Commit Graph

107 Commits

Author SHA1 Message Date
Hans Petter Selasky
07dbde6777 Add a kernel function to delist our kernel character devices, so that
the device name can be re-used right away in case we are destroying
the character devices in the background.

MFC after:	4 days
Reported by:	dchagin@
2015-01-14 14:04:29 +00:00
Hans Petter Selasky
b78e84d132 Resolve a special case deadlock: When two or more threads are
simultaneously detaching kernel drivers on the same USB device we can
get stuck in the "usb_wait_pending_ref_locked()" function because the
conditions needed for allowing detach are not met. The "destroy_dev()"
function waits for all system calls involving the given character
device to return. Character device system calls may lock the USB
enumeration lock, which is also held when "destroy_dev()" is
called. This can sometimes lead to a deadlock not noticed by
WITNESS. The current solution is to ensure the calling thread is the
only one holding the USB enumeration lock and prevent other threads
from getting refs while a USB device detach is ongoing. This turned
out not to be sufficient. To solve this deadlock we could use
"destroy_dev_sched()" to schedule the device destruction in the
background, but then we don't know when it is safe to free() the
private data of the character device. Instead a callback function is
executed by the USB explore process to kill off any leftover USB
character devices synchronously after the USB device explore code is
finished and the USB enumeration lock is no longer locked. This makes
porting easier and also ensures us that character devices must
eventually go away after a USB device detach.

While at it ensure that "flag_iserror" is only written when "priv_mtx"
is locked, which is protecting it.

MFC after:	5 days
2015-01-13 16:37:43 +00:00
Hans Petter Selasky
c3ccd2bf8c Fix for deadlock in USB device side mode.
MFC after:	3 days
2014-08-05 06:38:21 +00:00
Hans Petter Selasky
af3b2549c4 Pull in r267961 and r267973 again. Fix for issues reported will follow. 2014-06-28 03:56:17 +00:00
Glen Barber
37a107a407 Revert r267961, r267973:
These changes prevent sysctl(8) from returning proper output,
such as:

 1) no output from sysctl(8)
 2) erroneously returning ENOMEM with tools like truss(1)
    or uname(1)
 truss: can not get etype: Cannot allocate memory
2014-06-27 22:05:21 +00:00
Hans Petter Selasky
3da1cf1e88 Extend the meaning of the CTLFLAG_TUN flag to automatically check if
there is an environment variable which shall initialize the SYSCTL
during early boot. This works for all SYSCTL types both statically and
dynamically created ones, except for the SYSCTL NODE type and SYSCTLs
which belong to VNETs. A new flag, CTLFLAG_NOFETCH, has been added to
be used in the case a tunable sysctl has a custom initialisation
function allowing the sysctl to still be marked as a tunable. The
kernel SYSCTL API is mostly the same, with a few exceptions for some
special operations like iterating childrens of a static/extern SYSCTL
node. This operation should probably be made into a factored out
common macro, hence some device drivers use this. The reason for
changing the SYSCTL API was the need for a SYSCTL parent OID pointer
and not only the SYSCTL parent OID list pointer in order to quickly
generate the sysctl path. The motivation behind this patch is to avoid
parameter loading cludges inside the OFED driver subsystem. Instead of
adding special code to the OFED driver subsystem to post-load tunables
into dynamically created sysctls, we generalize this in the kernel.

Other changes:
- Corrected a possibly incorrect sysctl name from "hw.cbb.intr_mask"
to "hw.pcic.intr_mask".
- Removed redundant TUNABLE statements throughout the kernel.
- Some minor code rewrites in connection to removing not needed
TUNABLE statements.
- Added a missing SYSCTL_DECL().
- Wrapped two very long lines.
- Avoid malloc()/free() inside sysctl string handling, in case it is
called to initialize a sysctl from a tunable, hence malloc()/free() is
not ready when sysctls from the sysctl dataset are registered.
- Bumped FreeBSD version to indicate SYSCTL API change.

MFC after:	2 weeks
Sponsored by:	Mellanox Technologies
2014-06-27 16:33:43 +00:00
Marius Strobl
cabe79d9d8 Avoid the USB device disconnected and controller shutdown clutter on system
shutdown by putting the former under !rebooting and turning the latter into
debug messages.

Reviewed by:	hps
MFC after:	1 week
Sponsored by:	Bally Wulff Games & Entertainment GmbH
2014-06-10 08:20:00 +00:00
Hans Petter Selasky
d008478e28 Try to resolve a possible deadlock when detaching USB devices which
create character devices. The deadlock can happen if an application is
issuing IOCTLs which require USB refcounting, at the same time the USB
device is detaching.

There is already a counter in place in the USB device structure to
detect this situation, but it was not always checked ahead of invoking
functions that might destroy character devices, like detach, set
configuration, set alternate interface or detach active kernel driver.

Reported by:	Daniel O'Connor <doconnor@gsoft.com.au>
MFC after:	1 week
2014-03-20 13:53:24 +00:00
Hans Petter Selasky
f97da128ab Fix a possible memory use after free and leak situation associated
with USB device detach when using character device handles. This also
includes LibUSB. It turns out that "usb_close()" cannot always get a
reference to clean up its USB transfers and such, if called during the
kernel USB device detach.

Analysis by:	hselasky @
Reported by:	Juergen Lock <nox@jelal.kn-bremen.de>
MFC after:	1 week
2014-01-17 10:35:18 +00:00
Hans Petter Selasky
a0d53e0b38 Implement better error recovery for Transaction Translators, TTs,
found in High Speed USB HUBs which translate from High Speed USB into
FULL or LOW speed USB. In some rare cases SPLIT transactions might get
lost, which might leave the TT in an unknown state. Whenever we detect
such an error try to issue either a clear TT buffer request, or if
that is not possible reset the whole TT.

MFC after:	1 week
2014-01-13 15:21:11 +00:00
Hans Petter Selasky
e892b3fe36 USB method structures for USB controllers and USB pipes should be
constant and does not need to be modified. This also saves a small
amount of RAM.
2013-12-11 13:20:32 +00:00
Hans Petter Selasky
273c749c93 Fix some statical clang analyzer warnings. 2013-05-25 17:09:58 +00:00
Hans Petter Selasky
6c21893360 Fix compile warning. 2013-05-03 13:33:36 +00:00
Hans Petter Selasky
2c79a775ed - Add more defines to limit USB memory usage and number of allocations
in reduced memory systems.

- Split allocation and freeing of the configuration descriptor into a separate
function, so that the configuration descriptor can be made fixed size
to save memory allocations. This applies for both device and host mode.
2013-05-03 11:10:04 +00:00
Hans Petter Selasky
5b0752bbc1 Add some defines to limit USB memory usage in reduced memory systems. 2013-05-03 09:23:06 +00:00
Hans Petter Selasky
fa49bce566 Allow the default USB template to be specified at compile time. 2013-05-03 08:19:09 +00:00
Hans Petter Selasky
15d4edd3d4 Add missing ifdef's for reduced feature compilations. 2013-04-03 10:31:13 +00:00
Hans Petter Selasky
23de050b72 - Make quirk for reading device descriptor from broken USB devices.
Else they won't enumerate at all:
hw.usb.full_ddesc=1
- Reduce the USB descriptor read timeout from 1000ms to
500ms. Typical value for LOW speed devices is 50-100ms.
- Enumerate USB device a maximum of 3 times when a port
connection change event is detected, before giving up.

MFC after:	1 month
2013-03-13 15:38:01 +00:00
Hans Petter Selasky
a18a7a414a Resolve a LOR after r246616. Protect control requests using the USB device
enumeration lock. Make sure all callers of usbd_enum_lock() check the return
value. Remove the control transfer specific lock. Bump the FreeBSD version
number, hence external USB modules may need to be recompiled due to a USB
device structure change.

MFC after:	1 week
2013-02-13 12:35:17 +00:00
Hans Petter Selasky
6950c75f40 - Move scratch data from the USB bus structure to the USB device structure
so that simultaneous access cannot happen. Protect scratch area using
the enumeration lock. Also reduce stack usage in usbd_transfer_setup()
by moving some big stack members to the scratch area. This saves around
200 bytes of stack.
- Fix a whitespace.

MFC after:	1 week
2013-02-10 10:56:13 +00:00
Hans Petter Selasky
9b3a48ee6f Add defines to more easily allow a single threaded version of the FreeBSD
USB stack. This is useful for non-kernel purposes, like the loader.
2013-02-05 14:44:25 +00:00
Hans Petter Selasky
d2b99310b1 Modify the FreeBSD USB kernel code so that it can be compiled directly
into the FreeBSD boot loader, typically for non-USB aware BIOSes, EFI systems
or embedded platforms. This is also useful for out of the system compilation
of the FreeBSD USB stack for various purposes. The USB kernel files can
now optionally include a global header file which should include all needed
definitions required to compile the FreeBSD USB stack. When the global USB
header file is included, no other USB header files will be included by
default.

Add new file containing the USB stack configuration for the
FreeBSD loader build.

Replace some __FBSDID()'s by /* $FreeBSD$ */ comments. Now all
USB files follow the same style.

Use cases:
 - console in loader via USB
 - loading kernel via USB

Discussed with:		Hiroki Sato, hrs @ EuroBSDCon
2013-01-30 15:26:04 +00:00
Hans Petter Selasky
83cadd7dcc Add missing CTLFLAG_TUN flag to tunable sysctls in USB stack.
Rearrange the tunables and belonging sysctl declarations, so
that they are next to eachother.

Submitted by:	n_hibma @
MFC after:	1 week
2012-10-26 06:04:47 +00:00
Hans Petter Selasky
f61fc4bd7d Fix for missing locks due to recent change.
PR:		usb/170606
MFC after:	2 weeks
2012-08-13 18:34:04 +00:00
Hans Petter Selasky
a5cf1aaaff Add support for the so-called streams feature of BULK endpoints
in SUPER-speed mode, USB 3.0.

This feature has not been tested yet, due to lack of hardware.

This feature is useful when implementing protocols like UASP,
USB attached SCSI which promises higher USB mass storage throughput.

This patch also implements support for hardware processing of endpoints
for increased performance. The switching to hardware processing
of an endpoint is done via a callback to the USB controller driver. The
stream feature is implemented like a variant of a hardware USB protocol.

USB controller drivers implementing device mode needs to be updated to
implement the new "xfer_stall" USB controller method and remove the
"xfer" argument from the "set_stall" method.

The API's toward existing USB drivers are preserved. To setup a USB transfer
in stream mode, set the "stream_id" field of the USB config structure to
the desired value.

The maximum number of BULK streams is currently hardcoded and limited to 8
via a define in usb_freebsd.h.

All USB drivers should be re-compiled after this change.

LibUSB will be updated next week to support streams mode. A new IOCTL to
setup BULK streams as already been implemented. The ugen device nodes
currently only supports stream ID zero.

The FreeBSD version has been bumped.

MFC after:	2 weeks
2012-08-12 17:53:06 +00:00
Hans Petter Selasky
2203949434 Style.
MFC after:	2 weeks
2012-08-10 14:51:41 +00:00
Hans Petter Selasky
d9073c1e6a Improve support for detaching kernel drivers on a per interface basis.
MFC after:	1 week
2012-06-01 16:30:54 +00:00
Hans Petter Selasky
6d917491f5 Fix compiler warnings, mostly signed issues,
when USB modules are compiled with WARNS=9.

MFC after:	1 weeks
2012-04-02 10:50:42 +00:00
Hans Petter Selasky
9157ad4b17 Improve USB mass storage quirk auto detection.
MFC after:	3 days
2011-10-29 12:32:13 +00:00
Hans Petter Selasky
3f37fb622c Refactor auto-quirk solution so that we break as few external
drivers as possible.

PR:		usb/160299
Approved by:	re (kib)
Suggested by:	rwatson
MFC after:	0 days
2011-09-10 15:55:36 +00:00
Hans Petter Selasky
d46dc4ad74 This patch adds automatic detection of USB mass storage devices
which does not support the no synchronize cache SCSI command.

The __FreeBSD_version version macro has been bumped and
external kernel modules needs to be recompiled after
this patch.

Approved by:    re (kib)
MFC after:      1 week
PR:		usb/160299
2011-09-02 18:50:44 +00:00
Hans Petter Selasky
2ffd5fdcc4 Use synchronous device destruction instead of asynchronous, so that a new
device having the same name like a previous one is not created before the old
one is gone. This fixes some panics due to asserts in the devfs code which
were added recently.

Approved by:    re (kib)
MFC after:      1 week
2011-08-11 11:30:21 +00:00
Hans Petter Selasky
d43ffe9465 - Ensure that we get all the required nomatch devd events.
MFC after:	3 days
2011-06-24 19:02:56 +00:00
Hans Petter Selasky
a6b6015063 - Move execution of event handlers into the probe and attach function so that
dynamically loaded device drivers get a chance to run their event hooks.

- Decouple the USB suspend and resume lock from witness. It produces some
false warnings due to reusing the lock name among multiple devices.

MFC after:	3 days
2011-06-24 18:14:43 +00:00
Andriy Gapon
cd10bffa61 usb: change to one-pass probing of device drivers
This brings USB bus more in line with how newbus is supposed to be used.
Also, because of the two-pass probing the following message was produced
by devd in default configuration when almost any USB device was
connected:
	Unknown USB device: vendor <> product <> bus <>
This should be fixed now.

Note that many USB device drivers pass some information from probe
method to attach method via ivars.  For this to continue working we rely
on the fact that the subr_bus code calls probe method of a winning driver
again before calling its attach method in the case where multiple
drivers claim to support a device.  This is done because device
description is set in successful probe methods and we want to get a correct
device description from a winning driver.  So now this logic is re-used
for setting ivars too.

Reviewed by:	hselasky
MFC after:	1 month
2011-05-18 07:40:12 +00:00
Hans Petter Selasky
afa524e6cf Cleanup usb_notify_addq_compat(). It should not
be needed any more.

MFC after:	7 days
2011-05-08 08:22:11 +00:00
John Baldwin
58ccf5b41c Remove unneeded includes of <sys/linker_set.h>. Other headers that use
it internally contain nested includes.

Reviewed by:	bde
2011-01-11 13:59:06 +00:00
Hans Petter Selasky
02c4024199 We need to define a cdev variable associated with each USB device,
hence existing applications like webcamd are expecting that.
This problem was introduced by SVN change 214221 where cdev=
was replaced by ugen= by accident. Solve this problem by
redefining cdev= in devd notifications.

MFC after 3 days.

Approved by:	thompsa (mentor)
2010-11-30 08:25:57 +00:00
Nick Hibma
0834ed4757 Bugfix: Move the 'at <location string' to the beginning of the attach
notification. devd would stop evaluating at 'at' (not '<k>=<v>') and
hence prevent 'port=X' (and 'bus=<"on" string>) from making it into the
environment for the devd action.

Reviewed by:	hselasky
MFC after:	2 weeks
2010-11-05 08:30:16 +00:00
Nick Hibma
6df3eebb63 Don't terminate the notification with \n. This is done in
usb_device.c:devctl_notify_f().
2010-11-04 21:06:36 +00:00
Hans Petter Selasky
8427ed847d Add support for setting per-interface PnP information.
Submitted by:	Nick Hibma
Approved by:	thompsa (mentor)
2010-10-27 17:38:05 +00:00
Hans Petter Selasky
8bb77249db Add possibility to generate devctl notifications regardless of UGEN presence.
Submitted by:  Nick Hibma
Approved by:    thompsa (mentor)
2010-10-22 20:13:45 +00:00
Hans Petter Selasky
963169b4af This commit adds full support for USB 3.0 devices in host and device
mode in the USB core.  The patch mostly consists of updating the USB
HUB code to support USB 3.0 HUBs. This patch also add some more USB
controller methods to support more active-alike USB controllers like
the XHCI which needs to be informed about various device state events.

USB 3.0 HUBs are not tested yet, due to lack of hardware, but are
believed to work.

After this update the initial device descriptor is only read twice
when we know that the bMaxPacketSize is too small for a single packet
transfer of this descriptor.

Approved by:    thompsa (mentor)
2010-10-04 23:18:05 +00:00
Andrew Thompson
ae538d8533 Reduce the need to accesss struct usb_device by providing functions to access
the product, manufacturer and serial strings.

Submitted by:	Hans Petter Selasky
2010-09-02 04:39:45 +00:00
Andrew Thompson
d2d71ce7a8 Add support for power mode filtering as some USB hardware does not support
power saving.

Submitted by:	Hans Petter Selasky
2010-09-02 04:05:00 +00:00
Andriy Gapon
1bdfff2252 fix a few cases where a string is passed via format argument instead of
via %s

Most of the cases looked harmless, but this is done for the sake of
correctness.  In one case it even allowed to drop an intermediate buffer.

Found by:	clang
MFC after:	2 week
2010-06-11 19:27:21 +00:00
Andrew Thompson
2df1e9a62a If a USB device is suspended and a USB set config request is issued when the
USB enumeration lock is locked, then the USB stack fails to resume the device
because locking the USB enumeration lock is part of the resume procedure. To
solve this issue a new lock is introduced which only protects the suspend and
resume callbacks, which can be dropped inside the usbd_do_request_flags()
function, to allow suspend and resume during so-called enumeration operations.

Submitted by:	Hans Petter Selasky
2010-05-12 22:42:35 +00:00
Andrew Thompson
5b3bb704f7 Use a more obvious prefix for the USB control (endpoint 0) transfers rather
than default_*.
2010-04-22 22:15:08 +00:00
Andrew Thompson
91cd92400f Properly name the sxlocks, mutexes and condvars. 2010-04-22 22:00:16 +00:00
Andrew Thompson
0d92ac2a92 Use SX_DUPOK rather than making the string unique. 2010-04-22 21:41:50 +00:00