Commit Graph

469744 Commits

Author SHA1 Message Date
Petr Mladek
59d48b3f1f usb: hub: rename usb_kick_khubd() to usb_kick_hub_wq()
USB hub started to use a workqueue instead of kthread. Let's make it clear from
the function names.

Signed-off-by: Petr Mladek <pmladek@suse.cz>
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 22:33:19 -07:00
Petr Mladek
32a6958998 usb: hub: convert khubd into workqueue
There is no need to have separate kthread for handling USB hub events.
It is more elegant to use the workqueue framework.

The workqueue is allocated as freezable because the original thread was
freezable as well.

Also it is allocated as ordered because the code is not ready for parallel
processing of hub events, see choose_devnum().

struct usb_hub is passed via the work item. Therefore we do not need
hub_event_list.

Also hub_thread() is not longer needed. It would call only hub_event().
The rest of the code did manipulate the kthread and it is handled by the
workqueue framework now.

kick_khubd is renamed to kick_hub_wq() to make the function clear. And the
protection against races is done another way, see below.

hub_event_lock has been removed. It cannot longer be used to protect struct
usb_hub between hub_event() and hub_disconnect(). Instead we need to get
hub->kref already in kick_hub_wq().

The lock is not really needed for the other scenarios as well. queue_work()
returns whether it succeeded. We could revert the needed operations
accordingly. This is enough to avoid duplicity and inconsistencies.

Yes, the removed lock causes that there is not longer such a strong
synchronization between scheduling the work and manipulating
hub->disconnected.

But kick_hub_wq() must never be called together with hub_disconnect()
otherwise even the original code would have failed. Any callers are
responsible for this.

Therefore the only problem is that hub_disconnect() could be called in parallel
with hub_event(). But this was possible even in the past. struct usb_hub is
still guarded by hub->kref and released in hub_events() when needed.

Note that the source file is still full of the obsolete "khubd" strings.
Let's remove them in a follow up patch. This patch already is complex enough.

Thanks a lot Alan Stern <stern@rowland.harvard.edu> for code review, many useful
tips and guidance. Also thanks to Tejun Heo <tj@kernel.org> for hints how to
allocate the workqueue.

Signed-off-by: Petr Mladek <pmladek@suse.cz>
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 22:33:19 -07:00
Petr Mladek
eb6e292487 usb: hub: rename hub_events() to hub_event() and handle only one event there
We would like to convert khubd kthread to a workqueue. As a result hub_events()
will handle only one event per call.

In fact, we could do this already now because there is another cycle in
hub_thread(). It calls hub_events() until hub_event_list is empty.

This patch renames the function to hub_event(), removes the while cycle, and
renames the goto targets from loop* to out*.

When touching the code, it fixes also formatting of dev_err() and dev_dbg()
calls to make checkpatch.pl happy :-)

Signed-off-by: Petr Mladek <pmladek@suse.cz>
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 22:31:12 -07:00
Petr Mladek
5d14f32383 usb: hub: keep hub->dev reference all the time when struct usb_hub lives
This is just a small optimization of the fix from the commit c605f3cdff
("usb: hub: take hub->hdev reference when processing from eventlist).

We do not need to take the reference for each event. Instead we could get it
when struct usb_hub is allocated and put it when it is released. By other words,
we could handle it the same way as the reference for hub->intfdev.

The motivation is that it will make the life easier when switching from khubd
kthread to a workqueue.

Suggested-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Petr Mladek <pmladek@suse.cz>
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 22:31:11 -07:00
Mark Knibbs
3512e7bfea storage: Add quirk for another SCM-based USB-SCSI converter
There is apparently another SCM USB-SCSI converter with ID 04E6:000F. It
is listed along with 04E6:000B in the Windows INF file for the Startech
ICUSBSCSI2 as "eUSB SCSI Adapter (Bus Powered)". The quirk allows
devices with SCSI ID other than 0 to be accessed.

Also make a couple of existing SCM product IDs lower case to be
consistent with other entries.

Signed-off-by: Mark Knibbs <markk@clara.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 22:29:27 -07:00
Mark Knibbs
57cde01a7b storage: Add quirks for Castlewood and Double-H USB-SCSI converters
Castlewood Systems supplied various models of USB-SCSI converter with their
ORB external removable-media drive. The ORB Windows and Macintosh drivers
support six USB IDs:
 084B:A001     [VID 084B is Castlewood Systems]
 04E6:0002 (*) ORB USB Smart Cable P/N 88205-001 (generic SCM ID)
 2027:A001     Double-H Technology DH-2000SC
 1822:0001 (*) Ariston iConnect/iSCSI
 07AF:0004 (*) Microtech XpressSCSI (25-pin)
 07AF:0005 (*) Microtech XpressSCSI (50-pin)

*: quirk already in unusual-devs.h

[Apparently the official VID for Double-H Technology is 0x07EB = 2027
decimal. That's another hex/decimal mix-up with these SCM-based products
(in addition to the Ariston and Entrega ones). Perhaps the USB-IF informed
companies of their allocated VID in decimal, but they assumed it was hex?
It seems all Entrega products used VID 0x1645, not just the USB-SCSI
converter.]

Double-H Technology Co., Ltd. produced a USB-SCSI converter, model
DH-2000SC, which is probably the one supported by the ORB drivers. Perhaps
the Castlewood-bundled product had a different label or PID though?
Castlewood mentioned Conmate as being one type of USB-SCSI converter.
Conmate and Double-H seem related somehow; both company addresses in the
same road, and at one point the Conmate web site mentioned DH-2000H4,
DH-200D4/DH-2000C4 as models of USB hub (DH short for Double-H presumably).
Conmate did show a USB-SCSI converter model CM-660 on their web site at one
point. My guess is that was identical to the DH-2000SC.

Mention of the Double-H product:
  http://web.archive.org/web/20010221010141/http://www.doubleh.com.tw/dh-2000sc.htm
The only picture I could find is at
  http://jp.acesuppliers.com/catalog/j64/component/page03.html
The casing design looks the same as my ORB USB Smart Cable which has ID
04E6:0002.

Anyway, that's enough rambling. Here's the patch.

storage: Add quirks for Castlewood and Double-H USB-SCSI converters

Add quirks for two SCM-based USB-SCSI converters which were bundled with
some Castlewood ORB removable drives. Without the quirk only the (single)
drive with SCSI ID 0 can be accessed.

Signed-off-by: Mark Knibbs <markk@clara.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 22:29:27 -07:00
Mark Knibbs
bda9893c50 storage: Replace magic number with define in usb_stor_euscsi_init()
usb_stor_euscsi_init() calls usb_stor_control_msg() with timeout
argument 5000. USB_CTRL_SET_TIMEOUT is defined to be 5000 in usb.h, so
would it make sense to use that instead? Patch below if it would.

Signed-off-by: Mark Knibbs <markk@clara.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 22:29:27 -07:00
Greg Kroah-Hartman
5cb307c4c2 USB: quirks.h: use BIT()
Use the BIT macro instead of "open coding" bit fields.  This makes it
easier to actually see that the bits are not conflicting/overlapping.

Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 22:23:55 -07:00
Lu Baolu
ddbe1fca0b USB: Add device quirk for ASUS T100 Base Station keyboard
This full-speed USB device generates spurious remote wakeup event
as soon as USB_DEVICE_REMOTE_WAKEUP feature is set. As the result,
Linux can't enter system suspend and S0ix power saving modes once
this keyboard is used.

This patch tries to introduce USB_QUIRK_IGNORE_REMOTE_WAKEUP quirk.
With this quirk set, wakeup capability will be ignored during
device configure.

This patch could be back-ported to kernels as old as 2.6.39.

Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 22:20:59 -07:00
Tobias Jakobi
4463e15267 usb3503: clarify what the registers 'PDS' and 'CFG1' really do
The current comment sounds like you have to disable some of
the ports to be able to use self-powered mode. This is
misleading, so change the wording to reflect this.

Signed-off-by: Tobias Jakobi <tjakobi@math.uni-bielefeld.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 22:15:18 -07:00
Tobias Jakobi
3717c5c9b9 usb3503: fix typos in devicetree binding documentation
'availe' to 'available'
'desribed' to 'described'

Signed-off-by: Tobias Jakobi <tjakobi@math.uni-bielefeld.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 22:15:18 -07:00
Tobias Jakobi
6b0844214d usb3503: correct error message in probe ('connect' to 'interrupt')
intn is obviously the interrupt pin.

Signed-off-by: Tobias Jakobi <tjakobi@math.uni-bielefeld.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 22:15:18 -07:00
Mickael Maison
a6cd244b92 usb: Fixed a few typos
Fixed typos in comments of various drivers/usb files

Signed-off-by: Mickael Maison <mickael.maison@gmail.com>
Acked-by: Felipe Balbi <balbi@ti.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 22:15:18 -07:00
Andy Shevchenko
7adce46784 USB: storage: use %*ph specifier to dump small buffers
Instead of dereference each byte let's use %*ph specifier in the printk()
calls.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 22:15:18 -07:00
Chen Gang
13d79f223f drivers/usb/host/ehci-xilinx-of.c: Include "linux/of_irq.h" to avoid compiling error
Need include it for irq_of_parse_and_map(), the related error with
allmodconfig under microblaze:

  drivers/usb/host/ehci-xilinx-of.c: In function ‘ehci_hcd_xilinx_of_probe’:
  drivers/usb/host/ehci-xilinx-of.c:156:2: error: implicit declaration of function ‘irq_of_parse_and_map’ [-Werror=implicit-function-declaration]
    irq = irq_of_parse_and_map(dn, 0);
    ^

Signed-off-by: Chen Gang <gang.chen.5i5j@gmail.com>
Acked-by: Michal Simek <monstr@monstr.eu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 22:09:46 -07:00
Thomas Pugliese
275e517c30 usb: wusbcore: fix device disconnect on rekey timeout
If three or more wireless devices are connected and two of them
disconnect between 1-3 seconds apart, it can cause the HWA to disconnect
the remaining devices due to failing to see a DN_Alive message from
them.  This happens because when the HWA detects that the first device
is gone, it will attempt to rekey the remaining devices.  If one of the
devices is not responding because it has also been disconnected but not
yet timed out, the synchronous rekey operation running on the wusbd
workqueue can block for up to 5 seconds.  This will prevent the
KEEPALIVE timer from running and DN_Alive messages from being processed
because they are processed by the same workqueue.  This patch moves the
rekey operation to a separate workqueue since it is the only wusb work
item that needs to communicate directly with wireless devices.  The rest
of the WUSB work items either perform no device IO or communicate
directly with the host controller and should not be blocked out by a
non-responding wireless device.

Signed-off-by: Thomas Pugliese <thomas.pugliese@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 22:06:59 -07:00
Thomas Pugliese
b94be0db5b usb: wusbcore: skip done segs before completing aborted transfer
When completing an aborted transfer, skip done segs before calling
wa_complete_remaining_xfer_segs to avoid a runtime warning.  The warning
is harmless in this case but avoiding it prevents false error reports.

Signed-off-by: Thomas Pugliese <thomas.pugliese@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 22:06:59 -07:00
Thomas Pugliese
f55025289c usb: wusbcore: USB_WUSB_CBAF depends on USB
Add USB build dependency for USB_WUSB_CBAF.

Signed-off-by: Thomas Pugliese <thomas.pugliese@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 22:06:33 -07:00
Thomas Pugliese
e8f8ba4736 usb: wusbcore: remove USB_WUSB build dependency on PCI
Wireless USB does not require PCI so remove USB_WUSB build dependency on
PCI.

Signed-off-by: Thomas Pugliese <thomas.pugliese@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 22:06:33 -07:00
Thomas Pugliese
cbb671992b usb: hwa: add USB build dependency for USB_HWA_HCD
An HWA is a USB device so it depends on USB.

Signed-off-by: Thomas Pugliese <thomas.pugliese@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 22:06:33 -07:00
Thomas Pugliese
534574bb46 uwb: doc: add documentation for ASIE sysfs attribute
Document the /sys/class/uwb_rc/uwbN/ASIE sysfs attribute

Signed-off-by: Thomas Pugliese <thomas.pugliese@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 22:05:26 -07:00
Thomas Pugliese
d08e1ad994 uwb: add an ASIE sysfs attribute to uwb_rc devices
Allow user mode to add and remove application specific information
elements (ASIEs) to the beacon of a uwb_rc device.

Signed-off-by: Thomas Pugliese <thomas.pugliese@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 22:05:26 -07:00
Thomas Pugliese
005799d560 uwb: doc: bring uwb documentation up to date
The existing UWB documentation states that UWB devices show up in
/sys/class/uwb and /sys/bus/uwb/devices.  Up until the most recent
changes to add uwb devices to their own bus, neither of these statements
were actually true.  Now, UWB devices do show up in /sys/bus/uwb/devices
so the documentation has been updated to reflect that and removed
references to /sys/class/uwb.

The existing documentation also states that local UWB radio controller
(RC) devices also show up as UWB devices along with their wireless peer
UWB devices under /sys/class/uwb.  This is not true so that statement
has been removed.

Signed-off-by: Thomas Pugliese <thomas.pugliese@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 22:04:44 -07:00
Thomas Pugliese
848879340a uwb: create a uwb bus type and add in-range peer devices to it
Documentation/usb/WUSB-Design-overview.txt states that UWB devices seen
by a UWB radio controller are added to /sys/bus/uwb/devices, but this
was not actually being done.  This functionality is needed in order for
UWB peer devices to be enumerated by user mode tools.  This patch
creates a uwb bus type and adds UWB peer devices to it as they are
discovered by the radio controller.

Signed-off-by: Thomas Pugliese <thomas.pugliese@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 22:04:44 -07:00
Thomas Pugliese
58e4ab3eb0 uwb: remove UWB build dependency on PCI
UWB does not require PCI to be enabled so remove build dependency.

Signed-off-by: Thomas Pugliese <thomas.pugliese@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 22:04:44 -07:00
Thomas Pugliese
e53582bbb5 uwb: update uwb device prints to be more useful
Print info about the radio controller device instead of the its parent
when UWB devices connect and disconnect.

Signed-off-by: Thomas Pugliese <thomas.pugliese@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 22:04:44 -07:00
Thomas Pugliese
9282502044 uwb: line length cleanup
Fix line length in uwb-internal.h

Signed-off-by: Thomas Pugliese <thomas.pugliese@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 22:03:54 -07:00
Julia Lawall
4f8c0602ea wusb: delete double assignment
Delete successive assignments to the same location.

A simplified version of the semantic match that finds this problem is as
follows: (http://coccinelle.lip6.fr/)

// <smpl>
@@
expression i;
@@

*i = ...;
 i = ...;
// </smpl>

Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 22:03:54 -07:00
Bartlomiej Zolnierkiewicz
2b6127ddef usb: renesas_usbhs: fix driver dependencies
Renesas USBHS controller support should be available only on
Renesas ARM SoCs and SuperH architecture.

Signed-off-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com>
Acked-by: Kyungmin Park <kyungmin.park@samsung.com>
Cc: Magnus Damm <magnus.damm@gmail.com>
Acked-by: Geert Uytterhoeven <geert+renesas@glider.be>
Acked-by: Simon Horman <horms+renesas@verge.net.au>
Acked-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 21:57:22 -07:00
Vivek Gautam
2db941623d usb: host: ohci-exynos: Remove unnecessary usb-phy support
Now that we have completely moved from older USB-PHY drivers
to newer GENERIC-PHY drivers for PHYs available with USB controllers
on Exynos series of SoCs, we can remove the support for the same
in our host drivers too.

We also defer the probe for our host in case we end up getting
EPROBE_DEFER error when getting PHYs.

Signed-off-by: Vivek Gautam <gautam.vivek@samsung.com>
Acked-by: Jingoo Han <jg1.han@samsung.com>
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 21:53:48 -07:00
Arjun Sreedharan
113ad911ad usb: misc: yurex: remove useless casting of private_data
Signed-off-by: Arjun Sreedharan <arjun024@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 21:50:57 -07:00
Hans de Goede
cffb9be80f xhci: Log extra info on "ERROR Transfer event TRB DMA ptr not part of current TD"
Lately (with the use of uas / bulk-streams) we have been seeing several
cases where this error triggers (which should never happen).

Add some extra logging to make debugging these errors easier.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 21:46:11 -07:00
Hans de Goede
f85c9fb62c xhci: Remove "FIXME - check all the stream rings for pending cancellations"
Even though a Set TR deq ptr command operates on a ring, and an endpoint
can have multiple rings, we can have only one Set TR deq ptr command pending.

When an endpoint with streams halts or is stopped to unlink urbs, there
will only be at most one ring active / one td being executed (the td
stopped_td points to).

So when we reset the endpoint (for a halt), or the stop command completes, we
will queue one Set TR deq ptr command at most, cancelled urbs on other stream
rings then the one being executed will have there trbs turned to nops, and
once the hcd gets around to execute that stream ring they will be simply
skipped.

So the SET_DEQ_PENDING flag in the endpoint is sufficient protection against
starting the endpoing before all stream rings are cleaned up.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 21:46:11 -07:00
Hans de Goede
0d4976ec8e xhci: Always ring the doorbell for active eps when a Set TR deq ptr cmd completes
Even if the stream for which the command was intended has been freed in the
mean time. This ensures that things start rolling again after an unlink / halt.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 21:46:11 -07:00
Hans de Goede
d3a43e66e0 xhci: Fold queue_set_tr_deq into xhci_queue_new_dequeue_state
xhci_queue_new_dequeue_state is the only caller of queue_set_tr_deq
and queue_set_tr_deq checks for SET_DEQ_PENDING, where as
xhci_queue_new_dequeue_state sets it which is inconsistent.

Simply fold the 2 into one is a nice cleanup and fixes the inconsistency.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 21:46:11 -07:00
Hans de Goede
b7f9696bd1 xhci: xhci_ring_device: Ring stream ring bells for endpoints with streams
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 21:46:10 -07:00
Al Cooper
14e61a1bd9 usb: xhci_suspend is not stopping the root hub timer for the shared HCD
V2 - Restart polling (which will restart the timer) for the shared
HCD in xhci_resume().

xhci_suspend() will stop the primary HCD's root hub timer, but leaves
the shared HCD's timer running. This change adds stopping of the
shared HCD timer.

Signed-off-by: Al Cooper <alcooperx@gmail.com>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 21:46:10 -07:00
Hans de Goede
1e3452e3f0 xhci: Move allocating of command for new_dequeue_state to queue_set_tr_deq()
There are multiple reasons for this:

1) This fixes a missing check for xhci_alloc_command failing in
   xhci_handle_cmd_stop_ep()
2) This adds a warning when we cannot set the new dequeue state because of
   xhci_alloc_command failing
3) It puts the allocation of the command after the sanity checks in
   queue_set_tr_deq(), avoiding leaking the command if those fail
4) Since queue_set_tr_deq now owns the command it can free it if queue_command
   fails
5) It reduces code duplication

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 21:46:10 -07:00
Hans de Goede
fac1f48584 uas: Add response iu handling
If something goes wrong in our communication with an uas device we may get
a response iu in reaction to a cmnd, rather then a status iu. In this case
propagate an error upwards, rather then logging a bogus iu message.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 21:42:12 -07:00
Hans de Goede
ce39fe6fa1 uas: Log error codes when logging errors
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 21:42:12 -07:00
Hans de Goede
1ad7ed5af3 uas: Cleanup uas_log_cmd_state usage
Instead of doing:

uas_log_cmd_state(cmnd, __func__)
scmd_printk(KERN_ERR, cmnd, "error doing foo %d\n", err)

On error, resulting in 2 log calls for a single error, make uas_log_cmd_state
take a status code, and change calls like the above to:

uas_log_cmd_state(cmnd, "error doing foo", err)

Also change various sanity checks (which should never trigger) from:
"scmd_printk(KERN_ERR, cmnd, "sanity foo failed\n")" to calling the new
uas_log_cmd_state(), so that when they do trigger we get more info.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 21:42:12 -07:00
Hans de Goede
102c00cb91 uas: Remove protype hardware usb interface info
We've removed all hack from the driver for pre-production hardware.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 21:42:12 -07:00
Hans de Goede
5ad22cfc13 uas: Remove support for old sense ui as used in pre-production hardware
I've access to a number of different uas devices now, and none of them use
old style sense urbs. The only case where these code-paths trigger is with
the asm1051 and there they do the wrong thing, as the asm1051 sends 8 bytes
status iu-s when it does not have any sense data, but uses new style
sense iu-s regardless, as can be seen for scsi cmnds where there is sense
data.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 21:42:12 -07:00
Hans de Goede
eb7d664ae4 uas: Drop COMMAND_COMPLETED flag
It was only used to sanity check against completing the same cmnd twice,
but that is the case we're likely operating on free-ed memory, and doing
sanity checks on free-ed memory is not really helpful.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 21:42:12 -07:00
Hans de Goede
6dcd8ec240 uas: Use scsi_print_command
Use scsi_print_command to print commands during errors, rather then printing
the rather meaningless pointer to the command.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 21:42:11 -07:00
Hans de Goede
51b361737b uas: Do not log urb status error on cancellation
Check for both type of cancellation codes for sense and data urbs.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 21:42:11 -07:00
Hans de Goede
e5e558192f uas: Use streams on upcoming 10Gbps / 3.1 USB
Limit the no-streams case to speeds less then USB_SPEED_SUPER.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 21:42:11 -07:00
Hans de Goede
f9dc024a2d uas: pre_reset and suspend: Fix a few races
The purpose of uas_pre_reset is to:

1) Stop any new commands from being submitted while an externally triggered
   usb-device-reset is running
2) Wait for any pending commands to finish before allowing the usb-device-reset
   to continue

The purpose of uas_suspend is to:
2) Wait for any pending commands to finish before suspending

This commit fixes races in both paths:

1) For 1) we use scsi_block_requests, but the scsi midlayer calls queuecommand
   without holding any locks, so a queuecommand may already past the midlayer
   scsi_block_requests checks when we call it, add a check to uas_queuecommand
   to fix this

2) For 2) we were waiting for all sense-urbs to complete, there are 2 problems
   with this approach:
a) data-urbs may complete after the sense urb, so we need to check for those
   too
b) if a sense-urb completes with a iu id of READ/WRITE_READY a command is not
   yet done. We submit a new sense-urb immediately in this case, but that
   submit may fail (in which case it will get retried by uas_do_work), if this
   happens the sense_urbs anchor may become empty while the cmnd is not yet
   done

Also unblock requests on timeout, to avoid things getting stuck in that case.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 21:42:11 -07:00
Hans de Goede
4c5481efb4 uas: Fix memleak of non-submitted urbs
Not all urbs we've allocated are necessarily also submitted, non-submitted
urbs will not be free-ed by their completion handler. So we need to free
them manually.

There are 2 scenarios where this can happen:

1) We have failed to submit some urbs at abort / disconnect
2) When running over usb-2 we may have never tried to submit the data urbs
   when completing the scsi cmnd, because we never got a READ/WRITE_READY iu

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 21:42:11 -07:00
Hans de Goede
616f0e6cab uas: Drop all references to a scsi_cmnd once it has been aborted
Do not keep references around to a cmnd which is under error handling.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2014-09-23 21:42:11 -07:00