[patch 21/26] [libata] locking rewrite (== fix)
From: Greg KH
Date: Tue Dec 13 2005 - 03:33:00 EST
-stable review patch. If anyone has any objections, please let us know.
------------------
From: Jeff Garzik <jgarzik@xxxxxxxxx>
[libata] locking rewrite (== fix)
A lot of power packed into a little patch.
This change eliminates the sharing between our controller-wide spinlock
and the SCSI core's Scsi_Host lock. As the locking in libata was
already highly compartmentalized, always referencing our own lock, and
never scsi_host::host_lock.
As a side effect, this change eliminates a deadlock from calling
scsi_finish_command() while inside our spinlock.
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxx>
---
drivers/scsi/libata-core.c | 2 --
drivers/scsi/libata-scsi.c | 9 ++++++++-
2 files changed, 8 insertions(+), 3 deletions(-)
--- linux-2.6.14.3.orig/drivers/scsi/libata-core.c
+++ linux-2.6.14.3/drivers/scsi/libata-core.c
@@ -3916,8 +3916,6 @@ static void ata_host_init(struct ata_por
host->unique_id = ata_unique_id++;
host->max_cmd_len = 12;
- scsi_assign_lock(host, &host_set->lock);
-
ap->flags = ATA_FLAG_PORT_DISABLED;
ap->id = host->unique_id;
ap->host = host;
--- linux-2.6.14.3.orig/drivers/scsi/libata-scsi.c
+++ linux-2.6.14.3/drivers/scsi/libata-scsi.c
@@ -39,6 +39,7 @@
#include <scsi/scsi.h>
#include "scsi.h"
#include <scsi/scsi_host.h>
+#include <scsi/scsi_device.h>
#include <linux/libata.h>
#include <asm/uaccess.h>
@@ -1565,8 +1566,12 @@ int ata_scsi_queuecmd(struct scsi_cmnd *
struct ata_port *ap;
struct ata_device *dev;
struct scsi_device *scsidev = cmd->device;
+ struct Scsi_Host *shost = scsidev->host;
- ap = (struct ata_port *) &scsidev->host->hostdata[0];
+ ap = (struct ata_port *) &shost->hostdata[0];
+
+ spin_unlock(shost->host_lock);
+ spin_lock(&ap->host_set->lock);
ata_scsi_dump_cdb(ap, cmd);
@@ -1589,6 +1594,8 @@ int ata_scsi_queuecmd(struct scsi_cmnd *
ata_scsi_translate(ap, dev, cmd, done, atapi_xlat);
out_unlock:
+ spin_unlock(&ap->host_set->lock);
+ spin_lock(shost->host_lock);
return 0;
}
--
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/