kernel/linux.git/drivers/misc/mic, branch linux-4.20.y

mic: vop: Fix use-after-free on remove

2019-02-15T07:11:06+00:00

commit 70ed7148dadb812f2f7c9927e98ef3cf4869dfa9 upstream. KASAN detects a use-after-free when vop devices are removed. This problem was introduced by commit 0063e8bbd2b62d136 ("virtio_vop: don't kfree device on register failure"). That patch moved the freeing of the struct _vop_vdev to the release function, but failed to ensure that vop holds a reference to the device when it doesn't want it to go away. A kfree() was replaced with a put_device() in the unregistration path, but the last reference to the device is already dropped in unregister_virtio_device() so the struct is freed before vop is done with it. Fix it by holding a reference until cleanup is done. This is similar to the fix in virtio_pci in commit 2989be09a8a9d6 ("virtio_pci: fix use after free on release"). ================================================================== BUG: KASAN: use-after-free in vop_scan_devices+0xc6c/0xe50 [vop] Read of size 8 at addr ffff88800da18580 by task kworker/0:1/12 CPU: 0 PID: 12 Comm: kworker/0:1 Not tainted 5.0.0-rc4+ #53 Workqueue: events vop_hotplug_devices [vop] Call Trace: dump_stack+0x74/0xbb print_address_description+0x5d/0x2b0 ? vop_scan_devices+0xc6c/0xe50 [vop] kasan_report+0x152/0x1aa ? vop_scan_devices+0xc6c/0xe50 [vop] ? vop_scan_devices+0xc6c/0xe50 [vop] vop_scan_devices+0xc6c/0xe50 [vop] ? vop_loopback_free_irq+0x160/0x160 [vop_loopback] process_one_work+0x7c0/0x14b0 ? pwq_dec_nr_in_flight+0x2d0/0x2d0 ? do_raw_spin_lock+0x120/0x280 worker_thread+0x8f/0xbf0 ? __kthread_parkme+0x78/0xf0 ? process_one_work+0x14b0/0x14b0 kthread+0x2ae/0x3a0 ? kthread_park+0x120/0x120 ret_from_fork+0x3a/0x50 Allocated by task 12: kmem_cache_alloc_trace+0x13a/0x2a0 vop_scan_devices+0x473/0xe50 [vop] process_one_work+0x7c0/0x14b0 worker_thread+0x8f/0xbf0 kthread+0x2ae/0x3a0 ret_from_fork+0x3a/0x50 Freed by task 12: kfree+0x104/0x310 device_release+0x73/0x1d0 kobject_put+0x14f/0x420 unregister_virtio_device+0x32/0x50 vop_scan_devices+0x19d/0xe50 [vop] process_one_work+0x7c0/0x14b0 worker_thread+0x8f/0xbf0 kthread+0x2ae/0x3a0 ret_from_fork+0x3a/0x50 The buggy address belongs to the object at ffff88800da18008 which belongs to the cache kmalloc-2k of size 2048 The buggy address is located 1400 bytes inside of 2048-byte region [ffff88800da18008, ffff88800da18808) The buggy address belongs to the page: page:ffffea0000368600 count:1 mapcount:0 mapping:ffff88801440dbc0 index:0x0 compound_mapcount: 0 flags: 0x4000000000010200(slab|head) raw: 4000000000010200 ffffea0000378608 ffffea000037a008 ffff88801440dbc0 raw: 0000000000000000 00000000000d000d 00000001ffffffff 0000000000000000 page dumped because: kasan: bad access detected Memory state around the buggy address: ffff88800da18480: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ffff88800da18500: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb >ffff88800da18580: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ^ ffff88800da18600: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ffff88800da18680: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ================================================================== Fixes: 0063e8bbd2b62d136 ("virtio_vop: don't kfree device on register failure") Signed-off-by: Vincent Whitchurch Cc: stable Signed-off-by: Greg Kroah-Hartman

virtio: don't allocate vqs when names[i] = NULL

2019-01-22T20:09:52+00:00

commit a229989d975eb926076307c1f2f5e4c6111768e7 upstream. Some vqs may not need to be allocated when their related feature bits are disabled. So callers may pass in such vqs with "names = NULL". Then we skip such vq allocations. Signed-off-by: Wei Wang Signed-off-by: Michael S. Tsirkin Signed-off-by: Wei Wang Signed-off-by: Wei Wang Reviewed-by: Cornelia Huck Cc: stable@vger.kernel.org Fixes: 86a559787e6f ("virtio-balloon: VIRTIO_BALLOON_F_FREE_PAGE_HINT") Signed-off-by: Greg Kroah-Hartman

misc: mic/scif: fix copy-paste error in scif_create_remote_lookup

2018-11-27T08:00:38+00:00

gcc '-Wunused-but-set-variable' warning: drivers/misc/mic/scif/scif_rma.c: In function 'scif_create_remote_lookup': drivers/misc/mic/scif/scif_rma.c:373:25: warning: variable 'vmalloc_num_pages' set but not used [-Wunused-but-set-variable] 'vmalloc_num_pages' should be used to determine if the address is within the vmalloc range. Fixes: ba612aa8b487 ("misc: mic: SCIF memory registration and unregistration") Signed-off-by: YueHaibing Cc: stable Signed-off-by: Greg Kroah-Hartman

misc: mic: scif: remove set but not used variables 'src_dma_addr, dst_dma_addr'

2018-10-15T18:47:40+00:00

Fixes gcc '-Wunused-but-set-variable' warning: drivers/misc/mic/scif/scif_dma.c: In function 'scif_rma_list_dma_copy_wrapper': drivers/misc/mic/scif/scif_dma.c:1558:27: warning: variable 'dst_dma_addr' set but not used [-Wunused-but-set-variable] drivers/misc/mic/scif/scif_dma.c:1558:13: warning: variable 'src_dma_addr' set but not used [-Wunused-but-set-variable] They never used since introduction in commit 7cc31cd27752 ("misc: mic: SCIF DMA and CPU copy interface") Signed-off-by: YueHaibing Signed-off-by: Greg Kroah-Hartman

misc: mic: fix a DMA pool free failure

2018-10-15T18:47:40+00:00

In _scif_prog_signal(), the boolean variable 'x100' is used to indicate whether the MIC Coprocessor is X100. If 'x100' is true, the status descriptor will be used to write the value to the destination. Otherwise, a DMA pool will be allocated for this purpose. Specifically, if the DMA pool is allocated successfully, two memory addresses will be returned. One is for the CPU and the other is for the device to access the DMA pool. The former is stored to the variable 'status' and the latter is stored to the variable 'src'. After the allocation, the address in 'src' is saved to 'status->src_dma_addr', which is actually in the DMA pool, and 'src' is then modified. Later on, if an error occurs, the execution flow will transfer to the label 'dma_fail', which will check 'x100' and free up the allocated DMA pool if 'x100' is false. The point here is that 'status->src_dma_addr' is used for freeing up the DMA pool. As mentioned before, 'status->src_dma_addr' is in the DMA pool. And thus, the device is able to modify this data. This can potentially cause failures when freeing up the DMA pool because of the modified device address. This patch avoids the above issue by using the variable 'src' (with necessary calculation) to free up the DMA pool. Signed-off-by: Wenwen Wang Signed-off-by: Greg Kroah-Hartman

misc: mic: scif: remove redundant check on ret < 0

2018-10-11T10:12:55+00:00

The check for ret < 0 is redundant as any places prior to this point where ret is set to an error value the code will exit out of the loop to the error exit label 'err'. Remove this redundant dead code. Detected by CoverityScan, CID#1339528 ("Logically dead code") Signed-off-by: Colin Ian King Signed-off-by: Greg Kroah-Hartman

misc: mic: scif: Remove unused variable

2018-09-25T18:21:02+00:00

Clang warns when a variable is assigned to itself. drivers/misc/mic/scif/scif_dma.c:1577:12: warning: explicitly assigning value of variable of type 'bool' (aka '_Bool') to itself [-Wself-assign] dst_local = dst_local; ~~~~~~~~~ ^ ~~~~~~~~~ 1 warning generated. This is usually done to avoid an unused variable warning, which is the case here. dst_local is used nowhere in this function, which has been the case since the initial code drop in commit 7cc31cd27752 ("misc: mic: SCIF DMA and CPU copy interface") in 2015. Just remove the variable, it can be added back if it was intended to be used. Link: https://github.com/ClangBuiltLinux/linux/issues/107 Signed-off-by: Nathan Chancellor Signed-off-by: Greg Kroah-Hartman

mm, oom: distinguish blockable mode for mmu notifiers

2018-08-22T17:52:44+00:00

There are several blockable mmu notifiers which might sleep in mmu_notifier_invalidate_range_start and that is a problem for the oom_reaper because it needs to guarantee a forward progress so it cannot depend on any sleepable locks. Currently we simply back off and mark an oom victim with blockable mmu notifiers as done after a short sleep. That can result in selecting a new oom victim prematurely because the previous one still hasn't torn its memory down yet. We can do much better though. Even if mmu notifiers use sleepable locks there is no reason to automatically assume those locks are held. Moreover majority of notifiers only care about a portion of the address space and there is absolutely zero reason to fail when we are unmapping an unrelated range. Many notifiers do really block and wait for HW which is harder to handle and we have to bail out though. This patch handles the low hanging fruit. __mmu_notifier_invalidate_range_start gets a blockable flag and callbacks are not allowed to sleep if the flag is set to false. This is achieved by using trylock instead of the sleepable lock for most callbacks and continue as long as we do not block down the call chain. I think we can improve that even further because there is a common pattern to do a range lookup first and then do something about that. The first part can be done without a sleeping lock in most cases AFAICS. The oom_reaper end then simply retries if there is at least one notifier which couldn't make any progress in !blockable mode. A retry loop is already implemented to wait for the mmap_sem and this is basically the same thing. The simplest way for driver developers to test this code path is to wrap userspace code which uses these notifiers into a memcg and set the hard limit to hit the oom. This can be done e.g. after the test faults in all the mmu notifier managed memory and set the hard limit to something really small. Then we are looking for a proper process tear down. [akpm@linux-foundation.org: coding style fixes] [akpm@linux-foundation.org: minor code simplification] Link: http://lkml.kernel.org/r/20180716115058.5559-1-mhocko@kernel.org Signed-off-by: Michal Hocko Acked-by: Christian König # AMD notifiers Acked-by: Leon Romanovsky # mlx and umem_odp Reported-by: David Rientjes Cc: "David (ChunMing) Zhou" Cc: Paolo Bonzini Cc: Alex Deucher Cc: David Airlie Cc: Jani Nikula Cc: Joonas Lahtinen Cc: Rodrigo Vivi Cc: Doug Ledford Cc: Jason Gunthorpe Cc: Mike Marciniszyn Cc: Dennis Dalessandro Cc: Sudeep Dutt Cc: Ashutosh Dixit Cc: Dimitri Sivanich Cc: Boris Ostrovsky Cc: Juergen Gross Cc: "Jérôme Glisse" Cc: Andrea Arcangeli Cc: Felix Kuehling Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

misc: mic: SCIF Fix scif_get_new_port() error handling

2018-08-05T14:29:11+00:00

There are only 2 callers of scif_get_new_port() and both appear to get the error handling wrong. Both treat zero returns as error, but it actually returns negative error codes and >= 0 on success. Fixes: e9089f43c9a7 ("misc: mic: SCIF open close bind and listen APIs") Signed-off-by: Dan Carpenter Signed-off-by: Greg Kroah-Hartman

misc: mic: fix passing the current time

2018-07-07T15:44:52+00:00

I noticed that the mic driver passes a 'struct timespec64' as part of a message into an attached device, where it is used to set the current system time. This won't actually work if one of the two sides runs a 32-bit kernel and the other runs a 64-bit kernel, since the structure layout is different between the two. I found this while replacing calls to the deprecated do_settimeofday64() interface with the modern ktime_get_real_ts() variant, but it seems appropriate to address both at the same time here. To make sure we have a sane structure, let's define our own structure using the layout of the 64-bit kernel. Signed-off-by: Arnd Bergmann Signed-off-by: Greg Kroah-Hartman