kernel.git/drivers/infiniband, branch linux-2.6.30.y

IB/mlx4: Add strong ordering to local inval and fast reg work requests

2009-07-02T23:50:05Z

commit 2ac6bf4ddc87c3b6b609f8fa82f6ebbffeac12f4 upstream. The ConnectX Programmer's Reference Manual states that the "SO" bit must be set when posting Fast Register and Local Invalidate send work requests. When this bit is set, the work request will be executed only after all previous work requests on the send queue have been executed. (If the bit is not set, Fast Register and Local Invalidate WQEs may begin execution too early, which violates the defined semantics for these operations) This fixes the issue with NFS/RDMA reported in Signed-off-by: Jack Morgenstein Signed-off-by: Roland Dreier Signed-off-by: Greg Kroah-Hartman

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

2009-05-13T23:31:12Z

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband: IB/mlx4: Don't overwrite fast registration page list when posting work request RDMA/cxgb3: Don't complete flushed send work requests twice

Merge branches 'cxgb3' and 'mlx4' into for-linus

2009-05-13T22:16:17Z

Fix deadlock in ipathfs ->get_sb()

2009-05-09T14:49:40Z

forgot to unlock superblock before calling deactivate_super()... Signed-off-by: Al Viro

IB/mlx4: Don't overwrite fast registration page list when posting work request

2009-05-08T04:35:13Z

The low-level mlx4 driver modified the page-list addresses for fast register work requests post send to big-endian, and set a "present" bit. This caused problems later when the consumer attempted to unmap the pages using the page-list (using the list addresses which were assumed to be still in CPU-endian order). Fix the mlx4 driver to allocate two buffers and use a private buffer for the hardware-format bus addresses. This patch fixes , an NFS/RDMA server crash. The cause of the crash was found by Vu Pham of Mellanox. The fix is along the lines suggested by Steve Wise in comment #21 in bug 1571. Signed-off-by: Jack Morgenstein Signed-off-by: Roland Dreier

Merge git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-rc-fixes-2.6

2009-05-02T23:36:34Z

* git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-rc-fixes-2.6: (53 commits) [SCSI] libosd: OSD2r05: on-the-wire changes for latest OSD2 revision 5. [SCSI] libosd: OSD2r05: OSD_CRYPTO_KEYID_SIZE will grow 20 => 32 bytes [SCSI] libosd: OSD2r05: Prepare for rev5 attribute list changes [SCSI] libosd: fix potential ERR_PTR dereference in osd_initiator.c [SCSI] mpt2sas : bump driver version to 01.100.02.00 [SCSI] mpt2sas: fix hotplug event processing [SCSI] mpt2sas : release diagnotic buffers prior host reset [SCSI] mpt2sas : Broadcast Primative AEN bug fix [SCSI] mpt2sas : Identify Dell series-7 adapters at driver load time [SCSI] mpt2sas : driver name needs to be in the MPT2IOCINFO ioctl [SCSI] mpt2sas : running out of message frames [SCSI] mpt2sas : fix oops when firmware sends large sense buffer size [SCSI] mpt2sas : the sanity check in base_interrupt needs to be on dword boundary [SCSI] mpt2sas : unique ioctl magic number [SCSI] fix sign extension with 1.5TB usb-storage LBD=y [SCSI] ipr: Fix sleeping function called with interrupts disabled [SCSI] fcoe: fip: add multicast filter to receive FIP advertisements. [SCSI] libfc: Fix compilation warnings with allmodconfig [SCSI] fcoe: fix spelling typos and bad comments [SCSI] fcoe: don't export functions that are internal to fcoe ...

RDMA/cxgb3: Don't complete flushed send work requests twice

2009-04-29T22:15:59Z

When the SQ is flushed, mark the flushed entries as not signaled so the poll logic doesn't re-insert the CQ entry thinking its an out of order completion. The bug can cause the NFS/RDMA server to crash due to processing the same completed work request twice. Signed-off-by: Steve Wise Signed-off-by: Roland Dreier

Merge branches 'cxgb3', 'ipoib', 'mthca', 'mlx4' and 'nes' into for-linus

2009-04-28T23:01:31Z

RDMA/nes: Update iw_nes version

2009-04-27T20:46:29Z

Update version number to 1.5.0.0 Signed-off-by: Chien Tung Signed-off-by: Roland Dreier

RDMA/nes: Fix error path in nes_accept()

2009-04-27T20:45:19Z

If reg_phys_mem() fails, we need to free memory allocated for MPA frame with private data before returning the error. Also move nes_add_ref() after the reg_phys_mem() is successful. Signed-off-by: Faisal Latif Signed-off-by: Roland Dreier