This is a note to let you know that I've just added the patch titled RDMA/bnxt_re: Avoid calling wake_up threads from spin_lock context to the 5.15-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: rdma-bnxt_re-avoid-calling-wake_up-threads-from-spin.patch and it can be found in the queue-5.15 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable@xxxxxxxxxxxxxxx> know about it. commit f0a9e5b9f0de5ca752e28e3e2b6fc1e06bae7bb2 Author: Kashyap Desai <kashyap.desai@xxxxxxxxxxxx> Date: Fri Jun 9 04:01:39 2023 -0700 RDMA/bnxt_re: Avoid calling wake_up threads from spin_lock context [ Upstream commit 3099bcdc19b701f732f638ee45679858c08559bb ] bnxt_qplib_service_creq can be called from interrupt or tasklet or process context. So the function take irq variant of spin_lock. But when wake_up is invoked with the lock held, it is putting the calling context to sleep. [exception RIP: __wake_up_common+190] RIP: ffffffffb7539d7e RSP: ffffa73300207ad8 RFLAGS: 00000083 RAX: 0000000000000001 RBX: ffff91fa295f69b8 RCX: dead000000000200 RDX: ffffa733344af940 RSI: ffffa73336527940 RDI: ffffa73336527940 RBP: 000000000000001c R8: 0000000000000002 R9: 00000000000299c0 R10: 0000017230de82c5 R11: 0000000000000002 R12: ffffa73300207b28 R13: 0000000000000000 R14: ffffa733341bf928 R15: 0000000000000000 ORIG_RAX: ffffffffffffffff CS: 0010 SS: 0018 Call the wakeup after releasing the lock. Fixes: 1ac5a4047975 ("RDMA/bnxt_re: Add bnxt_re RoCE driver") Signed-off-by: Kashyap Desai <kashyap.desai@xxxxxxxxxxxx> Signed-off-by: Selvin Xavier <selvin.xavier@xxxxxxxxxxxx> Link: https://lore.kernel.org/r/1686308514-11996-3-git-send-email-selvin.xavier@xxxxxxxxxxxx Signed-off-by: Leon Romanovsky <leon@xxxxxxxxxx> Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx> diff --git a/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c b/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c index abeee7cc7dc6d..3b8cb46551bf2 100644 --- a/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c +++ b/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c @@ -299,7 +299,8 @@ static int bnxt_qplib_process_func_event(struct bnxt_qplib_rcfw *rcfw, } static int bnxt_qplib_process_qp_event(struct bnxt_qplib_rcfw *rcfw, - struct creq_qp_event *qp_event) + struct creq_qp_event *qp_event, + u32 *num_wait) { struct creq_qp_error_notification *err_event; struct bnxt_qplib_hwq *hwq = &rcfw->cmdq.hwq; @@ -308,6 +309,7 @@ static int bnxt_qplib_process_qp_event(struct bnxt_qplib_rcfw *rcfw, u16 cbit, blocked = 0; struct pci_dev *pdev; unsigned long flags; + u32 wait_cmds = 0; __le16 mcookie; u16 cookie; int rc = 0; @@ -367,9 +369,10 @@ static int bnxt_qplib_process_qp_event(struct bnxt_qplib_rcfw *rcfw, crsqe->req_size = 0; if (!blocked) - wake_up(&rcfw->cmdq.waitq); + wait_cmds++; spin_unlock_irqrestore(&hwq->lock, flags); } + *num_wait += wait_cmds; return rc; } @@ -383,6 +386,7 @@ static void bnxt_qplib_service_creq(struct tasklet_struct *t) struct creq_base *creqe; u32 sw_cons, raw_cons; unsigned long flags; + u32 num_wakeup = 0; /* Service the CREQ until budget is over */ spin_lock_irqsave(&hwq->lock, flags); @@ -401,7 +405,8 @@ static void bnxt_qplib_service_creq(struct tasklet_struct *t) switch (type) { case CREQ_BASE_TYPE_QP_EVENT: bnxt_qplib_process_qp_event - (rcfw, (struct creq_qp_event *)creqe); + (rcfw, (struct creq_qp_event *)creqe, + &num_wakeup); creq->stats.creq_qp_event_processed++; break; case CREQ_BASE_TYPE_FUNC_EVENT: @@ -429,6 +434,8 @@ static void bnxt_qplib_service_creq(struct tasklet_struct *t) rcfw->res->cctx, true); } spin_unlock_irqrestore(&hwq->lock, flags); + if (num_wakeup) + wake_up_nr(&rcfw->cmdq.waitq, num_wakeup); } static irqreturn_t bnxt_qplib_creq_irq(int irq, void *dev_instance)