This patch adds the 10ms TRSTRCY delay between a reset and the following
Set Address command that is required by the USB 2.0 specification to the
EHCI root hub driver. The generic_hub driver that's used for XHCI and
external hubs already included this delay. This is such a glaring
violation of the spec that I'm really amazed how many USB 2.0 devices
we tested before seemed perfectly fine with responding to a Set Address
within 2 microframes of the reset...
It also increases the port reset hold delay by one millisecond to avoid
an ugly race condition on Tegra SoCs: they decided to time the 50ms
themselves instead of relying on the CPU to do it (fair enough), and to
automatically transition Port Reset to 0 and Port Enable to 1 after that
(bad idea). If the CPU's read-modify-write to clear Port Reset races
exactly with the host controller setting Port Enable, we may end up
clearing the bit again and going into the companion controller handoff
path later on. The added millisecond shouldn't cause any problems for
other host controllers and is not a big deal compared to other delays in
this code path.
BUG=chrome-os-partner:26749
TEST=Run several dozen reboot loops with The USB Stick of Death (TM) (a
blue Patriot XT 13fe:5200 with bcdDevice = 1.00), make sure it always
gets detected correctly.
Original-Change-Id: Idd3329ae6d7e5e1c07a84a5475549b3459836b31
Original-Signed-off-by: Julius Werner <jwerner@chromium.org>
Original-Reviewed-on: https://chromium-review.googlesource.com/189872
Original-Reviewed-by: Gabe Black <gabeblack@chromium.org>
Original-Reviewed-by: Jim Lin <jilin@nvidia.com>
Original-Reviewed-by: Hung-Te Lin <hungte@chromium.org>
(cherry picked from commit 4deca38e9d79f6373f4418fcaf51a6945232c8b8)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: I68a29bfd2e0f30409fbfc330b2575f0f9f61a79d
Reviewed-on: http://review.coreboot.org/7221
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
This patch enables the OHCI driver to use DMA memory, which is necessary
for ARM systems where DMA devices are not cache coherent. I really only
need this to test some later OHCI changes, but it was easy enough...
copied almost verbatim from ehci.c.
Change-Id: Ia717eef28340bd6182a6782e83bfdd0693cf0db1
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/193730
Reviewed-by: Stefan Reinauer <reinauer@chromium.org>
(cherry picked from commit e46b6ebc439e86a00e13bf656d60cf6c186a3777)
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Reviewed-on: http://review.coreboot.org/7010
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
- Remove the call to clear_stall in xhci_reset_endpoint because we will
call clear_stall from the mass-storage driver.
- Remove the xhci_reset_endpoint call from xhci_bulk on STALL since we
will reset on the next transfer anyway.
- Remove the clear_halt parameter from xhci_bulk since it's now unused.
Signed-off-by: Shawn Nematbakhsh <shawnn@chromium.org>
Change-Id: I852b87621861109e596ec24b78a8f036d796ff14
Reviewed-on: https://chromium-review.googlesource.com/192866
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Julius Werner <jwerner@chromium.org>
(cherry picked from commit e67e4f0545cbdc074328c83c7edccf9e712cd7be)
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Reviewed-on: http://review.coreboot.org/7011
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Marc Jones <marc.jones@se-eng.com>
If a port is connected before and after an xhci controller reset, the
PORTSC CSC bit may not be asserted. Add an additional check in
xhci_rh_port_status_changed for the PRC bit so we can correctly handle
ports in such a state.
Signed-off-by: Shawn Nematbakhsh <shawnn@chromium.org>
Change-Id: I2d623aae647ab13711badd7211ab467afdc69548
Reviewed-on: https://chromium-review.googlesource.com/189394
Reviewed-by: Julius Werner <jwerner@chromium.org>
(cherry picked from commit ee7c3ea182b35bb6ce3c62f301c4515714f6e654)
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Reviewed-on: http://review.coreboot.org/7002
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
The generic roothub reset port function is overly broad and does some
things which may be undesirable, such as issuing multiple resets to a
port if the reset is deemed to have finished too quickly. Remove the
generic function and replace it with a controller-specific function,
currently only implemented for xhci.
Change-Id: Id46f73ea3341d4d01d2b517c6bf687402022d272
Signed-off-by: Shawn Nematbakhsh <shawnn@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/189495
Reviewed-by: Julius Werner <jwerner@chromium.org>
(cherry picked from commit 54e1da075b0106b0a1f736641fa52c39401d349d)
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Reviewed-on: http://review.coreboot.org/7001
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Forgot an asterisk and everything goes to hell. Sorry about that.
Change-Id: I6b2503ca3ea0f80d4e4e5d8b8c0e986fec5db2c9
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/173587
Reviewed-by: Stefan Reinauer <reinauer@google.com>
Commit-Queue: David James <davidjames@chromium.org>
(cherry picked from commit 2a357560a697b56cc6022a4dd3dda47b33568d83)
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Reviewed-on: http://review.coreboot.org/6854
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
The dump_td() debug function in the EHCI stack incorrectly masks the
amount of transferred bytes on output... the actual field is 15 bits
wide (30:16). Let's just use the mask constant we already have for all
the other code.
Change-Id: I28c6f0ec75cc613e38d53b670645d19bf9ffe1b9
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/174986
Reviewed-by: Stefan Reinauer <reinauer@chromium.org>
(cherry picked from commit 570077da7f16bbe2204b4a80790e4bd8fe1a2bd7)
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Reviewed-on: http://review.coreboot.org/6853
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
This patch represents a major overhaul of the USB enumeration code in
order to make it cleaner and much more robust to weird or malicious
devices. The main improvement is that it correctly parses the USB
descriptors even if there are unknown descriptors interspersed within,
which is perfectly legal and in particular present on all SuperSpeed
devices (due to the SuperSpeed Endpoint Companion Descriptor).
In addition, it gets rid of the really whacky and special cased
get_descriptor() function, which would read every descriptor twice
whether it made sense or not. The new code makes the callers allocate
descriptor memory and only read stuff twice when it's really necessary
(i.e. the device and configuration descriptors).
Finally, it also moves some more responsibilities into the
controller-specific set_address() function in order to make sure things
are initialized at the same stage for all controllers. In the new model
it initializes the device entry (which zeroes the endpoint array), sets
up endpoint 0 (including MPS), sets the device address and finally
returns the whole usbdev_t structure with that address correctly set.
Note that this should make SuperSpeed devices work, but SuperSpeed hubs
are a wholly different story and would require a custom hub driver
(since the hub descriptor and port status formats are different for USB
3.0 ports, and the whole issue about the same hub showing up as two
different devices on two different ports might present additional
challenges). The stack currently just issues a warning and refuses to
initialize this part of the hub, which means that 3.0 devices connected
through a 3.0 hub may not work correctly.
Change-Id: Ie0b82dca23b7a750658ccc1a85f9daae5fbc20e1
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/170666
Reviewed-by: Kees Cook <keescook@chromium.org>
(cherry picked from commit ecec80e062f7efe32a9a17479dcf8cb678a4a98b)
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Reviewed-on: http://review.coreboot.org/6780
Tested-by: build bot (Jenkins)
This patch removes the confusing concept of a special "xhci_speed" with
a different numeric value from the usual speed used throughout the USB
core (except for the places directly interacting with the xHC, which are
explicitly marked). It also moves the MPS0 decoding function into the
core and moves some definitions around in preparation of later changes
that will make the stack SuperSpeed-ready. It makes both set_address
implementations share a constant for the specification-defined
SetAddress() recovery delay and removes pointless additional delays from
the non-XHCI version.
Change-Id: I422379d05d4a502b12dae183504e5231add5466a
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/170664
Reviewed-by: Stefan Reinauer <reinauer@google.com>
Commit-Queue: Ronald Minnich <rminnich@chromium.org>
(cherry picked from commit f160d4439c0d7cea1d2e6b97207935d61dcbb2f2)
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Reviewed-on: http://review.coreboot.org/6776
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
This patch fixes a bug in the XHCI stack that occurs when a multi-TRB TD
times out before the last TRB is processed. The driver will correctly
issue a Stop Endpoint command in that case, but the xHC will still
preserve the transfer state and just pick up right after that on the
next doorbell ring. It will then process the leftover TRBs from the old
TD the next time a transfer is issued. (cf. XHCI 4.6.9)
We fix this by changing the existing xhci_reset_endpoint() calls in
transfer functions to not only trigger on Halted (2) and Error (4), but
also on Stopped (3). That function will not actually issue a Reset
Endpoint command in this case, but it will nuke the whole transfer ring
and issue a Set TR Dequeue Pointer command, which is sufficient (though
slightly overkill) to solve our problem.
Change-Id: I3abbe30ff9d4911a8af1f792324e018d427019e8
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/170833
Reviewed-by: Ronald Minnich <rminnich@chromium.org>
Reviewed-by: Kees Cook <keescook@chromium.org>
(cherry picked from commit f12424af0e29ac12963e8e5a7970fadcc0bb6cee)
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Reviewed-on: http://review.coreboot.org/6787
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Tested-by: build bot (Jenkins)
The existing USB_MEMORY mechanism to instantiate non-PCI host
controllers is clunky and inflexible... most importantly, it doesn't
allow multiple host controllers of the same kind. This patch replaces it
with a function that allows payloads to directly instantiate as many
host controllers of whatever type they need.
Change-Id: Ic21d2016a4ef92c67fa420bdc0f0d8a6508b69e5
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/169454
Reviewed-by: Stefan Reinauer <reinauer@google.com>
(cherry picked from commit b6e95c39dd91f654f0a345f17b3196f56adf4891)
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Reviewed-on: http://review.coreboot.org/6644
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
This patch fixes the following minor bugs in the USB stack:
1. Ensure that all dynamically allocated device structures are cleaned
on detachment, and that the device address is correctly released again.
2. Make sure MSC and HID drivers notice missing endpoints and actually
detach the device in that case (to prevent it from being used).
3. Make sure XHCI-specific set_address() cleans up all data structures
on failure.
4. Fix broken Slot ID range check that prevented XHCI devices from being
correctly cleaned up.
Change-Id: I7b2b9c8cd6c5e93cb19abcf01425bcd85d2e1f22
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/170665
Reviewed-by: Stefan Reinauer <reinauer@google.com>
Commit-Queue: Ronald Minnich <rminnich@chromium.org>
Tested-by: Ronald Minnich <rminnich@chromium.org>
(cherry picked from commit 9671472263ddd0c30400ae3b6da780a18cd21ded)
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Reviewed-on: http://review.coreboot.org/6701
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
The USB MSC device structure contains a "ready" state that can be either
"ready", "not ready" or "detached". The last one can only be assigned
when the device is completely unresponsive and gets forcefully logically
detached via usb_detach_device(). This call (at least in the current
version) also calls all destructors and frees the complete usbdev_t
structure (including the MSC specific part), which unfortunately makes
storing the "detached" state in that very structure a little pointless.
This patch reduces the "ready" value to a simple boolean and makes sure
that all detachment cases immediately return from the MSC driver,
carefully avoiding any use-after-free opportunities.
Change-Id: Iff1c0849f9ce7c95d399bb9a1a0a94469951194d
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/170667
(cherry picked from commit fd4529f37fdd1c93a8b902488ffeef7001b1a05a)
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Reviewed-on: http://review.coreboot.org/6654
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Tested-by: build bot (Jenkins)
This patch updates the libpayload XHCI stack to run on ARM CPUs (tested
with the DWC3 controller on an Exynos5420). Firstly, it adds support for
64-byte Slot/Endpoint Context sizes. Since the existing context handling
code represented the whole device context as a C struct (whose size has
to be known at compile time), it was necessary to refactor the input and
device context structures to consist of pointers to the actual contexts
instead.
Secondly, it moves all data structures that the xHC accesses through DMA
to cache-coherent memory. With a similar rationale as in the ARM patches
for EHCI, using explicit cache maintenance functions to correctly handle
the actual transfer buffers in all cases is presumably impossible.
Instead this patch also chooses to create a DMA bounce buffer in the
XHCI stack where transfer buffers which are not already cache-coherent
will be copied to/from.
Change-Id: I14e82fffb43b4d52d687b65415f2e33920e088de
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/169453
Reviewed-by: Stefan Reinauer <reinauer@google.com>
(cherry picked from commit 1fa9964063cce6cbd87ba68334806dde8aa2354c)
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Reviewed-on: http://review.coreboot.org/6643
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
This patch makes the EHCI driver work on ARM platforms which usually do
not support automatic cache snooping. It uses the new DMA memory
mechanism (which needs to be correctly set up in the Coreboot mainboard
code) to allocate all EHCI-internal communication structures in
cache-coherent memory, and cleans/invalidates the externally supplied
transfer buffers in Bulk and Control functions with explicit calls as
necessary.
Old-Change-Id: Ie8a62545d905b7a4fdd2a56b9405774be69779e5
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/167339
(cherry picked from commit 322338934add36a5372ffe7d2a45e61a4fdd4a54)
libpayload: ehci: Cache management is hard, let's go copying...
It turns out that my previous commit to make the EHCI stack cache aware
on ARM devices wasn't quite correct, and the problem is actually much
trickier than I thought. After having some fun with more weird transfer
problems that appear/disappear based on stack alignment, this is my
current worst-case threat model that any cache managing implementation
would need to handle correctly:
Some upper layer calls ehci_bulk() with a transfer buffer on its stack.
Due to stack alignment, it happens to start just at the top of a cache
line, so up to 64 - 4 bytes of ehci_bulk's stack will share that line.
ehci_bulk() calls dcache_clean() and initializes the USB transfer.
Between that point and the call to dcache_invalidate() at the end of
ehci_bulk(), any access to the stack variables in that cache line (even
a speculative prefetch) will refetch the line into the cache. Afterwards
any other access to a random memory location that just happens to get
aliased to the same cache line may evict it again, causing the processor
to write out stale data to the transfer buffer and possibly overwrite
data that has already been received over USB.
In short, any dcache_clean/dcache_invalidate-based implementation that
preserves correctness while allowing any arbitrary (non cache-aligned)
memory location as a transfer buffer is presumed to be impossible.
Instead, this patch causes all transfer data to be copied to/from a
cache-coherent bounce buffer. It will still transfer directly if the
supplied buffer is already cache-coherent, which can be used by callers
to optimize their transfers (and is true by default on x86).
Old-Change-Id: I112908410bdbc8ca028d44f2f5d388c529f8057f
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/169231
Reviewed-by: Stefan Reinauer <reinauer@chromium.org>
(cherry picked from commit 702dc50f1d56fe206442079fa443437f4336daed)
Squashed the initial commit and a follow up fix.
Change-Id: Idf7e5aa855b4f0221f82fa380a76049f273e4c88
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Reviewed-on: http://review.coreboot.org/6633
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
The readwrite_chunk was private to the usb mass storage driver, but wasn't
marked as static which was upsetting the compiler.
Change-Id: I0ef5c5f96a29f793dd43ff672a939902bad13c45
Signed-off-by: Gabe Black <gabeblack@google.com>
Reviewed-on: https://chromium-review.googlesource.com/169816
Reviewed-by: Julius Werner <jwerner@chromium.org>
Commit-Queue: Gabe Black <gabeblack@chromium.org>
Tested-by: Gabe Black <gabeblack@chromium.org>
(cherry picked from commit 8140e6145b3d072b7f12a924418570022207c065)
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Reviewed-on: http://review.coreboot.org/6648
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Currently, we wait for up to 30 seconds for a device to become ready to
respond to a TEST_UNIT_READY command. In practice, all media devices become
ready much sooner. But, certain devices do not function with libpayload's
USB driver, and always timeout. To provide a better user experience when
booting with such devices, reduce the timeout to 5 seconds.
Change-Id: Icceab99fa266cdf441847627087eaa5de9b88ecc
Signed-off-by: Shawn Nematbakhsh <shawnn@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/169209
(cherry picked from commit 9e55204e92adca0476d273565683f211d6803e7a)
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Reviewed-on: http://review.coreboot.org/6647
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
When bringing up media, we claim to wait for up to 30 seconds for a
device to respond to our TEST_UNIT_READY command. Actually, we can wait
far longer because we do not take into account execution delay.
To improve timeout accuracy, make use of gettimeofday(), which calculates
time based upon a CPU counter. This improves the user experience
slightly when certain non-working USB devices are used.
Change-Id: Id9605ecfc0a522d7a0b039fd8eac541232605082
Signed-off-by: Shawn Nematbakhsh <shawnn@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/169208
Reviewed-by: Julius Werner <jwerner@chromium.org>
(cherry picked from commit 1d3d535db83ff478c512e37f37015b43927b3efc)
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Reviewed-on: http://review.coreboot.org/6646
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Add a new function to split transfer requests into chunks of
64KB in order to be as compatible as possible with devices that
choke when sent large transfer requests.
Change-Id: Id11990bd149af14af5535de4af47bda21d1ab51e
Signed-off-by: Duncan Laurie <dlaurie@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/169170
Reviewed-by: Julius Werner <jwerner@chromium.org>
(cherry picked from commit 4c413b007aa23da830877127dd556c4c38b43042)
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Reviewed-on: http://review.coreboot.org/6636
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
The current USB hub code always clears the port status change after
checking it, regardless of whether it was set in the first place. Since
this check runs on every poll, it might create a race condition where
the port status changes right between the GET_PORT_STATUS and the
CLEAR_FEATURE(C_PORT_CONNECT), thus clearing the statrus change flag
before it was ever read. Let's add one extra if() to avoid that possible
headache.
Change-Id: Idd46c2199dc6c240bd9ef068fbe70cccc88bac42
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/168098
(cherry picked from commit f7f6f008f701ab3e4a4f785032d8024d676e11cb)
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Reviewed-on: http://review.coreboot.org/6617
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
The EHCI host controllers in Samsung Exynos SoC seem to be a little more
picky than Intel ones. When they reach the dummy_qh in the periodic
frame list, they try to access the next qTD pointer even though it's
NULL, and run into a HostSystemError. This patch explicitly sets the
Terminate bit on those pointers to mark them invalid.
Change-Id: I50fa79bbf1c5fab306d7885c01efd66b13e279b8
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/66884
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
(cherry picked from commit c575a5c958ce88732d28044352c89418bcd5ea86)
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Reviewed-on: http://review.coreboot.org/6608
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
The USB bulk and control transfer functions in libpayload currently
always return 0 for success and 1 for all errors. This is sufficient for
current use cases (essentially just mass storage), but other classes
(like certain Ethernet adapters) need to be able to tell if a transfer
reached the intended amount of bytes, or if it fell short.
This patch slightly changes that USB API to return -1 on errors, and the
amount of transferred bytes on successes. All drivers in the current
libpayload mainline are modified to conform to the new error detection
model. Any third party users of this API will need to adapt their
if (...<controller>->bulk/control(...)) checks to
if (...<controller>->bulk/control(...) < 0) as well.
The host controller drivers for OHCI and EHCI correctly implement the
new behavior. UHCI and the XHCI stub just comply with the new API by
returning 0 or -1, but do not actually count the returned bytes.
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/48308
Reviewed-by: Gabe Black <gabeblack@chromium.org>
Reviewed-by: Stefan Reinauer <reinauer@google.com>
Tested-by: Gabe Black <gabeblack@chromium.org>
Commit-Queue: Gabe Black <gabeblack@chromium.org>
Updated the patch to support XHCI as well.
Change-Id: Ic2ea2810c5edb992cbe185bc9711d2f8f557cae6
(cherry picked from commit e39e2d84762a3804653d950a228ed2269c651458)
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Reviewed-on: http://review.coreboot.org/6390
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
The current XHCI code only sets IOC on the last TRB of a TD, and
doesn't set ISP anywhere. On my Synopsys DesignWare3 controller, this
won't generate an event at all when we have a short transfer that is not
on the last TRB of a TD, resulting in event ring desync and everyone
having a bad time. However, just setting ISP on other TRBs doesn't
really make for a nice solution: we then need to do ugly special casing
to fish out the spurious second transfer event you get for short
packets, and we still need a way to figure out how many bytes were
transferred. Since the Short Packet transfer event only reports
untransferred bytes for the current TRB, we would have to manually walk
the rest of the unprocessed TRB chain and add up the bytes. Check out
U-Boot and the Linux kernel to see how complicated this looks in
practice.
Now what if we had a way to just tell the HC "I want an event at exactly
*this* point in the TD, I want it to have the right completion code for
the whole TD, and to contain the exact number of bytes written"? Enter
the Event Data TRB: this little gizmo really does pretty much exactly
what any sane XHCI driver would want, and I have no idea why it isn't
used more often. It solves both the short packet event generation and
counting the transferred bytes without requiring any special magic in
software.
Change-Id: Idab412d61edf30655ec69c80066bfffd80290403
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/170980
Reviewed-by: Stefan Reinauer <reinauer@google.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
(cherry picked from commit e512c8bcaa5b8e05cae3b9d04cd4947298de999d)
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Reviewed-on: http://review.coreboot.org/6516
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
When libpayload header files are included in the payload itself, it's possible
that the payloads config settings will conflict with the ones in libpayload.
It's also possible for the libpayload config settings to conflict with the
payloads. To avoid that, the libpayload config settings have _LP_ (for
libpayload) added to them. The symbols themselves as defined in the Config.in files
are still the same, but the prefix added to them is now CONFIG_LP_ instead of just
CONFIG_.
Change-Id: Ib8a46d202e7880afdeac7924d69a949bfbcc5f97
Signed-off-by: Gabe Black <gabeblack@google.com>
Reviewed-on: https://gerrit.chromium.org/gerrit/65303
Reviewed-by: Stefan Reinauer <reinauer@google.com>
Tested-by: Gabe Black <gabeblack@chromium.org>
Commit-Queue: Gabe Black <gabeblack@chromium.org>
(cherry picked from commit 23e866da20862cace0ed2a67d6fb74056bc9ea9a)
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Reviewed-on: http://review.coreboot.org/6427
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Marc Jones <marc.jones@se-eng.com>
This change makes it possible for vboot to avoid an
exploit that could cause involuntary switch to dev mode.
It gives depthcharge/vboot some information on the
type of input device that generated a key.
BUG=chrome-os-partner:21729
TEST=manually tested for panther
BRANCH=none
CQ-DEPEND=CL:182420,CL:182241,CL:182946
Change-Id: I87bdac34bfc50f3adb0b35a2c57a8f95f4fbc35b
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Signed-off-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-on: https://chromium-review.googlesource.com/182357
Reviewed-by: Luigi Semenzato <semenzato@chromium.org>
Tested-by: Luigi Semenzato <semenzato@chromium.org>
Commit-Queue: Luigi Semenzato <semenzato@chromium.org>
Reviewed-on: http://review.coreboot.org/6003
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
Setting of `controller->reg_base` is of no use here, as it is never read
(in another function) later. Looks like this pattern originated from uhci.c
where it makes sense.
By removing the indirection through `reg_base` we also fix a possible
truncation to u32.
Change-Id: I5c99c5bf1f5b1d6c04bd84d87fd3e275fd7d0411
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: http://review.coreboot.org/6251
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Fix a possible null-pointer dereference (hopefully) before anyone runs
into this. Also don't switch ports to xHCI if initialization failed.
Change-Id: I5dbaeb435a98ead0b50d27fde13c9f1433ea3e81
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: http://review.coreboot.org/6245
Tested-by: build bot (Jenkins)
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
As the controller structure is never fully cleared, this one wasn't
initialized for non-pci controllers (but checked for non-null later).
Change-Id: I852671c5f55650bdb6cd97f4ec74b1f95ee894c7
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: http://review.coreboot.org/6246
Tested-by: build bot (Jenkins)
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Using void* for physical addresses leads to much casting and confuses
developers when to convert from physical to virtual addresses or
the other way around. When using plain integers for physical addresses
and pointers for virtual addresses things become much cleaner and we
won't ever end up dereferencing a physical address.
Change-Id: I24cd53b81c7863b6d14f0cbb4ce8937728b37c1c
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: http://review.coreboot.org/6244
Tested-by: build bot (Jenkins)
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Remove a redundant phys_to_virt() that sneaked in the initialization of
PCI xHCI controllers. The use of casts from void* to u32 (and vice versa)
prompts for things going wrong here. That will be addressed in a later
commit.
Change-Id: Ibc71ed6ee7016529c0e3a51559aaec07aaaba315
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: http://review.coreboot.org/6243
Tested-by: build bot (Jenkins)
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
being a good citizen on the box, libpayload tries to return to EHCI
mode on shutdown, so a non-XHCI capable USB driver after it (eg. in
the OS) finds something to work with.
Change-Id: Id227d646e08a258b841c644263112f0815dd486c
Signed-off-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-on: http://review.coreboot.org/4547
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
The EHCI driver defines a maximum transfer timeout of two seconds. The
comments state that during tests the maximum amount of required transfer
time was for the SCSI TEST_UNIT_READY command on certain devices. We
have now observed a USB device (Patriot Memory 13fe:3100) that can NAK
this command for slightly more than two seconds. It will also completely
fail if the timeout hits, since it gets confused by the subsequent CSW
retry/recovery mechanism and starts producing babble errors. This patch
increases the timeout to three seconds to circumvent this problem.
To test, boot a Falco from a red-black RageXT USB stick.
Change-Id: I3c4fef468fb16eacc5a487d76d025a78fb450e27
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/63095
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Duncan Laurie <dlaurie@chromium.org>
Reviewed-by: Sameer Nanda <snanda@chromium.org>
Reviewed-on: http://review.coreboot.org/4379
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
Mass storage devices such as card readers show up as
as USB devices. However the media not be inserted. In those
situations the previous code would just fake a disk and
call usbcreate_disk. This is inappropriate because it forms
a 1:1 mapping of USB device to disk leading to the inability
to remove the disk and/or handle "hot plug" card insertion
and removals.
To alleviate this issue introduce the notion of ready to the
usbmsc structure. It tracks detached, not ready, and ready
states. The polling routine is then used to track not ready
to ready transitions thereby creating and removing disks
appropriately. This handles the case of inserting and removing
a card that shows up as a new disk.
Booted recovery mode. Able to observe inerstion and removal
of sdcard. Also able to insert valid USB flash drive to boot
as well.
Change-Id: I3eefbe537ec1b9c975744b8984b06c17ae236f40
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/57948
Reviewed-by: Duncan Laurie <dlaurie@chromium.org>
Reviewed-on: http://review.coreboot.org/4226
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
There is currently a hard-coded 30 sec delay in the mass storage
driver while waiting for each device to become ready. However, mass
storage card readers that are empty return an error code on the
TEST UNIT READY command. A REQUEST SENSE command then needs to be
issued and interrogate the data to determine if no media is present.
If no media determination is found to be true the USB device is no
longer considered a candidate to be a disk.
This code does lead to the fact that the media card reader needs to be
populated at enumeration time. I suspect this is not an issue as it
appears the storage stack in libpayload can't handle removable media
coming online later.
Booted recovery and dev modes. Noted that removable mass storage
devices with no media were ignored without any boot delay.
Change-Id: Ida7a45614d97c6e6fbfc9bb099765aad4df550fd
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/57828
Reviewed-by: Duncan Laurie <dlaurie@chromium.org>
Reviewed-on: http://review.coreboot.org/4225
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Restructure USB stack to not depend on PCI, and
make PCI stub available on x86, but provide fixed
BARs for ARM (Exynos 5)
Change-Id: Iee7c8b134c22b661a9a515e24943470c9dbadd1f
Signed-off-by: Stefan Reinauer <reinauer@google.com>
Reviewed-on: https://gerrit.chromium.org/gerrit/49970
Reviewed-on: http://review.coreboot.org/4175
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Two structures in the USB EHCI stack were pointing
to hardware but not marked attribute((packed)) hence
leaving it to GCC to correctly align the data structures.
Next, the number of reserved bytes in hc_op_t was wrong
(but implicitly aligned to the correct values on x86)
It seems this worked fine on x86, but on ARM it was doing
the wrong thing.
Signed-off-by: Stefan Reinauer <reinauer@google.com>
Change-Id: I94bed4850ded7d3f7bbc7ff3079c103c6054c22d
Reviewed-on: https://gerrit.chromium.org/gerrit/55555
Commit-Queue: Stefan Reinauer <reinauer@google.com>
Reviewed-by: Stefan Reinauer <reinauer@google.com>
Tested-by: Stefan Reinauer <reinauer@google.com>
Reviewed-on: http://review.coreboot.org/4174
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
On Intel's Panther Point the xHCI ports are shared with an EHCI
controller. Our xHCI driver switches them to xHCI, naturally. But
we forgot to switch them back on shutdown, which left them
unusable by a non-xHCI aware operating system.
Change-Id: I70ef08655a603b42ee939935d50cf77ea97878a3
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: http://review.coreboot.org/3791
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
Well, it turned out to be more as some gaps ;)
but we finally have xHCI running. It's well tested against a QM77 Ivy
Bridge board.
We have no SuperSpeed support (yet). On Ivy Bridge, SuperSpeed is not
advertised and USB 3 devices will just work at HighSpeed.
There are still some bit fields in xhci_private.h, so this might need
little more work to run on ARM.
Change-Id: I7a2cb3f226d24573659142565db38b13acdc218c
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Signed-off-by: Patrick Georgi <patrick.georgi@secunet.com>
Reviewed-on: http://review.coreboot.org/3452
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
This is mostly a rewrite, don't even try to read a diff.
Tested with an internal rate matching hub on a QM77 board and three hubs
integrated into DELL monitors.
Change-Id: Ib12fa2aa90af4e0f37143d2ed92c4a1705b6d774
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: http://review.coreboot.org/3451
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
The current drivers for external usb hubs and root hubs all follow
the same pattern. Before adding another one with 90% of the same code,
extract the common parts and rewrite them with a simple interface.
This also adds debouncing of new attachments. Current drivers just
waited 100ms before they reset the device. However, we should check
if the device becomes disconnected and reconnected during this period.
Porting of the current hub drivers will take place in separate
commits (when I have time to test the older HCIs).
Change-Id: I0c0ce0ac1b1cc51fb4cd009b3f9fcd1b9d2ba8fe
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: http://review.coreboot.org/3450
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Read bInterval from endpoint descriptors and store it in our endpoint_t
struct. The interval is encoded dependently on the device' speed and the
endpoint's type. Therefore, it will be normalized to the binary logarithm
of the number of microframes, i.e.
t = 125us * 2^interval
The interval attribute will be used in the xHCI driver.
Change-Id: I65a8eda6145faf34666800789f0292e640a8141b
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: http://review.coreboot.org/3449
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
xHCI requires special treatment of set_address since it determines
the device number itself (instead of the driver, as with the other
controllers). The controller also wants to validate a chosen device
configuration and we need to setup additional structures for the
device and the endpoints.
Therefore, we add three functions to the hci_t structure, namely:
set_address()
finish_device_config()
destroy_device()
Current implementation for the Set Address request moved into
generic_set_address() which is set_address() for the UHCI, OCHI and
EHCI drivers. The latter two are only provided as hooks for the xHCI
driver.
The Set Configuration request is moved after endpoint enumeration.
For all other controller drivers nothing changes, as there is no other
device communication between the lines where the set_configuration()
call moved.
Change-Id: I6127627b9367ef573aa1a1525782bc1304ea350d
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Signed-off-by: Patrick Georgi <patrick.georgi@secunet.com>
Reviewed-on: http://review.coreboot.org/3447
Tested-by: build bot (Jenkins)
Reviewed-by: Anton Kochkov <anton.kochkov@gmail.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
During device initialization, skip any non-endpoint descriptor before
reading the endpoint descriptors. By now, only HID descriptors were
skipped.
Change-Id: I190f3ae44b864aa71d5f32c3738097cf8f33a61b
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: http://review.coreboot.org/3446
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Anton Kochkov <anton.kochkov@gmail.com>
This change modifies the code in libpayload that scans the PCI hierarchy for
USB controllers. Previously, if a devices primary function (function 0) was a
bridge, then none of the other functions, if any, would be looked at. If one
of the other functions was a bridge, that wouldn't be handled either. The new
version looks at each function that's present no matter what, and if it
discovers that it's a bridge it scans the other side.
Change-Id: I37f269a4fe505fd32d9594e2daf17ddd78609c15
Signed-off-by: Gabe Black <gabeblack@google.com>
Reviewed-on: http://review.coreboot.org/2517
Reviewed-by: Marc Jones <marc.jones@se-eng.com>
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
The function dump_qh() was added a while back but never used.
Hide it behind USB_DEBUG so it doesn't cause warnings when not
debugging the USB stack.
Change-Id: Idb3c7bb214895ef82676d181836a578bf161e8e0
Signed-off-by: Stefan Reinauer <reinauer@google.com>
Reviewed-on: http://review.coreboot.org/2909
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martin.roth@se-eng.com>
The controller's shutdown function free()s the controller structure so
we shouldn't access it any more after calling shutdown.
As all controllers detach themself, i.e. unchain themself from usb_hcs,
just keep iterating over usb_hcs until it's NULL.
Change-Id: Ie85caba0f685494c3fe04c550a5a14bc4158a94e
Signed-off-by: Mathias Krause <minipli@googlemail.com>
Reviewed-on: http://review.coreboot.org/2900
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Tested-by: build bot (Jenkins)
Reviewed-by: Anton Kochkov <anton.kochkov@gmail.com>
It shouldn't be used any more as we're about to free() the memory behind
the controller -- therefore detach it.
Change-Id: I875322a9940570c51d412a7f3bfb6af4ea3b3764
Signed-off-by: Mathias Krause <minipli@googlemail.com>
Reviewed-on: http://review.coreboot.org/2899
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Nico Huber <nico.huber@secunet.com>
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
And include the new, split out version in drivers/keyboard.c and
drivers/usb/usbhid.c. Those files were including curses.h just for those
definitions, but the include path was only fixed up to to point to the
libpayload versions of those files if one of the variants of curses was
compiled in. If neither was, gcc would fall back to the system version of that
header which is wrong.
Change-Id: I8c2ee0baf5f0702bd8c713c8dd4613a4bb269ce5
Signed-off-by: Gabe Black <gabeblack@google.com>
Reviewed-on: http://review.coreboot.org/2762
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
This function is static and not used in that file. To avoid the compiler
complaining about that fact, put the two functions and the call to dump_ed
(currently #if 0) behind #ifdef USB_DEBUG
Change-Id: Ic373313b5fff81f09800f286b32238350ab699c6
Signed-off-by: Gabe Black <gabeblack@google.com>
Signed-off-by: Stefan Reinauer <reinauer@google.com>
Reviewed-on: http://review.coreboot.org/2716
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
It might be useful to provide a USB driver in the payload itself instead of in
libpayload. For example there are multiple payloads being built and linked
against the same libpayload, and they might not need or even want to have the
same set of drivers installed.
This change adds two new functions, usb_generic_create and usb_generic_remove,
which behave like the usbdisk_create and usbdisk_remove functions which are
defined for USB mass storage devices. If a USB device isn't recognized and
claimed by one of the built in USB class drivers (currently hub, hid, and msc)
and the create function is defined, then it will be called to give the payload
a chance to use the device. Once it's removed, if usb_generic_remove is
defined it will be called, effectively giving the payload notice.
Built and booted depthcharge on Link. Built depthcharge for Daisy. Built
a netbooting payload, called usb_poll() with those functions implemented, and
verified that they were called and that the devices they were told about were
reasonable and the same as what was reported by lsusb in the booted system.
Change-Id: Ief7c0a513b60849fbf2986ef4ae5c9e7825fef16
Signed-off-by: Gabe Black <gabeblack@google.com>
Reviewed-on: http://review.coreboot.org/2666
Tested-by: build bot (Jenkins)
Reviewed-by: Kimarie Hoot <kimarie.hoot@se-eng.com>
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Julius Werner