- 20 Nov, 2014 7 commits
-
-
James Ralston authored
commit 690000b9 upstream. This patch adds the AHCI-mode SATA Device IDs for the Intel Sunrise Point PCH. Signed-off-by:
James Ralston <james.d.ralston@intel.com> Signed-off-by:
Tejun Heo <tj@kernel.org> Signed-off-by:
Luis Henriques <luis.henriques@canonical.com>
-
Tejun Heo authored
commit 66a7cbc3 upstream. Samsung pci-e SSDs on macbooks failed miserably on NCQ commands, so 67809f85 ("ahci: disable NCQ on Samsung pci-e SSDs on macbooks") disabled NCQ on them. It turns out that NCQ is fine as long as MSI is not used, so let's turn off MSI and leave NCQ on. Signed-off-by:
Imre Kaloz <kaloz@openwrt.org> Fixes: 67809f85 ("ahci: disable NCQ on Samsung pci-e SSDs on macbooks") Signed-off-by:
-
Simon Horman authored
commit aa1cf258 upstream. Unlike other SATA R-Car r8a7790 controllers the r8a7790 ES1 SATA R-Car controller needs to be run with DIPM disabled. Signed-off-by:
Simon Horman <horms+renesas@verge.net.au> Signed-off-by:
-
Heinz Mauelshagen authored
commit 40d43c4b upstream. The dm-raid superblock (struct dm_raid_superblock) is padded to 512 bytes and that size is being used to read it in from the metadata device into one preallocated page. Reading or writing this on a 512-byte sector device works fine but on a 4096-byte sector device this fails. Set the dm-raid superblock's size to the logical block size of the metadata device, because IO at that size is guaranteed too work. Also add a size check to avoid silent partial metadata loss in case the superblock should ever grow past the logical block size or PAGE_SIZE. [includes pointer math fix from Dan Carpenter] Reported-by:
"Liuhua Wang" <lwang@suse.com> Signed-off-by:
Heinz Mauelshagen <heinzm@redhat.com> Signed-off-by:
Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by:
Mike Snitzer <snitzer@redhat.com> Signed-off-by:
-
Max Filippov authored
commit 2651cc69 upstream. Userspace actually passes single parameter (path name) to the umount syscall, so new umount just fails. Fix it by requesting old umount syscall implementation and re-wiring umount to it. Signed-off-by:
Max Filippov <jcmvbkbc@gmail.com> Signed-off-by:
-
Mikulas Patocka authored
commit 9d28eb12 upstream. The shrinker uses gfp flags to indicate what kind of operation can the driver wait for. If __GFP_IO flag is present, the driver can wait for block I/O operations, if __GFP_FS flag is present, the driver can wait on operations involving the filesystem. dm-bufio tested for __GFP_IO. However, dm-bufio can run on a loop block device that makes calls into the filesystem. If __GFP_IO is present and __GFP_FS isn't, dm-bufio could still block on filesystem operations if it runs on a loop block device. The change from __GFP_IO to __GFP_FS supposedly fixes one observed (though unreproducible) deadlock involving dm-bufio and loop device. Signed-off-by:
Mikulas Patocka <mpatocka@redhat.com> Signed-off-by:
-
Dan Carpenter authored
commit f2e323ec upstream. We need to add a limit check here so we don't overflow the buffer. Signed-off-by:
Mauro Carvalho Chehab <mchehab@osg.samsung.com> Cc: Moritz Mühlenhoff <jmm@inutil.org> Signed-off-by:
-
- 17 Nov, 2014 33 commits
-
-
Ilya Dryomov authored
commit 792c3a91 upstream. Need to use WQ_MEM_RECLAIM for our workqueues to prevent I/O lockups under memory pressure - we sit on the memory reclaim path. Signed-off-by:
Ilya Dryomov <idryomov@redhat.com> Tested-by:
Micha Krause <micha@krausam.de> Reviewed-by:
Sage Weil <sage@redhat.com> Signed-off-by:
-
Wei Yongjun authored
commit 255939e7 upstream. Fix to return -ENOMEM from the workqueue alloc error handling case instead of 0, as done elsewhere in this function. Reviewed-by:
Alex Elder <elder@linaro.org> Signed-off-by:
Wei Yongjun <yongjun_wei@trendmicro.com.cn> Cc: Ilya Dryomov <ilya.dryomov@inktank.com> Signed-off-by:
-
Ilya Dryomov authored
commit 58d1362b upstream. drivers/block/rbd.c: In function ‘rbd_dev_device_setup’: drivers/block/rbd.c:5090:19: warning: format not a string literal and no format arguments [-Wformat-security] Reported-by:
kbuild test robot <fengguang.wu@intel.com> Signed-off-by:
Ilya Dryomov <ilya.dryomov@inktank.com> Signed-off-by:
-
Olivier Gay authored
commit f974008f upstream. Add keyboard input assist controls usages from approved hid usage table request HUTTR42: http://www.usb.org/developers/hidpage/HUTRR42c.pdfSigned-off-by:
Olivier Gay <ogay@logitech.com> Acked-by:
Dmitry Torokhov <dmitry.torokhov@gmail.com> Signed-off-by:
Jiri Kosina <jkosina@suse.cz> Signed-off-by:
-
Adel Gadllah authored
commit 1af39588 upstream. This device needs the quirk as well. Tested-by:
Kevin Fenzi <kevin@scrye.com> Signed-off-by:
Adel Gadllah <adel.gadllah@gmail.com> Signed-off-by:
-
Adel Gadllah authored
commit 29d05c2e upstream. This device needs the quirk as well. Signed-off-by:
-
Oliver Neukum authored
commit 5235166f upstream. There is a second mouse sharing the same vendor strings but different IDs. Signed-off-by:
Oliver Neukum <oneukum@suse.de> Signed-off-by:
-
Oliver Neukum authored
commit 4980f957 upstream. This mouse keeps disconnecting in runlevel 3. It needs the ALWAYS_POLL quirk. Signed-off-by:
-
Johan Hovold authored
commit bfe3c873 upstream. Enable the always-poll quirk for Elan Touchscreens found on some recent Samsung laptops. Without this quirk the device keeps disconnecting from the bus (and is re-enumerated) unless opened (and kept open, should an input event occur). Note that while the device can be run-time suspended, the autosuspend timeout must be high enough to allow the device to be polled at least once before being suspended. Specifically, using autosuspend_delay_ms=0 will still cause the device to disconnect on input events. Signed-off-by:
Johan Hovold <johan@kernel.org> Signed-off-by:
-
Johan Hovold authored
commit 0b750b3b upstream. Add quirk to make sure that a device is always polled for input events even if it hasn't been opened. This is needed for devices that disconnects from the bus unless the interrupt endpoint has been polled at least once or when not responding to an input event (e.g. after having shut down X). Signed-off-by:
-
Jason Gerecke authored
commit 5989a55a upstream. The commit which introduced TransducerSerialNumber (368c9664) is missing two crucial implementation details. Firstly, the commit does not set the type/code/bit/max fields as expected later down the code which can cause the driver to crash when a tablet with this usage is connected. Secondly, the call to 'set_bit' causes MSC_PULSELED to be sent instead of the expected MSC_SERIAL. This commit addreses both issues. Signed-off-by:
Jason Gerecke <jason.gerecke@wacom.com> Reviewed-by:
Benjamin Tissoires <benjamin.tissoires@redhat.com> Reviewed-by:
Ping Cheng <pingc@wacom.com> Signed-off-by:
-
Eric Dumazet authored
commit 39bb5e62 upstream. Some drivers are unable to perform TX completions in a bound time. They instead call skb_orphan() Problem is skb_fclone_busy() has to detect this case, otherwise we block TCP retransmits and can freeze unlucky tcp sessions on mostly idle hosts. Signed-off-by:
Eric Dumazet <edumazet@google.com> Fixes: 1f3279ae ("tcp: avoid retransmits of TCP packets hanging in host queues") Signed-off-by:
David S. Miller <davem@davemloft.net> [ luis: backport to 3.16 provided by Eric Dumzet ] Signed-off-by:
-
Roger Pau Monné authored
commit 61cecca8 upstream. Fix leaking a page when a grant mapping has failed. Signed-off-by:
Roger Pau Monné <roger.pau@citrix.com> Reported-and-Tested-by:
Tao Chen <boby.chen@huawei.com> Signed-off-by:
Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Signed-off-by:
-
Richard Genoud authored
commit 1bf1890e upstream. I ran into this error after a ubiupdatevol, because I forgot to backport e9110361 UBI: fix the volumes tree sorting criteria. UBI error: process_pool_aeb: orphaned volume in fastmap pool UBI error: ubi_scan_fastmap: Attach by fastmap failed, doing a full scan! kmem_cache_destroy ubi_ainf_peb_slab: Slab cache still has objects CPU: 0 PID: 1 Comm: swapper Not tainted 3.14.18-00053-gf05cac8dbf85 #1 [<c000d298>] (unwind_backtrace) from [<c000baa8>] (show_stack+0x10/0x14) [<c000baa8>] (show_stack) from [<c01b7a68>] (destroy_ai+0x230/0x244) [<c01b7a68>] (destroy_ai) from [<c01b8fd4>] (ubi_attach+0x98/0x1ec) [<c01b8fd4>] (ubi_attach) from [<c01ade90>] (ubi_attach_mtd_dev+0x2b8/0x868) [<c01ade90>] (ubi_attach_mtd_dev) from [<c038b510>] (ubi_init+0x1dc/0x2ac) [<c038b510>] (ubi_init) from [<c0008860>] (do_one_initcall+0x94/0x140) [<c0008860>] (do_one_initcall) from [<c037aadc>] (kernel_init_freeable+0xe8/0x1b0) [<c037aadc>] (kernel_init_freeable) from [<c02730ac>] (kernel_init+0x8/0xe4) [<c02730ac>] (kernel_init) from [<c00093f0>] (ret_from_fork+0x14/0x24) UBI: scanning is finished Freeing the cache in the error path fixes the Slab error. Tested on at91sam9g35 (3.14.18+fastmap backports) Signed-off-by:
Richard Genoud <richard.genoud@gmail.com> Signed-off-by:
-
Dan Streetman authored
commit 10ccaf17 upstream. In powerpc pseries platform dlpar operations, use device_online() and device_offline() instead of cpu_up() and cpu_down(). Calling cpu_up/down() directly does not update the cpu device offline field, which is used to online/offline a cpu from sysfs. Calling device_online/offline() instead keeps the sysfs cpu online value correct. The hotplug lock, which is required to be held when calling device_online/offline(), is already held when dlpar_online/offline_cpu() are called, since they are called only from cpu_probe|release_store(). This patch fixes errors on phyp (PowerVM) systems that have cpu(s) added/removed using dlpar operations; without this patch, the /sys/devices/system/cpu/cpuN/online nodes do not correctly show the online state of added/removed cpus. Signed-off-by:
Dan Streetman <ddstreet@ieee.org> Cc: Nathan Fontenot <nfont@linux.vnet.ibm.com> Fixes: 0902a904 ("Driver core: Use generic offline/online for CPU offline/online") Signed-off-by:
Michael Ellerman <mpe@ellerman.id.au> Signed-off-by:
-
Sinclair Yeh authored
commit 9a72384d upstream. When screen objects are enabled, the bpp is assumed to be 32, otherwise it is set to 16. v2: * Use u32 instead of u64 for assumed_bpp. * Fixed mechanism to check for screen objects * Limit the back buffer size to VRAM. Signed-off-by:
Sinclair Yeh <syeh@vmware.com> Reviewed-by:
Thomas Hellstrom <thellstrom@vmware.com> Signed-off-by:
-
Kirill Tkhai authored
commit f1e3a093 upstream. Probability of use-after-free isn't zero in this place. Signed-off-by:
Kirill Tkhai <ktkhai@parallels.com> Signed-off-by:
Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Link: http://lkml.kernel.org/r/20140922183636.11015.83611.stgit@localhostSigned-off-by:
Ingo Molnar <mingo@kernel.org> [ luis: backported to 3.16: used peterz's backport to 3.17 ] Signed-off-by:
-
Stephen Boyd authored
commit 8b374399 upstream. Frank reports that after continuing in kgdb the RX stale event doesn't occur until after the RX fifo is filled up with exactly the amount of characters programmed for the RX watermark (in this case it's 48). To read a single character from the uartdm hardware we force a stale event so that any characters in the RX packing buffer are flushed into the RX fifo immediately instead of waiting for a stale timeout or for the fifo to fill. Forcing that stale event asserts the stale interrupt but we never clear that interrupt via UART_CR_CMD_RESET_STALE_INT in the polling functions. So when kgdb continues the stale interrupt is left pending in the hardware and we don't timeout with a stale event, like we usually would if a user typed one character on the console, until the reset stale interrupt and stale event commands are sent. Frank could get things working again by running handle_rx_dm(). By putting enough characters into the fifo he could trigger a watermark interrupt, and thus cause handle_rx_dm() to run finally resetting the stale interrupt and enabling the stale event so that single characters would cause timeouts again. The fix is to just do what the interrupt routine was doing all along and clear the stale interrupt and enable the event again. Doing this also smooths over any differences in the fifo behavior between v1.3 and v1.4 hardware allowing us to skip forcing the uart into single character mode. Reviewed-by:
Frank Rowand <frank.rowand@sonymobile.com> Tested-by:
Stephen Boyd <sboyd@codeaurora.org> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by:
-
Bryan O'Donoghue authored
commit aece118e upstream. Intel processors which don't report cache information via cpuid(2) or cpuid(4) need quirk code in the legacy_cache_size callback to report this data. For Intel that callback is is intel_size_cache(). This patch enables calling of cpu_detect_cache_sizes() inside of init_intel() and hence the calling of the legacy_cache callback in intel_size_cache(). Adding this call will ensure that PIII Tualatin currently in intel_size_cache() and Quark SoC X1000 being added to intel_size_cache() in this patch will report their respective cache sizes. This model of calling cpu_detect_cache_sizes() is consistent with AMD/Via/Cirix/Transmeta and Centaur. Also added is a string to idenitfy the Quark as Quark SoC X1000 giving better and more descriptive output via /proc/cpuinfo Adding cpu_detect_cache_sizes to init_intel() will enable calling of intel_size_cache() on Intel processors which currently no code can reach. Therefore this patch will also re-enable reporting of PIII Tualatin cache size information as well as add Quark SoC X1000 support. Comment text and cache flow logic suggested by Thomas Gleixner Signed-off-by:
Bryan O'Donoghue <pure.logic@nexus-software.ie> Cc: davej@redhat.com Cc: hmh@hmh.eng.br Link: http://lkml.kernel.org/r/1412641189-12415-3-git-send-email-pure.logic@nexus-software.ieSigned-off-by:
Thomas Gleixner <tglx@linutronix.de> Signed-off-by:
-
Axel Lin authored
commit 91ba0e59 upstream. Fix a copy-paste bug when converting to the control framework. Fixes: commit 5d478e0d ("[media] tda7432: convert to the control framework") Signed-off-by:
Axel Lin <axel.lin@ingics.com> Signed-off-by:
Mauro Carvalho Chehab <m.chehab@samsung.com> Signed-off-by:
-
Antti Palosaari authored
commit f538e085 upstream. Maximum satellite symbol rate used is 45000000Sps which overflows when multiplied by 135. As final calculation result is fraction, we could use mult_frac macro in order to keep calculation inside 32 bit number limits and prevent overflow. Original bug and fix was provided by Nibble Max. I decided to implement it differently as it is now. Reported-by:
Nibble Max <nibble.max@gmail.com> Tested-by:
Antti Palosaari <crope@iki.fi> Signed-off-by:
-
Mauro Carvalho Chehab authored
commit 29bbb7bd upstream. Add support for PCTV microStick (77e) device that uses a sms1140 chipset. Signed-off-by:
-
Paul Fertser authored
commit 62ea864f upstream. As reported on [1], this device needs this quirk to be able to reliably initialise the webcam. [1] http://ubuntuforums.org/showthread.php?t=2145996Signed-off-by:
Paul Fertser <fercerpav@gmail.com> Signed-off-by:
Laurent Pinchart <laurent.pinchart@ideasonboard.com> Signed-off-by:
-
Maciej Matraszek authored
commit 3bacc10c upstream. Fix clamp_align() used in v4l_bound_align_image() to prevent overflow when passed large value like UINT32_MAX. In the current implementation: clamp_align(UINT32_MAX, 8, 8192, 3) returns 8, because in line: x = (x + (1 << (align - 1))) & mask; x overflows to (-1 + 4) & 0x7 = 3, while expected value is 8192. v4l_bound_align_image() is heavily used in VIDIOC_S_FMT and VIDIOC_SUBDEV_S_FMT ioctls handlers, and documentation of the latter explicitly states that: "The modified format should be as close as possible to the original request." -- http://linuxtv.org/downloads/v4l-dvb-apis/vidioc-subdev-g-fmt.html Thus one would expect, that passing UINT32_MAX as format width and height will result in setting maximum possible resolution for the device. Particularly, when the driver doesn't support VIDIOC_ENUM_FRAMESIZES ioctl, which is common in the codebase. Fixes changeset: b0d3159bSigned-off-by:
Maciej Matraszek <m.matraszek@samsung.com> Acked-by:
Sakari Ailus <sakari.ailus@linux.intel.com> Signed-off-by:
-
Josh Boyer authored
commit e351943b upstream. The userspace drm.h include doesn't prefix the drm directory. This can lead to compile failures as /usr/include/drm/ isn't in the standard gcc include paths. Fix it to be <drm/drm.h>, which matches the rest of the driver drm header files that get installed into /usr/include/drm. Red Hat Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=1138759 Fixes: 1d7a5cbfReported-by:
Jeffrey Bastian <jbastian@redhat.com> Signed-off-by:
Josh Boyer <jwboyer@fedoraproject.org> Signed-off-by:
Dave Airlie <airlied@redhat.com> Signed-off-by:
-
Tvrtko Ursulin authored
commit c479f438 upstream. sg_alloc_table_from_pages() can build us a table with coalesced ranges which means we need to iterate over pages and not sg table entries when releasing page references. Signed-off-by:
Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: "Barbalho, Rafael" <rafael.barbalho@intel.com> Tested-by:
Rafael Barbalho <rafael.barbalho@intel.com> Reviewed-by:
Chris Wilson <chris@chris-wilson.co.uk> [danvet: Remove unused local variable sg.] Signed-off-by:
Daniel Vetter <daniel.vetter@intel.com> Signed-off-by:
-
Emil Velikov authored
commit b485a700 upstream. nv92 hardware has only 16 interrupt lines, while nv94 and later has 32. Accessing 0xe0c{0,4} registers on nv92 can lead to incorrect PDISP setup. This is a regression introduced with commit 9d0f5ec9ee0fd5dc5fc1cc2cf559286431e406e3 Author: Ben Skeggs <bskeggs@redhat.com> Date: Mon May 12 15:22:42 2014 +1000 gpio: split g92 class from nv50 Reported-by: estece on #nouveau Signed-off-by:
Emil Velikov <emil.l.velikov@gmail.com> Signed-off-by:
Ben Skeggs <bskeggs@redhat.com> Signed-off-by:
-
Benjamin Herrenschmidt authored
commit 1e99cfa8 upstream. The translation from the X driver to the KMS one typo'ed a couple of array indices, causing the HW cursor to look weird (blocky with leaking edge colors). This fixes it. Signed-off-by:
Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by:
-
Hans de Goede authored
commit 9ff84a17 upstream. Without this the aux port does not get detected, and consequently the touchpad will not work. https://bugzilla.redhat.com/show_bug.cgi?id=1110011Signed-off-by:
Hans de Goede <hdegoede@redhat.com> Signed-off-by:
-
Dmitry Torokhov authored
commit aa972409 upstream. Unfortunately, ForcePad capability is not actually exported over PS/2, so we have to resort to DMI checks. Reported-by:
Nicole Faerber <nicole.faerber@kernelconcepts.de> Signed-off-by:
-
Mikulas Patocka authored
commit f74a289b upstream. The framebuffer code uses the current background color to fill the border when switching consoles, however, this results in inconsistent behavior. For example: - start Midnigh Commander - the border is black - switch to another console and switch back - the border is cyan - type something into the command line in mc - the border is cyan - switch to another console and switch back - the border is black - press F9 to go to menu - the border is black - switch to another console and switch back - the border is dark blue When switching to a console with Midnight Commander, the border is random color that was left selected by the slang subsystem. This patch fixes this inconsistency by always using black as the background color when switching consoles. Signed-off-by:
Tomi Valkeinen <tomi.valkeinen@ti.com> Signed-off-by:
-
Mikulas Patocka authored
commit 5b789da8 upstream. The function bitcpy_rev has a bug that may result in screen corruption. The bug happens under these conditions: * the end of the destination area of a copy operation is aligned on a long word boundary * the end of the source area is not aligned on a long word boundary * we are copying more than one long word In this case, the variable shift is non-zero and the variable first is zero. The statements FB_WRITEL(comp(d0, FB_READL(dst), first), dst) reads the last long word of the destination and writes it back unchanged (because first is zero). Correctly, we should write the variable d0 to the last word of the destination in this case. This patch fixes the bug by introducing and extra test if first is zero. The patch also removes the references to fb_memmove in the code that is commented out because fb_memmove was removed from framebuffer subsystem. Signed-off-by:
-
Prarit Bhargava authored
commit d3051b48 upstream. A panic was seen in the following sitation. There are two threads running on the system. The first thread is a system monitoring thread that is reading /proc/modules. The second thread is loading and unloading a module (in this example I'm using my simple dummy-module.ko). Note, in the "real world" this occurred with the qlogic driver module. When doing this, the following panic occurred: ------------[ cut here ]------------ kernel BUG at kernel/module.c:3739! invalid opcode: 0000 [#1] SMP Modules linked in: binfmt_misc sg nfsv3 rpcsec_gss_krb5 nfsv4 dns_resolver nfs fscache intel_powerclamp coretemp kvm_intel kvm crct10dif_pclmul crc32_pclmul crc32c_intel ghash_clmulni_intel aesni_intel lrw igb gf128mul glue_helper iTCO_wdt iTCO_vendor_support ablk_helper ptp sb_edac cryptd pps_core edac_core shpchp i2c_i801 pcspkr wmi lpc_ich ioatdma mfd_core dca ipmi_si nfsd ipmi_msghandler auth_rpcgss nfs_acl lockd sunrpc xfs libcrc32c sr_mod cdrom sd_mod crc_t10dif crct10dif_common mgag200 syscopyarea sysfillrect sysimgblt i2c_algo_bit drm_kms_helper ttm isci drm libsas ahci libahci scsi_transport_sas libata i2c_core dm_mirror dm_region_hash dm_log dm_mod [last unloaded: dummy_module] CPU: 37 PID: 186343 Comm: cat Tainted: GF O-------------- 3.10.0+ #7 Hardware name: Intel Corporation S2600CP/S2600CP, BIOS RMLSDP.86I.00.29.D696.1311111329 11/11/2013 task: ffff8807fd2d8000 ti: ffff88080fa7c000 task.ti: ffff88080fa7c000 RIP: 0010:[<ffffffff810d64c5>] [<ffffffff810d64c5>] module_flags+0xb5/0xc0 RSP: 0018:ffff88080fa7fe18 EFLAGS: 00010246 RAX: 0000000000000003 RBX: ffffffffa03b5200 RCX: 0000000000000000 RDX: 0000000000001000 RSI: ffff88080fa7fe38 RDI: ffffffffa03b5000 RBP: ffff88080fa7fe28 R08: 0000000000000010 R09: 0000000000000000 R10: 0000000000000000 R11: 000000000000000f R12: ffffffffa03b5000 R13: ffffffffa03b5008 R14: ffffffffa03b5200 R15: ffffffffa03b5000 FS: 00007f6ae57ef740(0000) GS:ffff88101e7a0000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000404f70 CR3: 0000000ffed48000 CR4: 00000000001407e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 Stack: ffffffffa03b5200 ffff8810101e4800 ffff88080fa7fe70 ffffffff810d666c ffff88081e807300 000000002e0f2fbf 0000000000000000 ffff88100f257b00 ffffffffa03b5008 ffff88080fa7ff48 ffff8810101e4800 ffff88080fa7fee0 Call Trace: [<ffffffff810d666c>] m_show+0x19c/0x1e0 [<ffffffff811e4d7e>] seq_read+0x16e/0x3b0 [<ffffffff812281ed>] proc_reg_read+0x3d/0x80 [<ffffffff811c0f2c>] vfs_read+0x9c/0x170 [<ffffffff811c1a58>] SyS_read+0x58/0xb0 [<ffffffff81605829>] system_call_fastpath+0x16/0x1b Code: 48 63 c2 83 c2 01 c6 04 03 29 48 63 d2 eb d9 0f 1f 80 00 00 00 00 48 63 d2 c6 04 13 2d 41 8b 0c 24 8d 50 02 83 f9 01 75 b2 eb cb <0f> 0b 66 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 55 48 89 e5 41 RIP [<ffffffff810d64c5>] module_flags+0xb5/0xc0 RSP <ffff88080fa7fe18> Consider the two processes running on the system. CPU 0 (/proc/modules reader) CPU 1 (loading/unloading module) CPU 0 opens /proc/modules, and starts displaying data for each module by traversing the modules list via fs/seq_file.c:seq_open() and fs/seq_file.c:seq_read(). For each module in the modules list, seq_read does op->start() <-- this is a pointer to m_start() op->show() <- this is a pointer to m_show() op->stop() <-- this is a pointer to m_stop() The m_start(), m_show(), and m_stop() module functions are defined in kernel/module.c. The m_start() and m_stop() functions acquire and release the module_mutex respectively. ie) When reading /proc/modules, the module_mutex is acquired and released for each module. m_show() is called with the module_mutex held. It accesses the module struct data and attempts to write out module data. It is in this code path that the above BUG_ON() warning is encountered, specifically m_show() calls static char *module_flags(struct module *mod, char *buf) { int bx = 0; BUG_ON(mod->state == MODULE_STATE_UNFORMED); ... The other thread, CPU 1, in unloading the module calls the syscall delete_module() defined in kernel/module.c. The module_mutex is acquired for a short time, and then released. free_module() is called without the module_mutex. free_module() then sets mod->state = MODULE_STATE_UNFORMED, also without the module_mutex. Some additional code is called and then the module_mutex is reacquired to remove the module from the modules list: /* Now we can delete it from the lists */ mutex_lock(&module_mutex); stop_machine(__unlink_module, mod, NULL); mutex_unlock(&module_mutex); This is the sequence of events that leads to the panic. CPU 1 is removing dummy_module via delete_module(). It acquires the module_mutex, and then releases it. CPU 1 has NOT set dummy_module->state to MODULE_STATE_UNFORMED yet. CPU 0, which is reading the /proc/modules, acquires the module_mutex and acquires a pointer to the dummy_module which is still in the modules list. CPU 0 calls m_show for dummy_module. The check in m_show() for MODULE_STATE_UNFORMED passed for dummy_module even though it is being torn down. Meanwhile CPU 1, which has been continuing to remove dummy_module without holding the module_mutex, now calls free_module() and sets dummy_module->state to MODULE_STATE_UNFORMED. CPU 0 now calls module_flags() with dummy_module and ... static char *module_flags(struct module *mod, char *buf) { int bx = 0; BUG_ON(mod->state == MODULE_STATE_UNFORMED); and BOOM. Acquire and release the module_mutex lock around the setting of MODULE_STATE_UNFORMED in the teardown path, which should resolve the problem. Testing: In the unpatched kernel I can panic the system within 1 minute by doing while (true) do insmod dummy_module.ko; rmmod dummy_module.ko; done and while (true) do cat /proc/modules; done in separate terminals. In the patched kernel I was able to run just over one hour without seeing any issues. I also verified the output of panic via sysrq-c and the output of /proc/modules looks correct for all three states for the dummy_module. dummy_module 12661 0 - Unloading 0xffffffffa03a5000 (OE-) dummy_module 12661 0 - Live 0xffffffffa03bb000 (OE) dummy_module 14015 1 - Loading 0xffffffffa03a5000 (OE+) Signed-off-by:
Prarit Bhargava <prarit@redhat.com> Reviewed-by:
Oleg Nesterov <oleg@redhat.com> Signed-off-by:
Rusty Russell <rusty@rustcorp.com.au> Signed-off-by:
-
