kernel/git/stable/linux.git/drivers/ntb, branch linux-4.12.y

ntb: transport shouldn't disable link due to bogus values in SPADs

2017-08-30T08:26:42+00:00

commit f3fd2afed8eee91620d05b69ab94c14793c849d7 upstream. It seems that under certain scenarios the SPAD can have bogus values caused by an agent (i.e. BIOS or other software) that is not the kernel driver, and that causes memory window setup failure. This should not cause the link to be disabled because if we do that, the driver will never recover again. We have verified in testing that this issue happens and prevents proper link recovery. Signed-off-by: Dave Jiang Acked-by: Allen Hubbe Signed-off-by: Jon Mason Fixes: 84f766855f61 ("ntb: stop link work when we do not have memory") Signed-off-by: Greg Kroah-Hartman

ntb: no sleep in ntb_async_tx_submit

2017-06-19T18:24:41+00:00

Do not sleep in ntb_async_tx_submit, which could deadlock. This reverts commit "8c874cc140d667f84ae4642bb5b5e0d6396d2ca4" Fixes: 8c874cc140d6 ("NTB: Address out of DMA descriptor issue with NTB") Reported-by: Jia-Ju Bai Signed-off-by: Allen Hubbe Acked-by: Dave Jiang Signed-off-by: Jon Mason

ntb: ntb_hw_intel: Skylake doorbells should be 32bits, not 64bits

2017-06-19T18:24:41+00:00

Fixing doorbell register length to 32bits per spec. On Skylake NTB, the doorbell registers are 32bit write only registers. The source for the doorbell is a 64bit register that shows the interrupt bits. Signed-off-by: Dave Jiang Fixes: 783dfa6cc41b ("ntb: Adding Skylake Xeon NTB support") Acked-by: Allen Hubbe Signed-off-by: Jon Mason

ntb_transport: fix bug calculating num_qps_mw

2017-06-19T18:24:41+00:00

A divide by zero error occurs if qp_count is less than mw_count because num_qps_mw is calculated to be zero. The calculation appears to be incorrect. The requirement is for num_qps_mw to be set to qp_count / mw_count with any remainder divided among the earlier mws. For example, if mw_count is 5 and qp_count is 12 then mws 0 and 1 will have 3 qps per window and mws 2 through 4 will have 2 qps per window. Thus, when mw_num < qp_count % mw_count, num_qps_mw is 1 higher than when mw_num >= qp_count. Signed-off-by: Logan Gunthorpe Fixes: e26a5843f7f5 ("NTB: Split ntb_hw_intel and ntb_transport drivers") Acked-by: Allen Hubbe Signed-off-by: Jon Mason

ntb_transport: fix qp count bug

2017-06-19T18:24:41+00:00

In cases where there are more mw's than spads/2-2, the mw count gets reduced to match the limitation. ntb_transport also tries to ensure that there are fewer qps than mws but uses the full mw count instead of the reduced one. When this happens, the math in 'ntb_transport_setup_qp_mw' will get confused and result in a kernel paging request bug. This patch fixes the bug by reducing qp_count to the reduced mw count instead of the full mw count. Signed-off-by: Logan Gunthorpe Fixes: e26a5843f7f5 ("NTB: Split ntb_hw_intel and ntb_transport drivers") Acked-by: Allen Hubbe Signed-off-by: Jon Mason

ntb: Correct modinfo usage statement for ntb_perf

2017-06-19T18:24:41+00:00

The order parameters are powers of 2; adjust the usage information to use correct mathematical representations. Signed-off-by: Gary R Hook Fixes: 8a7b6a778a85 ("ntb: ntb perf tool") Acked-by: Dave Jiang Signed-off-by: Jon Mason

ntb: ntb_hw_intel: link_poll isn't clearing the pending status properly

2017-02-17T04:11:26+00:00

On Skylake hardware, the link_poll isn't clearing the pending interrupt bit. Adding a new function for SKX that handles clearing of status bit the right way. Signed-off-by: Dave Jiang Fixes: 783dfa6c ("ntb: Adding Skylake Xeon NTB support") Signed-off-by: Jon Mason

ntb_transport: Pick an unused queue

2017-02-17T04:11:26+00:00

Fix typo causing ntb_transport_create_queue to select the first queue every time, instead of using the next free queue. Signed-off-by: Thomas VanSelus Signed-off-by: Aaron Sierra Acked-by: Allen Hubbe Fixes: fce8a7bb5 ("PCI-Express Non-Transparent Bridge Support") Signed-off-by: Jon Mason

ntb: ntb_perf missing dmaengine_unmap_put

2017-02-17T04:11:26+00:00

In the normal I/O execution path, ntb_perf is missing a call to dmaengine_unmap_put() after submission. That causes us to leak unmap objects. Signed-off-by: Dave Jiang Fixes: 8a7b6a77 ("ntb: ntb perf tool") Signed-off-by: Jon Mason

NTB: ntb_transport: fix debugfs_remove_recursive

2017-02-17T04:10:52+00:00

The call to debugfs_remove_recursive(qp->debugfs_dir) of the sub-level directory must not be later than debugfs_remove_recursive(nt_debugfs_dir) of the top-level directory. Otherwise, the sub-level directory will not exist, and it would be invalid (panic) to attempt to remove it. This removes the top-level directory last, after sub-level directories have been cleaned up. Signed-off-by: Allen Hubbe Fixes: e26a5843f ("NTB: Split ntb_hw_intel and ntb_transport drivers") Signed-off-by: Jon Mason