
Signed-off-by: Kuo-Jung Su <dantesu@faraday-tech.com>
CC: Marek Vasut <marex@denx.de>
Since hardware revision 1.11.0, the following interrupt status
registers are now W1C (i.e., write 1 clear):
1. Interrupt Source Group 0 Register (0x144) (EP0 Abort: BIT5)
2. Interrupt Source Group 2 Register (0x14C) (All bits)
And before revision 1.11.0, these registers are all R/W.
Which means software must write a 0 to clear the status.
Signed-off-by: Kuo-Jung Su <dantesu@faraday-tech.com>
CC: Marek Vasut <marex@denx.de>
It is necessary to deter the host from sending subsequent DFU_GETSTATUS
request in the case of e.g. writing the buffer to medium.
Here the timeout is increased when we fill up the whole buffer. This delay
allows eMMC memory to perform its internal operations.
Otherwise we end up with HOST's error regarding GET_STATUS receive timeout.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Before switching to the real Kbuild, drivers/usb/gadget/Makefile
must be fixed.
If none of CONFIG_USB_GADGET, CONFIG_USB_ETHER, CONFIG_USB_DEVICE
is defined, both obj- and obj-y get empty.
We need non-empty obj- or obj-y on each Makefile
to generate built-in.o on the real Kbuild.
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
This patch add new defines for usb phy for Exynos4x12.
Signed-off-by: Piotr Wilczek <p.wilczek@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
add the possibility to set the bcdDevice number board specific.
Therefore the weak function g_dnl_get_board_bcd_device_number()
is introduced. Used on the siemens boards.
Signed-off-by: Heiko Schocher <hs@denx.de>
Acked-by: Lukasz Majewski <l.majewski@samsung.com>
Cc: Marek Vasut <marek.vasut@gmail.com>
Cc: Kyungmin Park <kyungmin.park@samsung.com>
add the possibility to set the iSerialNumber board specific.
Default value for iSerialNumber is 0x0. This value can
changed board specific through the new function
g_dnl_set_serialnumber() which must be called from the
board specific function g_dnl_bind_fixup().
Signed-off-by: Heiko Schocher <hs@denx.de>
Cc: Marek Vasut <marek.vasut@gmail.com>
Cc: Lukasz Majewski <l.majewski@samsung.com>
Cc: Kyungmin Park <kyungmin.park@samsung.com>
Tested-by: Lukasz Majewski <l.majewski@samsung.com>
This patch allows exiting from UMS mode to u-boot prompt
by detaching usb cable or by pressing ctrl+c.
Add new config: CONFIG_USB_CABLE_CHECK. If defined then board
file should provide function: usb_cable_connected() (include/usb.h)
that return 1 if cable is connected and 0 otherwise.
Changes v2:
- add a note to the README
Signed-off-by: Przemyslaw Marczak <p.marczak@samsung.com>
Cc: Marek Vasut <marex@denx.de>
This patch prevents:
- ums disk capacity miscalculation because of integer overflow
Changes v2:
- Prevents passing zero size disk capacity to ums gadget driver
- Change function ums_get_capacity() to ums_disk_init() and do ums disk
initialization before gadget init
- Remove unnecessary code from mass storage driver
Signed-off-by: Przemyslaw Marczak <p.marczak@samsung.com>
Cc: Marek Vasut <marex@denx.de>
This patch introduces some cleanups to ums code. Changes:
ums common:
- introduce UMS_START_SECTOR and UMS_NUM_SECTORS as defined in
usb_mass_storage.h both default values as 0 if board config
doesn't define them
common cleanup changes:
- change name of struct "ums_board_info" to "ums"
- "ums_device" fields are moved to struct ums and "dev_num" is removed
- change function name: board_ums_init to ums_init
- remove "extern" prefixes from usb_mass_storage.h
cmd_usb_mass_storage:
- change error() to printf() if need to print info message
- change return values to command_ret_t type at ums command code
- add command usage string
Changes v2:
ums common:
- always returns number of read/write sectors
- coding style clean-up
ums gadget:
- calculate amount of read/write from device returned value.
Signed-off-by: Przemyslaw Marczak <p.marczak@samsung.com>
Cc: Marek Vasut <marek.vasut@gmail.com>
Move defines only needed by mv_udc.c to a file
in the same directory.
This allows usbtty to compile for mv_udc,
but it still doesn't link.
Signed-off-by: Troy Kisky <troy.kisky@boundarydevices.com>
Only perform one copy, either in the bounce
routine for IN transfers, or the debounce
rtn for OUT transfer.
On out transfers, only copy the number
of bytes received from the bounce buffer
Signed-off-by: Troy Kisky <troy.kisky@boundarydevices.com>
i.mx6 has 1 otg controller, and 3 host ports. So,
CONFIG_USB_MAX_CONTROLLER_COUNT can be greater than 1
even though only 1 device mode controller is supported.
Signed-off-by: Troy Kisky <troy.kisky@boundarydevices.com>
Implementation of USB download function which supports THOR protocol.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Cc: Marek Vasut <marex@denx.de>
New parameter, namely *name has been added to g_dnl_bind_fixup().
It is necessary (for compatibility reasons) to assign new USB idProduct
and idVendor for different usb functions.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Cc: Marek Vasut <marex@denx.de>
When the usb_configuration structure is declared as static, it is very
hard to assure, that relevant fields (as e.g. config->interfaces[]) are
cleared out before new call to g_dnl related functions.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
The s3c udc driver sends data in a max packet size. Therefore the dcache
invalidate range shall be equal to max packet, not the entire
DMA_BUFFER_SIZE.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Cc: Marek Vasut <marex@denx.de>
Make sure the transfer descriptor is flushed
before the queue is updated so that the controller
will not see old information.
Signed-off-by: Troy Kisky <troy.kisky@boundarydevices.com>
This prevents a crash if tftpboot is given a bad filename.
rx_req will be released by eth_reset_config
which is called by eth_disconnect,
which is called using the .disconnect member of usb_gadget_driver by mv_pullup in mv_udc
which is called using the .pullup member of usb_gadget_ops by usb_gadget_disconnect
which is called by usb_eth_halt
which is called using the .halt member of eth_device by eth_halt
which is called by TftpHandler when TFTP_ERR_FILE_NOT_FOUND or TFTP_ERR_ACCESS_DENIED occurs
I trigger this with the following commands
setenv ipaddr 10.0.0.2 && setenv netmask 255.255.255.0 && setenv serverip 10.0.0.1
setenv usbnet_devaddr 00:11:22:33:44:55 && setenv usbnet_hostaddr 00:aa:bb:cc:dd:ee
setenv ethprime usb_ether && setenv ethact usb_ether && setenv ncip 10.0.0.1
tftpboot 10800000 10.0.0.1:missing_file
Signed-off-by: Troy Kisky <troy.kisky@boundarydevices.com>
set wMaxPacketSize for full speed descriptors
fs_source_desc, fs_sink_desc to 64.
Full-speed bulk endpoint can have a maximum packet size of
8, 16, 32, or 64 bytes, so choice 64.
The hs_source_desc, hs_sink_desc, already have their wMaxPacketSize
set to 512. That is the only legal value for high speed bulk endpoints.
Strictly speaking, this patch is not needed because
usb_ep_autoconfig will call ep_matches which will
override wMaxPacketSize for BULK endpoints only with
the size associated with the endpoint setup by the udc driver.
But if you want to rely on this, you may as well combine the
full speed descriptor with the high speed descriptor to
minimize confusion.
Signed-off-by: Troy Kisky <troy.kisky@boundarydevices.com>
The download gadget code and DFU function lacks of proper declarations
for the case when a target board wants to use only one of available usb
functions.
Moreover the relevant declarations have been moved to consistent
localization (like <dfu.h>).
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Cc: Marek Vasut <marex@denx.de>
Only the <linux/usb/gadget.h> requires error.h include. Hence, several
includes of error.h at USB gadget functions are not needed.
Moreover unnecessary malloc.h includes were also removed.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Cc: Marek Vasut <marex@denx.de>
The mass storage composite function is now compiled in only when
CONFIG_USB_GADGET_MASS_STORAGE is defined.
Such change provides binary size reduction for boards which use USB
download gadget (like am335x_evm) with DFU, but don't use UMS.
For example at am335x_evm board reduction is more than 2KiB for
text and around 120B for data.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Cc: Marek Vasut <marex@denx.de>
Add atmel usba udc driver support, porting from Linux kernel
The original code in Linux Kernel information is as following
commit e01ee9f509a927158f670408b41127d4166db1c7
Author: Jingoo Han <jg1.han@samsung.com>
Date: Tue Jul 30 17:00:51 2013 +0900
usb: gadget: use dev_get_platdata()
Use the wrapper function for retrieving the platform data instead of
accessing dev->platform_data directly.
Signed-off-by: Bo Shen <voice.shen@atmel.com>
As seen with codesourcery compiler 2010q1, the buf pointer in
usb_request structure is not aligned on 4 bytes boundary causing
data aborts in eth_setup -> conf_buf -> usb_gadget_config_buf.
Make it as align access to fix this issue.
Signed-off-by: Troy Kisky <troy.kisky@boundarydevices.com>
[voice.shen@atmel.com: add commit message]
Signed-off-by: Bo Shen <voice.shen@atmel.com>
As seen on GCC 4.6 Linaro compiler, control_req buffer is not aligned
on 4 byte boundaray causing data aborts in eth_setup -> conf_buf
during dhcp boot over usb_ether. Fix the issue my aligning control_req
buffer using DEFINE_CACHE_ALIGN_BUFFER.
Tested on am335x_evm platform (beaglebone).
Applies on 2013.10-rc1 branch.
Cc: Tom Rini <trini@ti.com>
Cc: Marek Vasut <marex@denx.de>
Signed-off-by: Joel Fernandes <joelf@ti.com>
The manufacturer and product IDs are dynamically assigned when gadget is
bind.
Now the IDs aren't assigned at struct g_dnl_string_defs definition.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Cc: Marek Vasut <marex@denx.de>
Cc: "Egli, Samuel" <samuel.egli@siemens.com>
USB composite gadget (g_dnl) supports only one configuration. Due to that
the corresponding field - bConfigurationValue has been changed.
Moreover more descriptive names were chosen for relevant fields.
Windows XP setup:
- Thesyscon USB Descriptor Dumper
- zadig_xp program for WinUSB installation (which is required by dfu-util)
- dfu-util for windows (version 0.6)
- TRATS target connected via USB hub to test Win XP machine.
Tested at: Trats - Exynos4210
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Cc: Marek Vasut <marex@denx.de>
Cc: "Egli, Samuel" <samuel.egli@siemens.com>
This patch extends dfu code to support transmission with -R switch
specified at dfu-util.
When -R is specified, the extra USB_REQ_DFU_DETACH request is sent after
successful data transmission. Then dfu resources are released and reset
command is issued.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
The requests sent to the controller are not properly cache aligned
most of the time, thus implement a simple bounce buffer to avoid
problem with cache.
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Fabio Estevam <fabio.estevam@freescale.com>
Cc: Lei Wen <leiwen@marvell.com>
Cc: Otavio Salvador <otavio@ossystems.com.br>
Cc: Stefano Babic <sbabic@denx.de>
Implement functions to flush/invalidate dcache over QH and qTDs
and make use of them where appropriate. Also use them to replace
the old incorrect cache management attempt. This is the first step
towards making this driver work with data cache enabled.
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Fabio Estevam <fabio.estevam@freescale.com>
Cc: Lei Wen <leiwen@marvell.com>
Cc: Otavio Salvador <otavio@ossystems.com.br>
Cc: Stefano Babic <sbabic@denx.de>
The code for retrieving qTD item for particular endpoint is hard
to understand, moreover it's duplicated all over the driver. Move
the code into single nice and documented function.
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Fabio Estevam <fabio.estevam@freescale.com>
Cc: Lei Wen <leiwen@marvell.com>
Cc: Otavio Salvador <otavio@ossystems.com.br>
Cc: Stefano Babic <sbabic@denx.de>
Allocate the qTD items all at once instead of allocating them
separately. Moreover, make sure each qTD is properly aligned
to 32-bytes boundary and that cache can be safely flushed over
each qTD touple.
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Fabio Estevam <fabio.estevam@freescale.com>
Cc: Lei Wen <leiwen@marvell.com>
Cc: Otavio Salvador <otavio@ossystems.com.br>
Cc: Stefano Babic <sbabic@denx.de>
The code for retrieving QH for particular endpoint is hard to understand,
moreover it's duplicated all over the driver. Move the code into single
nice and documented function.
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Fabio Estevam <fabio.estevam@freescale.com>
Cc: Lei Wen <leiwen@marvell.com>
Cc: Otavio Salvador <otavio@ossystems.com.br>
Cc: Stefano Babic <sbabic@denx.de>
Check the length of system cacheline at compile-time and fail
if the system uses too long cachelines.
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Fabio Estevam <fabio.estevam@freescale.com>
Cc: Lei Wen <leiwen@marvell.com>
Cc: Otavio Salvador <otavio@ossystems.com.br>
Cc: Stefano Babic <sbabic@denx.de>
The endpoint QH list has to be aligned to 10-bit boundary. We also have
to make sure the list is aligned on a cacheline boundary. Make sure it
is. Furthermore, check if the memory allocation for the QH list didn't
fail. Moveover, improve the comment about the QH list structure.
Finally, the qTD item list has to be aligned only to 5-bit boundary, not
10-bit as it is now, fix this as well.
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Fabio Estevam <fabio.estevam@freescale.com>
Cc: Lei Wen <leiwen@marvell.com>
Cc: Otavio Salvador <otavio@ossystems.com.br>
Cc: Stefano Babic <sbabic@denx.de>
Both the endpoint queue head and the endpoint item list is a controller
specific thing. Move them both into controller private data.
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Fabio Estevam <fabio.estevam@freescale.com>
Cc: Lei Wen <leiwen@marvell.com>
Cc: Otavio Salvador <otavio@ossystems.com.br>
Cc: Stefano Babic <sbabic@denx.de>
There is no need to init this field at runtime, so init it statically.
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Fabio Estevam <fabio.estevam@freescale.com>
Cc: Lei Wen <leiwen@marvell.com>
Cc: Otavio Salvador <otavio@ossystems.com.br>
Cc: Stefano Babic <sbabic@denx.de>
The QH_MAXNUM is used in absolutelly incorrect manner and is not
even needed. Remove it and correctly replace it's occurance with
2 * NUM_ENDPOINTS .
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Fabio Estevam <fabio.estevam@freescale.com>
Cc: Lei Wen <leiwen@marvell.com>
Cc: Otavio Salvador <otavio@ossystems.com.br>
Cc: Stefano Babic <sbabic@denx.de>
Clean up the code that checks the validity of a USB gadget driver
in usb_gadget_register_driver(). Moreover, limit the speed of the
driver to either FULL or HIGH, this is more precise and once we
have xHCI support, also more correct.
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Fabio Estevam <fabio.estevam@freescale.com>
Cc: Lei Wen <leiwen@marvell.com>
Cc: Otavio Salvador <otavio@ossystems.com.br>
Cc: Stefano Babic <sbabic@denx.de>
The usb_lowlevel_init() call already fills and passes back struct
ehci_ctrl , which readily contains correctly determined address of
the port register block address computed from values from controller
configuration registers. Leverage this and make use of this value
as this makes the code mode universal, but also gets us rid of the
CONFIG_USB_REG_BASE configuration option.
Moreover, this patch cleans up the usb_gadget_register_driver() call
a little by correcting the error handling. Note the usb_lowlevel_init()
and mvudc_probe() are now called in reversed order, but this has no
impact on the code.
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Fabio Estevam <fabio.estevam@freescale.com>
Cc: Lei Wen <leiwen@marvell.com>
Cc: Otavio Salvador <otavio@ossystems.com.br>
Cc: Stefano Babic <sbabic@denx.de>
Move the constant values that are programmed into mv_ep.ep into
separate static const structure so they can be memcpy()'d when
the initialization happens.
Moveover, we only every init NUM_ENDPOINTS, not 2 * NUM_ENDPOINTS,
so fix this bug as well.
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Fabio Estevam <fabio.estevam@freescale.com>
Cc: Lei Wen <leiwen@marvell.com>
Cc: Otavio Salvador <otavio@ossystems.com.br>
Cc: Stefano Babic <sbabic@denx.de>
The endpoints are operated on a per-controller basis, move the
endpoint array into controller's private data. Also shuffle the
struct mv_ep structure definition just above the definition of
the struct mv_drv so they're well grouped together.
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Fabio Estevam <fabio.estevam@freescale.com>
Cc: Lei Wen <leiwen@marvell.com>
Cc: Otavio Salvador <otavio@ossystems.com.br>
Cc: Stefano Babic <sbabic@denx.de>
The mv_udc driver is broken for a while and doesn't even compile.
This patch fixes the issues and gets the driver into working state
again. This driver was tested on Freescale i.MX233/i.MX28 .
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Fabio Estevam <fabio.estevam@freescale.com>
Cc: Lei Wen <leiwen@marvell.com>
Cc: Otavio Salvador <otavio@ossystems.com.br>
Cc: Stefano Babic <sbabic@denx.de>
Newer gcc versions warn about unused variables. This patch corrects a few of
those warnings that popped up in a build for the palmtreo680 board.
Signed-off-by: Mike Dunn <mikedunn@newsguy.com>
The DFU's state machine original implementation author and copyright were
missing.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Acked-by: Stefan Schmidt <stefan@datenfreihafen.org>
This subtle change fix problem with too small amount of allocated
memory to store DFU function pointers.
One needs to allocate extra space for sentinel NULL pointer in this array
of function pointers.
With the previous code, the NULL value overwrites malloc internal data
and afterwards free(f_dfu->function) crashes.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Cc: Marek Vasut <marex@denx.de>
add a weak dummy function g_dnl_fixup to add the possibility to update
the device_desc board specific. Used on the upcoming siemens board
support, where idVendor and idProduct is stored in an eeprom.
Signed-off-by: Heiko Schocher <hs@denx.de>
Cc: Marek Vasut <marek.vasut@gmail.com>
Cc: Lukasz Majewski <l.majewski@samsung.com>
Cc: Kyungmin Park <kyungmin.park@samsung.com>
Reviewed-by: Lukasz Majewski <l.majewski@samsung.com>
- set in to_dfu_mode() f_dfu->dfu_state = DFU_STATE_dfuIDLE
as after every to_dfu_mode call this is done, so move
this into to_dfu_mode
- switch in dfu_bind() into dfu mode:
This fixes wrong "dfu-util -l" output, when calling
"dfu-util -l" after a board reset, without doing a
download before. See also discussion here:
http://lists.denx.de/pipermail/u-boot/2013-June/157272.html
Signed-off-by: Heiko Schocher <hs@denx.de>
Cc: Lukasz Majewski <l.majewski@samsung.com>
Cc: Pantelis Antoniou <panto@antoniou-consulting.com>
Cc: Tom Rini <trini@ti.com>
Cc: Kyungmin Park <kyungmin.park@samsung.com>
Cc: Marek Vasut <marex@denx.de>
Cc: Wolfgang Denk <wd@denx.de>
Cc: Samuel Egli <samuel.egli@siemens.com>
Acked-by: Lukasz Majewski <l.majewski@samsung.com>
Tested-by: Lukasz Majewski <l.majewski@samsung.com>
This patch memcpy is used instead of an assignment to
avoid unaligned access execption on some ARM platforms.
Signed-off-by: Piotr Wilczek <p.wilczek@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
CC: Lukasz Majewski <l.majewski@samsung.com>
Tested-by: Heiko Schocher <hs@denx.de>
Acked-by: Heiko Schocher <hs@denx.de>
This patch use get|put_unaligned_le16 to access structure data
to avoid data abort on some ARM platforms.
Signed-off-by: Piotr Wilczek <p.wilczek@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
CC: Marek Vasut <marex@denx.de>
Without this, second usb_composite_register() call fails always
with -EINVAL.
Signed-off-by: Heiko Schocher <hs@denx.de>
Cc: Lukasz Majewski <l.majewski@samsung.com>
Cc: Kyungmin Park <kyungmin.park@samsung.com>
Cc: Marek Vasut <marex@denx.de>
The Faraday FOTG210 is an OTG chip which could operate
as either an EHCI Host or a USB Device at a time.
Signed-off-by: Kuo-Jung Su <dantesu@faraday-tech.com>
CC: Marek Vasut <marex@denx.de>
Use get_unaligned() while fetching wMaxPacketSize to avoid
voilating any alignment rules.
Signed-off-by: Vivek Gautam <gautam.vivek@samsung.com>
Cc: Lukasz Majewski <l.majewski@samsung.com>
Cc: Piotr Wilczek <p.wilczek@samsung.com>
Cc: Kyungmin Park <kyungmin.park@samsung.com>
Cc: Lukasz Dalek <luk0104@gmail.com>
Cc: Marek Vasut <marex@denx.de>
If CONFIG_USB_DEV_PULLUP_GPIO is defined, a link error occurs because the
set_GPIO_mode() helper function is not implemented. This function doesn't do
much except make the code a little more readable, so I just manually coded its
equivalent and removed the prototype from the header file. It is invoked no
where else in the code.
While I was at it, I noticed that two other function prototypes in the same
header file are also neither implemented nor invoked anywhere, so I removed them
as well.
Signed-off-by: Mike Dunn <mikedunn@newsguy.com>
This patch adds the USB Mass Storage Gadget to u-boot
New command called "ums" is implemented to provide access
to on-device embedded persistent memory.
USB Mass Storage is supposed to work on top of the USB
Gadget framework
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Signed-off-by: Piotr Wilczek <p.wilczek@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
CC: Marek Vasut <marek.vasut@gmail.com>
The storage_common.c source file from v2.6.36 Linux kernel.
commit d26a6aa08b9f12b44fb1ee65625e7480d3d5bb81
Author: Michal Nazarewicz <m.nazarewicz@samsung.com>
Date: Mon Nov 9 14:15:23 2009 +0100
USB: g_mass_storage: code cleaned up and comments updated
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Signed-off-by: Andrzej Pietrasiewicz <andrzej.p@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
CC: Marek Vasut <marek.vasut@gmail.com>
This commit fixes problems with some non-standard requests send with
device address instead of interface address (bmRequestType.Receipent field).
This happens with dfu-util (debian version: 0.5), which address non standard
requests (like w_value=0x21 and bRequest=GET_DESCRIPTOR) to device.
Without this fix, the above request is STALLED, and hence causes dfu-util
to assume some standard configuration (packet size = 1024B instead of 4096B)
In turn it displays following errors:
Error obtaining DFU functional descriptor
Warning: Assuming DFU version 1.0
Warning: Transfer size can not be detected
...
Warning: Trying default transfer size 1024
This fix allows passing non-standard request to function setup code, where
it shall be handled.
Tested at: Trats (exynos4210)
Tested with:DFU and UMS gadgets
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
When both CONFIG_USB_GADGET & CONFIG_USB_ETHER are defined
the makefile links objects twice.
This patch uses a Makefile specific idiom of
'if defined(CONFIG_USB_GADGET) || defined(CONFIG_USB_ETHER)'
to handle the case.
Signed-off-by: Pantelis Antoniou <panto@antoniou-consulting.com>
Zero out timeout value; letting it filled with undefined values
ends up with the dfu host hanging.
Signed-off-by: Pantelis Antoniou <panto@antoniou-consulting.com>
Well, not terminating the list causes very interesting crashes.
As in changing the vendor & product ID crashes. Fun.
Signed-off-by: Pantelis Antoniou <panto@antoniou-consulting.com>
Existing U-Boot musb driver has no support for the new gadget framework
and also seems to have other limitations. As gadget framework is ported
from Linux it seems pretty natural to port musb gadget driver as well.
This driver supports both host and peripheral modes.
This is not a replacement for current musb driver (at least now) as
there are still some consumers of the old UDC interface.
No DMA operation support included, CONFIG_MUSB_PIO_ONLY should be
defined.
Virtual root hub device is not implemented.
Known problems: with no devices connected usb_lowlevel_start() fails.
Signed-off-by: Ilya Yanok <ilya.yanok@cogentembedded.com>
The gadget ethernet driver needs to keep copies of the MAC address (at
both endpoints) as strings so it needs a custom function for validation
of the MAC. It was not however performing a totally correct check and
also was emitting a warning about a set but unused variable. The
solution to both is that after checking the string contents we use the
standard test for a valid MAC.
Cc: Joe Hershberger <joe.hershberger@ni.com>
Cc: Marek Vasut <marex@denx.de>
Signed-off-by: Tom Rini <trini@ti.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Since commit 957731ed (ARM: remove broken "lpd7a40x" boards),
lh7a40x cpu and serial driver have become unused. Remove them.
Signed-off-by: Albert ARIBAUD <albert.u.boot@aribaud.net>
Add to pxa25x based devices support for USB ethernet gadget. This is a
port of pxa25x UDC driver from Linux kernel.
Signed-off-by: Lukasz Dalek <luk0104@gmail.com>
Removed DEV_CONFIG_CDC and DEV_CONFIG_SUBSET and replaced it with
CONFIG_USB_ETH_CDC and CONFIG_USB_ETH_SUBSET.
Signed-off-by: Lukasz Dalek <luk0104@gmail.com>
Introduced CONFIG_USB_ETH_CDC and CONFIG_USB_ETH_SUBSET as preparation
for removal DEV_CONFIG_CDC and DEV_CONFIG_SUBSET
Signed-off-by: Lukasz Dalek <luk0104@gmail.com>
PXA25X chips don't support alternate settings so driver uses non-CDC
driver.
But only code defined between DEV_CONFIG_CDC signals that network is up.
This patch is fixing this bug by signaling that network is up after USB
SET_INTERFACE request.
Signed-off-by: Lukasz Dalek <luk0104@gmail.com>
Support for f_dfu USB function.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Cc: Marek Vasut <marex@denx.de>
Composite USB download gadget support (g_dnl) for download functions.
This code works on top of composite gadget.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Cc: Marek Vasut <marex@denx.de>
These are all the files which use the API incorrectly but did not get
built using MAKEALL -a powerpc|arm. I have no compiler for them, but
the remaining issues should be far less than without this patch.
Any outstanding issues are left to the maintainers of boards that use
these drivers.
Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>
This commit adds support for storing private data to Samsung's UDC
driver. This data is afterward used by usb gadget.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Cc: Marek Vasut <marex@denx.de>
USB Composite gadget implementation for u-boot. It builds on top
of USB UDC drivers.
This commit is based on following files from Linux Kernel v2.6.36:
./include/linux/usb/composite.h
./drivers/usb/gadget/composite.c
SHA1: d187abb9a83e6c6b6e9f2ca17962bdeafb4bc903
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Cc: Marek Vasut <marex@denx.de>
This patch cleans up the Samsung's UDC driver. It replaces several
DEBUG_* macros with debug_cond().
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Cc: Marek Vasut <marex@denx.de>
Remove the repeated USB descriptor code and use usbdescriptors.h file.
ch9.h file has been copied from linux and is needed for USB gadget
related work.
Now usbdescriptors.h and ch9.h shall be used together.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Cc: Marek Vasut <marex@denx.de>
Merge our duplicate definitions with the common header.
Also fix drivers/usb/gadget/s3c_udc_otg_xfer_dma.c to
use min() instead of min_t() since we remove the latter
from compat.h.
Additionally use memalign() directly as the lin_gadget
specific kmalloc() macro is removed from lin_gadget_compat.h
by this patch.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Anatolij Gustschin <agust@denx.de>
Cc: Lukasz Majewski <l.majewski@samsung.com>
This patch adds the support for high speed in usb device framework and usbtty
driver. This feature has been kept within a macro CONFIG_USBD_HS, so the board
configuration files have to define this macro to enable high speed support.
Along with that specific peripheral drivers also need to define a function to
let the framework know that the enumeration has happened at high speed.
This function prototype is "int is_usbd_high_speed(void)"
Signed-off-by: Vipin Kumar <vipin.kumar@st.com>
Signed-off-by: Amit Virdi <amit.virdi@st.com>
The dev->req_config flag was indicating that the forwarded
request needs to perform the usb gadget delayed status.
This is however not needed anymore, so it can be removed.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Cc: Marek Vasut <marex@denx.de>
While receiving packets from FIFO sometimes the buffer provided was
nonaligned. Fix this by taking a temporary aligned buffer and then
copying the content to nonaligned buffer.
Signed-off-by: Shiraz Hashim <shiraz.hashim@st.com>
Signed-off-by: Amit Virdi <amit.virdi@st.com>
This patch adds the support for usb device high speed for designware peripheral.
Signed-off-by: Vipin Kumar <vipin.kumar@st.com>
Signed-off-by: Amit Virdi <amit.virdi@st.com>
This patch fixes a few bugs in USB device controller driver.
The fixes are as follows
1. Adding error condition checks eg. NULL return
2. Endpoint other than endpoint 0 (control endpoint) are initialized
only if usb state machine reaches STATE_ADDRESSED or above
3. Zero length packet handling corrected
4. Dead code removed
5. Bulk out endpoint returns after servicing 1 interrupt and returns
back to service if more interrupts are pending
Signed-off-by: Vipin Kumar <vipin.kumar@st.com>
Signed-off-by: Amit Virdi <amit.virdi@st.com>
Acked-by: Marek Vasut <marex@denx.de>
The earlier usb device controller driver was specific to spear platforms. This
patch implements the usb device controller driver as a generic controller which
can be reused by other platforms using this peripheral.
Signed-off-by: Vipin Kumar <vipin.kumar@st.com>
Signed-off-by: Amit Virdi <amit.virdi@st.com>
Acked-by: Marek Vasut <marex@denx.de>
Fix:
In file included from s3c_udc_otg.c:216:0:
s3c_udc_otg_xfer_dma.c: In function 'complete_tx':
s3c_udc_otg_xfer_dma.c:280:33: warning: variable 'is_short' set but not used
s3c_udc_otg_xfer_dma.c:280:6: warning: variable 'ep_tsr' set but not used
s3c_udc_otg_xfer_dma.c: In function 's3c_udc_irq':
s3c_udc_otg_xfer_dma.c:469:16: warning: variable 'flags' set but not used
s3c_udc_otg_xfer_dma.c:468:18: warning: variable 'gintmsk' set but not used
s3c_udc_otg_xfer_dma.c: In function 's3c_queue':
s3c_udc_otg_xfer_dma.c:582:14: warning: variable 'gintsts' set but not used
s3c_udc_otg_xfer_dma.c:581:16: warning: variable 'flags' set but not used
s3c_udc_otg_xfer_dma.c: In function 's3c_ep0_read':
s3c_udc_otg_xfer_dma.c:778:6: warning: variable 'ret' set but not used
s3c_udc_otg_xfer_dma.c: In function 's3c_udc_set_halt':
s3c_udc_otg_xfer_dma.c:1020:16: warning: variable 'flags' set but not used
s3c_udc_otg_xfer_dma.c: In function 's3c_ep0_setup':
s3c_udc_otg_xfer_dma.c:1258:13: warning: initialization from incompatible
pointer type [enabled by default]
s3c_udc_otg_xfer_dma.c:1239:16: warning: variable 'is_in' set but not used
s3c_udc_otg_xfer_dma.c:1239:9: warning: variable 'bytes' set but not used
s3c_udc_otg.c: In function 'usb_gadget_register_driver':
s3c_udc_otg.c:292:16: warning: variable 'flags' set but not used
s3c_udc_otg.c: In function 'usb_gadget_unregister_driver':
s3c_udc_otg.c:338:16: warning: variable 'flags' set but not used
s3c_udc_otg.c: In function 's3c_ep_enable':
s3c_udc_otg.c:582:16: warning: variable 'flags' set but not used
s3c_udc_otg.c: In function 's3c_ep_disable':
s3c_udc_otg.c:646:16: warning: variable 'flags' set but not used
s3c_udc_otg.c: In function 's3c_dequeue':
s3c_udc_otg.c:704:16: warning: variable 'flags' set but not used
Signed-off-by: Anatolij Gustschin <agust@denx.de>
Cc: Lukasz Majewski <l.majewski@samsung.com>
Cc: Remy Bohmer <linux@bohmer.net>
Commit bdbcdc89 "pxa: convert pxa27x_udc to use read and write
functions" added a number of C++ comments. Fix these.
Signed-off-by: Wolfgang Denk <wd@denx.de>
In 9792987721 Stefan describes a usecase
where the previous behavior of leaving wMaxPacketSize be unaligned
caused fatal problems. The initial fix for this problem was incomplete
however as it showed another cases of non-aligned access that previously
worked implicitly. This switches to making sure that all access of
wMaxPacketSize are done via (get|put)_unaligned.
In order to maintain a level of readability to the code in some cases
we now use a variable for the value of wMaxPacketSize and in others, a
macro.
Cc: Minkyu Kang <mk7.kang@samsung.com>
Cc: Remy Bohmer <linux@bohmer.net>
OpenRISC:
Tested-by: Stefan Kristiansson <stefan.kristiansson@saunalahti.fi>
Beagleboard xM, Pandaboard run-tested, s5p_goni build-tested.
Signed-off-by: Tom Rini <trini@ti.com>
This commit provides UDC driver support for Samsung's SoC
family of processors.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Cc: Minkyu Kang <mk7.kang@samsung.com>
Cc: Remy Bohmer <linux@bohmer.net>
Fix:
ep0.c: In function 'ep0_get_descriptor':
ep0.c:187:8: warning: variable 'cp' set but not used
[-Wunused-but-set-variable]
Signed-off-by: Anatolij Gustschin <agust@denx.de>
Cc: Remy Bohmer <linux@bohmer.net>
Acked-by: Remy Bohmer <linux@bohmer.net>
Acked-by: Marek Vasut <marek.vasut@gmail.com>
Fix:
core.c: In function 'usbd_device_event_irq':
core.c:596:21: warning: variable 'state' set but not used
[-Wunused-but-set-variable]
Signed-off-by: Anatolij Gustschin <agust@denx.de>
Cc: Remy Bohmer <linux@bohmer.net>
Acked-by: Remy Bohmer <linux@bohmer.net>
Acked-by: Marek Vasut <marek.vasut@gmail.com>
Add waiting for receiving Ethernet gadget state on the Windows host
side before dropping pullup, but keep it for debug.
Signed-off-by: Vitaly Kuzmichev <vkuzmichev@mvista.com>
Port USB gadget RNDIS protocol support from linux-2.6.26
(.27 gadget stack actually has composite drivers).
Signed-off-by: Vitaly Kuzmichev <vkuzmichev@mvista.com>
Disconnecting USB gadget with pending interrupt may cause its wrong
handling in the next time when interface will be started again
(especially actual for RNDIS). This interrupt may force the gadget
to queue unexpected response before setup stage.
Despite the fact that such interrupt handled after dropped pullup
also may add pending response, this will not bring to any issues due to
usb_ep_disable (which clears the queue) called on gadget unregistering.
Signed-off-by: Vitaly Kuzmichev <vkuzmichev@mvista.com>
Calling eth_bind at usb_eth_init time causes renaming of the network
device from 'usb_ether' to 'usb0'. Fixing this to keep the first name.
Signed-off-by: Vitaly Kuzmichev <vkuzmichev@mvista.com>
Since the ether may not be the only one usb gadget would be used
in the uboot, it is neccessary to do the register each time the
eth begin to work to make usb gadget driver less confussed when
we want to use two different usb gadget at the same time.
Usb gadget driver could simple ignore the register operation, if
it find the driver has been registered already.
Signed-off-by: Lei Wen <leiwen@marvell.com>
Before this commit, weak symbols were not overridden by non-weak symbols
found in archive libraries when linking with recent versions of
binutils. As stated in the System V ABI, "the link editor does not
extract archive members to resolve undefined weak symbols".
This commit changes all Makefiles to use partial linking (ld -r) instead
of creating library archives, which forces all symbols to participate in
linking, allowing non-weak symbols to override weak symbols as intended.
This approach is also used by Linux, from which the gmake function
cmd_link_o_target (defined in config.mk and used in all Makefiles) is
inspired.
The name of each former library archive is preserved except for
extensions which change from ".a" to ".o". This commit updates
references accordingly where needed, in particular in some linker
scripts.
This commit reveals board configurations that exclude some features but
include source files that depend these disabled features in the build,
resulting in undefined symbols. Known such cases include:
- disabling CMD_NET but not CMD_NFS;
- enabling CONFIG_OF_LIBFDT but not CONFIG_QE.
Signed-off-by: Sebastien Carlier <sebastien.carlier@gmail.com>
After gadget reinitializaton (after tftp has been done once)
packet_received may become equal to 1 due to nuking OUT_EP
while disabling it in eth_reset_config.
rx_submit called from usb_eth_init queues rx_req first time.
But the first call of usb_eth_recv from NetLoop queues rx_req
again due to packet_received = 1.
The following flow shows the path of functions calls when
this happens:
net/net.c:NetLoop
|
+-net/eth.c:eth_init
| ether.c:usb_eth_init
| |
| +-udc_driver:usb_gadget_handle_interrupts
| | udc_driver:...
| | ether.c:eth_setup
| | ether.c:eth_set_config
| | ether.c:eth_reset_config
| | udc_driver:usb_ep_disable
| | udc_driver:nuke
| | ether.c:rx_complete
| | ether.c: packet_received = 1;
| |
| +-ether.c:rx_submit
| udc_driver:usb_ep_queue --- The first time when rx_req is queued
|
+-net/eth.c:eth_rx
ether.c:usb_eth_recv
|
+-udc_driver:usb_gadget_handle_interrupts
| udc_driver:... --- no interrupts, returning
+-ether.c: if (packet_received) { ...
ether.c:rx_submit
udc_driver:usb_ep_queue --- The second time!
Signed-off-by: Vitaly Kuzmichev <vkuzmichev@mvista.com>
Remove and fix needless and destructive operations with tx/rx_req.
1) 'req' in rx_complete is always not NULL and always equals to rx_req
2) Free allocated tx_req if rx_req allocation has failed
3) Do not zero out tx/rx_req in usb_eth_init, leave this for
eth_reset_config which will be called at the next use of usb0
4) Additional check in usb_eth_recv is not required
Signed-off-by: Vitaly Kuzmichev <vkuzmichev@mvista.com>
Fixes most of checkpatch warnings and errors in USB gadget stack.
The most frequently encountered problems are:
1) "(foo*)", "foo * bar", "foo* bar"
2) C99 // comments
3) No spaces before/after/around '?', ':', '=', '==', ',', '&', '('
4) Spaces before '['
5) Spaces between function names and '('
6) Block braces in wrong places
7) Spaces before tabs
8) Macros with complex values not enclosed in parenthesis
9) Multiline comments start just after /*
Signed-off-by: Vitaly Kuzmichev <vkuzmichev@mvista.com>
The patch removes an endless loop in the usb_eth_send
if the tx_complete is not called before going
in the loop. The driver interrupt routine is called
allowing the driver to check if the TX is completed.
Signed-off-by: Stefano Babic <sbabic@denx.de>
The buffer for the status request must be word aligned
because it is accessed with 32 bit pointer in the
eth_status_complete function.
Signed-off-by: Stefano Babic <sbabic@denx.de>
The patch is based on commit bb9496c6f7e853e5d4edd5397c9d45f1968d623c
(done by Julia Lawall <julia@diku.dk>) from linux-2.6.git.
Since num is unsigned, it would seem better to use simple_strtoul that
simple_strtol.
Signed-off-by: Vitaly Kuzmichev <vkuzmichev@mvista.com>
Fix potential oops on rare error path.
The patch is based on commit e7b13ec9235b9fded90f826ceeb8c34548631351
(done by David Brownell <david-b@pacbell.net>) from linux-2.6.git.
Description of the issue taken from linux kernel bugzilla:
(https://bugzilla.kernel.org/show_bug.cgi?id=9594)
The potential error can be tracked down as follows:
(1) line 807: let the second conjunct on the "if" statment be false
meaning "dev->status_ep" is null. This means the "if" evaluates
to false.
follow thru the code until...
(2) line 808: usb_ep_disable(dev->status_ep) passes in a null argument,
however "usb_ep_disable" cannot handle that:
(from include/linux/usb/gadget.h)
191 static inline int
192 usb_ep_disable (struct usb_ep *ep)
193 {
194 return ep->ops->disable (ep);
195 }
--
Signed-off-by: Vitaly Kuzmichev <vkuzmichev@mvista.com>
Fix possible oops on stat_req->buf initialization and fix ep0 and
status_ep confusion (last one is just intended for stat_req keeping).
Signed-off-by: Vitaly Kuzmichev <vkuzmichev@mvista.com>
Signed-off-by: Stefano Babic <sbabic@denx.de>
Fix in_ep and out_ep confusion (rx_req was allocated from out_ep, not
from in_ep) and add lost dev->req freeing.
Signed-off-by: Vitaly Kuzmichev <vkuzmichev@mvista.com>
Add lost 'qmult' definition for High Speed devices and make it
configurable through CONFIG_USB_ETH_QMULT.
Signed-off-by: Vitaly Kuzmichev <vkuzmichev@mvista.com>
Prohibit simultaneous usage of both old and new gadget stacks and
allow UDC drivers to be dependent on CONFIG_USB_ETHER.
Signed-off-by: Vitaly Kuzmichev <vkuzmichev@mvista.com>
I executed 'find . -name "*.[chS]" -perm 755 -exec chmod 644 {} \;'
Signed-off-by: Thomas Weber <swirl@gmx.li>
Add some more: neither Makefile nor config.mk need execute permissions.
Signed-off-by: Wolfgang Denk <wd@denx.de>
SPEAr SoCs contain a synopsys usb device controller.
USB Device IP can work in 2 modes
- DMA mode
- Slave mode
The driver adds support only for slave mode operation of usb
device IP. This driver is used along with standard USBTTY
driver to obtain a tty interface over USB on the host
Signed-off-by: Vipin <vipin.kumar@st.com>
This Patch adds Support for PXA27X UDC.
(Rebased to drivers/usb reorganisation)
Signed-off-by: Vivek Kutal <vivek.kutal@azingo.com>
Signed-off-by: Remy Bohmer <linux@bohmer.net>
move to linux usb driver organisation
as following
drivers/usb/gadget
drivers/usb/host
drivers/usb/musb
Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Signed-off-by: Remy Bohmer <linux@bohmer.net>