kernel.git/drivers/dma/ioat/init.c, branch linux-4.9.y

dmaengine: ioat: fix prototype of ioat_enumerate_channels

2019-11-25T08:53:35Z

[ Upstream commit f4d34aa8c887a8a2d23ef546da0efa10e3f77241 ] Signed-off-by: Rami Rosen Signed-off-by: Vinod Koul Signed-off-by: Sasha Levin

driver/dma/ioat: Call del_timer_sync() without holding prep_lock

2018-11-13T19:16:53Z

[ Upstream commit cfb03be6c7e8a1591285849c361d67b09f5149f7 ] The following lockdep splat was observed: [ 1222.241750] ====================================================== [ 1222.271301] WARNING: possible circular locking dependency detected [ 1222.301060] 4.16.0-10.el8+5.x86_64+debug #1 Not tainted [ 1222.326659] ------------------------------------------------------ [ 1222.356565] systemd-shutdow/1 is trying to acquire lock: [ 1222.382660] ((&ioat_chan->timer)){+.-.}, at: [<00000000f71e1a28>] del_timer_sync+0x5/0xf0 [ 1222.422928] [ 1222.422928] but task is already holding lock: [ 1222.451743] (&(&ioat_chan->prep_lock)->rlock){+.-.}, at: [<000000008ea98b12>] ioat_shutdown+0x86/0x100 [ioatdma] : [ 1223.524987] Chain exists of: [ 1223.524987] (&ioat_chan->timer) --> &(&ioat_chan->cleanup_lock)->rlock --> &(&ioat_chan->prep_lock)->rlock [ 1223.524987] [ 1223.594082] Possible unsafe locking scenario: [ 1223.594082] [ 1223.622630] CPU0 CPU1 [ 1223.645080] ---- ---- [ 1223.667404] lock(&(&ioat_chan->prep_lock)->rlock); [ 1223.691535] lock(&(&ioat_chan->cleanup_lock)->rlock); [ 1223.728657] lock(&(&ioat_chan->prep_lock)->rlock); [ 1223.765122] lock((&ioat_chan->timer)); [ 1223.784095] [ 1223.784095] *** DEADLOCK *** [ 1223.784095] [ 1223.813492] 4 locks held by systemd-shutdow/1: [ 1223.834677] #0: (reboot_mutex){+.+.}, at: [<0000000056d33456>] SYSC_reboot+0x10f/0x300 [ 1223.873310] #1: (&dev->mutex){....}, at: [<00000000258dfdd7>] device_shutdown+0x1c8/0x660 [ 1223.913604] #2: (&dev->mutex){....}, at: [<0000000068331147>] device_shutdown+0x1d6/0x660 [ 1223.954000] #3: (&(&ioat_chan->prep_lock)->rlock){+.-.}, at: [<000000008ea98b12>] ioat_shutdown+0x86/0x100 [ioatdma] In the ioat_shutdown() function: spin_lock_bh(&ioat_chan->prep_lock); set_bit(IOAT_CHAN_DOWN, &ioat_chan->state); del_timer_sync(&ioat_chan->timer); spin_unlock_bh(&ioat_chan->prep_lock); According to the synchronization rule for the del_timer_sync() function, the caller must not hold locks which would prevent completion of the timer's handler. The timer structure has its own lock that manages its synchronization. Setting the IOAT_CHAN_DOWN bit should prevent other CPUs from trying to use that device anyway, there is probably no need to call del_timer_sync() while holding the prep_lock. So the del_timer_sync() call is now moved outside of the prep_lock critical section to prevent the circular lock dependency. Signed-off-by: Waiman Long Reviewed-by: Dave Jiang Signed-off-by: Vinod Koul Signed-off-by: Sasha Levin Signed-off-by: Greg Kroah-Hartman

dmaengine: ioat: Fix error handling path

2018-02-25T10:05:48Z

[ Upstream commit 5c9afbda911ce20b3f2181d1e440a0222e1027dd ] If the last test in 'ioat_dma_self_test()' fails, we must release all the allocated resources and not just part of them. Signed-off-by: Christophe JAILLET Acked-by: Dave Jiang Signed-off-by: Vinod Koul Signed-off-by: Sasha Levin Signed-off-by: Greg Kroah-Hartman

dmaengine: ioatdma: workaround SKX ioatdma version

2017-08-07T01:59:46Z

[ Upstream commit 34a31f0af84158955a9747fb5c6712da5bbb5331 ] The Skylake ioatdma is technically CBDMA 3.2+ and contains the same hardware bits with some additional 3.3 features, but it's not really 3.3 where the driver is concerned. Signed-off-by: Dave Jiang Signed-off-by: Vinod Koul Signed-off-by: Sasha Levin Signed-off-by: Greg Kroah-Hartman

dmaengine: ioatdma: Add Skylake PCI Dev ID

2017-08-07T01:59:46Z

[ Upstream commit 1594c18fd297a8edcc72bc4b161f3f52603ebb92 ] Adding Skylake Xeon PCI device ids for ioatdma and related bits. Signed-off-by: Dave Jiang Signed-off-by: Vinod Koul Signed-off-by: Sasha Levin Signed-off-by: Greg Kroah-Hartman

dmaengine: iota: ioat_alloc_chan_resources should not perform sleeping allocations.

2017-03-22T11:43:35Z

commit 21d25f6a4217e755906cb548b55ddab39d0e88b9 upstream. On a kernel with DEBUG_LOCKS, ioat_free_chan_resources triggers an in_interrupt() warning. With PROVE_LOCKING, it reports detecting a SOFTIRQ-safe to SOFTIRQ-unsafe lock ordering in the same code path. This is because dma_generic_alloc_coherent() checks if the GFP flags permit blocking. It allocates from different subsystems if blocking is permitted. The free path knows how to return the memory to the correct allocator. If GFP_KERNEL is specified then the alloc and free end up going through cma_alloc(), which uses mutexes. Given that ioat_free_chan_resources() can be called in interrupt context, ioat_alloc_chan_resources() must specify GFP_NOWAIT so that the allocations do not block and instead use an allocator that uses spinlocks. Signed-off-by: Krister Johansen Acked-by: Dave Jiang Signed-off-by: Vinod Koul Signed-off-by: Greg Kroah-Hartman

dmaengine: ioatdma: fix uninitialized array usage

2016-08-08T02:32:00Z

Static analysis showed that unitialized array is being used for compare. At line 850 when a dma_mapping_error() occurs, it jumps to dma_unmap. At this point, dma_srcs has not been initialized. However, the code after dma_unmap label checks dma_srcs for a comparison and thus is comparing to random garbage in the array. Given that when dest_dma is being mapped this is the first instance of mapping DMA memory and failed, there is really nothing to be cleaned up and thus should jump to free_resources label instead. Signed-off-by: Dave Jiang Signed-off-by: Vinod Koul

dmaengine: ioat: statify symbol

2016-07-23T10:37:29Z

Sparse warns: drivers/dma/ioat/init.c:1215:6: warning: symbol 'ioat_resume' was not declared. Should it be static? Signed-off-by: Vinod Koul Acked-by: Dave Jiang

dmaengine: ioatdma: disable relaxed ordering for ioatdma

2016-05-14T08:06:52Z

ioatdma by default is in snoop mode. Relaxed ordering according to spec does not do anything in snoop mode. However, it causes hang or significant performance degrade when tested with NTB. Disabling in the driver due to some BIOS do not configure it correctly. Signed-off-by: Dave Jiang Signed-off-by: Vinod Koul

dmaengine: ioatdma: Use dma_pool_zalloc

2016-05-03T06:53:12Z

Dma_pool_zalloc combines dma_pool_alloc and memset 0. The semantic patch that makes this transformation is as follows: (http://coccinelle.lip6.fr/) // @@ expression d,e; statement S; @@ d = - dma_pool_alloc + dma_pool_zalloc (...); if (!d) S - memset(d, 0, sizeof(*d)); // Signed-off-by: Julia Lawall Signed-off-by: Vinod Koul