This is a note to let you know that I've just added the patch titled IB/rxe: Clear queue buffer when modifying QP to reset to the 4.8-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary The filename of the patch is: ib-rxe-clear-queue-buffer-when-modifying-qp-to-reset.patch and it can be found in the queue-4.8 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable@xxxxxxxxxxxxxxx> know about it. >From aa75b07b478a774b1432e2df1be5cd8ae834de0f Mon Sep 17 00:00:00 2001 From: Yonatan Cohen <yonatanc@xxxxxxxxxxxx> Date: Wed, 16 Nov 2016 10:39:17 +0200 Subject: IB/rxe: Clear queue buffer when modifying QP to reset From: Yonatan Cohen <yonatanc@xxxxxxxxxxxx> commit aa75b07b478a774b1432e2df1be5cd8ae834de0f upstream. RXE resets the send-q only once in rxe_qp_init_req() when QP is created, but when the QP is reused after QP reset, the send-q holds previous garbage data. This garbage data wrongly fails CQEs that otherwise should have completed successfully. Fixes: 8700e3e7c485 ("Soft RoCE driver") Signed-off-by: Yonatan Cohen <yonatanc@xxxxxxxxxxxx> Reviewed-by: Moni Shoua <monis@xxxxxxxxxxxx> Signed-off-by: Leon Romanovsky <leon@xxxxxxxxxx> Signed-off-by: Doug Ledford <dledford@xxxxxxxxxx> Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx> --- drivers/infiniband/sw/rxe/rxe_qp.c | 1 + drivers/infiniband/sw/rxe/rxe_queue.c | 9 +++++++++ drivers/infiniband/sw/rxe/rxe_queue.h | 2 ++ 3 files changed, 12 insertions(+) --- a/drivers/infiniband/sw/rxe/rxe_qp.c +++ b/drivers/infiniband/sw/rxe/rxe_qp.c @@ -522,6 +522,7 @@ static void rxe_qp_reset(struct rxe_qp * if (qp->sq.queue) { __rxe_do_task(&qp->comp.task); __rxe_do_task(&qp->req.task); + rxe_queue_reset(qp->sq.queue); } /* cleanup attributes */ --- a/drivers/infiniband/sw/rxe/rxe_queue.c +++ b/drivers/infiniband/sw/rxe/rxe_queue.c @@ -84,6 +84,15 @@ err1: return -EINVAL; } +inline void rxe_queue_reset(struct rxe_queue *q) +{ + /* queue is comprised from header and the memory + * of the actual queue. See "struct rxe_queue_buf" in rxe_queue.h + * reset only the queue itself and not the management header + */ + memset(q->buf->data, 0, q->buf_size - sizeof(struct rxe_queue_buf)); +} + struct rxe_queue *rxe_queue_init(struct rxe_dev *rxe, int *num_elem, unsigned int elem_size) --- a/drivers/infiniband/sw/rxe/rxe_queue.h +++ b/drivers/infiniband/sw/rxe/rxe_queue.h @@ -84,6 +84,8 @@ int do_mmap_info(struct rxe_dev *rxe, size_t buf_size, struct rxe_mmap_info **ip_p); +void rxe_queue_reset(struct rxe_queue *q); + struct rxe_queue *rxe_queue_init(struct rxe_dev *rxe, int *num_elem, unsigned int elem_size); Patches currently in stable-queue which might be from yonatanc@xxxxxxxxxxxx are queue-4.8/ib-rxe-fix-handling-of-erroneous-wr.patch queue-4.8/ib-rxe-update-qp-state-for-user-query.patch queue-4.8/ib-rxe-clear-queue-buffer-when-modifying-qp-to-reset.patch queue-4.8/ib-rxe-fix-kernel-panic-in-udp-tunnel-with-gro-and-rx-checksum.patch -- To unsubscribe from this list: send the line "unsubscribe stable" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html