Modify the XHCI drivers to accommodate for handling multiple event rings in
case there are multiple interrupters. Add the required APIs so clients are
able to allocate/request for an interrupter ring, and pass this information
back to the client driver. This allows for users to handle the resource
accordingly, such as passing the event ring base address to an audio DSP.
There is no actual support for multiple MSI/MSI-X vectors.
[export xhci_initialize_ring_info() -wcheng]
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Signed-off-by: Wesley Cheng <quic_wcheng@quicinc.com>
Link: https://lore.kernel.org/r/20240102214549.22498-2-quic_wcheng@quicinc.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
- The HCD address_device callback now accepts a user-defined timeout value
in milliseconds, providing better control over command execution times.
- The default timeout value for the address_device command has been set
to 5000 ms, aligning with the USB 3.2 specification. However, this
timeout can be adjusted as needed.
- The xhci_setup_device function has been updated to accept the timeout
value, allowing it to specify the maximum wait time for the command
operation to complete.
- The hub driver has also been updated to accommodate the newly added
timeout parameter during the SET_ADDRESS request.
Signed-off-by: Hardik Gajjar <hgajjar@de.adit-jv.com>
Reviewed-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20231027152029.104363-1-hgajjar@de.adit-jv.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In some situations where xhci removal happens parallel to xhci_handshake,
we encounter a scenario where the xhci_handshake can't succeed, and it
polls until timeout.
If xhci_handshake runs until timeout it can on some platforms result in
a long wait which might lead to a watchdog timeout.
Add a helper that checks xhci status during the handshake, and exits if
set state is entered. Use this helper in places where xhci_handshake is
called unlocked and has a long timeout. For example xhci command timeout
and xhci reset.
[commit message and code comment rewording -Mathias]
Signed-off-by: Udipto Goswami <quic_ugoswami@quicinc.com>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20231019102924.2797346-18-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Increase the event ring dequeue pointer for port change events in the
same way as other event types. No need to handle it separately.
This only touches the driver side tracking of event ring dequeue.
Note: this does move forward the event ring dequeue increase for port
change events a bit.
Previously the dequeue was increased before temporarily dropping
the xhci lock while kicking roothub polling.
Now dequeue is increased after re-aquiring the lock.
This should not matter as event ring dequeue is not touched at all by
hub thread. It's only touched in xhci interrupt handler.
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20231019102924.2797346-14-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
No matter what type of event we receive we want to increase the event
ring dequeue pointer one step for every event that is handled.
For unknown reasons the event ring dequeue increase is done inside the
transfer event handler and port event handler.
As the transfer event handler got more complex and can now loop through
several transfer TRBs on a transfer ring, there were additinal checks
added to avoid increasing event ring dequeue more than one step.
No need for elaborate checks to avoid increasing event ring dequeue
in case the transfer event handler goes through a loop.
Just increasing the event ring dequeue outside the transfer event
handler.
End goal is to increase event ring dequeue in just one place.
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20231019102924.2797346-13-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Mathias notes that the ERST_PTR_MASK macro is named as if it's masking
the Event Ring Dequeue Pointer in the ERDP register, but in actuality
it's masking the inverse.
Invert the macro's value for clarity.
Migrate it to the modern GENMASK_ULL() syntax to avoid u64 casts.
Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20231019102924.2797346-10-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When using more than one Event Ring segment (ERSTSZ > 1), software shall
set the DESI bits in the ERDP register to the number of the segment to
which the upper ERDP bits are pointing. The xHC may use the DESI bits
as a shortcut to determine whether it needs to check for an Event Ring
Full condition: If it's enqueueing events in a different segment, it
need not compare its internal Enqueue Pointer with the Dequeue Pointer
in the upper bits of the ERDP register (sec 5.5.2.3.3).
Not setting the DESI bits correctly can result in the xHC enqueueing
events past the Dequeue Pointer. On Renesas uPD720201 host controllers,
incorrect DESI bits cause an interrupt storm. For comparison, VIA VL805
host controllers do not exhibit such problems. Perhaps they do not take
advantage of the optimization afforded by the DESI bits.
To fix the issue, assign the segment number to each struct xhci_segment
in xhci_segment_alloc(). When advancing the Dequeue Pointer in
xhci_update_erst_dequeue(), write the segment number to the DESI bits.
On driver probe, set the DESI bits to zero in xhci_set_hc_event_deq() as
processing starts in segment 0. Likewise on driver teardown, clear the
DESI bits to zero in xhci_free_interrupter() when clearing the upper
bits of the ERDP register. Previously those functions (incorrectly)
treated the DESI bits as if they're declared RsvdP.
Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20231019102924.2797346-5-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The next_trb() helper relies on a link TRB at the end of a ring segment
to know a segment ends. This works well with transfer rings that use
link trbs, but not with event rings.
Event rings segments are always filled by host to segment size
before moving to next segment. It does not use link TRBs
Check for both link trb and full segment in next_trb() helper to
support event rings.
Useful if several interrupters with several event rings are supported.
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20231019102924.2797346-4-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
We want to trace other port structure members than just port number
so pass entire port structure as parameter instead of just port number.
Dig the port number from the port structure.
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20231019102924.2797346-2-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The Event Handler Busy bit shall be cleared by software when the Event
Ring is empty. The xHC is thereby informed that it may raise another
interrupt once it has enqueued new events (sec 4.17.2).
However since commit dc0ffbea57 ("usb: host: xhci: update event ring
dequeue pointer on purpose"), the EHB bit is already cleared after half
a segment has been processed.
As a result, spurious interrupts may occur:
- xhci_irq() processes half a segment, clears EHB, continues processing
remaining events.
- xHC enqueues new events. Because EHB has been cleared, xHC sets
Interrupt Pending bit. Interrupt moderation countdown begins.
- Meanwhile xhci_irq() continues processing events. Interrupt
moderation countdown reaches zero, so an MSI interrupt is signaled.
- xhci_irq() empties the Event Ring, clears EHB again and is done.
- Because an MSI interrupt has been signaled, xhci_irq() is run again.
It discovers there's nothing to do and returns IRQ_NONE.
Avoid by clearing the EHB bit only at the end of xhci_irq().
Fixes: dc0ffbea57 ("usb: host: xhci: update event ring dequeue pointer on purpose")
Signed-off-by: Lukas Wunner <lukas@wunner.de>
Cc: stable@vger.kernel.org # v5.5+
Cc: Peter Chen <peter.chen@kernel.org>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20230915143108.1532163-4-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
As mentioned in:
commit 474ed23a62 ("xhci: align the last trb before link if it is
easily splittable.")
A bounce buffer is utilized for ensuring that transfers that span across
ring segments are aligned to the EP's max packet size. However, the device
that is used to map the DMA buffer to is currently using the XHCI HCD,
which does not carry any DMA operations in certain configrations.
Migration to using the sysdev entry was introduced for DWC3 based
implementations where the IOMMU operations are present.
Replace the reference to the controller device to sysdev instead. This
allows the bounce buffer to be properly mapped to any implementations that
have an IOMMU involved.
cc: stable@vger.kernel.org
Fixes: 4c39d4b949 ("usb: xhci: use bus->sysdev for DMA configuration")
Signed-off-by: Wesley Cheng <quic_wcheng@quicinc.com>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20230915143108.1532163-2-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Trying to keep track of free trbs in a ring by adding and subtracting
deltas each time a enqueue or dequeue is increased or moved has proven to
be buggy and complicated, especially over long periods of time.
Recently a bug in counting free trbs was fixed, now taking into account
cancelled URBs that were turned into no-ops, preventing free_trbs to
slowly wander off causing unnecessary ring expansion. See
commit fe82f16aaf ("xhci: Fix incorrect tracking of free space on
transfer rings")
Turns out its a lot easier to just calculate the numer of free TRB
based on ring size and the current enqueue and dequeue pointer values.
This is currently only needed for the command ring as multi segment
transfer rings already ensures there is enough room the ring during
the ring expansion check.
We could get rid of the ring->num_trbs_free entry completely, but as
the xhci DbC code also uses it we don't clean that up in this patch.
Reported-by: Miller Hunter <MillerH@hearthnhome.com>
Closes: https://bugzilla.kernel.org/show_bug.cgi?id=217242
Tested-by: Miller Hunter <MillerH@hearthnhome.com>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Message-ID: <20230602144009.1225632-8-mathias.nyman@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The amount of new TRBs needed is calculated incorrectly when expanding a
transfer ring.
The room_on_ring() helper will correctly report that the ring needs
expansion if the enqueue pointer is about to reach the dequeue segment.
If enqueue reaches the dequeue segment then there is no easy way
to expand the ring by adding new segments between enqueue and dequeue.
This leads to ring expansion even if num_trbs_free is larger than
num_trbs we are queueing.
As a result we try to store a negative number in a unsigned int, leading
to a huge percieved trb need, and doubling of ring size.
Rework and rename the room_on_ring() to a helper that checks if ring
needs expansion, and return number of new segments needed. Don't rely on
the tracked ring->num_trbs_free value as turns out it has been unreliable.
Use ring enqueue and dequeue positions to determine expansion need.
The unsigned int issue was first reported first Chao zeng, and a bit
later seen in a real world bug.
Reported-by: chao zeng <chao.zengup@gmail.com>
Closes: https://bugzilla.kernel.org/show_bug.cgi?id=217242
Tested-by: Miller Hunter <MillerH@hearthnhome.com>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Message-ID: <20230602144009.1225632-7-mathias.nyman@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This incorrect tracking caused unnecessary ring expansion in some
usecases which over days of use consume a lot of memory.
xhci driver tries to keep track of free transfer blocks (TRBs) on the
ring buffer, but failed to add back some cancelled transfers that were
turned into no-op operations instead of just moving past them.
This can happen if there are several queued pending transfers which
then are cancelled in reverse order.
Solve this by counting the numer of steps we move the dequeue pointer
once we complete a transfer, and add it to the number of free trbs
instead of just adding the trb number of the current transfer.
This way we ensure we count the no-op trbs on the way as well.
Fixes: 55f6153d8c ("xhci: remove extra loop in interrupt context")
Cc: stable@vger.kernel.org
Reported-by: Miller Hunter <MillerH@hearthnhome.com>
Closes: https://bugzilla.kernel.org/show_bug.cgi?id=217242
Tested-by: Miller Hunter <MillerH@hearthnhome.com>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20230515134059.161110-3-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Move functions to setup msi from xhci.c to xhci-pci.c to decouple
PCI specific code from generic xhci code.
No functional changes, functions are an exact copy
[commit message rewording -Mathias]
Signed-off-by: Josue David Hernandez Gutierrez <josue.d.hernandez.gutierrez@intel.com>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20230317154715.535523-11-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
resume_done is just a timestamp, avoid confusing it with completions
related to port state transitions that are named *_done
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20230202150505.618915-10-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Now that we have a port structure for each port it makes sense to
move per port variables, timestamps and completions there.
Get rid of storing bitfileds and arrays of port specific items per bus.
Move
unsigned long resume_done;
insigned long rexit_ports
struct completion rexit_done;
struct completion u3exit_done;
Rename rexit_ports to rexit_active, and remove a redundant hcd
speed check while checking if rexit_active is set.
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20230202150505.618915-8-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
xHC supports several interrupters, each with its own mmio register set,
event ring and MSI/MSI-X vector. Transfers can be assigned different
interrupters when queued. See xhci 4.17 for details.
Current driver only supports one interrupter.
Create a xhci_interrupter structure containing an event ring, pointer to
mmio registers for this interrupter, variables to store registers over s3
suspend, erst, etc. Add functions to create and free an interrupter, and
pass an interrupter pointer to functions that deal with events.
Secondary interrupters are also useful without having an interrupt vector.
One use case is the xHCI audio sideband offloading where a DSP can take
care of specific audio endpoints.
When all transfer events of an offloaded endpoint can be mapped to a
separate interrupter event ring the DSP can poll this ring, and we can mask
these events preventing waking up the CPU.
Only minor functional changes such as clearing some of the interrupter
registers when freeing the interrupter.
Still create only one primary interrupter.
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20230202150505.618915-4-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
We need the USB fixes in here and this resolves merge conflicts as
reported in linux-next in the following files:
drivers/usb/host/xhci.c
drivers/usb/host/xhci.h
drivers/usb/typec/ucsi/ucsi.c
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When the host controller is not responding, all URBs queued to all
endpoints need to be killed. This can cause a kernel panic if we
dereference an invalid endpoint.
Fix this by using xhci_get_virt_ep() helper to find the endpoint and
checking if the endpoint is valid before dereferencing it.
[233311.853271] xhci-hcd xhci-hcd.1.auto: xHCI host controller not responding, assume dead
[233311.853393] Unable to handle kernel NULL pointer dereference at virtual address 00000000000000e8
[233311.853964] pc : xhci_hc_died+0x10c/0x270
[233311.853971] lr : xhci_hc_died+0x1ac/0x270
[233311.854077] Call trace:
[233311.854085] xhci_hc_died+0x10c/0x270
[233311.854093] xhci_stop_endpoint_command_watchdog+0x100/0x1a4
[233311.854105] call_timer_fn+0x50/0x2d4
[233311.854112] expire_timers+0xac/0x2e4
[233311.854118] run_timer_softirq+0x300/0xabc
[233311.854127] __do_softirq+0x148/0x528
[233311.854135] irq_exit+0x194/0x1a8
[233311.854143] __handle_domain_irq+0x164/0x1d0
[233311.854149] gic_handle_irq.22273+0x10c/0x188
[233311.854156] el1_irq+0xfc/0x1a8
[233311.854175] lpm_cpuidle_enter+0x25c/0x418 [msm_pm]
[233311.854185] cpuidle_enter_state+0x1f0/0x764
[233311.854194] do_idle+0x594/0x6ac
[233311.854201] cpu_startup_entry+0x7c/0x80
[233311.854209] secondary_start_kernel+0x170/0x198
Fixes: 50e8725e7c ("xhci: Refactor command watchdog and fix split string.")
Cc: stable@vger.kernel.org
Signed-off-by: Jimmy Hu <hhhuuu@google.com>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Message-ID: <0fe978ed-8269-9774-1c40-f8a98c17e838@linux.intel.com>
Link: https://lore.kernel.org/r/20230116142216.1141605-3-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The list API provides the list_count_nodes() to help with counting
existing nodes in the list. Utilise it.
Acked-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://lore.kernel.org/r/20221130134838.23805-4-andriy.shevchenko@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Make sure to also limit the amount of soft reset retries for transaction
errors on streams in cases where the transaction error event doesn't point
to any specific TRB.
In these cases we don't know the TRB or stream ring, but we do know which
endpoint had the error.
To keep error counting simple and functional, move the current err_count
from ring structure to endpoint structure.
Cc: stable@vger.kernel.org
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20221130091944.2171610-6-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When HCE(Host Controller Error) is set, it means that the xhci hardware
controller has an error at this time, but the current xhci driver
software does not log this event.
By adding an HCE event detection in the xhci interrupt processing
interface, a warning log is output to the system, which is convenient
for system device status tracking.
Signed-off-by: Longfang Liu <liulongfang@huawei.com>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20221130091944.2171610-2-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The list API now provides the list_count() to help with counting
existing nodes in the list. Utilise it.
Acked-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://lore.kernel.org/r/20221123144901.40493-4-andriy.shevchenko@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The XHCI_RESET_EP_QUIRK was added in 2009 to support prototype xHC
hardware from Fresco Logic that needed an additional configure endpoint
command after a reset endpoint.
That hardware should not have made it to the market.
Now, 13 years later its about time we get rid of it.
quirk was added in commit ac9d8fe7c6 ("USB: xhci: Add quirk for Fresco
Logic xHCI hardware.")
Print a debug message about the removed quirk if against all odds we run
into this controller.
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20220511220450.85367-9-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The 'stop endpoint' command timer was started when a 'stop endpoint'
command was added to the command queue.
This can trigger unwanted timeouts if there are several pending commands
in the queue that xHC needs to handle first.
The generic command timer, which was added later than the 'stop endpoint'
timeout timer, times each command currently being handled by xHC hardware.
A timed out stop endpoint command was treated as a more severe issue than
other failed commands, so the separate stop endpoint timer was left
unchanged.
Use the generic command timer for stop endpoint commands. Identify if
the timed out command was a stop endpoint command in the generic handler,
and treat it with the same severity as earlier.
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20220511220450.85367-7-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In some situations software handles TRB events slower than adding TRBs.
If the number of TRB events to be processed in a given interrupt is exactly
the same as the event ring size 256, then the local variable
"event_ring_deq" that holds the initial dequeue position is equal to
software_dequeue after handling all 256 interrupts.
It will cause driver to not update ERDP to hardware,
Software dequeue pointer is out of sync with ERDP on interrupt exit.
On the next interrupt, the event ring may full but driver will not
update ERDP as software_dequeue is equal to ERDP.
[ 536.377115] xhci_hcd 0000:00:12.0: ERROR unknown event type 37
[ 566.933173] sd 8:0:0:0: [sdb] tag#27 uas_eh_abort_handler 0 uas-tag 7 inflight: CMD OUT
[ 566.933181] sd 8:0:0:0: [sdb] tag#27 CDB: Write(10) 2a 00 17 71 e6 78 00 00 08 00
[ 572.041186] xhci_hcd On some situataions,the0000:00:12.0: xHCI host not responding to stop endpoint command.
[ 572.057193] xhci_hcd 0000:00:12.0: Host halt failed, -110
[ 572.057196] xhci_hcd 0000:00:12.0: xHCI host controller not responding, assume dead
[ 572.057236] sd 8:0:0:0: [sdb] tag#26 uas_eh_abort_handler 0 uas-tag 6 inflight: CMD
[ 572.057240] sd 8:0:0:0: [sdb] tag#26 CDB: Write(10) 2a 00 38 eb cc d8 00 00 08 00
[ 572.057244] sd 8:0:0:0: [sdb] tag#25 uas_eh_abort_handler 0 uas-tag 5 inflight: CMD
Hardware ERDP is updated mid event handling if there are more than 128
events in an interrupt (half of ring size).
Fix this by updating the software local variable at the same time as
hardware ERDP.
[commit message rewording -Mathias]
Fixes: dc0ffbea57 ("usb: host: xhci: update event ring dequeue pointer on purpose")
Reviewed-by: Peter Chen <peter.chen@kernel.org>
Signed-off-by: Weitao Wang <WeitaoWang-oc@zhaoxin.com>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20220408134823.2527272-2-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Make xhci_disable_slot() synchronous, thus ensuring it, and
xhci_free_dev() calling it return after xHC controller completes
the disable slot command.
Otherwise the roothub and xHC host may runtime suspend, and clear the
command ring while the disable slot command is being processed.
This causes a command completion mismatch as the completion event can't
be mapped to the correct command.
Command ring gets out of sync and commands time out.
Driver finally assumes host is unresponsive and bails out.
usb 2-4: USB disconnect, device number 10
xhci_hcd 0000:00:0d.0: ERROR mismatched command completion event
...
xhci_hcd 0000:00:0d.0: xHCI host controller not responding, assume dead
xhci_hcd 0000:00:0d.0: HC died; cleaning up
Cc: <stable@vger.kernel.org>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20211210141735.1384209-3-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Turns out some xHC controllers require all 64 bits in the CRCR register
to be written to execute a command abort.
The lower 32 bits containing the command abort bit is written first.
In case the command ring stops before we write the upper 32 bits then
hardware may use these upper bits to set the commnd ring dequeue pointer.
Solve this by making sure the upper 32 bits contain a valid command
ring dequeue pointer.
The original patch that only wrote the first 32 to stop the ring went
to stable, so this fix should go there as well.
Fixes: ff0e50d356 ("xhci: Fix command ring pointer corruption while aborting a command")
Cc: stable@vger.kernel.org
Tested-by: Pavankumar Kondeti <quic_pkondeti@quicinc.com>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20211126122340.1193239-2-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The command ring pointer is located at [6:63] bits of the command
ring control register (CRCR). All the control bits like command stop,
abort are located at [0:3] bits. While aborting a command, we read the
CRCR and set the abort bit and write to the CRCR. The read will always
give command ring pointer as all zeros. So we essentially write only
the control bits. Since we split the 64 bit write into two 32 bit writes,
there is a possibility of xHC command ring stopped before the upper
dword (all zeros) is written. If that happens, xHC updates the upper
dword of its internal command ring pointer with all zeros. Next time,
when the command ring is restarted, we see xHC memory access failures.
Fix this issue by only writing to the lower dword of CRCR where all
control bits are located.
Cc: stable@vger.kernel.org
Signed-off-by: Pavankumar Kondeti <pkondeti@codeaurora.org>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20211008092547.3996295-5-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Seen on a VLI VL805 PCIe to USB controller. For non-stream endpoints
at least, if the xHC halts on a particular TRB due to an error then
the DCS field in the Out Endpoint Context maintained by the hardware
is not updated with the current cycle state.
Using the quirk XHCI_EP_CTX_BROKEN_DCS and instead fetch the DCS bit
from the TRB that the xHC stopped on.
[ bjorn: rebased to v5.14-rc2 ]
Link: https://github.com/raspberrypi/linux/issues/3060
Cc: stable@vger.kernel.org
Signed-off-by: Jonathan Bell <jonathan@raspberrypi.org>
Signed-off-by: Bjørn Mork <bjorn@mork.no>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20211008092547.3996295-3-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
As we register two usb buses for each xHC, and systems with several
hosts are more and more common it is getting hard to follow the
flow of debug messages without knowing which bus they belong to
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20210820123503.2605901-7-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Add more debugging messages to follow what happends to a URB internally
in special cases like URB cancel, halted endpoints and endpoint reset.
Helps tracking issues like URB never given back by host.
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20210820123503.2605901-6-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Only TDs with status TD_CLEARING_CACHE will be given back after
cache is cleared with a set TR deq command.
xhci_invalidate_cached_td() failed to set the TD_CLEARING_CACHE status
for some cancelled TDs as it assumed an endpoint only needs to clear the
TD it stopped on.
This isn't always true. For example with streams enabled an endpoint may
have several stream rings, each stopping on a different TDs.
Note that if an endpoint has several stream rings, the current code
will still only clear the cache of the stream pointed to by the last
cancelled TD in the cancel list.
This patch only focus on making sure all canceled TDs are given back,
avoiding hung task after device removal.
Another fix to solve clearing the caches of all stream rings with
cancelled TDs is needed, but not as urgent.
This issue was simultanously discovered and debugged by
by Tao Wang, with a slightly different fix proposal.
Fixes: 674f8438c1 ("xhci: split handling halted endpoints into two steps")
Cc: <stable@vger.kernel.org> #5.12
Reported-by: Tao Wang <wat@codeaurora.org>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20210820123503.2605901-4-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Removes static char buffer usage in the following decode functions:
xhci_decode_ctrl_ctx()
xhci_decode_slot_context()
xhci_decode_usbsts()
xhci_decode_doorbell()
xhci_decode_ep_context()
Caller must provide a buffer to use.
In tracing use __get_str() as recommended to pass buffer.
Minor changes are needed in other xhci code as these functions are also
used elsewhere
Cc: <stable@vger.kernel.org>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20210820123503.2605901-3-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Save a bit of power by not interrupting so often by default if
XHCI_AVOID_BEI quirk is set.
In normal cases the xhci driver will only generate an interrupt on the last
isochronous TRB of an URB. In a common UVC webcam usecase there are 32 TRBs
per URB.
if AVOID_BEI flag is set then xhci driver will force an interrupt every 8th
isoc TRB to make sure the event ring doesn't get too full.
This is however way too frequent in common single webcam use cases, causing
1000 interrupts/sec and thus poor powermanagement performance.
Instead start with interrupting every 32 isoc TRB, and halve it in case
event ring becomes half-full. Stop halving when reaching a rate of every
8th trb.
This is a one way solution. If interrupt rate is increased it will stay
high until driver is reloaded. The highest rate is the same as the old
default rate.
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20210617150354.1512157-3-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If endpoints halts due to a stall then the dequeue pointer read from
hardware may already be set ahead of the stalled TRB.
After commit 674f8438c1 ("xhci: split handling halted endpoints into two
steps") in 5.12 xhci driver won't issue a Set TR Dequeue if hardware
dequeue pointer is already in the right place.
Turns out the "Set TR Dequeue pointer" command is anyway needed as it in
addition to moving the dequeue pointer also clears endpoint state and
cache.
Fixes: 674f8438c1 ("xhci: split handling halted endpoints into two steps")
Cc: <stable@vger.kernel.org> # 5.12
Reported-by: Peter Ganzhorn <peter.ganzhorn@googlemail.com>
Tested-by: Peter Ganzhorn <peter.ganzhorn@googlemail.com>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20210525074100.1154090-3-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
5.12 kernel changes how xhci handles cancelled URBs and halted
endpoints. Among these changes cancelled and stalled URBs are no longer
given back before they are cleared from xHC hardware cache.
These changes unfortunately cleared the -EPIPE status of a stalled
transfer in one case before giving bak the URB, causing a USB card reader
to fail from working.
Fixes: 674f8438c1 ("xhci: split handling halted endpoints into two steps")
Cc: <stable@vger.kernel.org> # 5.12
Reported-by: Peter Ganzhorn <peter.ganzhorn@googlemail.com>
Tested-by: Peter Ganzhorn <peter.ganzhorn@googlemail.com>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20210525074100.1154090-2-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit 9ebf300078 ("xhci: Fix halted endpoint at stop endpoint command
completion") in 5.12 changes how cancelled URBs are given back.
To cancel a URB xhci driver needs to stop the endpoint first.
To clear a halted endpoint xhci driver needs to reset the endpoint.
In rare cases when an endpoint halt (error) races with a endpoint stop we
need to clear the reset before removing, and giving back the cancelled URB.
The above change in 5.12 takes care of this, but it also relies on the
reset endpoint completion handler to give back the cancelled URBs.
There are cases when driver refuses to queue reset endpoint commands,
for example when a link suddenly goes to an inactive error state.
In this case the cancelled URB is never given back.
Fix this by giving back the URB in the stop endpoint if queuing a reset
endpoint command fails.
Fixes: 9ebf300078 ("xhci: Fix halted endpoint at stop endpoint command completion")
CC: <stable@vger.kernel.org> # 5.12
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20210512080816.866037-3-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The same values are parsed several times from transfer and event
TRBs by different functions in the same call path, all while processing
one transfer event.
As the TRBs are in DMA memory and can be accessed by the xHC host we want
to avoid this to prevent double-fetch issues.
To resolve this pass the already parsed values to the different functions
in the path of parsing a transfer event
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20210406070208.3406266-5-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Force-threaded interrupt handlers used to run with interrupts enabled,
something which could lead to deadlocks in case a threaded handler
shared a lock with code running in hard interrupt context (e.g. timer
callbacks) and did not explicitly disable interrupts.
Since commit 81e2073c17 ("genirq: Disable interrupts for force
threaded handlers") interrupt handlers always run with interrupts
disabled on non-RT so that drivers no longer need to do handle forced
threading ("threadirqs").
Drop the now obsolete workaround added by commit 63aea0dbab ("USB:
xhci: fix lock-inversion problem").
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: Bart Van Assche <bart.vanassche@sandisk.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Johan Hovold <johan@kernel.org>
Link: https://lore.kernel.org/r/20210322111140.32056-1-johan@kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
On some systems rt2800usb and mt7601u devices are unable to operate since
commit f8f80be501 ("xhci: Use soft retry to recover faster from
transaction errors")
Seems that some xHCI controllers can not perform Soft Retry correctly,
affecting those devices.
To avoid the problem add xhci->quirks flag that restore pre soft retry
xhci behaviour for affected xHCI controllers. Currently those are
AMD_PROMONTORYA_4 and AMD_PROMONTORYA_2, since it was confirmed
by the users: on those xHCI hosts issue happen and is gone after
disabling Soft Retry.
[minor commit message rewording for checkpatch -Mathias]
Fixes: f8f80be501 ("xhci: Use soft retry to recover faster from transaction errors")
Cc: <stable@vger.kernel.org> # 4.20+
Reported-by: Bernhard <bernhard.gebetsberger@gmx.at>
Tested-by: Bernhard <bernhard.gebetsberger@gmx.at>
Signed-off-by: Stanislaw Gruszka <stf_xl@wp.pl>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=202541
Link: https://lore.kernel.org/r/20210311115353.2137560-2-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Currently xhci-hcd.ko building depends on USB_XHCI_MTK, this
is not flexible for some cases. For example:
USB_XHCI_HCD is y, and USB_XHCI_MTK is m, then we can't
implement extended functions if only update xhci-mtk.ko
This patch is used to remove the dependence.
Acked-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Signed-off-by: Chunfeng Yun <chunfeng.yun@mediatek.com>
Link: https://lore.kernel.org/r/0b62e21ddfacc1c2874726dd27ccab80c993f303.1615170625.git.chunfeng.yun@mediatek.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
xhci driver may in some special cases need to copy small amounts
of payload data to a bounce buffer in order to meet the boundary
and alignment restrictions set by the xHCI specification.
In the majority of these cases the data is in a sg list, and
driver incorrectly assumed data is always in urb->sg when using
the bounce buffer.
If data instead is contiguous, and in urb->transfer_buffer, we may still
need to bounce buffer a small part if data starts very close (less than
packet size) to a 64k boundary.
Check if sg list is used before copying data to/from it.
Fixes: f9c589e142 ("xhci: TD-fragment, align the unsplittable case with a bounce buffer")
Cc: stable@vger.kernel.org
Reported-by: Andreas Hartmann <andihartmann@01019freenet.de>
Tested-by: Andreas Hartmann <andihartmann@01019freenet.de>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20210203113702.436762-2-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
