This reverts commit 35773dac5f. It's a
hack that caused regressions in the usb-storage and userspace USB
drivers that use usbfs and libusb. Commit 70cabb7d992f "xhci 1.0: Limit
arbitrarily-aligned scatter gather." should fix the issues seen with the
ax88179_178a driver on xHCI 1.0 hosts, without causing regressions.
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
Cc: stable@vger.kernel.org # 3.12
Often, usb drivers need some driver_info to get a device to work. To
have access to driver_info when using new_id, allow to pass a reference
vendor:product tuple from which new_id will inherit driver_info.
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Section 4.11.7.1 of rev 1.0 of the xhci specification states that a link TRB
can only occur at a boundary between underlying USB frames (512 bytes for
high speed devices).
If this isn't done the USB frames aren't formatted correctly and, for example,
the USB3 ethernet ax88179_178a card will stop sending (while still receiving)
when running a netperf tcp transmit test with (say) and 8k buffer.
This should be a candidate for stable, the ax88179_178a driver defaults to
gso and tso enabled so it passes a lot of fragmented skb to the USB stack.
Notes from Sarah:
Discussion: http://marc.info/?l=linux-usb&m=138384509604981&w=2
This patch fixes a long-standing xHCI driver bug that was revealed by a
change in 3.12 in the usb-net driver. Commit
638c5115a7 "USBNET: support DMA SG" added
support to use bulk endpoint scatter-gather (urb->sg). Only the USB
ethernet drivers trigger this bug, because the mass storage driver sends
sg list entries in page-sized chunks.
This patch only fixes the issue for bulk endpoint scatter-gather. The
problem will still occur for periodic endpoints, because hosts will
interpret no-op transfers as a request to skip a service interval, which
is not what we want.
Luckily, the USB core isn't set up for scatter-gather on isochronous
endpoints, and no USB drivers use scatter-gather for interrupt
endpoints. Document this known limitation so that developers won't try
to use urb->sg for interrupt endpoints until this issue is fixed. The
more comprehensive fix would be to allow link TRBs in the middle of the
endpoint ring and revert this patch, but that fix would touch too much
code to be allowed in for stable.
This patch should be backported to kernels as old as 3.12, that contain
the commit 638c5115a7 "USBNET: support DMA
SG". Without this patch, the USB network device gets wedged, and stops
sending packets. Mark Lord confirms this patch fixes the regression:
http://marc.info/?l=linux-netdev&m=138487107625966&w=2
Signed-off-by: David Laight <david.laight@aculab.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
Tested-by: Mark Lord <mlord@pobox.com>
Cc: stable@vger.kernel.org
How it's supposed to work:
--------------------------
USB 2.0 Link PM is a lower power state that some newer USB 2.0 devices
support. USB 3.0 devices certified by the USB-IF are required to
support it if they are plugged into a USB 2.0 only port, or a USB 2.0
cable is used. USB 2.0 Link PM requires both a USB device and a host
controller that supports USB 2.0 hardware-enabled LPM.
USB 2.0 Link PM is designed to be enabled once by software, and the host
hardware handles transitions to the L1 state automatically. The premise
of USB 2.0 Link PM is to be able to put the device into a lower power
link state when the bus is idle or the device NAKs USB IN transfers for
a specified amount of time.
...but hardware is broken:
--------------------------
It turns out many USB 3.0 devices claim to support USB 2.0 Link PM (by
setting the LPM bit in their USB 2.0 BOS descriptor), but they don't
actually implement it correctly. This manifests as the USB device
refusing to respond to transfers when it is plugged into a USB 2.0 only
port under the Haswell-ULT/Lynx Point LP xHCI host.
These devices pass the xHCI driver's simple test to enable USB 2.0 Link
PM, wait for the port to enter L1, and then bring it back into L0. They
only start to break when L1 entry is interleaved with transfers.
Some devices then fail to respond to the next control transfer (usually
a Set Configuration). This results in devices never enumerating.
Other mass storage devices (such as a later model Western Digital My
Passport USB 3.0 hard drive) respond fine to going into L1 between
control transfers. They ACK the entry, come out of L1 when the host
needs to send a control transfer, and respond properly to those control
transfers. However, when the first READ10 SCSI command is sent, the
device NAKs the data phase while it's reading from the spinning disk.
Eventually, the host requests to put the link into L1, and the device
ACKs that request. Then it never responds to the data phase of the
READ10 command. This results in not being able to read from the drive.
Some mass storage devices (like the Corsair Survivor USB 3.0 flash
drive) are well behaved. They ACK the entry into L1 during control
transfers, and when SCSI commands start coming in, they NAK the requests
to go into L1, because they need to be at full power.
Not all USB 3.0 devices advertise USB 2.0 link PM support. My Point
Grey USB 3.0 webcam advertises itself as a USB 2.1 device, but doesn't
have a USB 2.0 BOS descriptor, so we don't enable USB 2.0 Link PM. I
suspect that means the device isn't certified.
What do we do about it?
-----------------------
There's really no good way for the kernel to test these devices.
Therefore, the kernel needs to disable USB 2.0 Link PM by default, and
distros will have to enable it by writing 1 to the sysfs file
/sys/bus/usb/devices/../power/usb2_hardware_lpm. Rip out the xHCI Link
PM test, since it's not sufficient to detect these buggy devices, and
don't automatically enable LPM after the device is addressed.
This patch should be backported to kernels as old as 3.11, that
contain the commit a558ccdcc7 "usb: xhci:
add USB2 Link power management BESL support". Without this fix, some
USB 3.0 devices will not enumerate or work properly under USB 2.0 ports
on Haswell-ULT systems.
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
Cc: stable@vger.kernel.org
usb_wait_anchor_empty_timeout() should wait till the completion handler
has run. Both the zd1211rw driver and the uas driver (in its task mgmt) depend
on the completion handler having completed when usb_wait_anchor_empty_timeout()
returns, as they read state set by the completion handler after an
usb_wait_anchor_empty_timeout() call.
But __usb_hcd_giveback_urb() calls usb_unanchor_urb before calling the
completion handler. This is necessary as the completion handler may
re-submit and re-anchor the urb. But this introduces a race where the state
these drivers want to read has not been set yet by the completion handler
(this race is easily triggered with the uas task mgmt code).
I've considered adding an anchor_count to struct urb, which would be
incremented on anchor and decremented on unanchor, and then only actually
do the anchor / unanchor on 0 -> 1 and 1 -> 0 transtions, combined with
moving the unanchor call in hcd_giveback_urb to after calling the completion
handler. But this will only work if urb's are only re-anchored to the same
anchor as they were anchored to before the completion handler ran.
And at least one driver re-anchors to another anchor from the completion
handler (rtlwifi).
So I have come up with this patch instead, which adds the ability to
suspend wakeups of usb_wait_anchor_empty_timeout() waiters to the usb_anchor
functionality, and uses this in __usb_hcd_giveback_urb() to delay wake-ups
until the completion handler has run.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Oliver Neukum <oliver@neukum.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
And do so in a way which ensures that any fields added in the future will
also get properly zero-ed.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Oliver Neukum <oliver@neukum.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The hcd-driver free_streams method can return an error, so lets properly
propagate that.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Some host controllers(such as xHCI) can support building
packet from discontinuous buffers, so introduce one flag
and helper for this kind of host controllers, then the
feature can help some applications(such as usbnet) by
supporting arbitrary length of sg buffers.
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Ming Lei <ming.lei@canonical.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When building the htmldocs (in verbose mode), scripts/kernel-doc reports the
following type of warnings:
Warning(drivers/usb/core/usb.c:76): No description found for return value of
'usb_find_alt_setting'
Fix them by:
- adding some missing descriptions of return values
- using "Return" sections for those descriptions
Signed-off-by: Yacine Belkadi <yacine.belkadi.1@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
bInterval must be within the range 1 - 16
when running at High/Super speed, and within
the range 1 - 255 when running at Full/Low speed.
In order to catch drivers passing a too
large bInterval on Super/High speed scenarios
(thus overflowing urb->interval), let's clamp()
the argument to the allowed ranges.
Signed-off-by: Felipe Balbi <balbi@ti.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
USB spec stats that short packet can only appear at the end
of transfer. Because lost of HC(EHCI/UHCI/OHCI/...) can't
build a full packet from discontinuous buffers, we introduce
the limit in usb_submit_urb() to avoid such kind of bad sg buffers
coming from driver.
The limit might be a bit strict:
- platform has iommu to do sg list mapping
- some host controllers may support to build full packet from
discontinuous buffers.
But considered that most of HCs don't support that, and driver
need work well or keep consistent on different HCs and ARCHs, we
have to introduce the limit.
Currently, only usbtest is reported to pass such sg buffers to HC,
and other users(mass storage, usbfs) don't have the problem.
We don't check it on USB wireless device, because:
- wireless devices can't be attached to common USB
bus(EHCI/UHCI/OHCI/...)
- the max packet size of endpoint may be odd, and often can't
devide 4KB which is a typical usage in usb mass storage application
Reported-by: Konstantin Filatov <kfilatov@parallels.com>
Reported-by: Denis V. Lunev <den@openvz.org>
Cc: Felipe Balbi <balbi@ti.com>
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Ming Lei <ming.lei@canonical.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The USB_MAXCHILDREN symbol is used in include/uapi/linux/usb/ch11.h, a
user-mode header, even though it is defined in include/linux/usb.h,
which is kernel-only. This causes compile-time errors when user
programs try to #include linux/usb/ch11.h.
This patch fixes the problem by moving the definition of USB_MAXCHILDREN
into ch11.h. It also gets rid of unneeded parentheses.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Adds abitilty to tune L1 timeout (inactivity timer for usb2 link sleep)
and BESL (best effort service latency)via sysfs.
This also adds a new usb2_lpm_parameters structure with those variables to
struct usb_device.
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
usb 2.0 devices with link power managment (LPM) can describe their idle link
timeouts either in BESL or HIRD format, so far xHCI has only supported HIRD but
later xHCI errata add BESL support as well
BESL timeouts need to inform exit latency changes with an evaluate
context command the same way USB 3.0 link PM code does.
The same xhci_change_max_exit_latency() function is used as with USB3
but code is pulled out from #ifdef CONFIG_PM as USB2.0 BESL LPM
funcionality does not depend on CONFIG_PM.
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
The current EHCI code sleeps a flat 110ms in the resume path if there
was a USB 1.1 device connected to its companion controller during
suspend, waiting for the device to reappear and reset so that it can be
handed back to the companion. This is necessary if the device uses
persist, so that the companion controller can actually see it during its
own resume path.
However, if the device doesn't use persist, this is entirely
unnecessary. We might just as well ignore it and have the normal device
detection/reset/handoff code handle it asynchronously when it eventually
shows up. As USB 1.1 devices are almost exclusively HIDs these days (for
which persist has no value), this can allow distros to shave another
tenth of a second off their resume time.
In order to enable this optimization, the patch also adds a new
usb_for_each_dev() iterator that is exported by the USB core and wraps
bus_for_each_dev() with the logic to differentiate between struct
usb_device and struct usb_interface on the usb_bus_type bus.
Signed-off-by: Julius Werner <jwerner@chromium.org>
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch (as1675) removes the CONFIG_USB_SUSPEND option, essentially
replacing it everywhere with CONFIG_PM_RUNTIME (except for one place
in hub.c, where it is replaced with CONFIG_PM because the code needs
to be used in both runtime and system PM). The net result is code
shrinkage and simplification.
There's very little point in keeping CONFIG_USB_SUSPEND because almost
everybody enables it. The few that don't will find that the usbcore
module has gotten somewhat bigger and they will have to take active
measures if they want to prevent hubs from being runtime suspended.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
CC: Peter Chen <peter.chen@freescale.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch adds comments on interface driver suspend callback
to emphasize that the failure return value is ignored by
USB core in system sleep context, so do not try to recover
device for this case and let resume/reset_resume callback
handle the suspend failure if needed.
Also kerneldoc for usb_suspend_both() is updated with the
fact.
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Ming Lei <ming.lei@canonical.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch (as1649) adds a mechanism for host controller drivers to
inform usbcore when they have begun or ended resume signalling on a
particular root-hub port. The core will then make sure that the root
hub does not get runtime-suspended while the port resume is going on.
Since commit 596d789a21 (USB: set hub's
default autosuspend delay as 0), the system tries to suspend hubs
whenever they aren't in use. While a root-hub port is being resumed,
the root hub does not appear to be in use. Attempted runtime suspends
fail because of the ongoing port resume, but the PM core just keeps on
trying over and over again. We want to prevent this wasteful effort.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Tested-by: Ming Lei <ming.lei@canonical.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Matching on device and interface class with with unspecified
subclass and protocol is sometimes useful. This is slightly
different from USB_DEVICE_AND_INTERFACE_INFO which requires
the full interface class/subclass/protocol triplet.
Signed-off-by: Bjørn Mork <bjorn@mork.no>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch (as1620) speeds up USB root-hub resumes in the common case
where every enabled port has its suspend feature set (which currently
will be true for every runtime resume of the root hub). If all the
enabled ports are suspended then resuming the root hub won't resume
any of the downstream devices. In this case there's no need for a
Resume Recovery delay, because that delay is meant to give devices a
chance to get ready for active use.
To keep track of the port suspend features, the patch adds a
"port_is_suspended" flag to struct usb_device. This has to be tracked
separately from the device's state; it's entirely possible for a USB-2
device to be suspended while the suspend feature on its parent port is
clear. The reason is that devices will go into suspend whenever their
parent hub does.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Reviewed-by: Peter Chen <peter.chen@freescale.com>
Tested-by: Peter Chen <peter.chen@freescale.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch (as1619) improves the interface to the "hub_for_each_child"
macro. The name clearly suggests that the macro iterates over child
devices; it does not suggest that the loop will also iterate over
unnconnected ports.
The patch changes the macro so that it will skip over unconnected
ports and iterate only the actual child devices. The two existing
call sites are updated to avoid testing for a NULL child pointer,
which is now unnecessary.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch (as1611) updates the USB documentation and kerneldoc to
give a more precise meaning for the URB_ISO_ASAP flag and to explain
more of the details of scheduling for isochronous URBs.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
