Linux kernel stable tree https://git.rulkc.org/pub/scm/linux/kernel/git/stable/linux-stable.git/atom?h=linux-3.16.y 2019-12-19T15:58:19+00:00 2019-12-19T15:58:19+00:00 Johan Hovold johan@kernel.org 2019-10-09T17:09:42+00:00 urn:sha1:90461e68ee390441684ef9d28feeaa5e34c3863c commit bed5ef230943863b9abf5eae226a20fad9a8ff71 upstream. The driver was using its struct usb_interface pointer as an inverted disconnected flag and was setting it to NULL before making sure all completion handlers had run. This could lead to NULL-pointer dereferences in the dev_err() statements in the completion handlers which relies on said pointer. Fix this by using a dedicated disconnected flag. Note that this is also addresses a NULL-pointer dereference at release() and a struct usb_interface reference leak introduced by a recent runtime PM fix, which depends on and should have been submitted together with this patch. Fixes: 4212cd74ca6f ("USB: usb-skeleton.c: remove err() usage") Fixes: 5c290a5e42c3 ("USB: usb-skeleton: fix runtime PM after driver unbind") Signed-off-by: Johan Hovold <johan@kernel.org> Link: https://lore.kernel.org/r/20191009170944.30057-2-johan@kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Ben Hutchings <ben@decadent.org.uk> 2019-12-19T15:58:08+00:00 Johan Hovold johan@kernel.org 2019-10-01T08:49:05+00:00 urn:sha1:9e3c26ddbcce4d4205b3cdbdfb74413c1e1f5a26 commit 5c290a5e42c3387e82de86965784d30e6c5270fd upstream. Since commit c2b71462d294 ("USB: core: Fix bug caused by duplicate interface PM usage counter") USB drivers must always balance their runtime PM gets and puts, including when the driver has already been unbound from the interface. Leaving the interface with a positive PM usage counter would prevent a later bound driver from suspending the device. Fixes: c2b71462d294 ("USB: core: Fix bug caused by duplicate interface PM usage counter") Signed-off-by: Johan Hovold <johan@kernel.org> Link: https://lore.kernel.org/r/20191001084908.2003-2-johan@kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Ben Hutchings <ben@decadent.org.uk> 2014-01-08T23:01:39+00:00 Paul Gortmaker paul.gortmaker@windriver.com 2014-01-08T16:08:26+00:00 urn:sha1:803a536243b3a1ed2289f41897b11b72bd083309 None of these files are actually using any __init type directives and hence don't need to include <linux/init.h>. Most are just a left over from __devinit and __cpuinit removal, or simply due to code getting copied from one driver to the next. Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> 2013-07-25T19:01:13+00:00 Chen Wang unicornxx.wang@gmail.com 2013-07-19T02:15:18+00:00 urn:sha1:140983c283b302167b387646218a97f112cc0c03 Updated skel_read() in usb-skeleton.c. When there is no data in the buffer, we would allow retry for both blocking and nonblocking cases. Original logic give retry only for blocking case. Actually we can also allow retry for nonblocking case. This will reuse the existing retry logic and handle the return of -EAGAIN in one place. Also if the data to be read is short and can be retrieved in quick time, we can also give a chance for nonblocking case and may catch the data and copy it back to userspace in one read() call too. Signed-off-by: Chen Wang <unicornxx.wang@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> 2013-03-25T20:32:20+00:00 Du Xing duxing2007@gmail.com 2013-03-20T12:47:46+00:00 urn:sha1:c79041a44045a40329d9ada3f8679c4b30c5b76b In skel_read,the reader blocked in wait_for_completion before submit bulk in urb. Using processed_urb is for retaining the completion in the case that previous interruptible wait in skel_read was interrupted and complete before next skel_read. Replacing completion with waitqueue can avoid working around the counting nature of completions and fix the bug. Signed-off-by: Du Xing duxing2007@gmail.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> 2012-10-24T21:40:50+00:00 Constantine Shulyupin const@MakeLinux.com 2012-10-10T14:14:00+00:00 urn:sha1:e8cebb9cde3716800219ea8473306d431e83154b Fixing compilaton error. Incrementing usage counter only on successful execution of skel_open. Removing redundant locking Some last changes in function skel_open and finally commit 52a7499 Revert "USB: usb-skeleton.c: fix open/disconnect race" introduced a bug in function skel_open, which this patch fixes. Changes since v2: - refactoring - Removing redundant mutex synchronization. Changes since v1: - Fixed accordingly feedback of Oliver Neukum oneukum@suse.de: also need to drop the lock. Signed-off-by: Constantine Shulyupin <const@MakeLinux.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> 2012-04-27T18:24:45+00:00 Greg Kroah-Hartman gregkh@linuxfoundation.org 2012-04-27T18:24:45+00:00 urn:sha1:4212cd74ca6f0c42f5a0065797dd1c310a485ad8 err() was a very old USB-specific macro that I thought had gone away. This patch removes it from being used in the driver and uses dev_err() instead. Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> 2012-01-24T20:02:38+00:00 Greg Kroah-Hartman gregkh@suse.de 2012-01-24T20:02:38+00:00 urn:sha1:52a749992ca6a0fd304609af40ed3bfd6cef4660 This reverts commit 26c71a79cade5ccad80e0752cd82f3518df48fb3. It's not needed, to quote Ming Lei: Looks you have queued the patch into your tree, but just now I find the patch is not needed at all, since we have had minor_rwsem(drivers/usb/core/file.c) for this purpose, please drop the patch, sorry for it. Cc: Ming Lei <tom.leiming@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> 2012-01-09T20:09:47+00:00 Linus Torvalds torvalds@linux-foundation.org 2012-01-09T20:09:47+00:00 urn:sha1:55b81e6f2795484ea8edf5805c95c007cacfa736 * 'usb-next' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb: (232 commits) USB: Add USB-ID for Multiplex RC serial adapter to cp210x.c xhci: Clean up 32-bit build warnings. USB: update documentation for usbmon usb: usb-storage doesn't support dynamic id currently, the patch disables the feature to fix an oops drivers/usb/class/cdc-acm.c: clear dangling pointer drivers/usb/dwc3/dwc3-pci.c: introduce missing kfree drivers/usb/host/isp1760-if.c: introduce missing kfree usb: option: add ZD Incorporated HSPA modem usb: ch9: fix up MaxStreams helper USB: usb-skeleton.c: cleanup open_count USB: usb-skeleton.c: fix open/disconnect race xhci: Properly handle COMP_2ND_BW_ERR USB: remove dead code from suspend/resume path USB: add quirk for another camera drivers: usb: wusbcore: Fix dependency for USB_WUSB xhci: Better debugging for critical host errors. xhci: Be less verbose during URB cancellation. xhci: Remove debugging about ring structure allocation. xhci: Remove debugging about toggling cycle bits. xhci: Remove debugging for individual transfers. ... 2012-01-04T23:52:41+00:00 Ming Lei tom.leiming@gmail.com 2011-12-16T14:20:44+00:00 urn:sha1:e28dbb0661ca49ada2d6a307a1ec93dd75d515e0 It is not necessary to use the 'open_count' for handling runtime pm only, because runtinme pm has built-in counter to handle this, so remove it to make code clean. Signed-off-by: Ming Lei <tom.leiming@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>