From: Douglas Gilbert <dgilbert@interlog.com>
To: linux-scsi@vger.kernel.org
Cc: martin.petersen@oracle.com, jejb@linux.vnet.ibm.com, hare@suse.de
Subject: [RFC v2 4/6] scsi: improve scsi_device_lookup
Date: Sun, 24 May 2020 11:58:12 -0400 [thread overview]
Message-ID: <20200524155814.5895-5-dgilbert@interlog.com> (raw)
In-Reply-To: <20200524155814.5895-1-dgilbert@interlog.com>
When the __scsi_device_lookup() function is given a "ctl" (i.e. the
latter part of "hctl" tuple), improve the loop to find a matching
device (LU) in the given host. Rather than loop over all LUs in the
host, first loop over all targets to find a match on "ct" then, if
found, loop over all LUs in that target for a match on "l". These
nested loops are better since they don't visit LUs belonging to
non-matching targets. This improvement flows through to the locked
version of this function, namely scsi_device_lookup().
Remove a 21 year old comment by the author that no longer seem
relevant.
Signed-off-by: Douglas Gilbert <dgilbert@interlog.com>
---
drivers/scsi/scsi.c | 26 ++++++++++++++++++++------
1 file changed, 20 insertions(+), 6 deletions(-)
diff --git a/drivers/scsi/scsi.c b/drivers/scsi/scsi.c
index 0fb650aebcfb..9e7658aebdb7 100644
--- a/drivers/scsi/scsi.c
+++ b/drivers/scsi/scsi.c
@@ -35,7 +35,6 @@
*
* Jiffies wrap fixes (host->resetting), 3 Dec 1998 Andrea Arcangeli
*
- * out_of_space hacks, D. Gilbert (dpg) 990608
*/
#include <linux/module.h>
@@ -789,16 +788,31 @@ EXPORT_SYMBOL(scsi_device_lookup_by_target);
struct scsi_device *__scsi_device_lookup(struct Scsi_Host *shost,
uint channel, uint id, u64 lun)
{
- unsigned long l_idx;
+ unsigned long l_idx, m_idx;
+ struct scsi_target *starg;
struct scsi_device *sdev;
- xa_for_each_marked(&shost->__devices, l_idx, sdev,
+ if (xa_empty(&shost->__devices))
+ return NULL;
+ if (xa_empty(&shost->__targets))
+ goto inconsistent;
+ xa_for_each(&shost->__targets, l_idx, starg) {
+ if (!(starg->id == id && starg->channel == channel))
+ continue;
+ xa_for_each_marked(&starg->devices, m_idx, sdev,
+ SCSI_XA_NON_SDEV_DEL) {
+ if (sdev->lun == lun)
+ return sdev;
+ }
+ }
+ return NULL;
+inconsistent:
+ xa_for_each_marked(&shost->__devices, m_idx, sdev,
SCSI_XA_NON_SDEV_DEL) {
- if (sdev->channel == channel && sdev->id == id &&
- sdev->lun ==lun)
+ if (sdev->id == id && sdev->channel == channel &&
+ sdev->lun == lun)
return sdev;
}
-
return NULL;
}
EXPORT_SYMBOL(__scsi_device_lookup);
--
2.25.1
next prev parent reply other threads:[~2020-05-24 15:58 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-05-24 15:58 [RFC v2 0/6] scsi: rework mid-layer with xarrays Douglas Gilbert
2020-05-24 15:58 ` [RFC v2 1/6] scsi: xarray hctl Douglas Gilbert
2020-05-25 6:57 ` Hannes Reinecke
2020-05-25 16:30 ` Douglas Gilbert
2020-05-25 17:40 ` Matthew Wilcox
2020-05-26 2:01 ` Douglas Gilbert
2020-05-26 3:01 ` Matthew Wilcox
2020-05-26 7:24 ` Hannes Reinecke
2020-05-26 6:21 ` Hannes Reinecke
2020-05-26 14:27 ` Matthew Wilcox
2020-05-26 17:53 ` Hannes Reinecke
2020-05-26 18:39 ` Matthew Wilcox
2020-05-26 19:28 ` James Bottomley
2020-05-26 19:10 ` Douglas Gilbert
2020-05-26 20:27 ` Douglas Gilbert
2020-05-27 2:53 ` Douglas Gilbert
2020-05-24 15:58 ` [RFC v2 2/6] scsi: xarray, iterations on scsi_target Douglas Gilbert
2020-05-25 7:06 ` Hannes Reinecke
2020-05-24 15:58 ` [RFC v2 3/6] scsi: xarray mark sdev_del state Douglas Gilbert
2020-05-25 7:00 ` Hannes Reinecke
2020-05-25 16:47 ` Douglas Gilbert
2020-05-24 15:58 ` Douglas Gilbert [this message]
2020-05-25 7:07 ` [RFC v2 4/6] scsi: improve scsi_device_lookup Hannes Reinecke
2020-05-24 15:58 ` [RFC v2 5/6] scsi: add __scsi_target_lookup Douglas Gilbert
2020-05-24 15:58 ` [RFC v2 6/6] scsi: count number of targets Douglas Gilbert
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=20200524155814.5895-5-dgilbert@interlog.com \
--to=dgilbert@interlog.com \
--cc=hare@suse.de \
--cc=jejb@linux.vnet.ibm.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).