- 12 Oct, 2017 29 commits
-
-
Greg Kroah-Hartman authored
commit bd7a3fe7 upstream. Andrey Konovalov reported a possible out-of-bounds problem for a USB interface association descriptor. He writes: It seems there's no proper size check of a USB_DT_INTERFACE_ASSOCIATION descriptor. It's only checked that the size is >= 2 in usb_parse_configuration(), so find_iad() might do out-of-bounds access to intf_assoc->bInterfaceCount. And he's right, we don't check for crazy descriptors of this type very well, so resolve this problem. Yet another issue found by syzkaller... Reported-by:
Andrey Konovalov <andreyknvl@google.com> Tested-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Dmitry Fleytman authored
commit b2a542bb upstream. Commit e0429362 ("usb: Add device quirk for Logitech HD Pro Webcams C920 and C930e") introduced quirk to workaround an issue with some Logitech webcams. The workaround is introducing delay for some USB operations. According to our testing, delay introduced by original commit is not long enough and in rare cases we still see issues described by the aforementioned commit. This patch increases delays introduced by original commit. Having this patch applied we do not see those problems anymore. Signed-off-by:
Dmitry Fleytman <dmitry@daynix.com> Signed-off-by:
-
Greg Kroah-Hartman authored
commit 2e1c4239 upstream. Andrey Konovalov reported a possible out-of-bounds problem for the cdc_parse_cdc_header function. He writes: It looks like cdc_parse_cdc_header() doesn't validate buflen before accessing buffer[1], buffer[2] and so on. The only check present is while (buflen > 0). So fix this issue up by properly validating the buffer length matches what the descriptor says it is. Reported-by:
-
Alan Stern authored
commit 786de92b upstream. The uas driver has a subtle bug in the way it handles alternate settings. The uas_find_uas_alt_setting() routine returns an altsetting value (the bAlternateSetting number in the descriptor), but uas_use_uas_driver() then treats that value as an index to the intf->altsetting array, which it isn't. Normally this doesn't cause any problems because the various alternate settings have bAlternateSetting values 0, 1, 2, ..., so the value is equal to the index in the array. But this is not guaranteed, and Andrey Konovalov used the syzkaller fuzzer with KASAN to get a slab-out-of-bounds error by violating this assumption. This patch fixes the bug by making uas_find_uas_alt_setting() return a pointer to the altsetting entry rather than either the value or the index. Pointers are less subject to misinterpretation. Signed-off-by:
Alan Stern <stern@rowland.harvard.edu> Reported-by:
-
Martin K. Petersen authored
commit 77082ca5 upstream. A user may lower the max_sectors_kb setting in sysfs to accommodate certain workloads. Previously we would always set the max I/O size to either the block layer default or the optional preferred I/O size reported by the device. Keep the current heuristics for the initial setting of max_sectors_kb. For subsequent invocations, only update the current queue limit if it exceeds the capabilities of the hardware. Reported-by:
Don Brace <don.brace@microsemi.com> Reviewed-by:
Martin Wilck <mwilck@suse.com> Tested-by:
Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by:
-
Luca Coelho authored
commit 01a9c948 upstream. The OTP in some SKUs have erroneously allowed 40MHz and 80MHz channels in the 5.2GHz band. The firmware has been modified to not allow this in those SKUs, so the driver needs to do the same otherwise the firmware will assert when we try to use it. Signed-off-by:
Luca Coelho <luciano.coelho@intel.com> Signed-off-by:
-
Adrian Salido authored
commit 8320caee upstream. The buffer allocation is not currently accounting for an extra byte for the report id. This can cause an out of bounds access in function i2c_hid_set_or_send_report() with reportID > 15. Signed-off-by:
Adrian Salido <salidoa@google.com> Reviewed-by:
Benson Leung <bleung@chromium.org> Signed-off-by:
Guenter Roeck <groeck@chromium.org> Signed-off-by:
Dmitry Torokhov <dmitry.torokhov@gmail.com> Signed-off-by:
Jiri Kosina <jkosina@suse.cz> Signed-off-by:
-
Shu Wang authored
commit 2b0b8499 upstream. The trampoline allocated by function tracer was overwriten by function_graph tracer, and caused a memory leak. The save_global_trampoline should have saved the previous trampoline in register_ftrace_graph() and restored it in unregister_ftrace_graph(). But as it is implemented, save_global_trampoline was only used in unregister_ftrace_graph as default value 0, and it overwrote the previous trampoline's value. Causing the previous allocated trampoline to be lost. kmmeleak backtrace: kmemleak_vmalloc+0x77/0xc0 __vmalloc_node_range+0x1b5/0x2c0 module_alloc+0x7c/0xd0 arch_ftrace_update_trampoline+0xb5/0x290 ftrace_startup+0x78/0x210 register_ftrace_function+0x8b/0xd0 function_trace_init+0x4f/0x80 tracing_set_tracer+0xe6/0x170 tracing_set_trace_write+0x90/0xd0 __vfs_write+0x37/0x170 vfs_write+0xb2/0x1b0 SyS_write+0x55/0xc0 do_syscall_64+0x67/0x180 return_from_SYSCALL_64+0x0/0x6a [ Looking further into this, I found that this was left over from when the function and function graph tracers shared the same ftrace_ops. But in commit 5f151b24 ("ftrace: Fix function_profiler and function tracer together"), the two were separated, and the save_global_trampoline no longer was necessary (and it may have been broken back then too). -- Steven Rostedt ] Link: http://lkml.kernel.org/r/20170912021454.5976-1-shuwang@redhat.com Fixes: 5f151b24 ("ftrace: Fix function_profiler and function tracer together") Signed-off-by:
Shu Wang <shuwang@redhat.com> Signed-off-by:
Steven Rostedt (VMware) <rostedt@goodmis.org> Signed-off-by:
-
Alexander Shishkin authored
commit fd085bb1 upstream. For reasons unknown, the stm_source removal path uses device_destroy() to kill the underlying device object. Because device_destroy() uses devt to look for the device to destroy and the fact that stm_source devices don't have one (or all have the same one), it just picks the first device in the class, which may well be the wrong one. That is, loading stm_console and stm_heartbeat and then removing both will die in dereferencing a freed object. Since this should have been device_unregister() in the first place, use it instead of device_destroy(). Signed-off-by:
Alexander Shishkin <alexander.shishkin@linux.intel.com> Fixes: 7bd1d409 ("stm class: Introduce an abstraction for System Trace Module devices") Signed-off-by:
-
Olaf Hering authored
commit 549e658a upstream. Till recently the expected length of bytes read by the daemon did depend on the context. It was either hv_start_fcopy or hv_do_fcopy. The daemon had a buffer size of two pages, which was much larger than needed. Now the expected length of bytes read by the daemon changed slightly. For START_FILE_COPY it is still the size of hv_start_fcopy. But for WRITE_TO_FILE and the other operations it is as large as the buffer that arrived via vmbus. In case of WRITE_TO_FILE that is slightly larger than a struct hv_do_fcopy. Since the buffer in the daemon was still larger everything was fine. Currently, the daemon reads only what is actually needed. The new buffer layout is as large as a struct hv_do_fcopy, for the WRITE_TO_FILE operation. Since the kernel expects a slightly larger size, hvt_op_read will return -EINVAL because the daemon will read slightly less than expected. Address this by restoring the expected buffer size in case of WRITE_TO_FILE. Fixes: 'c7e490fc ("Drivers: hv: fcopy: convert to hv_utils_transport")' Fixes: '3f2baa8a ("Tools: hv: update buffer handling in hv_fcopy_daemon")' Signed-off-by:
Olaf Hering <olaf@aepfle.de> Signed-off-by:
K. Y. Srinivasan <kys@microsoft.com> Signed-off-by:
-
Nicolai Stange authored
commit bf563b01 upstream. When printing the driver_override parameter when it is 4095 and 4094 bytes long, the printing code would access invalid memory because we need count+1 bytes for printing. Reject driver_override values of these lengths in driver_override_store(). This is in close analogy to commit 4efe874a ("PCI: Don't read past the end of sysfs "driver_override" buffer") from Sasha Levin. Fixes: 3d713e0e ("driver core: platform: add device binding path 'driver_override'") Signed-off-by:
Nicolai Stange <nstange@suse.de> Signed-off-by:
-
Takashi Iwai authored
commit 7682e399 upstream. The usx2y driver allocates the stream read/write buffers in continuous pages depending on the stream setup, and this may spew the kernel warning messages with a stack trace like: WARNING: CPU: 1 PID: 1846 at mm/page_alloc.c:3883 __alloc_pages_slowpath+0x1ef2/0x2d70 Modules linked in: CPU: 1 PID: 1846 Comm: kworker/1:2 Not tainted .... It may confuse user as if it were any serious error, although this is no fatal error and the driver handles the error case gracefully. Since the driver has already some sanity check of the given size (128 and 256 pages), it can't pass any crazy value. So it's merely page fragmentation. This patch adds __GFP_NOWARN to each caller for suppressing such kernel warnings. The original issue was spotted by syzkaller. Reported-by:
Takashi Iwai <tiwai@suse.de> Signed-off-by:
-
Guneshwor Singh authored
commit a931b9ce upstream. Commit 04c5d5a4 ("ALSA: compress: Embed struct device") removed the statement that used 'str' but didn't remove the variable itself. So remove it. [Adding stable to Cc since pr_debug() may refer to the uninitialized buffer -- tiwai] Fixes: 04c5d5a4 ("ALSA: compress: Embed struct device") Signed-off-by:
Guneshwor Singh <guneshwor.o.singh@intel.com> Signed-off-by:
-
Casey Schaufler authored
commit 57e7ba04 upstream. security_inode_getsecurity() provides the text string value of a security attribute. It does not provide a "secctx". The code in xattr_getsecurity() that calls security_inode_getsecurity() and then calls security_release_secctx() happened to work because SElinux and Smack treat the attribute and the secctx the same way. It fails for cap_inode_getsecurity(), because that module has no secctx that ever needs releasing. It turns out that Smack is the one that's doing things wrong by not allocating memory when instructed to do so by the "alloc" parameter. The fix is simple enough. Change the security_release_secctx() to kfree() because it isn't a secctx being returned by security_inode_getsecurity(). Change Smack to allocate the string when told to do so. Note: this also fixes memory leaks for LSMs which implement inode_getsecurity but not release_secctx, such as capabilities. Signed-off-by:
Casey Schaufler <casey@schaufler-ca.com> Reported-by:
Konstantin Khlebnikov <khlebnikov@yandex-team.ru> Signed-off-by:
James Morris <james.l.morris@oracle.com> Signed-off-by:
-
Alan Stern authored
commit 1fbbb78f upstream. As a holdover from the old g_file_storage gadget, the g_mass_storage legacy gadget driver attempts to unregister itself when its main operating thread terminates (if it hasn't been unregistered already). This is not strictly necessary; it was never more than an attempt to have the gadget fail cleanly if something went wrong and the main thread was killed. However, now that the UDC core manages gadget drivers independently of UDC drivers, this scheme doesn't work any more. A simple test: modprobe dummy-hcd modprobe g-mass-storage file=... rmmod dummy-hcd ends up in a deadlock with the following backtrace: sysrq: SysRq : Show Blocked State task PC stack pid father file-storage D 0 1130 2 0x00000000 Call Trace: __schedule+0x53e/0x58c schedule+0x6e/0x77 schedule_preempt_disabled+0xd/0xf __mutex_lock.isra.1+0x129/0x224 ? _raw_spin_unlock_irqrestore+0x12/0x14 __mutex_lock_slowpath+0x12/0x14 mutex_lock+0x28/0x2b usb_gadget_unregister_driver+0x29/0x9b [udc_core] usb_composite_unregister+0x10/0x12 [libcomposite] msg_cleanup+0x1d/0x20 [g_mass_storage] msg_thread_exits+0xd/0xdd7 [g_mass_storage] fsg_main_thread+0x1395/0x13d6 [usb_f_mass_storage] ? __schedule+0x573/0x58c kthread+0xd9/0xdb ? do_set_interface+0x25c/0x25c [usb_f_mass_storage] ? init_completion+0x1e/0x1e ret_from_fork+0x19/0x24 rmmod D 0 1155 683 0x00000000 Call Trace: __schedule+0x53e/0x58c schedule+0x6e/0x77 schedule_timeout+0x26/0xbc ? __schedule+0x573/0x58c do_wait_for_common+0xb3/0x128 ? usleep_range+0x81/0x81 ? wake_up_q+0x3f/0x3f wait_for_common+0x2e/0x45 wait_for_completion+0x17/0x19 fsg_common_put+0x34/0x81 [usb_f_mass_storage] fsg_free_inst+0x13/0x1e [usb_f_mass_storage] usb_put_function_instance+0x1a/0x25 [libcomposite] msg_unbind+0x2a/0x42 [g_mass_storage] __composite_unbind+0x4a/0x6f [libcomposite] composite_unbind+0x12/0x14 [libcomposite] usb_gadget_remove_driver+0x4f/0x77 [udc_core] usb_del_gadget_udc+0x52/0xcc [udc_core] dummy_udc_remove+0x27/0x2c [dummy_hcd] platform_drv_remove+0x1d/0x31 device_release_driver_internal+0xe9/0x16d device_release_driver+0x11/0x13 bus_remove_device+0xd2/0xe2 device_del+0x19f/0x221 ? selinux_capable+0x22/0x27 platform_device_del+0x21/0x63 platform_device_unregister+0x10/0x1a cleanup+0x20/0x817 [dummy_hcd] SyS_delete_module+0x10c/0x197 ? ____fput+0xd/0xf ? task_work_run+0x55/0x62 ? prepare_exit_to_usermode+0x65/0x75 do_fast_syscall_32+0x86/0xc3 entry_SYSENTER_32+0x4e/0x7c What happens is that removing the dummy-hcd driver causes the UDC core to unbind the gadget driver, which it does while holding the udc_lock mutex. The unbind routine in g_mass_storage tells the main thread to exit and waits for it to terminate. But as mentioned above, when the main thread exits it tries to unregister the mass-storage function driver. Via the composite framework this ends up calling usb_gadget_unregister_driver(), which tries to acquire the udc_lock mutex. The result is deadlock. The simplest way to fix the problem is not to be so clever: The main thread doesn't have to unregister the function driver. The side effects won't be so terrible; if the gadget is still attached to a USB host when the main thread is killed, it will appear to the host as though the gadget's firmware has crashed -- a reasonably accurate interpretation, and an all-too-common occurrence for USB mass-storage devices. In fact, the code to unregister the driver when the main thread exits is specific to g-mass-storage; it is not used when f-mass-storage is included as a function in a larger composite device. Therefore the entire mechanism responsible for this (the fsg_operations structure with its ->thread_exits method, the fsg_common_set_ops() routine, and the msg_thread_exits() callback routine) can all be eliminated. Even the msg_registered bitflag can be removed, because now the driver is unregistered in only one place rather than in two places. Signed-off-by:
Felipe Balbi <felipe.balbi@linux.intel.com> Acked-by:
Michal Nazarewicz <mina86@mina86.com> Signed-off-by:
-
Li Jun authored
commit 8e55d303 upstream. If there is no UDC available, the msg register will fail and this flag will not be set, but the driver is already added into pending driver list, then the module removal modprobe -r can not remove the driver from the pending list. Signed-off-by:
Li Jun <jun.li@nxp.com> Signed-off-by:
-
Dan Carpenter authored
commit fa1ed74e upstream. The user buffer has "uurb->buffer_length" bytes. If the kernel has more information than that, we should truncate it instead of writing past the end of the user's buffer. I added a WARN_ONCE() to help the user debug the issue. Reported-by:
Dan Carpenter <dan.carpenter@oracle.com> Acked-by:
-
Alan Stern authored
commit 7dbd8f4c upstream. A recent change to the synchronization in dummy-hcd was incorrect. The issue was that dummy_udc_stop() contained no locking and therefore could race with various gadget driver callbacks, and the fix was to add locking and issue the callbacks with the private spinlock held. UDC drivers aren't supposed to do this. Gadget driver callback routines are allowed to invoke functions in the UDC driver, and these functions will generally try to acquire the private spinlock. This would deadlock the driver. The correct solution is to drop the spinlock before issuing callbacks, and avoid races by emulating the synchronize_irq() call that all real UDC drivers must perform in their ->udc_stop() routines after disabling interrupts. This involves adding a flag to dummy-hcd's private structure to keep track of whether interrupts are supposed to be enabled, and adding a counter to keep track of ongoing callbacks so that dummy_udc_stop() can wait for them all to finish. A real UDC driver won't receive disconnect, reset, suspend, resume, or setup events once it has disabled interrupts. dummy-hcd will receive them but won't try to issue any gadget driver callbacks, which should be just as good. Signed-off-by:
-
Alan Stern authored
commit 0173a68b upstream. The dummy-hcd HCD/UDC emulator tries not to do too much work during each timer interrupt. But it doesn't try very hard; currently all it does is limit the total amount of bulk data transferred. Other transfer types aren't limited, and URBs that transfer no data (because of an error, perhaps) don't count toward the limit, even though on a real USB bus they would consume at least a minimum overhead. This means it's possible to get the driver stuck in an infinite loop, for example, if the host class driver resubmits an URB every time it completes (which is common for interrupt URBs). Each time the URB is resubmitted it gets added to the end of the pending-URBs list, and dummy-hcd doesn't stop until that list is empty. Andrey Konovalov was able to trigger this failure mode using the syzkaller fuzzer. This patch fixes the infinite-loop problem by restricting the URBs handled during each timer interrupt to those that were already on the pending list when the interrupt routine started. Newly added URBs won't be processed until the next timer interrupt. The problem of properly accounting for non-bulk bandwidth (as well as packet and transaction overhead) is not addressed here. Signed-off-by:
-
Alan Stern authored
commit fe659bcc upstream. The dummy-hcd UDC driver is not careful about the way it handles connection speeds. It ignores the module parameter that is supposed to govern the maximum connection speed and it doesn't set the HCD flags properly for the case where it ends up running at full speed. The result is that in many cases, gadget enumeration over dummy-hcd fails because the bMaxPacketSize byte in the device descriptor is set incorrectly. For example, the default settings call for a high-speed connection, but the maxpacket value for ep0 ends up being set for a Super-Speed connection. This patch fixes the problem by initializing the gadget's max_speed and the HCD flags correctly. Signed-off-by:
-
Jim Dickerson authored
commit 114ec3a6 upstream. Servers were emitting failed handoff messages but were not waiting the full 1 second as designated in section 4.22.1 of the eXtensible Host Controller Interface specifications. The handshake was using wrong units so calls were made with milliseconds not microseconds. Comments referenced 5 seconds not 1 second as in specs. The wrong units were also corrected in a second handshake call. Signed-off-by:
Jim Dickerson <jim.dickerson@hpe.com> Signed-off-by:
Mathias Nyman <mathias.nyman@linux.intel.com> Signed-off-by:
-
Takashi Iwai authored
commit bfc81a8b upstream. When a USB-audio device receives a maliciously adjusted or corrupted buffer descriptor, the USB-audio driver may access an out-of-bounce value at its parser. This was detected by syzkaller, something like: BUG: KASAN: slab-out-of-bounds in usb_audio_probe+0x27b2/0x2ab0 Read of size 1 at addr ffff88006b83a9e8 by task kworker/0:1/24 CPU: 0 PID: 24 Comm: kworker/0:1 Not tainted 4.14.0-rc1-42251-gebb2c243 #224 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Bochs 01/01/2011 Workqueue: usb_hub_wq hub_event Call Trace: __dump_stack lib/dump_stack.c:16 dump_stack+0x292/0x395 lib/dump_stack.c:52 print_address_description+0x78/0x280 mm/kasan/report.c:252 kasan_report_error mm/kasan/report.c:351 kasan_report+0x22f/0x340 mm/kasan/report.c:409 __asan_report_load1_noabort+0x19/0x20 mm/kasan/report.c:427 snd_usb_create_streams sound/usb/card.c:248 usb_audio_probe+0x27b2/0x2ab0 sound/usb/card.c:605 usb_probe_interface+0x35d/0x8e0 drivers/usb/core/driver.c:361 really_probe drivers/base/dd.c:413 driver_probe_device+0x610/0xa00 drivers/base/dd.c:557 __device_attach_driver+0x230/0x290 drivers/base/dd.c:653 bus_for_each_drv+0x161/0x210 drivers/base/bus.c:463 __device_attach+0x26e/0x3d0 drivers/base/dd.c:710 device_initial_probe+0x1f/0x30 drivers/base/dd.c:757 bus_probe_device+0x1eb/0x290 drivers/base/bus.c:523 device_add+0xd0b/0x1660 drivers/base/core.c:1835 usb_set_configuration+0x104e/0x1870 drivers/usb/core/message.c:1932 generic_probe+0x73/0xe0 drivers/usb/core/generic.c:174 usb_probe_device+0xaf/0xe0 drivers/usb/core/driver.c:266 really_probe drivers/base/dd.c:413 driver_probe_device+0x610/0xa00 drivers/base/dd.c:557 __device_attach_driver+0x230/0x290 drivers/base/dd.c:653 bus_for_each_drv+0x161/0x210 drivers/base/bus.c:463 __device_attach+0x26e/0x3d0 drivers/base/dd.c:710 device_initial_probe+0x1f/0x30 drivers/base/dd.c:757 bus_probe_device+0x1eb/0x290 drivers/base/bus.c:523 device_add+0xd0b/0x1660 drivers/base/core.c:1835 usb_new_device+0x7b8/0x1020 drivers/usb/core/hub.c:2457 hub_port_connect drivers/usb/core/hub.c:4903 hub_port_connect_change drivers/usb/core/hub.c:5009 port_event drivers/usb/core/hub.c:5115 hub_event+0x194d/0x3740 drivers/usb/core/hub.c:5195 process_one_work+0xc7f/0x1db0 kernel/workqueue.c:2119 worker_thread+0x221/0x1850 kernel/workqueue.c:2253 kthread+0x3a1/0x470 kernel/kthread.c:231 ret_from_fork+0x2a/0x40 arch/x86/entry/entry_64.S:431 This patch adds the checks of out-of-bounce accesses at appropriate places and bails out when it goes out of the given buffer. Reported-by:
-
Yoshihiro Shimoda authored
commit 0a2ce62b upstream. This patch fixes an issue that the usbhsf_fifo_clear() is possible to cause 10 msec delay if the pipe is RX direction and empty because the FRDY bit will never be set to 1 in such case. Fixes: e8d548d5 ("usb: renesas_usbhs: fifo became independent from pipe.") Signed-off-by:
Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com> Signed-off-by:
-
Yoshihiro Shimoda authored
commit 6124607a upstream. This patch fixes an issue that the driver sets the BCLR bit of {C,Dn}FIFOCTR register to 1 even when it's non-DCP pipe and the FRDY bit of {C,Dn}FIFOCTR register is set to 1. Fixes: e8d548d5 ("usb: renesas_usbhs: fifo became independent from pipe.") Signed-off-by:
-
Alan Stern authored
commit 113f6eb6 upstream. Kris Lindgren reports that without the NO_WP_DETECT flag, his Seagate external disk drive fails all write accesses. This regresssion dates back approximately to the start of the 4.x kernel releases. Signed-off-by:
Kris Lindgren <kris.lindgren@gmail.com> Signed-off-by:
-
Nicolas Ferre authored
commit 6baeda12 upstream. The driver triggers actions on both edges of the vbus signal. The former PIO controller was triggering IRQs on both falling and rising edges by default. Newer PIO controller don't, so it's better to set it explicitly to IRQF_TRIGGER_FALLING | IRQF_TRIGGER_RISING. Without this patch we may trigger the connection with host but only on some bouncing signal conditions and thus lose connecting events. Acked-by:
Ludovic Desroches <ludovic.desroches@microchip.com> Signed-off-by:
Nicolas Ferre <nicolas.ferre@microchip.com> Signed-off-by:
-
Alan Stern authored
commit 6e76c01e upstream. The gadgetfs driver as a long-outstanding FIXME, regarding a call of copy_to_user() made while holding a spinlock. This patch fixes the issue by dropping the spinlock and using the dev->udc_usage mechanism introduced by another recent patch to guard against status changes while the lock isn't held. Signed-off-by:
-
Alan Stern authored
commit 520b72fc upstream. The gadgetfs driver (drivers/usb/gadget/legacy/inode.c) was written before the UDC and composite frameworks were adopted; it is a legacy driver. As such, it expects that once bound to a UDC controller, it will not be unbound until it unregisters itself. However, the UDC framework does unbind function drivers while they are still registered. When this happens, it can cause the gadgetfs driver to misbehave or crash. For example, userspace can cause a crash by opening the device file and doing an ioctl call before setting up a configuration (found by Andrey Konovalov using the syzkaller fuzzer). This patch adds checks and synchronization to prevent these bad behaviors. It adds a udc_usage counter that the driver increments at times when it is using a gadget interface without holding the private spinlock. The unbind routine waits for this counter to go to 0 before returning, thereby ensuring that the UDC is no longer in use. The patch also adds a check in the dev_ioctl() routine to make sure the driver is bound to a UDC before dereferencing the gadget pointer, and it makes destroy_ep_files() synchronize with the endpoint I/O routines, to prevent the user from accessing an endpoint data structure after it has been removed. Signed-off-by:
-
David Eccher authored
commit b7bd98b7 upstream. Fix bad unlock balance: ep0_write enter with the locks locked from inode.c:1769, hence it must exit with spinlock held to avoid double unlock in dev_config. Signed-off-by:
David Eccher <d.eccher@gmail.com> Signed-off-by:
Felipe Balbi <balbi@ti.com> Signed-off-by:
-
- 08 Oct, 2017 11 commits
-
-
Greg Kroah-Hartman authored
-
Arnd Bergmann authored
commit 69d3973a upstream. gcc-7.0.1 warns about old code in ttpci: In file included from drivers/media/pci/ttpci/av7110.c:63:0: In function 'irdebi.isra.2', inlined from 'start_debi_dma' at drivers/media/pci/ttpci/av7110.c:376:3, inlined from 'gpioirq' at drivers/media/pci/ttpci/av7110.c:659:3: drivers/media/pci/ttpci/av7110_hw.h:406:3: warning: 'memcpy': specified size between 18446744071562067968 and 18446744073709551615 exceeds maximum object size 9223372036854775807 [-Wstringop-overflow=] memcpy(av7110->debi_virt, (char *) &res, count); In function 'irdebi.isra.2', inlined from 'start_debi_dma' at drivers/media/pci/ttpci/av7110.c:376:3, inlined from 'gpioirq' at drivers/media/pci/ttpci/av7110.c:668:3: drivers/media/pci/ttpci/av7110_hw.h:406:3: warning: 'memcpy': specified size between 18446744071562067968 and 18446744073709551615 exceeds maximum object size 9223372036854775807 [-Wstringop-overflow=] memcpy(av7110->debi_virt, (char *) &res, count); Apparently, 'count' can be negative here, which will then get turned into a giant size argument for memcpy. Changing the sizes to 'unsigned int' instead seems safe as we already check for maximum sizes, and it also simplifies the code a bit. Signed-off-by:
Arnd Bergmann <arnd@arndb.de> Signed-off-by:
Mauro Carvalho Chehab <mchehab@s-opensource.com> Signed-off-by:
-
Arnd Bergmann authored
commit 13f99ebd upstream. The latest gcc-7.0.1 snapshot points out that we if nr_ch is zero, we never initialize some variables: sound/pci/au88x0/au88x0_core.c: In function 'vortex_adb_allocroute': sound/pci/au88x0/au88x0_core.c:2304:68: error: 'mix[0]' may be used uninitialized in this function [-Werror=maybe-uninitialized] sound/pci/au88x0/au88x0_core.c:2305:58: error: 'src[0]' may be used uninitialized in this function [-Werror=maybe-uninitialized] I assume this can never happen in practice, but adding a check here doesn't hurt either and avoids the warning. The code has been unchanged since the start of git history. Signed-off-by:
-
Arnd Bergmann authored
commit 0527873b upstream. gcc-7 warns about some declarations that are more 'const' than necessary: arch/arm/mach-at91/pm.c:338:34: error: duplicate 'const' declaration specifier [-Werror=duplicate-decl-specifier] static const struct of_device_id const ramc_ids[] __initconst = { arch/arm/mach-bcm/bcm_kona_smc.c:36:34: error: duplicate 'const' declaration specifier [-Werror=duplicate-decl-specifier] static const struct of_device_id const bcm_kona_smc_ids[] __initconst = { arch/arm/mach-spear/time.c:207:34: error: duplicate 'const' declaration specifier [-Werror=duplicate-decl-specifier] static const struct of_device_id const timer_of_match[] __initconst = { arch/arm/mach-omap2/prm_common.c:714:34: error: duplicate 'const' declaration specifier [-Werror=duplicate-decl-specifier] static const struct of_device_id const omap_prcm_dt_match_table[] __initconst = { arch/arm/mach-omap2/vc.c:562:35: error: duplicate 'const' declaration specifier [-Werror=duplicate-decl-specifier] static const struct i2c_init_data const omap4_i2c_timing_data[] __initconst = { The ones in arch/arm were apparently all introduced accidentally by one commit that correctly marked a lot of variables as __initconst. Fixes: 19c233b7 ("ARM: appropriate __init annotation for const data") Acked-by:
Alexandre Belloni <alexandre.belloni@free-electrons.com> Acked-by:
Tony Lindgren <tony@atomide.com> Acked-by:
Nicolas Pitre <nico@linaro.org> Acked-by:
Florian Fainelli <f.fainelli@gmail.com> Acked-by:
Viresh Kumar <viresh.kumar@linaro.org> Acked-by:
Krzysztof Hałasa <khalasa@piap.pl> Signed-off-by:
-
Arnd Bergmann authored
commit f6aafac1 upstream. aarch64-linux-gcc-7 complains about code it doesn't fully understand: drivers/infiniband/hw/qib/qib_iba7322.c: In function 'qib_7322_txchk_change': include/asm-generic/bitops/non-atomic.h:105:35: error: 'shadow' may be used uninitialized in this function [-Werror=maybe-uninitialized] The code is right, and despite trying hard, I could not come up with a version that I liked better than just adding a fake initialization here to shut up the warning. Fixes: f931551b ("IB/qib: Add new qib driver for QLogic PCIe InfiniBand adapters") Signed-off-by:
Ira Weiny <ira.weiny@intel.com> Signed-off-by:
Doug Ledford <dledford@redhat.com> Signed-off-by:
-
Jisheng Zhang authored
commit 1d2d8de4 upstream. This is to fix below sparse warning: drivers/firmware/psci.c:mmm:nn: warning: duplicate const Signed-off-by:
Jisheng Zhang <jszhang@marvell.com> Signed-off-by:
Lorenzo Pieralisi <lorenzo.pieralisi@arm.com> Signed-off-by:
-
Gwendal Grignou authored
[ Upstream commit d85fc67d ] Without this patch, failed probe would not free resources like irq. ata port tdev object currently hold a reference to the ata port object. Therefore the ata port object release function will not get called until the ata_tport_release is called. But that would never happen, releasing the last reference of ata port dev is done by scsi_host_release, which is called by ata_host_release when the ata port object is released. The ata device objects actually do not need to explicitly hold a reference to their real counterpart, given the transport objects are the children of these objects and device_add() is call for each child. We know the parent will not be deleted until we call the child's device_del(). Reported-by:
Matthew Whitehead <tedheadster@gmail.com> Tested-by:
Tejun Heo <tj@kernel.org> Signed-off-by:
Gwendal Grignou <gwendal@chromium.org> Signed-off-by:
Sasha Levin <alexander.levin@verizon.com> Signed-off-by:
-
Darrick J. Wong authored
[ Upstream commit 08b005f1 ] The sole remaining caller of kmem_zalloc_greedy is bulkstat, which uses it to grab 1-4 pages for staging of inobt records. The infinite loop in the greedy allocation function is causing hangs[1] in generic/269, so just get rid of the greedy allocator in favor of kmem_zalloc_large. This makes bulkstat somewhat more likely to ENOMEM if there's really no pages to spare, but eliminates a source of hangs. [1] http://lkml.kernel.org/r/20170301044634.rgidgdqqiiwsmfpj%40XZHOUW.usersys.redhat.comSigned-off-by:
Darrick J. Wong <darrick.wong@oracle.com> Reviewed-by:
Christoph Hellwig <hch@lst.de> Signed-off-by:
-
Heiner Kallweit authored
[ Upstream commit 3b0277f1 ] Most likely a copy & paste error. Signed-off-by:
Heiner Kallweit <hkallweit1@gmail.com> Acked-by:
Jerome Brunet <jbrunet@baylibre.com> Signed-off-by:
Wolfram Sang <wsa@the-dreams.de> Fixes: 30021e37 ("i2c: add support for Amlogic Meson I2C controller") Signed-off-by:
-
Shaohua Li authored
[ Upstream commit 6d399783 ] Commit 57c67df4(md/raid10: submit IO from originating thread instead of md thread) submits bio directly for normal disks but not for replacement disks. There is no point we shouldn't do this for replacement disks. Cc: NeilBrown <neilb@suse.com> Signed-off-by:
Shaohua Li <shli@fb.com> Signed-off-by:
-
Zhu Yanjun authored
[ Upstream commit 3b12f73a ] In the function rds_ib_setup_qp, the error handle is missing. When some error occurs, it is possible that memory leak occurs. As such, error handle is added. Cc: Joe Jin <joe.jin@oracle.com> Reviewed-by:
Junxiao Bi <junxiao.bi@oracle.com> Reviewed-by:
Guanglei Li <guanglei.li@oracle.com> Signed-off-by:
Zhu Yanjun <yanjun.zhu@oracle.com> Acked-by:
Santosh Shilimkar <santosh.shilimkar@oracle.com> Signed-off-by:
David S. Miller <davem@davemloft.net> Signed-off-by:
-
