All of lore.kernel.org
 help / color / mirror / Atom feed
From: Hannes Reinecke <hare@suse.de>
To: "Martin K. Petersen" <martin.petersen@oracle.com>
Cc: Christoph Hellwig <hch@lst.de>,
	James Bottomley <james.bottomley@hansenpartnership.com>,
	linux-scsi@vger.kernel.org, Hannes Reinecke <hare@suse.de>
Subject: [PATCH 08/10] scsi_error: iterate over list of failed commands in scsi_eh_bus_device_reset()
Date: Mon, 23 Oct 2023 11:28:35 +0200	[thread overview]
Message-ID: <20231023092837.33786-9-hare@suse.de> (raw)
In-Reply-To: <20231023092837.33786-1-hare@suse.de>

Iterating over all devices in scsi_eh_bus_device_reset() is
inefficient as not all devices may be affected during SCSI EH.
There also is no reason to open-code scsi_eh_test_devices()
as for FAST_IO_FAIL we really should just return the commands
without further ado.
So rewrite the loop in scsi_eh_bus_device_reset() to match the
loop in scsi_eh_target_reset() and scsi_eh_bus_reset().

Signed-off-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Christoph Hellwig <hch@lst.de>
---
 drivers/scsi/scsi_error.c | 58 ++++++++++++++++++---------------------
 1 file changed, 27 insertions(+), 31 deletions(-)

diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c
index 63b762d5d66f..7c655d08a305 100644
--- a/drivers/scsi/scsi_error.c
+++ b/drivers/scsi/scsi_error.c
@@ -1570,54 +1570,50 @@ static int scsi_eh_bus_device_reset(struct Scsi_Host *shost,
 				    struct list_head *work_q,
 				    struct list_head *done_q)
 {
-	struct scsi_cmnd *scmd, *bdr_scmd, *next;
-	struct scsi_device *sdev;
-	enum scsi_disposition rtn;
+	struct scsi_cmnd *scmd, *next;
+	LIST_HEAD(tmp_list);
+	LIST_HEAD(check_list);
+
+	list_splice_init(work_q, &tmp_list);
+
+	while (!list_empty(&tmp_list)) {
+		struct scsi_device *sdev;
+		enum scsi_disposition rtn;
+
+		scmd = list_entry(tmp_list.next, struct scsi_cmnd, eh_entry);
+		sdev = scmd->device;
 
-	shost_for_each_device(sdev, shost) {
 		if (scsi_host_eh_past_deadline(shost)) {
+			/* push back on work queue for further processing */
+			list_splice_init(&check_list, work_q);
+			list_splice_init(&tmp_list, work_q);
 			SCSI_LOG_ERROR_RECOVERY(3,
 				sdev_printk(KERN_INFO, sdev,
 					    "%s: skip BDR, past eh deadline\n",
 					     current->comm));
-			scsi_device_put(sdev);
-			break;
+			return list_empty(work_q);
 		}
-		bdr_scmd = NULL;
-		list_for_each_entry(scmd, work_q, eh_entry)
-			if (scmd->device == sdev) {
-				bdr_scmd = scmd;
-				break;
-			}
-
-		if (!bdr_scmd)
-			continue;
-
 		SCSI_LOG_ERROR_RECOVERY(3,
 			sdev_printk(KERN_INFO, sdev,
 				     "%s: Sending BDR\n", current->comm));
 		rtn = scsi_try_bus_device_reset(sdev);
-		if (rtn == SUCCESS || rtn == FAST_IO_FAIL) {
-			if (!scsi_device_online(sdev) ||
-			    rtn == FAST_IO_FAIL ||
-			    !scsi_eh_tur(bdr_scmd)) {
-				list_for_each_entry_safe(scmd, next,
-							 work_q, eh_entry) {
-					if (scmd->device == sdev &&
-					    scsi_eh_action(scmd, rtn) != FAILED) {
-						set_host_byte(scmd, DID_RESET);
-						scsi_eh_finish_cmd(scmd, done_q);
-					}
-				}
-			}
-		} else {
+		if (rtn != SUCCESS && rtn != FAST_IO_FAIL)
 			SCSI_LOG_ERROR_RECOVERY(3,
 				sdev_printk(KERN_INFO, sdev,
 					    "%s: BDR failed\n", current->comm));
+		list_for_each_entry_safe(scmd, next, work_q, eh_entry) {
+			if (scmd->device != sdev)
+				continue;
+			if (rtn == SUCCESS)
+				list_move_tail(&scmd->eh_entry, &check_list);
+			else if (rtn == FAST_IO_FAIL) {
+				set_host_byte(scmd, DID_TRANSPORT_FAILFAST);
+				scsi_eh_finish_cmd(scmd, done_q);
+			}
 		}
 	}
 
-	return list_empty(work_q);
+	return scsi_eh_test_devices(&check_list, work_q, done_q, 0);
 }
 
 /**
-- 
2.35.3


  parent reply	other threads:[~2023-10-23  9:29 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-10-23  9:28 [PATCHv8 00/10] scsi: EH rework, main part Hannes Reinecke
2023-10-23  9:28 ` [PATCH 01/10] scsi: Use Scsi_Host as argument for eh_host_reset_handler Hannes Reinecke
2023-10-25 11:40   ` Benjamin Block
2023-10-23  9:28 ` [PATCH 02/10] scsi: Use Scsi_Host and channel number as argument for eh_bus_reset_handler() Hannes Reinecke
2023-10-25 13:33   ` Benjamin Block
2023-10-25 13:36     ` Hannes Reinecke
2023-10-25 15:15       ` Benjamin Block
2023-10-23  9:28 ` [PATCH 03/10] scsi: Use scsi_target as argument for eh_target_reset_handler() Hannes Reinecke
2023-10-25 15:11   ` Benjamin Block
2023-10-25 15:35     ` Hannes Reinecke
2023-10-25 15:49       ` Benjamin Block
2023-10-23  9:28 ` [PATCH 04/10] scsi: Use scsi_device as argument to eh_device_reset_handler() Hannes Reinecke
2023-10-26 12:24   ` Benjamin Block
2023-10-26 12:39     ` Hannes Reinecke
2023-10-26 12:44       ` Benjamin Block
2023-10-23  9:28 ` [PATCH 05/10] scsi: set host byte after EH completed Hannes Reinecke
2023-10-26 12:58   ` Benjamin Block
2023-10-23  9:28 ` [PATCH 06/10] scsi_error: iterate over list of failed commands in scsi_eh_bus_reset() Hannes Reinecke
2023-10-26 13:19   ` Benjamin Block
2023-10-23  9:28 ` [PATCH 07/10] scsi: Do not allocate scsi command in scsi_ioctl_reset() Hannes Reinecke
2023-10-26 13:47   ` Benjamin Block
2023-10-23  9:28 ` Hannes Reinecke [this message]
2023-10-26 13:54   ` [PATCH 08/10] scsi_error: iterate over list of failed commands in scsi_eh_bus_device_reset() Benjamin Block
2023-10-23  9:28 ` [PATCH 09/10] scsi_error: map FAST_IO_FAIL to -EAGAIN in SCSI EH Hannes Reinecke
2023-10-26 14:08   ` Benjamin Block
2023-10-26 16:50   ` Mike Christie
2023-10-27  5:05     ` Hannes Reinecke
2023-10-23  9:28 ` [PATCH 10/10] scsi: remove SUBMITTED_BY_SCSI_RESET_IOCTL Hannes Reinecke
2023-10-26 14:12   ` Benjamin Block
2023-10-24 17:30 ` [PATCHv8 00/10] scsi: EH rework, main part Benjamin Block
2023-10-24 17:40   ` Hannes Reinecke
2023-10-24 17:57     ` Benjamin Block
2024-03-06 13:40 ` Wenchao Hao
2024-11-05  9:10 ` Niklas Cassel
2024-11-05 15:22   ` Hannes Reinecke
2024-11-05 15:47     ` Niklas Cassel

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20231023092837.33786-9-hare@suse.de \
    --to=hare@suse.de \
    --cc=hch@lst.de \
    --cc=james.bottomley@hansenpartnership.com \
    --cc=linux-scsi@vger.kernel.org \
    --cc=martin.petersen@oracle.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.