From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mail-pz2-f42.google.com (mail-pz2-f42.google.com [74.125.228.42]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id E3E7E509F10 for ; Fri, 18 Sep 2026 18:17:16 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=74.125.228.42 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1789755438; cv=none; b=Vawydzxt6XIG6Hx1heTh2L5gMUgELO5qqZ6JjBQEIhwdefbvIq9GiQ85/rIg2ZacG5x/0YVYF5R/mqDe4HwYKVW0xenJdpS7k0HFI9yUir+ao/bl12Ur6/cVev5wyjiTY6zoabSoJQ3AjLWU1Huh8GMgoDd6oYGTUHELOmGSCgg= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1789755438; c=relaxed/simple; bh=brszAfCb8xPeY2++BmqbzkUD7Gs991VIWtKQ+a7lhTo=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=DR2F8kH7wxSCT0ameWja696b92CJd5L+o19gYl7tHcmTs5Jz5L3OqdwLpeTqm1tgUgsnFMBEWIU9vHtyL80AKGeBgvKtaq9145Vii0KgjEAGOYSiTo76lEW7AST9YI/prSL7IRt2aNb1w73v7TBTYrnEvdcN/vViaAEfLcNs57A= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=purestorage.com; spf=pass smtp.mailfrom=purestorage.com; dkim=pass (2048-bit key) header.d=purestorage.com header.i=@purestorage.com header.b=MlZkJLwS; arc=none smtp.client-ip=74.125.228.42 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=purestorage.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=purestorage.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=purestorage.com header.i=@purestorage.com header.b="MlZkJLwS" Received: by mail-pz2-f42.google.com with SMTP id 41be03b00d2f7-cc4c3304784so902063a12.3 for ; Fri, 18 Sep 2026 11:17:16 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=purestorage.com; s=google2022; t=1789755436; x=1790360236; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to:content-type; bh=7jjZqkSxtF056SbAAw5d/uzyUYxqOIAwzrV4C8XoM0U=; b=MlZkJLwSaWcYt7RTPzQ66mLjPxFpJSQQlugav351cPBDj42DLfYJrWSPtmxlaW1fdp CwrewX4Y0cTPqmGRutpWYS9Mj5zVOFJyQ10WwrREYcr7DNirT1XGDWuYSBfxheGnn87Z GCCwIq91llCnnHm98kBtQ4iELbGkDQfQbZwvGqY+rblmhP+HdY2sKs0XCVzSwLLvw3nr HVdPkBsRlvQ/82Cf+swcSKilengahHMbCVMVGa0KV+xNzx+KhB+QqEneaEfkPu6S7z8O rLvE2bs8oPOPzenSbGk28/YJAFw9ITfeKo+f4GxT0GuyO8yhcfaFKRJUZOcbET2J7j/5 dyTQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20260707; t=1789755436; x=1790360236; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-gg:x-gm-message-state:from :to:cc:subject:date:message-id:reply-to:content-type; bh=7jjZqkSxtF056SbAAw5d/uzyUYxqOIAwzrV4C8XoM0U=; b=WTgHRFVILs2Oe86NUrBnb68L123WQASqvKt/5ftJuAbhMvix+IKXa+59idtPRGgRxI LDmxcQ2UXillvrokDP1QoSn2YXmVvddPM/Dc7EEHXHDP2Uex1J4RvBG0S3Rda5wBCtso ibA0KqXIcs2lJ5RYh/ij5r6ouRTbu845SuPjkwin2bAUF12m9n8yfIKhqJjdxPROKrUI Cok74jkMdioWDiEqx0i/vtuMZ+y0huEoaN8RQ4h4bZ8Ke2C4yQsPc0Y93C2DQfJbaEBt CzdIQUbLZsHihMf9r9OVVSeRPF8RtQDIh8hv5tdx0P3SPtedlMkEFCnNhwaAoA1juBI6 Qaag== X-Forwarded-Encrypted: i=1; AKwUvBxgkde2Ii3HG9BrhaW0vZJBFpS8hMNxonXuQKmwUa2lHSCHk4es3+Mf96TQIf2kaNlLJTlxvdt0OCbdQuM=@vger.kernel.org X-Gm-Message-State: AFuF++ksfWNk8d4tFUDShP8qVE5xmPmD2sbny8M+Uq+ihsJVKhowG1gd dTTP/GaqqpvgLEgGK4oX9Kbs2QXklnMbu6rrpe5FAMXm4zoILbxL7JIRniivwebqFyw= X-Gm-Gg: AYBFou0j4lkDPLmTYYifJNystAAQuoA5OzWfjlPoOr0b8zdxV8EsvsoUN2bGJsHea9z BsehWx+r4fFwBFpiXQqQphFY5TbnP1UBxKcYf5VVX+R4XrXLlhwTuMaJfWf+zh7vPQnT7Es3dXg tGdNuCyYRRzsBAmuyl3BfkIdl7Sfu6XC5mLA99rPgizmTgjhUnCts02jytRvx6g80wK7jJGCQfV 8hos3yoD++bdroDiSFeaTeDZ1fg+UYmgedg6zQdjKZrJx9tU5Y9UU2q7M3jvxj+SPVCQRrNazVU Yh20ZVKnmSUdYfF+leeUECjTGYnHLZQ4souCXpAz8aTVH/3tEUaN38sS8O3uj1j3lYvopTGLWv7 IwxGHTFJG3RPdCOQ/JAyij7TrwQq4gZJ0chguzIM9RAQmbBdUA1myA7fR1SQm/r+I4hEfgSo+BA RMHx61wue/fsUODTfkX8FAWsVfOUrXGihNMgAKwJJDXqwkU8gPWfZreoMN6PKEyAfQOZ5xetGkh DllYywvlNeAXBeolX30Jg== X-Received: by 2002:a17:90b:4b82:b0:39d:f06f:79be with SMTP id 98e67ed59e1d1-39e5502546cmr10108721a91.15.1789755435735; Fri, 18 Sep 2026 11:17:15 -0700 (PDT) Received: from apollo.purestorage.com ([208.88.152.253]) by smtp.googlemail.com with ESMTPSA id 5a478bee46e88-33c331aeeddsm335107eec.24.2026.09.18.11.17.14 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Fri, 18 Sep 2026 11:17:15 -0700 (PDT) From: Mohamed Khalfella To: Keith Busch , Jens Axboe , Christoph Hellwig , Sagi Grimberg Cc: Justin Tee , Naresh Gottumukkala , Paul Ely , Hannes Reinecke , Chaitanya Kulkarni , James Smart , Randy Jennings , Mohamed Khalfella , linux-nvme@lists.infradead.org, linux-kernel@vger.kernel.org Subject: [PATCH 15/18] nvme-fc: Hold inflight requests while in FENCING state Date: Fri, 18 Sep 2026 11:14:15 -0700 Message-ID: <20260918181614.3947933-16-mkhalfella@purestorage.com> X-Mailer: git-send-email 2.55.0 In-Reply-To: <20260918181614.3947933-1-mkhalfella@purestorage.com> References: <20260918181614.3947933-1-mkhalfella@purestorage.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: 8bit With the association torn down at the start of nvme_fc_fencing_work(), canceling the held requests inside nvme_fc_delete_association() would complete them at fence START, letting them fail over to other paths before CCR/CQT guarantees the target can no longer execute the original commands - defeating the hold this patch introduces. Split nvme_fc_delete_association() into __nvme_fc_teardown_association() (LLDD-facing teardown, no request completion) and nvme_fc_flush_held_requests() (cancel tagsets + unquiesce queues). nvme_fc_fencing_work() now runs only the teardown half up front and flushes the held requests on the FENCING -> FENCED transition. The queues stay quiesced across the fence so nothing new reaches the dead association. Update nvme_fc_fcpio_done() to not complete aborted requests or requests with transport errors. Held requests cannot be forced out mid-fence by the block layer, nvme_fc_timeout() always returns BLK_EH_RESET_TIMER and nvme_fc_start_ioerr_recovery() is a no-op in FENCING state. Every held op is FCPOP_STATE_COMPLETE before the fence starts (aborts drained, synchronize_rcu() done), so the flush never touches the LLDD. Signed-off-by: Mohamed Khalfella --- drivers/nvme/host/fc.c | 116 +++++++++++++++++++++++++++++++++-------- 1 file changed, 94 insertions(+), 22 deletions(-) diff --git a/drivers/nvme/host/fc.c b/drivers/nvme/host/fc.c index 42671cca439f..a6e0fa2dc341 100644 --- a/drivers/nvme/host/fc.c +++ b/drivers/nvme/host/fc.c @@ -172,7 +172,7 @@ struct nvme_fc_ctrl { struct kref ref; unsigned long flags; - u32 iocnt; + atomic_t iocnt; wait_queue_head_t ioabort_wait; struct nvme_fc_fcp_op aen_ops[NVME_NR_AEN_COMMANDS]; @@ -991,7 +991,8 @@ static void nvme_fc_ctrl_put(struct nvme_fc_ctrl *); static int nvme_fc_ctrl_get(struct nvme_fc_ctrl *); static void nvme_fc_error_recovery(struct nvme_fc_ctrl *ctrl); -static void nvme_fc_delete_association(struct nvme_fc_ctrl *ctrl); +static void __nvme_fc_teardown_association(struct nvme_fc_ctrl *ctrl); +static void nvme_fc_flush_held_requests(struct nvme_fc_ctrl *ctrl); static void __nvme_fc_finish_ls_req(struct nvmefc_ls_req_op *lsop) @@ -1826,7 +1827,7 @@ __nvme_fc_abort_op(struct nvme_fc_ctrl *ctrl, struct nvme_fc_fcp_op *op) atomic_set(&op->state, opstate); else if (test_bit(FCCTRL_TERMIO, &ctrl->flags)) { op->flags |= FCOP_FLAGS_TERMIO; - ctrl->iocnt++; + atomic_inc(&ctrl->iocnt); } spin_unlock_irqrestore(&ctrl->lock, flags); @@ -1856,20 +1857,29 @@ nvme_fc_abort_aen_ops(struct nvme_fc_ctrl *ctrl) } static inline void +__nvme_fc_fcpop_count_one_down(struct nvme_fc_ctrl *ctrl) +{ + if (atomic_dec_return(&ctrl->iocnt) == 0) + wake_up(&ctrl->ioabort_wait); +} + +static inline bool __nvme_fc_fcpop_chk_teardowns(struct nvme_fc_ctrl *ctrl, struct nvme_fc_fcp_op *op, int opstate) { unsigned long flags; + bool ret = false; if (opstate == FCPOP_STATE_ABORTED) { spin_lock_irqsave(&ctrl->lock, flags); if (test_bit(FCCTRL_TERMIO, &ctrl->flags) && op->flags & FCOP_FLAGS_TERMIO) { - if (!--ctrl->iocnt) - wake_up(&ctrl->ioabort_wait); + ret = true; } spin_unlock_irqrestore(&ctrl->lock, flags); } + + return ret; } static void nvme_fc_fencing_work(struct work_struct *work) @@ -1885,12 +1895,14 @@ static void nvme_fc_fencing_work(struct work_struct *work) */ nvme_stop_keep_alive(ctrl); cancel_work_sync(&ctrl->async_event_work); - nvme_fc_delete_association(fc_ctrl); + if (test_and_clear_bit(ASSOC_ACTIVE, &fc_ctrl->flags)) + __nvme_fc_teardown_association(fc_ctrl); rem = nvme_fence_ctrl(ctrl); if (rem) dev_info(ctrl->device, "CCR failed, starting error recovery\n"); + nvme_fc_flush_held_requests(fc_ctrl); nvme_change_ctrl_state(ctrl, NVME_CTRL_FENCED); if (nvme_change_ctrl_state(ctrl, NVME_CTRL_RESETTING)) queue_work(nvme_reset_wq, &fc_ctrl->ioerr_work); @@ -2003,7 +2015,8 @@ nvme_fc_fcpio_done(struct nvmefc_fcp_req *req) struct nvme_command *sqe = &op->cmd_iu.sqe; __le16 status = cpu_to_le16(NVME_SC_SUCCESS << 1); union nvme_result result; - bool terminate_assoc = true; + bool op_term, terminate_assoc = true; + enum nvme_ctrl_state state; int opstate; /* @@ -2043,6 +2056,9 @@ nvme_fc_fcpio_done(struct nvmefc_fcp_req *req) * association to be terminated. */ + /* Pairs with synchronize_rcu() in __nvme_fc_teardown_association() */ + rcu_read_lock(); + opstate = atomic_xchg(&op->state, FCPOP_STATE_COMPLETE); fc_dma_sync_single_for_cpu(ctrl->lport->dev, op->fcp_req.rspdma, @@ -2136,20 +2152,44 @@ nvme_fc_fcpio_done(struct nvmefc_fcp_req *req) done: if (op->flags & FCOP_FLAGS_AEN) { nvme_complete_async_event(&queue->ctrl->ctrl, status, &result); - __nvme_fc_fcpop_chk_teardowns(ctrl, op, opstate); + if (__nvme_fc_fcpop_chk_teardowns(ctrl, op, opstate)) + __nvme_fc_fcpop_count_one_down(ctrl); atomic_set(&op->state, FCPOP_STATE_IDLE); op->flags = FCOP_FLAGS_AEN; /* clear other flags */ nvme_fc_ctrl_put(ctrl); goto check_error; } - __nvme_fc_fcpop_chk_teardowns(ctrl, op, opstate); + /* + * We can not access op after the request is completed because it can + * be reused immediately. At the same time we want to wakeup the thread + * waiting for ongoing IOs _after_ requests are completed. This is + * necessary because that thread will start canceling inflight IOs + * and we want to avoid request completion racing with cancellation. + */ + op_term = __nvme_fc_fcpop_chk_teardowns(ctrl, op, opstate); + + /* + * If we are going to terminate associations and the controller is + * LIVE or FENCING, then do not complete this request now. Let error + * recovery cancel this request when it is safe to do so. + */ + state = nvme_ctrl_state(&ctrl->ctrl); + if (terminate_assoc && + (state == NVME_CTRL_LIVE || state == NVME_CTRL_FENCING)) + goto check_op_term; + if (!nvme_try_complete_req(rq, status, result)) nvme_fc_complete_rq(rq); +check_op_term: + if (op_term) + __nvme_fc_fcpop_count_one_down(ctrl); check_error: if (terminate_assoc) nvme_fc_start_ioerr_recovery(ctrl, "io error"); + + rcu_read_unlock(); } static int @@ -2802,7 +2842,8 @@ nvme_fc_start_fcp_op(struct nvme_fc_ctrl *ctrl, struct nvme_fc_queue *queue, * cmd with the csn was supposed to arrive. */ opstate = atomic_xchg(&op->state, FCPOP_STATE_COMPLETE); - __nvme_fc_fcpop_chk_teardowns(ctrl, op, opstate); + if (__nvme_fc_fcpop_chk_teardowns(ctrl, op, opstate)) + __nvme_fc_fcpop_count_one_down(ctrl); if (!(op->flags & FCOP_FLAGS_AEN)) { nvme_fc_unmap_data(ctrl, op->rq, op); @@ -3260,25 +3301,25 @@ nvme_fc_create_association(struct nvme_fc_ctrl *ctrl) return ret; } - /* - * This routine stops operation of the controller on the host side. - * On the host os stack side: Admin and IO queues are stopped, - * outstanding ios on them terminated via FC ABTS. - * On the link side: the association is terminated. + * This routine tears down the association on the link side and releases + * all LLDD resources: outstanding ios are terminated via FC ABTS, hw + * queues are deleted and the controller is marked inactive on the + * rport. Requests that nvme_fc_fcpio_done() decided to hold are NOT + * completed here, the caller must follow up with + * nvme_fc_flush_held_requests() once it is safe to release them. + * + * The caller must own the ASSOC_ACTIVE bit (test_and_clear_bit()). */ static void -nvme_fc_delete_association(struct nvme_fc_ctrl *ctrl) +__nvme_fc_teardown_association(struct nvme_fc_ctrl *ctrl) { struct nvmefc_ls_rcv_op *disls = NULL; unsigned long flags; - if (!test_and_clear_bit(ASSOC_ACTIVE, &ctrl->flags)) - return; - spin_lock_irqsave(&ctrl->lock, flags); set_bit(FCCTRL_TERMIO, &ctrl->flags); - ctrl->iocnt = 0; + atomic_set(&ctrl->iocnt, 0); spin_unlock_irqrestore(&ctrl->lock, flags); __nvme_fc_abort_outstanding_ios(ctrl, false); @@ -3287,11 +3328,18 @@ nvme_fc_delete_association(struct nvme_fc_ctrl *ctrl) nvme_fc_abort_aen_ops(ctrl); /* wait for all io that had to be aborted */ + wait_event(ctrl->ioabort_wait, atomic_read(&ctrl->iocnt) == 0); spin_lock_irq(&ctrl->lock); - wait_event_lock_irq(ctrl->ioabort_wait, ctrl->iocnt == 0, ctrl->lock); clear_bit(FCCTRL_TERMIO, &ctrl->flags); spin_unlock_irq(&ctrl->lock); + /* + * Wait for a started nvme_fc_fcpio_done() to complete before we + * proceed with cancelling inflight requests. This guarantees a + * request is completed by one of the two codepaths, not both. + */ + synchronize_rcu(); + nvme_fc_term_aen_ops(ctrl); /* @@ -3323,13 +3371,37 @@ nvme_fc_delete_association(struct nvme_fc_ctrl *ctrl) __nvme_fc_delete_hw_queue(ctrl, &ctrl->queues[0], 0); nvme_fc_free_queue(&ctrl->queues[0]); + nvme_fc_ctlr_inactive_on_rport(ctrl); +} + +/* Complete the requests held in __nvme_fc_teardown_association() */ +static void +nvme_fc_flush_held_requests(struct nvme_fc_ctrl *ctrl) +{ + nvme_cancel_tagset(&ctrl->ctrl); + nvme_cancel_admin_tagset(&ctrl->ctrl); + /* re-enable the admin_q so anything new can fast fail */ nvme_unquiesce_admin_queue(&ctrl->ctrl); /* resume the io queues so that things will fast fail */ nvme_unquiesce_io_queues(&ctrl->ctrl); +} - nvme_fc_ctlr_inactive_on_rport(ctrl); +/* + * This routine stops operation of the controller on the host side. + * On the host os stack side: Admin and IO queues are stopped, + * outstanding ios on them terminated via FC ABTS. + * On the link side: the association is terminated. + */ +static void +nvme_fc_delete_association(struct nvme_fc_ctrl *ctrl) +{ + if (!test_and_clear_bit(ASSOC_ACTIVE, &ctrl->flags)) + return; + + __nvme_fc_teardown_association(ctrl); + nvme_fc_flush_held_requests(ctrl); } static void -- 2.55.0