If the host controller stops responding to commands, we need to kill all
the URBs that were queued to all endpoints. The current code would only
kill URBs that had been queued to the endpoint rings. ep->ring is set
to NULL if streams has been enabled for the endpoint, which means URBs
submitted with a non-zero stream_id would never get killed. Fix this.
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
In preparation for fixing this function for streams endpoints, refactor
code in the command watchdog timeout function into two new functions.
One kills all URBs on a ring (either stream or endpoint), the other
kills all URBs associated with an endpoint. Fix a split string while
we're at it.
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
Although an interesting concept, I don't think that this is a good idea:
-This will result in lots of "virtual" scsi controllers confusing users
-If we get a scsi-bus-reset we will now need to do a usb-device-reset of all
uas devices on the same usb bus, which is something to avoid if possible
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
At the kernel-summit Sarah Sharp asked me if I was willing to become the
uas maintainer. I said yes, and here is a patch to make this official.
Also remove Matthew Wilcox and Sarah Sharp as maintainers at their request.
I've also added myself to the module's author tag, so that if people look there
rather then in maintainers they will know they should bug me about uas too.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
xhci streams support is fixed, unblock usb attached scsi.
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
Copy the sg alignment trick from the usb-storage driver, without this I'm
seeing intermittent errors when using uas devices with an ehci controller.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
The scsi error handling path re-uses previously queued up (and errored-out)
cmds. If such a re-used cmd had a data-phase then cmdinfo will have
data_in_urb / data_out_urb still set to the free-ed urbs from the errored-out
cmd, and they will get free-ed a second time when the error handling cmd
completes, corrupting the kernel heap.
Clearing cmdinfo on command queue-ing fixes this, and seems like a good idea
in general.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
The scsi-host structure is refcounted, scsi_remove_host tears down the
scsi-host but does not decrement the refcount, so we need to call
scsi_put_host on disconnect to get the underlying memory to be freed.
After calling scsi_remove_host, the scsi-core may still hold a reference to
the scsi-host, iow we may still get called after uas_disconnect, but we
do our own life cycle management of uas_devinfo, freeing it on disconnect,
and thus may end up using devinfo after it has been freed. Switch to letting
scsi_host_alloc allocate and manage the memory for us.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
cmds are either on the inflight list or on the dead list, never both, so
we only need one list head.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
Before this commit the uas driver would keep track of scsi commands which still
need to have some urbs submitted to the device, and complete this with an
ABORT result code on bus-reset or disconnect, but in flight scsi commands
which have all their urbs submitted, and thus are not part of the work list,
would never get their done callback called.
The problem is killed sense urbs don't have any tag info, so it is impossible
to tell which scsi cmd they belong to, so merely making sure all the urbs
have completed one way or the other is not enough.
This commit fixes this by changing the work list to an inflight list, which
keeps tracks of all inflight scsi cmnds, using the IS_IN_WORK_LIST flag to
determine if actual work needs to be done in uas_do_work(), and by moving
marking all inflight scsi commands as aborted and moving them to the dead list
on bus-reset or disconnect.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
uas_alloc_data_urb always gets called with a stream_id value of 0 when not
using streams. Removing the check makes it consistent with uas_alloc_sense_urb.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
For USB-2 connections the stream-id must always be 0.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
Some BIOS-es will hang on reboot when an uas device is attached and left in
uas mode on reboot.
This commit adds a shutdown handler which on reboot puts the device back into
usb-storage mode, fixing the hang on reboot on these systems.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
We can sleep in our own workqueue (which is the whole reason for having
it), and scsi error handlers are also always called from a context which
may sleep.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
Since we use a fixed tag / stream for tasks we cannot allow more then one
to run at the same time. This could happen before this time if a task timed
out.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
The fixed endpoint config code was only necessary to deal with an early
uas prototype which has never been released, so lets drop it and enforce
proper uas endpoint descriptors.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
The loop uses up to 3 bytes of the endpoint extra data.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
This is a preparation patch for adding better descriptor validation.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
Falling back from uas to usb-storage requires coordination between uas and
usb-storage, so use usb-storage's quirks module parameter, rather then
requiring the user to pass a param to 2 different modules.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
uas devices have 2 alternative settings on their usb-storage interface,
one for usb-storage and one for uas. Using the uas driver is preferred, so if
the uas driver is enabled, and the device has an uas alt setting, don't bind.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
Once we start supporting uas hardware, and as more and more uas devices
become available, we will likely start seeing broken devices. This patch
prepares for the inevitable need for blacklisting those devices from
using the uas driver (they will use usb-storage instead).
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
On disconnect USB3 protocol ports transit from U0 to SS.Inactive to Rx.Detect,
on a recoverable error, the port stays in SS.Inactive and we recover from it by
doing a warm-reset (through usb_device_reset if we have a udev for the port).
If this really is a disconnect we may end up trying the warm-reset anyways,
since khubd may run before the SS.Inactive to Rx.Detect transition, or it
may get skipped if the transition to Rx.Detect happens before khubd gets run.
With a loose connector, or in the case which actually led me to debugging this
bad ACPI firmware toggling Vbus off and on in quick succession, the port
may transition from Rx.Detect to U0 again before khubd gets run. In this case
the device state is unknown really, but khubd happily goes into the resuscitate
an existing device path, and the device driver never gets notified about the
device state being messed up.
If the above scenario happens with a streams using device, as soon as an urb
is submitted to an endpoint with streams, the following appears in dmesg:
ERROR Transfer event for disabled endpoint or incorrect stream ring
@0000000036807420 00000000 00000000 04000000 04078000
Notice how the TRB address is all zeros. I've seen this both on Intel
Pantherpoint and Nec xhci hosts.
Luckily we can detect the U0 to SS.Inactive to Rx.Detect to U0 all having
happened before khubd runs case since the C_LINK_STATE bit gets set in the
portchange bits on the U0 -> SS.Inactive change. This bit will also be set on
suspend / resume, but then it gets cleared by port_hub_init before khubd runs.
So if the C_LINK_STATE bit is set and a warm-reset is not needed, iow the port
is not still in SS.Inactive, and the port still has a connection, then the
device needs to be reset to put it back in a known state.
I've verified that doing the device reset also fixes the transfer event with
all zeros address issue.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
If streams are still allocated on device-reset or set-interface then the hcd
code implictly frees the streams. Clear host_endpoint->streams in this case
so that if a driver later tries to re-allocate them it won't run afoul of the
device already having streams check in usb_alloc_streams().
Note normally streams still being allocated at reset / set-intf would be a
driver bug, but this can happen without it being a driver bug on reset-resume.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
If we align segment dma pool memory to 64 bytes, then a segment can be located
at 0x10000040 - 0x1000043f, and a segment from another ring at 0x10000440 -
0x1000083f. The last trb in the first segment at 0x10000430 will then translate
to the same radix tree key as the first trb of the second segment, while they
are in different rings!
This patches fixes this by changing the alignment of the dma pool to be 1KB
rather then 64 bytes. An alternative fix would be to reduce the shift used
to calculate the radix tree keys, but that would (slighlty) grow the radix
trees so I believe this is the better fix.
Note this patch is mostly theoretical since in practice I've not seen
the dma_pool actually return not 1KB aligned memory.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
cmd_ring_reserved_trbs gets decremented by xhci_free_stream_info(), so set it
to 0 after freeing all rings, otherwise it wraps around to a very large value
when rings with streams are free-ed.
Before this patch the wrap-around could be triggered when xhci_resume
calls xhci_mem_cleanup if the controller resume fails.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
This is a preparation patch for teaching usb-storage to not bind to
uas devices.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
This is a preparation patch for teaching usb-storage to not bind to
uas devices.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
If we get ie 16 streams we can use stream-id 1-16, not 1-15.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
The iu struct definitions are usb packet definitions, so no alignment should
happen. Notice that assuming 32 bit alignment this does not make any
difference at all.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
The response iu struct before this patch has a size of 7 bytes (discounting
padding), which is weird since all other iu-s are explictly padded to
a multiple of 4 bytes.
More over submitting a 7 byte bulk transfer to the status endpoint when
expecting a response iu results in an USB babble error, as the device
actually sends 8 bytes.
Up on closer reading of the UAS spec:
http://www.t10.org/cgi-bin/ac.pl?t=f&f=uas2r00.pdf
The reason for this becomes clear, the 2 entries in "Table 17 — RESPONSE IU"
are numbered 4 and 6, looking at other iu definitions in the spec, esp.
multi-byte fields, this indicates that the ADDITIONAL RESPONSE INFORMATION
field is not a 2 byte field as one might assume at a first look, but is
a multi-byte field containing 3 bytes.
This also aligns with the SCSI Architecture Model 4 spec, which UAS is based
on which states in paragraph "7.1 Task management function procedure calls"
that the "Additional Response Information" output argument for a Task
management function procedure call is 3 bytes.
Last but not least I've verified this by sending a logical unit reset task
management call with an invalid lun to an actual uasp device, and received
back a response-iu with byte 6 being 0, and byte 7 being 9, which is the
responce code for an invalid iu, which confirms that the response code is
being reported in byte 7 of the response iu rather then in byte 6.
Things were working before despite this error in the response iu struct
definition because the additional response info field is normally filled
with zeros, and 0 is the response code value for success.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
Handle usb-device resets not triggered from uas_eh_bus_reset_handler(), when
this happens, disable cmd queuing during the reset, and wait for existing
requests to finish in pre_reset.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
Fix the uas_eh_bus_reset_handler not properly taking the usbdev lock
before calling usb_device_reset, the usb-core expects this lock to be
taken when usb_device_reset is called.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
I thought it would be a good idea to also test uas with usb-2, and it turns out
it was, as it did not work. The problem is that the uas driver was passing the
bEndpointAddress' direction bit to usb_rcvbulkpipe, the xhci code seems to not
care about this, but with the ehci code this causes usb_submit_urb failure.
With this fixed the uas code works nicely with an uas device plugged into
an ehci port.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
The cmd endpoint never has streams, so the stream_id parameter is unused.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
All callers of unlink_data_urbs drop devinfo->lock before calling it, and
then immediately take it again after the call. And the first thing
unlink_data_urbs does is take the lock again, and the last thing it does
is drop it. This commit removes all the unnecessary lock dropping and taking.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
- Rename labels to properly reflect this
- Don't skip free-ing the streams when scsi_init_shared_tag_map fails
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
Otherwise they may complete before they get anchored and thus never get
unanchored (as the unanchoring is done by the usb core on completion).
This commit also remove the usb_get_urb / usb_put_urb around cmd submission +
anchoring, since if done in the proper order this is not necessary.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
Sarah Sharp