linux/drivers/infiniband/core, branch v3.4.76

infiniband: pass rdma_cm module to netlink_dump_start

2012-10-28T17:14:15Z

[ Upstream commit 809d5fc9bf6589276a12bd4fd611e4c7ff9940c3 ] set netlink_dump_control.module to avoid panic. Signed-off-by: Gao feng Cc: Roland Dreier Cc: Sean Hefty Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman

IB/core: Fix mismatch between locked and pinned pages

2012-06-01T07:18:17Z

commit c4870eb874ac16dccef40e1bc7a002c7e9156adc upstream. Commit bc3e53f682d9 ("mm: distinguish between mlocked and pinned pages") introduced a separate counter for pinned pages and used it in the IB stack. However, in ib_umem_get() the pinned counter is incremented, but ib_umem_release() wrongly decrements the locked counter. Fix this. Signed-off-by: Yishai Hadas Reviewed-by: Christoph Lameter Signed-off-by: Roland Dreier Signed-off-by: Greg Kroah-Hartman

Merge tag 'ib-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband

2012-04-26T22:35:35Z

Pull infiniband fixes from Roland Dreier: "A few fixes for regressions introduced in 3.4-rc1: - fix memory leak in mlx4 - fix two problems with new MAD response generation code" * tag 'ib-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband: IB/mlx4: Fix memory leaks in ib_link_query_port() IB/mad: Don't send response for failed MADs IB/mad: Set 'D' bit in response for unhandled MADs

IB/mad: Don't send response for failed MADs

2012-04-24T23:08:57Z

Commit 0b307043049f ("IB/mad: Return error response for unsupported MADs") does not failed MADs (eg those that return IB_MAD_RESULT_FAILURE) properly -- these MADs should be silently discarded. (We should not force the lower-layer drivers to return SUCCESS | CONSUMED in this case, since the MAD is NOT successful). Unsupported MADs are not failures -- they return SUCCESS, but with an "unsupported error" status value inside the response MAD. Reviewed-by: Hal Rosenstock Signed-off-by: Jack Morgenstein Signed-off-by: Roland Dreier

IB/mad: Set 'D' bit in response for unhandled MADs

2012-04-24T23:06:50Z

Commit 0b307043049f ("IB/mad: Return error response for unsupported MADs") does not handle directed-route MADs properly -- it fails to set the 'D' bit in the response MAD status field. This is a problem for SmInfo MADs when the receiver does not have an SM running. Reviewed-by: Hal Rosenstock Signed-off-by: Jack Morgenstein Signed-off-by: Roland Dreier

IB/core: Don't return EINVAL from sysfs rate attribute for invalid speeds

2012-04-02T17:57:31Z

Commit e9319b0cb00d ("IB/core: Fix SDR rates in sysfs") changed our sysfs rate attribute to return EINVAL to userspace if the underlying device driver returns an invalid rate. Apparently some drivers do this when the link is down and some userspace pukes if it gets an error when reading this attribute, so avoid a regression by not return an error to match the old code. Signed-off-by: Roland Dreier

Merge tag 'rdma-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband

2012-03-21T17:33:42Z

Pull InfiniBand/RDMA changes for the 3.4 merge window from Roland Dreier: "Nothing big really stands out; by patch count lots of fixes to the mlx4 driver plus some cleanups and fixes to the core and other drivers." * tag 'rdma-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband: (28 commits) mlx4_core: Scale size of MTT table with system RAM mlx4_core: Allow dynamic MTU configuration for IB ports IB/mlx4: Fix info returned when querying IBoE ports IB/mlx4: Fix possible missed completion event mlx4_core: Report thermal error events mlx4_core: Fix one more static exported function IB: Change CQE "csum_ok" field to a bit flag RDMA/iwcm: Reject connect requests if cmid is not in LISTEN state RDMA/cxgb3: Don't pass irq flags to flush_qp() mlx4_core: Get rid of redundant ext_port_cap flags RDMA/ucma: Fix AB-BA deadlock IB/ehca: Fix ilog2() compile failure IB: Use central enum for speed instead of hard-coded values IB/iser: Post initial receive buffers before sending the final login request IB/iser: Free IB connection resources in the proper place IB/srp: Consolidate repetitive sysfs code IB/srp: Use pr_fmt() and pr_err()/pr_warn() IB/core: Fix SDR rates in sysfs mlx4: Enforce device max FMR maps in FMR alloc IB/mlx4: Set bad_wr for invalid send opcode ...

Merge branches 'cma', 'cxgb3', 'cxgb4', 'ehca', 'iser', 'mad', 'nes', 'qib', 'srp' and 'srpt' into for-next

2012-03-19T16:50:33Z

RDMA/iwcm: Reject connect requests if cmid is not in LISTEN state

2012-03-07T23:14:53Z

When destroying a listening cmid, the iwcm first marks the state of the cmid as DESTROYING, then releases the lock and calls into the iWARP provider to destroy the endpoint. Since the cmid is not locked, its possible for the iWARP provider to pass a connection request event to the iwcm, which will be silently dropped by the iwcm. This causes the iWARP provider to never free up the resources from this connection because the assumption is the iwcm will accept or reject this connection. The solution is to reject these connection requests. Signed-off-by: Steve Wise Signed-off-by: Roland Dreier

RDMA/ucma: Fix AB-BA deadlock

2012-03-05T20:27:57Z

When we destroy a cm_id, we must purge associated events from the event queue. If the cm_id is for a listen request, we also purge corresponding pending connect requests. This requires destroying the cm_id's associated with the connect requests by calling rdma_destroy_id(). rdma_destroy_id() blocks until all outstanding callbacks have completed. The issue is that we hold file->mut while purging events from the event queue. We also acquire file->mut in our event handler. Calling rdma_destroy_id() while holding file->mut can lead to a deadlock, since the event handler callback cannot acquire file->mut, which prevents rdma_destroy_id() from completing. Fix this by moving events to purge from the event queue to a temporary list. We can then release file->mut and call rdma_destroy_id() outside of holding any locks. Bug report by Or Gerlitz : [ INFO: possible circular locking dependency detected ] 3.3.0-rc5-00008-g79f1e43-dirty #34 Tainted: G I tgtd/9018 is trying to acquire lock: (&id_priv->handler_mutex){+.+.+.}, at: [] rdma_destroy_id+0x33/0x1f0 [rdma_cm] but task is already holding lock: (&file->mut){+.+.+.}, at: [] ucma_free_ctx+0xb6/0x196 [rdma_ucm] which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #1 (&file->mut){+.+.+.}: [] lock_acquire+0xf0/0x116 [] mutex_lock_nested+0x64/0x2e6 [] ucma_event_handler+0x148/0x1dc [rdma_ucm] [] cma_ib_handler+0x1a7/0x1f7 [rdma_cm] [] cm_process_work+0x32/0x119 [ib_cm] [] cm_work_handler+0xfb8/0xfe5 [ib_cm] [] process_one_work+0x2bd/0x4a6 [] worker_thread+0x1d6/0x350 [] kthread+0x84/0x8c [] kernel_thread_helper+0x4/0x10 -> #0 (&id_priv->handler_mutex){+.+.+.}: [] __lock_acquire+0x10d5/0x1752 [] lock_acquire+0xf0/0x116 [] mutex_lock_nested+0x64/0x2e6 [] rdma_destroy_id+0x33/0x1f0 [rdma_cm] [] ucma_free_ctx+0x117/0x196 [rdma_ucm] [] ucma_close+0x77/0xb4 [rdma_ucm] [] fput+0x117/0x1cf [] filp_close+0x6d/0x78 [] put_files_struct+0xbd/0x17d [] exit_files+0x46/0x4e [] do_exit+0x299/0x75d [] do_group_exit+0x7e/0xa9 [] get_signal_to_deliver+0x536/0x555 [] do_signal+0x39/0x634 [] do_notify_resume+0x27/0x69 [] retint_signal+0x46/0x83 other info that might help us debug this: Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(&file->mut); lock(&id_priv->handler_mutex); lock(&file->mut); lock(&id_priv->handler_mutex); *** DEADLOCK *** 1 lock held by tgtd/9018: #0: (&file->mut){+.+.+.}, at: [] ucma_free_ctx+0xb6/0x196 [rdma_ucm] stack backtrace: Pid: 9018, comm: tgtd Tainted: G I 3.3.0-rc5-00008-g79f1e43-dirty #34 Call Trace: [] ? console_unlock+0x18e/0x207 [] print_circular_bug+0x28e/0x29f [] __lock_acquire+0x10d5/0x1752 [] lock_acquire+0xf0/0x116 [] ? rdma_destroy_id+0x33/0x1f0 [rdma_cm] [] mutex_lock_nested+0x64/0x2e6 [] ? rdma_destroy_id+0x33/0x1f0 [rdma_cm] [] ? trace_hardirqs_on_caller+0x11e/0x155 [] ? trace_hardirqs_on+0xd/0xf [] rdma_destroy_id+0x33/0x1f0 [rdma_cm] [] ucma_free_ctx+0x117/0x196 [rdma_ucm] [] ucma_close+0x77/0xb4 [rdma_ucm] [] fput+0x117/0x1cf [] filp_close+0x6d/0x78 [] put_files_struct+0xbd/0x17d [] ? put_files_struct+0x22/0x17d [] exit_files+0x46/0x4e [] do_exit+0x299/0x75d [] do_group_exit+0x7e/0xa9 [] get_signal_to_deliver+0x536/0x555 [] ? trace_hardirqs_on+0xd/0xf [] do_signal+0x39/0x634 [] ? printk+0x3c/0x45 [] ? trace_hardirqs_on_caller+0x11e/0x155 [] ? trace_hardirqs_on+0xd/0xf [] ? _raw_spin_unlock_irq+0x2b/0x40 [] ? set_current_blocked+0x44/0x49 [] ? retint_signal+0x11/0x83 [] do_notify_resume+0x27/0x69 [] ? trace_hardirqs_on_thunk+0x3a/0x3f [] retint_signal+0x46/0x83 Signed-off-by: Sean Hefty Signed-off-by: Roland Dreier