- 26 Oct, 2013 40 commits
-
-
Miklos Szeredi authored
commit efeb9e60 upstream. Userspace can add names containing a slash character to the directory listing. Don't allow this as it could cause all sorts of trouble. Signed-off-by:
Miklos Szeredi <mszeredi@suse.cz> [bwh: Backported to 3.2: drop changes to parse_dirplusfile() which we don't have] Signed-off-by:
Ben Hutchings <ben@decadent.org.uk>
-
Alexey Khoroshilov authored
commit 2e923a05 upstream. free_buff_list and rec_buff_list are initialized in the middle of hdpvr_probe(), but if something bad happens before that, error handling code calls hdpvr_delete(), which contains iteration over the lists (via hdpvr_free_buffers()). The patch moves the lists initialization to the beginning and by the way fixes goto label in error handling of registering videodev. Found by Linux Driver Verification project (linuxtesting.org). Signed-off-by:
Alexey Khoroshilov <khoroshilov@ispras.ru> Signed-off-by:
Hans Verkuil <hans.verkuil@cisco.com> Signed-off-by:
Mauro Carvalho Chehab <m.chehab@samsung.com> [bwh: Backported to 3.2: adjust filename] Signed-off-by:
-
Hans Verkuil authored
commit 280847b5 upstream. Video nodes can be used at once after registration, so make sure the full initialization is done before registering them. Signed-off-by:
Mauro Carvalho Chehab <mchehab@redhat.com> [bwh: Backported to 3.2: adjust filename, context] Signed-off-by:
-
Maxim Patlasov authored
commit 06a7c3c2 upstream. The way how fuse calls truncate_pagecache() from fuse_change_attributes() is completely wrong. Because, w/o i_mutex held, we never sure whether 'oldsize' and 'attr->size' are valid by the time of execution of truncate_pagecache(inode, oldsize, attr->size). In fact, as soon as we released fc->lock in the middle of fuse_change_attributes(), we completely loose control of actions which may happen with given inode until we reach truncate_pagecache. The list of potentially dangerous actions includes mmap-ed reads and writes, ftruncate(2) and write(2) extending file size. The typical outcome of doing truncate_pagecache() with outdated arguments is data corruption from user point of view. This is (in some sense) acceptable in cases when the issue is triggered by a change of the file on the server (i.e. externally wrt fuse operation), but it is absolutely intolerable in scenarios when a single fuse client modifies a file without any external intervention. A real life case I discovered by fsx-linux looked like this: 1. Shrinking ftruncate(2) comes to fuse_do_setattr(). The latter sends FUSE_SETATTR to the server synchronously, but before getting fc->lock ... 2. fuse_dentry_revalidate() is asynchronously called. It sends FUSE_LOOKUP to the server synchronously, then calls fuse_change_attributes(). The latter updates i_size, releases fc->lock, but before comparing oldsize vs attr->size.. 3. fuse_do_setattr() from the first step proceeds by acquiring fc->lock and updating attributes and i_size, but now oldsize is equal to outarg.attr.size because i_size has just been updated (step 2). Hence, fuse_do_setattr() returns w/o calling truncate_pagecache(). 4. As soon as ftruncate(2) completes, the user extends file size by write(2) making a hole in the middle of file, then reads data from the hole either by read(2) or mmap-ed read. The user expects to get zero data from the hole, but gets stale data because truncate_pagecache() is not executed yet. The scenario above illustrates one side of the problem: not truncating the page cache even though we should. Another side corresponds to truncating page cache too late, when the state of inode changed significantly. Theoretically, the following is possible: 1. As in the previous scenario fuse_dentry_revalidate() discovered that i_size changed (due to our own fuse_do_setattr()) and is going to call truncate_pagecache() for some 'new_size' it believes valid right now. But by the time that particular truncate_pagecache() is called ... 2. fuse_do_setattr() returns (either having called truncate_pagecache() or not -- it doesn't matter). 3. The file is extended either by write(2) or ftruncate(2) or fallocate(2). 4. mmap-ed write makes a page in the extended region dirty. The result will be the lost of data user wrote on the fourth step. The patch is a hotfix resolving the issue in a simplistic way: let's skip dangerous i_size update and truncate_pagecache if an operation changing file size is in progress. This simplistic approach looks correct for the cases w/o external changes. And to handle them properly, more sophisticated and intrusive techniques (e.g. NFS-like one) would be required. I'd like to postpone it until the issue is well discussed on the mailing list(s). Changed in v2: - improved patch description to cover both sides of the issue. Signed-off-by:
Maxim Patlasov <mpatlasov@parallels.com> Signed-off-by:
-
Anand Avati authored
commit d331a415 upstream. Calls like setxattr and removexattr result in updation of ctime. Therefore invalidate inode attributes to force a refresh. Signed-off-by:
Anand Avati <avati@redhat.com> Reviewed-by:
Brian Foster <bfoster@redhat.com> Signed-off-by:
-
Maxim Patlasov authored
commit 4a4ac4eb upstream. The patch fixes a race between ftruncate(2), mmap-ed write and write(2): 1) An user makes a page dirty via mmap-ed write. 2) The user performs shrinking truncate(2) intended to purge the page. 3) Before fuse_do_setattr calls truncate_pagecache, the page goes to writeback. fuse_writepage_locked fills FUSE_WRITE request and releases the original page by end_page_writeback. 4) fuse_do_setattr() completes and successfully returns. Since now, i_mutex is free. 5) Ordinary write(2) extends i_size back to cover the page. Note that fuse_send_write_pages do wait for fuse writeback, but for another page->index. 6) fuse_writepage_locked proceeds by queueing FUSE_WRITE request. fuse_send_writepage is supposed to crop inarg->size of the request, but it doesn't because i_size has already been extended back. Moving end_page_writeback to the end of fuse_writepage_locked fixes the race because now the fact that truncate_pagecache is successfully returned infers that fuse_writepage_locked has already called end_page_writeback. And this, in turn, infers that fuse_flush_writepages has already called fuse_send_writepage, and the latter used valid (shrunk) i_size. write(2) could not extend it because of i_mutex held by ftruncate(2). Signed-off-by:
-
Anssi Hannula authored
commit 18e39186 upstream. hdmi_channel_allocation() tries to find a HDMI channel allocation that matches the number channels in the playback stream and contains only speakers that the HDMI sink has reported as available via EDID. If no such allocation is found, 0 (stereo audio) is used. Using CA 0 causes the audio causes the sink to discard everything except the first two channels (front left and front right). However, the sink may be capable of receiving more channels than it has speakers (and then perform downmix or discard the extra channels), in which case it is preferable to use a CA that contains extra channels than to use CA 0 which discards all the non-stereo channels. Additionally, it seems that HBR (HD) passthrough output does not work on Intel HDMI codecs when CA is set to 0 (possibly the codec zeroes channels not present in CA). This happens with all receivers that report a 5.1 speaker mask since a HBR stream is carried on 8 channels to the codec. Add a fallback in the CA selection so that the CA channel count at least matches the stream channel count, even if the stream contains channels not present in the sink speaker descriptor. Thanks to GrimGriefer at OpenELEC forums for discovering that changing the sink speaker mask allowed HBR output. Reported-by: GrimGriefer Reported-by: Ashecrow Reported-by:
Frank Zafka <kafkaesque1978@gmail.com> Reported-by:
Peter Frühberger <fritsch@xbmc.org> Signed-off-by:
Anssi Hannula <anssi.hannula@iki.fi> Signed-off-by:
Takashi Iwai <tiwai@suse.de> [bwh: Backported to 3.2: adjust context] Signed-off-by:
-
Alex Deucher authored
commit fb93df1c upstream. The table has the following format: typedef struct _ATOM_SRC_DST_TABLE_FOR_ONE_OBJECT //usSrcDstTableOffset pointing to this structure { UCHAR ucNumberOfSrc; USHORT usSrcObjectID[1]; UCHAR ucNumberOfDst; USHORT usDstObjectID[1]; }ATOM_SRC_DST_TABLE_FOR_ONE_OBJECT; usSrcObjectID[] and usDstObjectID[] are variably sized, so we can't access them directly. Use pointers and update the offset appropriately when accessing the Dst members. Signed-off-by:
Alex Deucher <alexander.deucher@amd.com> Signed-off-by:
-
Alex Deucher authored
commit acf88deb upstream. Setting MC_MISC_CNTL.GART_INDEX_REG_EN causes hangs on some boards on resume. The systems seem to work fine without touching this bit so leave it as is. v2: read-modify-write the GART_INDEX_REG_EN bit. I suspect the problem is that we are losing the other settings in the register. fixes: https://bugs.freedesktop.org/show_bug.cgi?id=52952Reported-by:
Ondrej Zary <linux@rainbow-software.org> Tested-by:
Daniel Tobias <dan.g.tob@gmail.com> Signed-off-by:
-
Alex Deucher authored
commit 95663948 upstream. If the LCD table contains an EDID record, properly account for the edid size when walking through the records. This should fix error messages about unknown LCD records. Signed-off-by:
-
Alex Deucher authored
commit 0b31e023 upstream. We need to allocate line buffer to each display when setting up the watermarks. Failure to do so can lead to a blank screen. This fixes blank screen problems on dce4.1/5 asics. Based on an initial fix from: Jay Cornwall <jay.cornwall@amd.com> Signed-off-by:
-
Alex Deucher authored
commit 118bdbd8 upstream. This LCD monitor (1280x1024 native) has a completely bogus detailed timing (640x350@70hz). User reports that 1280x1024@60 has waves so prefer 1280x1024@75. Manufacturer: MED Model: 7b8 Serial#: 99188 Year: 2005 Week: 5 EDID Version: 1.3 Analog Display Input, Input Voltage Level: 0.700/0.700 V Sync: Separate Max Image Size [cm]: horiz.: 34 vert.: 27 Gamma: 2.50 DPMS capabilities: Off; RGB/Color Display First detailed timing is preferred mode redX: 0.645 redY: 0.348 greenX: 0.280 greenY: 0.605 blueX: 0.142 blueY: 0.071 whiteX: 0.313 whiteY: 0.329 Supported established timings: 720x400@70Hz 640x480@60Hz 640x480@72Hz 640x480@75Hz 800x600@56Hz 800x600@60Hz 800x600@72Hz 800x600@75Hz 1024x768@60Hz 1024x768@70Hz 1024x768@75Hz 1280x1024@75Hz Manufacturer's mask: 0 Supported standard timings: Supported detailed timing: clock: 25.2 MHz Image Size: 337 x 270 mm h_active: 640 h_sync: 688 h_sync_end 784 h_blank_end 800 h_border: 0 v_active: 350 v_sync: 350 v_sync_end 352 v_blanking: 449 v_border: 0 Monitor name: MD30217PG Ranges: V min: 56 V max: 76 Hz, H min: 30 H max: 83 kHz, PixClock max 145 MHz Serial No: 501099188 EDID (in hex): 00ffffffffffff0034a4b80774830100 050f010368221b962a0c55a559479b24 125054afcf00310a0101010101018180 000000000000d60980a0205e63103060 0200510e1100001e000000fc004d4433 3032313750470a202020000000fd0038 4c1e530e000a202020202020000000ff 003530313039393138380a2020200078 Signed-off-by:
-
Alan Stern authored
commit 9d892429 upstream. This patch fixes a build error that occurs when CONFIG_PM is enabled and CONFIG_PM_SLEEP isn't: >> drivers/usb/host/ohci-pci.c:294:10: error: 'usb_hcd_pci_pm_ops' undeclared here (not in a function) .pm = &usb_hcd_pci_pm_ops Since the usb_hcd_pci_pm_ops structure is defined and used when CONFIG_PM is enabled, its declaration should not be protected by CONFIG_PM_SLEEP. Signed-off-by:
Alan Stern <stern@rowland.harvard.edu> Reported-by:
kbuild test robot <fengguang.wu@intel.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by:
-
Grant Likely authored
commit 0640332e upstream. Any calls to dt_alloc() need to be zeroed. This is a temporary fix, but the allocation function itself needs to zero memory before returning it. This is a follow up to patch 9e401275, "of: fdt: fix memory initialization for expanded DT" which fixed one call site but missed another. Signed-off-by:
Grant Likely <grant.likely@linaro.org> Acked-by:
Wladislav Wiebe <wladislav.kw@gmail.com> Signed-off-by:
-
Kees Cook authored
commit 43622021 upstream. The "Report ID" field of a HID report is used to build indexes of reports. The kernel's index of these is limited to 256 entries, so any malicious device that sets a Report ID greater than 255 will trigger memory corruption on the host: [ 1347.156239] BUG: unable to handle kernel paging request at ffff88094958a878 [ 1347.156261] IP: [<ffffffff813e4da0>] hid_register_report+0x2a/0x8b CVE-2013-2888 Signed-off-by:
Kees Cook <keescook@chromium.org> Signed-off-by:
Jiri Kosina <jkosina@suse.cz> [bwh: Backported to 3.2: use dbg_hid() not hid_err()] Signed-off-by:
-
Lan Tianyu authored
commit 524f42fa upstream. The ECDT of ASUSTEK L4R doesn't provide correct command and data I/O ports. The DSDT provides the correct information instead. For this reason, add this machine to quirk list for ECDT validation and use the EC information from the DSDT. [rjw: Changelog] References: https://bugzilla.kernel.org/show_bug.cgi?id=60765Reported-and-tested-by:
Daniele Esposti <expo@expobrain.net> Signed-off-by:
Lan Tianyu <tianyu.lan@intel.com> Signed-off-by:
Rafael J. Wysocki <rafael.j.wysocki@intel.com> [bwh: Backported to 3.2: adjust context] Signed-off-by:
-
Alan Stern authored
commit 69820e01 upstream. Since ohci-hcd supports runtime PM, the .pm field in its pci_driver structure should be protected by CONFIG_PM rather than CONFIG_PM_SLEEP. Without this change, OHCI controllers won't do runtime suspend if system suspend or hibernation isn't enabled. Signed-off-by:
-
Shawn Nematbakhsh authored
commit c8476fb8 upstream. If a USB controller with XHCI_RESET_ON_RESUME goes to runtime suspend, a reset will be performed upon runtime resume. Any previously suspended devices attached to the controller will be re-enumerated at this time. This will cause problems, for example, if an open system call on the device triggered the resume (the open call will fail). Note that this change is only relevant when persist_enabled is not set for USB devices. This patch should be backported to kernels as old as 3.0, that contain the commit c877b3b2 "xhci: Add reset on resume quirk for asrock p67 host". Signed-off-by:
Shawn Nematbakhsh <shawnn@chromium.org> Signed-off-by:
Sarah Sharp <sarah.a.sharp@linux.intel.com> Signed-off-by:
-
Stanislaw Gruszka authored
commit 6e956da2 upstream. We should not do temperature compensation on devices without EXTERNAL_TX_ALC bit set (called DynamicTxAgcControl on vendor driver). Such devices can have totally bogus TSSI parameters on the EEPROM, but still threaded by us as valid and result doing wrong TX power calculations. This fix inability to connect to AP on slightly longer distance on some Ralink chips/devices. Reported-and-tested-by:
Fabien ADAM <id2ndr@crocobox.org> Signed-off-by:
Stanislaw Gruszka <sgruszka@redhat.com> Signed-off-by:
John W. Linville <linville@tuxdriver.com> [bwh: Backported to 3.2: use rt2x00_eeprom_read()] Signed-off-by:
-
Sergei Shtylyov authored
commit f936f9b6 upstream. I'm testing SH-Mobile SDHI driver in DMA mode with a new DMA controller using 'bonnie++' and getting DMA error after which the tmio_mmc_dma.c code falls back to PIO but all commands time out after that. It turned out that the fallback code calls tmio_mmc_enable_dma() with RX/TX channels already freed and pointers to them cleared, so that the function bails out early instead of clearing the DMA bit in the CTL_DMA_ENABLE register. The regression was introduced by commit 162f43e3 (mmc: tmio: fix a deadlock). Moving tmio_mmc_enable_dma() calls to the top of the PIO fallback code in tmio_mmc_start_dma_{rx|tx}() helps. Signed-off-by:
Sergei Shtylyov <sergei.shtylyov@cogentembedded.com> Acked-by:
Guennadi Liakhovetski <g.liakhovetski@gmx.de> Signed-off-by:
Chris Ball <cjb@laptop.org> Signed-off-by:
-
Daniel Mack authored
commit f375fc52 upstream. Commit 7e8d5cd9 ("USB: Add EHCI support for MX27 and MX31 based boards") introduced code that could potentially lead to a NULL pointer dereference on driver removal. Fix this by checking for the value of pdata before dereferencing it. Signed-off-by:
Daniel Mack <zonque@gmail.com> Reported-by:
Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by:
-
Dan Carpenter authored
commit 2c4283ca upstream. In dt282x_ai_insn_read() we call this macro like: wait_for(!mux_busy(), comedi_error(dev, "timeout\n"); return -ETIME;); Because the if statement doesn't have curly braces it means we always return -ETIME and the function never succeeds. Signed-off-by:
Ian Abbott <abbotti@mev.co.uk> Signed-off-by:
-
Johan Hovold authored
commit 3b716caf upstream. Fix endianess bugs in parallel-port code which caused corrupt control-requests to be issued on big-endian machines. Reported-by:
Johan Hovold <jhovold@gmail.com> Signed-off-by:
-
Dan Carpenter authored
commit d0bd9a41 upstream. The write_parport_reg_nonblock() function shouldn't sleep because it's called with spinlocks held. Signed-off-by:
-
Tejun Heo authored
commit c34ac00c upstream. list_first_or_null() should test whether the list is empty and return pointer to the first entry if not in a RCU safe manner. It's broken in several ways. * It compares __kernel @__ptr with __rcu @__next triggering the following sparse warning. net/core/dev.c:4331:17: error: incompatible types in comparison expression (different address spaces) * It doesn't perform rcu_dereference*() and computes the entry address using container_of() directly from the __rcu pointer which is inconsitent with other rculist interface. As a result, all three in-kernel users - net/core/dev.c, macvlan, cgroup - are buggy. They dereference the pointer w/o going through read barrier. * While ->next dereference passes through list_next_rcu(), the compiler is still free to fetch ->next more than once and thus nullify the "__ptr != __next" condition check. Fix it by making list_first_or_null_rcu() dereference ->next directly using ACCESS_ONCE() and then use list_entry_rcu() on it like other rculist accessors. v2: Paul pointed out that the compiler may fetch the pointer more than once nullifying the condition check. ACCESS_ONCE() added on ->next dereference. v3: Restored () around macro param which was accidentally removed. Spotted by Paul. Signed-off-by:
Tejun Heo <tj@kernel.org> Reported-by:
Paul E. McKenney <paulmck@linux.vnet.ibm.com> Reviewed-by:
Josh Triplett <josh@joshtriplett.org> Signed-off-by:
-
Mike Dyer authored
commit 85fa532b upstream. Bit 9 of PLL2,3 and 4 is reserved as '0'. The 24bit fractional part should be split across each register in 8bit chunks. Signed-off-by:
Mike Dyer <mike.dyer@md-soft.co.uk> Signed-off-by:
Mark Brown <broonie@linaro.org> Signed-off-by:
-
Felix Fietkau authored
commit a1c781bb upstream. They are not implemented, and accessing them might trigger errors Signed-off-by:
Felix Fietkau <nbd@openwrt.org> Signed-off-by:
-
Felix Fietkau authored
commit e96542e5 upstream. Similar to a race condition that exists in the tx path, the hardware might re-read the 'next' pointer of a descriptor of the last completed frame. This only affects non-EDMA (pre-AR93xx) devices. To deal with this race, defer clearing and re-linking a completed rx descriptor until the next one has been processed. Signed-off-by:
-
Alex Williamson authored
commit 3269ee0b upstream. At best the current code only seems to free the leaf pagetables and the root. If you're unlucky enough to have a large gap (like any QEMU guest with more than 3G of memory), only the first chunk of leaf pagetables are freed (plus the root). This is a massive memory leak. This patch re-writes the pagetable freeing function to use a recursive algorithm and manages to not only free all the pagetables, but does it without any apparent performance loss versus the current broken version. Signed-off-by:
Alex Williamson <alex.williamson@redhat.com> Reviewed-by:
Marcelo Tosatti <mtosatti@redhat.com> Signed-off-by:
Joerg Roedel <joro@8bytes.org> Signed-off-by:
-
Anton Blanchard authored
commit 230aef7a upstream. Normally when we haven't implemented an alignment handler for a load or store instruction the process will be terminated. The alignment handler uses the DSISR (or a pseudo one) to locate the right handler. Unfortunately ldbrx and stdbrx overlap lfs and stfs so we incorrectly think ldbrx is an lfs and stdbrx is an stfs. This bug is particularly nasty - instead of terminating the process we apply an incorrect fixup and continue on. With more and more overlapping instructions we should stop creating a pseudo DSISR and index using the instruction directly, but for now add a special case to catch ldbrx/stdbrx. Signed-off-by:
Anton Blanchard <anton@samba.org> Signed-off-by:
Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by:
-
Oliver Neukum authored
commit 6dd433e6 upstream. Both could want to submit the same URB. Some checks of the flag intended to prevent that were missing. Signed-off-by:
Oliver Neukum <oneukum@suse.de> Signed-off-by:
-
Hans de Goede authored
commit b4f17a48 upstream. While reading the config parsing code I noticed this check is missing, without this check config->desc.wTotalLength can end up with a value larger then the dev->rawdescriptors length for the config, and when userspace then tries to get the rawdescriptors bad things may happen. Signed-off-by:
Hans de Goede <hdegoede@redhat.com> Signed-off-by:
-
majianpeng authored
commit 73d9f7ee upstream. For nofail == false request, if __map_request failed, the caller does cleanup work, like releasing the relative pages. It doesn't make any sense to retry this request. Signed-off-by:
Jianpeng Ma <majianpeng@gmail.com> Reviewed-by:
Sage Weil <sage@inktank.com> [bwh: Backported to 3.2: adjust indentation] Signed-off-by:
-
Felix Fietkau authored
commit 026d5b07 upstream. Otherwise in some cases, EAPOL frames might be filtered during the initial handshake, causing delays and assoc failures. Signed-off-by:
-
Roger Pau Monne authored
commit 5f338d90 upstream. With the current implementation, the callback in the tail of the list can be added twice, because the check done in gnttab_request_free_callback is bogus, callback->next can be NULL if it is the last callback in the list. If we add the same callback twice we end up with an infinite loop, were callback == callback->next. Replace this check with a proper one that iterates over the list to see if the callback has already been added. Signed-off-by:
Roger Pau Monné <roger.pau@citrix.com> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: David Vrabel <david.vrabel@citrix.com> Signed-off-by:
Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Acked-by:
Matt Wilson <msw@amazon.com> Reviewed-by:
David Vrabel <david.vrabel@citrix.com> Signed-off-by:
-
Manoj Chourasia authored
commit 212a871a upstream. This changes puts the commit 4fe9f8e2 back in place with the fixes for slab corruption because of the commit. When a device is unplugged, wait for all processes that have opened the device to close before deallocating the device. This commit was solving kernel crash because of the corruption in rb tree of vmalloc. The rootcause was the device data pointer was geting excessed after the memory associated with hidraw was freed. The commit 4fe9f8e2 was buggy as it was also freeing the hidraw first and then calling delete operation on the list associated with that hidraw leading to slab corruption. Signed-off-by:
Manoj Chourasia <mchourasia@nvidia.com> Tested-by:
Peter Wu <lekensteyn@gmail.com> Signed-off-by:
-
Jiri Kosina authored
commit df0cfd69 upstream. This basically reverts commit 4fe9f8e2. It causes multiple problems, namely: - after rmmod/modprobe cycle of bus driver, the input is not claimed any more. This is likely because of misplaced hid_hw_close() - it causes memory corruption on hidraw_list As original patch author is not responding to requests to fix his patch, and the original deallocation mechanism is not exposing any problems, I am reverting back to it. Signed-off-by:
-
Eric Dumazet authored
[ Upstream commit 55432d2b ] commit 5faa5df1 (inetpeer: Invalidate the inetpeer tree along with the routing cache) added a race : Before freeing an inetpeer, we must respect a RCU grace period, and make sure no user will attempt to increase refcnt. inetpeer_invalidate_tree() waits for a RCU grace period before inserting inetpeer tree into gc_list and waking the worker. At that time, no concurrent lookup can find a inetpeer in this tree. Signed-off-by:
Eric Dumazet <edumazet@google.com> Cc: Steffen Klassert <steffen.klassert@secunet.com> Acked-by:
Steffen Klassert <steffen.klassert@secunet.com> Signed-off-by:
David S. Miller <davem@davemloft.net> Signed-off-by:
-
Steffen Klassert authored
[ Upstream commit 5faa5df1 ] We initialize the routing metrics with the values cached on the inetpeer in rt_init_metrics(). So if we have the metrics cached on the inetpeer, we ignore the user configured fib_metrics. To fix this issue, we replace the old tree with a fresh initialized inet_peer_base. The old tree is removed later with a delayed work queue. Signed-off-by:
-
Ying Xue authored
[ Upstream commit 4225a398 ] When the lockdep validator is enabled, it will report the below warning when we enable a TIPC bearer: [ INFO: possible irq lock inversion dependency detected ] --------------------------------------------------------- Possible interrupt unsafe locking scenario: CPU0 CPU1 ---- ---- lock(ptype_lock); local_irq_disable(); lock(tipc_net_lock); lock(ptype_lock); <Interrupt> lock(tipc_net_lock); *** DEADLOCK *** the shortest dependencies between 2nd lock and 1st lock: -> (ptype_lock){+.+...} ops: 10 { [...] SOFTIRQ-ON-W at: [<c1089418>] __lock_acquire+0x528/0x13e0 [<c108a360>] lock_acquire+0x90/0x100 [<c1553c38>] _raw_spin_lock+0x38/0x50 [<c14651ca>] dev_add_pack+0x3a/0x60 [<c182da75>] arp_init+0x1a/0x48 [<c182dce5>] inet_init+0x181/0x27e [<c1001114>] do_one_initcall+0x34/0x170 [<c17f7329>] kernel_init+0x110/0x1b2 [<c155b6a2>] kernel_thread_helper+0x6/0x10 [...] ... key at: [<c17e4b10>] ptype_lock+0x10/0x20 ... acquired at: [<c108a360>] lock_acquire+0x90/0x100 [<c1553c38>] _raw_spin_lock+0x38/0x50 [<c14651ca>] dev_add_pack+0x3a/0x60 [<c8bc18d2>] enable_bearer+0xf2/0x140 [tipc] [<c8bb283a>] tipc_enable_bearer+0x1ba/0x450 [tipc] [<c8bb3a04>] tipc_cfg_do_cmd+0x5c4/0x830 [tipc] [<c8bbc032>] handle_cmd+0x42/0xd0 [tipc] [<c148e802>] genl_rcv_msg+0x232/0x280 [<c148d3f6>] netlink_rcv_skb+0x86/0xb0 [<c148e5bc>] genl_rcv+0x1c/0x30 [<c148d144>] netlink_unicast+0x174/0x1f0 [<c148ddab>] netlink_sendmsg+0x1eb/0x2d0 [<c1456bc1>] sock_aio_write+0x161/0x170 [<c1135a7c>] do_sync_write+0xac/0xf0 [<c11360f6>] vfs_write+0x156/0x170 [<c11361e2>] sys_write+0x42/0x70 [<c155b0df>] sysenter_do_call+0x12/0x38 [...] } -> (tipc_net_lock){+..-..} ops: 4 { [...] IN-SOFTIRQ-R at: [<c108953a>] __lock_acquire+0x64a/0x13e0 [<c108a360>] lock_acquire+0x90/0x100 [<c15541cd>] _raw_read_lock_bh+0x3d/0x50 [<c8bb874d>] tipc_recv_msg+0x1d/0x830 [tipc] [<c8bc195f>] recv_msg+0x3f/0x50 [tipc] [<c146a5fa>] __netif_receive_skb+0x22a/0x590 [<c146ab0b>] netif_receive_skb+0x2b/0xf0 [<c13c43d2>] pcnet32_poll+0x292/0x780 [<c146b00a>] net_rx_action+0xfa/0x1e0 [<c103a4be>] __do_softirq+0xae/0x1e0 [...] } >From the log, we can see three different call chains between CPU0 and CPU1: Time 0 on CPU0: kernel_init()->inet_init()->dev_add_pack() At time 0, the ptype_lock is held by CPU0 in dev_add_pack(); Time 1 on CPU1: tipc_enable_bearer()->enable_bearer()->dev_add_pack() At time 1, tipc_enable_bearer() first holds tipc_net_lock, and then wants to take ptype_lock to register TIPC protocol handler into the networking stack. But the ptype_lock has been taken by dev_add_pack() on CPU0, so at this time the dev_add_pack() running on CPU1 has to be busy looping. Time 2 on CPU0: netif_receive_skb()->recv_msg()->tipc_recv_msg() At time 2, an incoming TIPC packet arrives at CPU0, hence tipc_recv_msg() will be invoked. In tipc_recv_msg(), it first wants to hold tipc_net_lock. At the moment, below scenario happens: On CPU0, below is our sequence of taking locks: lock(ptype_lock)->lock(tipc_net_lock) On CPU1, our sequence of taking locks looks like: lock(tipc_net_lock)->lock(ptype_lock) Obviously deadlock may happen in this case. But please note the deadlock possibly doesn't occur at all when the first TIPC bearer is enabled. Before enable_bearer() -- running on CPU1 does not hold ptype_lock, so the TIPC receive handler (i.e. recv_msg()) is not registered successfully via dev_add_pack(), so the tipc_recv_msg() cannot be called by recv_msg() even if a TIPC message comes to CPU0. But when the second TIPC bearer is registered, the deadlock can perhaps really happen. To fix it, we will push the work of registering TIPC protocol handler into workqueue context. After the change, both paths taking ptype_lock are always in process contexts, thus, the deadlock should never occur. Signed-off-by:
Ying Xue <ying.xue@windriver.com> Signed-off-by:
Jon Maloy <jon.maloy@ericsson.com> Signed-off-by:
Paul Gortmaker <paul.gortmaker@windriver.com> Signed-off-by:
-
