Merge tag 'nvme-5.16-2021-11-25' of git://git.infradead.org/nvme into block-5.16

Pull NVMe fixes from Christoph:

"nvme fixes for Linux 5.16

 - add a NO APST quirk for a Kioxia device (Enzo Matsumiya)
 - fix write zeroes pi (Klaus Jensen)
 - various TCP transport fixes (Maurizio Lombardi and Varun Prakash)
 - ignore invalid fast_io_fail_tmo values (Maurizio Lombardi)
 - use IOCB_NOWAIT only if the filesystem supports it (Maurizio Lombardi)"

* tag 'nvme-5.16-2021-11-25' of git://git.infradead.org/nvme:
  nvmet: use IOCB_NOWAIT only if the filesystem supports it
  nvme: fix write zeroes pi
  nvme-fabrics: ignore invalid fast_io_fail_tmo values
  nvme-pci: add NO APST quirk for Kioxia device
  nvme-tcp: fix memory leak when freeing a queue
  nvme-tcp: validate R2T PDU in nvme_tcp_handle_r2t()
  nvmet-tcp: fix incomplete data digest send
  nvmet-tcp: fix memory leak when performing a controller reset
  nvmet-tcp: add an helper to free the cmd buffers
  nvmet-tcp: fix a race condition between release_queue and io_work

Author: axboe

drivers/nvme/host/core.c

@@ -895,10 +895,19 @@ static inline blk_status_t nvme_setup_write_zeroes(struct nvme_ns *ns,
 		cpu_to_le64(nvme_sect_to_lba(ns, blk_rq_pos(req)));
 	cmnd->write_zeroes.length =
 		cpu_to_le16((blk_rq_bytes(req) >> ns->lba_shift) - 1);
-	if (nvme_ns_has_pi(ns))
+
+	if (nvme_ns_has_pi(ns)) {
 		cmnd->write_zeroes.control = cpu_to_le16(NVME_RW_PRINFO_PRACT);
-	else
-		cmnd->write_zeroes.control = 0;
+
+		switch (ns->pi_type) {
+		case NVME_NS_DPS_PI_TYPE1:
+		case NVME_NS_DPS_PI_TYPE2:
+			cmnd->write_zeroes.reftag =
+				cpu_to_le32(t10_pi_ref_tag(req));
+			break;
+		}
+	}
+
 	return BLK_STS_OK;
 }
 
@@ -2469,6 +2478,20 @@ static const struct nvme_core_quirk_entry core_quirks[] = {
 		.vid = 0x14a4,
 		.fr = "22301111",
 		.quirks = NVME_QUIRK_SIMPLE_SUSPEND,
+	},
+	{
+		/*
+		 * This Kioxia CD6-V Series / HPE PE8030 device times out and
+		 * aborts I/O during any load, but more easily reproducible
+		 * with discards (fstrim).
+		 *
+		 * The device is left in a state where it is also not possible
+		 * to use "nvme set-feature" to disable APST, but booting with
+		 * nvme_core.default_ps_max_latency=0 works.
+		 */
+		.vid = 0x1e0f,
+		.mn = "KCD6XVUL6T40",
+		.quirks = NVME_QUIRK_NO_APST,
 	}
 };
 

drivers/nvme/host/fabrics.c

@@ -698,6 +698,9 @@ static int nvmf_parse_options(struct nvmf_ctrl_options *opts,
 			if (token >= 0)
 				pr_warn("I/O fail on reconnect controller after %d sec\n",
 					token);
+			else
+				token = -1;
+
 			opts->fast_io_fail_tmo = token;
 			break;
 		case NVMF_OPT_HOSTNQN:

drivers/nvme/host/tcp.c

@@ -572,7 +572,7 @@ static int nvme_tcp_handle_comp(struct nvme_tcp_queue *queue,
 	return ret;
 }
 
-static int nvme_tcp_setup_h2c_data_pdu(struct nvme_tcp_request *req,
+static void nvme_tcp_setup_h2c_data_pdu(struct nvme_tcp_request *req,
 		struct nvme_tcp_r2t_pdu *pdu)
 {
 	struct nvme_tcp_data_pdu *data = req->pdu;
@@ -581,32 +581,11 @@ static int nvme_tcp_setup_h2c_data_pdu(struct nvme_tcp_request *req,
 	u8 hdgst = nvme_tcp_hdgst_len(queue);
 	u8 ddgst = nvme_tcp_ddgst_len(queue);
 
+	req->state = NVME_TCP_SEND_H2C_PDU;
+	req->offset = 0;
 	req->pdu_len = le32_to_cpu(pdu->r2t_length);
 	req->pdu_sent = 0;
 
-	if (unlikely(!req->pdu_len)) {
-		dev_err(queue->ctrl->ctrl.device,
-			"req %d r2t len is %u, probably a bug...\n",
-			rq->tag, req->pdu_len);
-		return -EPROTO;
-	}
-
-	if (unlikely(req->data_sent + req->pdu_len > req->data_len)) {
-		dev_err(queue->ctrl->ctrl.device,
-			"req %d r2t len %u exceeded data len %u (%zu sent)\n",
-			rq->tag, req->pdu_len, req->data_len,
-			req->data_sent);
-		return -EPROTO;
-	}
-
-	if (unlikely(le32_to_cpu(pdu->r2t_offset) < req->data_sent)) {
-		dev_err(queue->ctrl->ctrl.device,
-			"req %d unexpected r2t offset %u (expected %zu)\n",
-			rq->tag, le32_to_cpu(pdu->r2t_offset),
-			req->data_sent);
-		return -EPROTO;
-	}
-
 	memset(data, 0, sizeof(*data));
 	data->hdr.type = nvme_tcp_h2c_data;
 	data->hdr.flags = NVME_TCP_F_DATA_LAST;
@@ -622,15 +601,14 @@ static int nvme_tcp_setup_h2c_data_pdu(struct nvme_tcp_request *req,
 	data->command_id = nvme_cid(rq);
 	data->data_offset = pdu->r2t_offset;
 	data->data_length = cpu_to_le32(req->pdu_len);
-	return 0;
 }
 
 static int nvme_tcp_handle_r2t(struct nvme_tcp_queue *queue,
 		struct nvme_tcp_r2t_pdu *pdu)
 {
 	struct nvme_tcp_request *req;
 	struct request *rq;
-	int ret;
+	u32 r2t_length = le32_to_cpu(pdu->r2t_length);
 
 	rq = nvme_find_rq(nvme_tcp_tagset(queue), pdu->command_id);
 	if (!rq) {
@@ -641,13 +619,28 @@ static int nvme_tcp_handle_r2t(struct nvme_tcp_queue *queue,
 	}
 	req = blk_mq_rq_to_pdu(rq);
 
-	ret = nvme_tcp_setup_h2c_data_pdu(req, pdu);
-	if (unlikely(ret))
-		return ret;
+	if (unlikely(!r2t_length)) {
+		dev_err(queue->ctrl->ctrl.device,
+			"req %d r2t len is %u, probably a bug...\n",
+			rq->tag, r2t_length);
+		return -EPROTO;
+	}
 
-	req->state = NVME_TCP_SEND_H2C_PDU;
-	req->offset = 0;
+	if (unlikely(req->data_sent + r2t_length > req->data_len)) {
+		dev_err(queue->ctrl->ctrl.device,
+			"req %d r2t len %u exceeded data len %u (%zu sent)\n",
+			rq->tag, r2t_length, req->data_len, req->data_sent);
+		return -EPROTO;
+	}
 
+	if (unlikely(le32_to_cpu(pdu->r2t_offset) < req->data_sent)) {
+		dev_err(queue->ctrl->ctrl.device,
+			"req %d unexpected r2t offset %u (expected %zu)\n",
+			rq->tag, le32_to_cpu(pdu->r2t_offset), req->data_sent);
+		return -EPROTO;
+	}
+
+	nvme_tcp_setup_h2c_data_pdu(req, pdu);
 	nvme_tcp_queue_request(req, false, true);
 
 	return 0;
@@ -1232,6 +1225,7 @@ static int nvme_tcp_alloc_async_req(struct nvme_tcp_ctrl *ctrl)
 
 static void nvme_tcp_free_queue(struct nvme_ctrl *nctrl, int qid)
 {
+	struct page *page;
 	struct nvme_tcp_ctrl *ctrl = to_tcp_ctrl(nctrl);
 	struct nvme_tcp_queue *queue = &ctrl->queues[qid];
 
@@ -1241,6 +1235,11 @@ static void nvme_tcp_free_queue(struct nvme_ctrl *nctrl, int qid)
 	if (queue->hdr_digest || queue->data_digest)
 		nvme_tcp_free_crypto(queue);
 
+	if (queue->pf_cache.va) {
+		page = virt_to_head_page(queue->pf_cache.va);
+		__page_frag_cache_drain(page, queue->pf_cache.pagecnt_bias);
+		queue->pf_cache.va = NULL;
+	}
 	sock_release(queue->sock);
 	kfree(queue->pdu);
 	mutex_destroy(&queue->send_mutex);

drivers/nvme/target/io-cmd-file.c

@@ -8,6 +8,7 @@
 #include <linux/uio.h>
 #include <linux/falloc.h>
 #include <linux/file.h>
+#include <linux/fs.h>
 #include "nvmet.h"
 
 #define NVMET_MAX_MPOOL_BVEC		16
@@ -266,7 +267,8 @@ static void nvmet_file_execute_rw(struct nvmet_req *req)
 
 	if (req->ns->buffered_io) {
 		if (likely(!req->f.mpool_alloc) &&
-				nvmet_file_execute_io(req, IOCB_NOWAIT))
+		    (req->ns->file->f_mode & FMODE_NOWAIT) &&
+		    nvmet_file_execute_io(req, IOCB_NOWAIT))
 			return;
 		nvmet_file_submit_buffered_io(req);
 	} else

drivers/nvme/target/tcp.c

@@ -166,6 +166,8 @@ static struct workqueue_struct *nvmet_tcp_wq;
 static const struct nvmet_fabrics_ops nvmet_tcp_ops;
 static void nvmet_tcp_free_cmd(struct nvmet_tcp_cmd *c);
 static void nvmet_tcp_finish_cmd(struct nvmet_tcp_cmd *cmd);
+static void nvmet_tcp_free_cmd_buffers(struct nvmet_tcp_cmd *cmd);
+static void nvmet_tcp_unmap_pdu_iovec(struct nvmet_tcp_cmd *cmd);
 
 static inline u16 nvmet_tcp_cmd_tag(struct nvmet_tcp_queue *queue,
 		struct nvmet_tcp_cmd *cmd)
@@ -297,6 +299,16 @@ static int nvmet_tcp_check_ddgst(struct nvmet_tcp_queue *queue, void *pdu)
 	return 0;
 }
 
+static void nvmet_tcp_free_cmd_buffers(struct nvmet_tcp_cmd *cmd)
+{
+	WARN_ON(unlikely(cmd->nr_mapped > 0));
+
+	kfree(cmd->iov);
+	sgl_free(cmd->req.sg);
+	cmd->iov = NULL;
+	cmd->req.sg = NULL;
+}
+
 static void nvmet_tcp_unmap_pdu_iovec(struct nvmet_tcp_cmd *cmd)
 {
 	struct scatterlist *sg;
@@ -306,6 +318,8 @@ static void nvmet_tcp_unmap_pdu_iovec(struct nvmet_tcp_cmd *cmd)
 
 	for (i = 0; i < cmd->nr_mapped; i++)
 		kunmap(sg_page(&sg[i]));
+
+	cmd->nr_mapped = 0;
 }
 
 static void nvmet_tcp_map_pdu_iovec(struct nvmet_tcp_cmd *cmd)
@@ -387,7 +401,7 @@ static int nvmet_tcp_map_data(struct nvmet_tcp_cmd *cmd)
 
 	return 0;
 err:
-	sgl_free(cmd->req.sg);
+	nvmet_tcp_free_cmd_buffers(cmd);
 	return NVME_SC_INTERNAL;
 }
 
@@ -632,10 +646,8 @@ static int nvmet_try_send_data(struct nvmet_tcp_cmd *cmd, bool last_in_batch)
 		}
 	}
 
-	if (queue->nvme_sq.sqhd_disabled) {
-		kfree(cmd->iov);
-		sgl_free(cmd->req.sg);
-	}
+	if (queue->nvme_sq.sqhd_disabled)
+		nvmet_tcp_free_cmd_buffers(cmd);
 
 	return 1;
 
@@ -664,8 +676,7 @@ static int nvmet_try_send_response(struct nvmet_tcp_cmd *cmd,
 	if (left)
 		return -EAGAIN;
 
-	kfree(cmd->iov);
-	sgl_free(cmd->req.sg);
+	nvmet_tcp_free_cmd_buffers(cmd);
 	cmd->queue->snd_cmd = NULL;
 	nvmet_tcp_put_cmd(cmd);
 	return 1;
@@ -700,10 +711,11 @@ static int nvmet_try_send_r2t(struct nvmet_tcp_cmd *cmd, bool last_in_batch)
 static int nvmet_try_send_ddgst(struct nvmet_tcp_cmd *cmd, bool last_in_batch)
 {
 	struct nvmet_tcp_queue *queue = cmd->queue;
+	int left = NVME_TCP_DIGEST_LENGTH - cmd->offset;
 	struct msghdr msg = { .msg_flags = MSG_DONTWAIT };
 	struct kvec iov = {
 		.iov_base = (u8 *)&cmd->exp_ddgst + cmd->offset,
-		.iov_len = NVME_TCP_DIGEST_LENGTH - cmd->offset
+		.iov_len = left
 	};
 	int ret;
 
@@ -717,6 +729,10 @@ static int nvmet_try_send_ddgst(struct nvmet_tcp_cmd *cmd, bool last_in_batch)
 		return ret;
 
 	cmd->offset += ret;
+	left -= ret;
+
+	if (left)
+		return -EAGAIN;
 
 	if (queue->nvme_sq.sqhd_disabled) {
 		cmd->queue->snd_cmd = NULL;
@@ -1406,8 +1422,7 @@ static void nvmet_tcp_finish_cmd(struct nvmet_tcp_cmd *cmd)
 {
 	nvmet_req_uninit(&cmd->req);
 	nvmet_tcp_unmap_pdu_iovec(cmd);
-	kfree(cmd->iov);
-	sgl_free(cmd->req.sg);
+	nvmet_tcp_free_cmd_buffers(cmd);
 }
 
 static void nvmet_tcp_uninit_data_in_cmds(struct nvmet_tcp_queue *queue)
@@ -1417,7 +1432,10 @@ static void nvmet_tcp_uninit_data_in_cmds(struct nvmet_tcp_queue *queue)
 
 	for (i = 0; i < queue->nr_cmds; i++, cmd++) {
 		if (nvmet_tcp_need_data_in(cmd))
-			nvmet_tcp_finish_cmd(cmd);
+			nvmet_req_uninit(&cmd->req);
+
+		nvmet_tcp_unmap_pdu_iovec(cmd);
+		nvmet_tcp_free_cmd_buffers(cmd);
 	}
 
 	if (!queue->nr_cmds && nvmet_tcp_need_data_in(&queue->connect)) {
@@ -1437,7 +1455,9 @@ static void nvmet_tcp_release_queue_work(struct work_struct *w)
 	mutex_unlock(&nvmet_tcp_queue_mutex);
 
 	nvmet_tcp_restore_socket_callbacks(queue);
-	flush_work(&queue->io_work);
+	cancel_work_sync(&queue->io_work);
+	/* stop accepting incoming data */
+	queue->rcv_state = NVMET_TCP_RECV_ERR;
 
 	nvmet_tcp_uninit_data_in_cmds(queue);
 	nvmet_sq_destroy(&queue->nvme_sq);