rpms/kernel/devel linux-2.6-scsi-async-double-add.patch, NONE, 1.1 kernel.spec, 1.120, 1.121

[Dave Jones (davej)]

Author: davej

Update of /cvs/pkgs/rpms/kernel/devel
In directory cvs-int.fedora.redhat.com:/tmp/cvs-serv17735

Modified Files:
	kernel.spec 
Added Files:
	linux-2.6-scsi-async-double-add.patch 
Log Message:
* Sun Aug 26 2007 Dave Jones <davej at redhat.com>
- Fix async scanning double-add problems.


linux-2.6-scsi-async-double-add.patch:

--- NEW FILE linux-2.6-scsi-async-double-add.patch ---
From: Matthew Wilcox <matthew at wil.cx>
Date: Tue, 26 Jun 2007 21:18:51 +0000 (-0600)
Subject: [SCSI] Fix async scanning double-add problems
X-Git-Url: http://git.kernel.org/?p=linux%2Fkernel%2Fgit%2Fjejb%2Fscsi-misc-2.6.git;a=commitdiff_plain;h=a93a091df8232fad60867d41fbc3be855a0b78f2

[SCSI] Fix async scanning double-add problems

Stress-testing and some thought has revealed some places where
asynchronous scanning needs some more attention to locking.

 - Since async_scan is a bit, we need to hold the host_lock while
   modifying it to prevent races against other CPUs modifying the word
   that bit is in.  This is probably a theoretical race for the moment,
   but other patches may change that.
 - The async_scan bit means not only that this host is being scanned
   asynchronously, but that all the devices attached to this host are not
   yet added to sysfs.  So we must ensure that this bit is always in sync.
   I've chosen to do this with the scan_mutex since it's already acquired
   in most of the right places.
 - If the host changes state to deleted while we're in the middle of
   a scan, we'll end up with some devices on the host's list which must
   be deleted.  Add a check to scsi_sysfs_add_devices() to ensure the
   host is still running.
 - To avoid the async_scan bit being protected by three locks, the
   async_scan_lock now only protects the scanning_list.

Signed-off-by: Matthew Wilcox <matthew at wil.cx>
Signed-off-by: James Bottomley <James.Bottomley at SteelEye.com>
---

diff --git a/drivers/scsi/scsi_scan.c b/drivers/scsi/scsi_scan.c
index a86e62f..309b224 100644
--- a/drivers/scsi/scsi_scan.c
+++ b/drivers/scsi/scsi_scan.c
@@ -121,6 +121,7 @@ MODULE_PARM_DESC(inq_timeout,
 		 "Timeout (in seconds) waiting for devices to answer INQUIRY."
 		 " Default is 5. Some non-compliant devices need more.");
 
+/* This lock protects only this list */
 static DEFINE_SPINLOCK(async_scan_lock);
 static LIST_HEAD(scanning_hosts);
 
@@ -1466,14 +1467,14 @@ struct scsi_device *__scsi_add_device(struct Scsi_Host *shost, uint channel,
 	if (strncmp(scsi_scan_type, "none", 4) == 0)
 		return ERR_PTR(-ENODEV);
 
-	if (!shost->async_scan)
-		scsi_complete_async_scans();
-
 	starget = scsi_alloc_target(parent, channel, id);
 	if (!starget)
 		return ERR_PTR(-ENOMEM);
 
 	mutex_lock(&shost->scan_mutex);
+	if (!shost->async_scan)
+		scsi_complete_async_scans();
+
 	if (scsi_host_scan_allowed(shost))
 		scsi_probe_and_add_lun(starget, lun, NULL, &sdev, 1, hostdata);
 	mutex_unlock(&shost->scan_mutex);
@@ -1586,10 +1587,10 @@ void scsi_scan_target(struct device *parent, unsigned int channel,
 	if (strncmp(scsi_scan_type, "none", 4) == 0)
 		return;
 
+	mutex_lock(&shost->scan_mutex);
 	if (!shost->async_scan)
 		scsi_complete_async_scans();
 
-	mutex_lock(&shost->scan_mutex);
 	if (scsi_host_scan_allowed(shost))
 		__scsi_scan_target(parent, channel, id, lun, rescan);
 	mutex_unlock(&shost->scan_mutex);
@@ -1634,15 +1635,15 @@ int scsi_scan_host_selected(struct Scsi_Host *shost, unsigned int channel,
 		"%s: <%u:%u:%u>\n",
 		__FUNCTION__, channel, id, lun));
 
-	if (!shost->async_scan)
-		scsi_complete_async_scans();
-
 	if (((channel != SCAN_WILD_CARD) && (channel > shost->max_channel)) ||
 	    ((id != SCAN_WILD_CARD) && (id >= shost->max_id)) ||
 	    ((lun != SCAN_WILD_CARD) && (lun > shost->max_lun)))
 		return -EINVAL;
 
 	mutex_lock(&shost->scan_mutex);
+	if (!shost->async_scan)
+		scsi_complete_async_scans();
+
 	if (scsi_host_scan_allowed(shost)) {
 		if (channel == SCAN_WILD_CARD)
 			for (channel = 0; channel <= shost->max_channel;
@@ -1661,7 +1662,8 @@ static void scsi_sysfs_add_devices(struct Scsi_Host *shost)
 {
 	struct scsi_device *sdev;
 	shost_for_each_device(sdev, shost) {
-		if (scsi_sysfs_add_sdev(sdev) != 0)
+		if (!scsi_host_scan_allowed(shost) ||
+		    scsi_sysfs_add_sdev(sdev) != 0)
 			scsi_destroy_sdev(sdev);
 	}
 }
@@ -1679,6 +1681,7 @@ static void scsi_sysfs_add_devices(struct Scsi_Host *shost)
 static struct async_scan_data *scsi_prep_async_scan(struct Scsi_Host *shost)
 {
 	struct async_scan_data *data;
+	unsigned long flags;
 
 	if (strncmp(scsi_scan_type, "sync", 4) == 0)
 		return NULL;
@@ -1698,8 +1701,13 @@ static struct async_scan_data *scsi_prep_async_scan(struct Scsi_Host *shost)
 		goto err;
 	init_completion(&data->prev_finished);
 
-	spin_lock(&async_scan_lock);
+	mutex_lock(&shost->scan_mutex);
+	spin_lock_irqsave(shost->host_lock, flags);
 	shost->async_scan = 1;
+	spin_unlock_irqrestore(shost->host_lock, flags);
+	mutex_unlock(&shost->scan_mutex);
+
+	spin_lock(&async_scan_lock);
 	if (list_empty(&scanning_hosts))
 		complete(&data->prev_finished);
 	list_add_tail(&data->list, &scanning_hosts);
@@ -1723,11 +1731,15 @@ static struct async_scan_data *scsi_prep_async_scan(struct Scsi_Host *shost)
 static void scsi_finish_async_scan(struct async_scan_data *data)
 {
 	struct Scsi_Host *shost;
+	unsigned long flags;
 
 	if (!data)
 		return;
 
 	shost = data->shost;
+
+	mutex_lock(&shost->scan_mutex);
+
 	if (!shost->async_scan) {
 		printk("%s called twice for host %d", __FUNCTION__,
 				shost->host_no);
@@ -1739,8 +1751,13 @@ static void scsi_finish_async_scan(struct async_scan_data *data)
 
 	scsi_sysfs_add_devices(shost);
 
-	spin_lock(&async_scan_lock);
+	spin_lock_irqsave(shost->host_lock, flags);
 	shost->async_scan = 0;
+	spin_unlock_irqrestore(shost->host_lock, flags);
+
+	mutex_unlock(&shost->scan_mutex);
+
+	spin_lock(&async_scan_lock);
 	list_del(&data->list);
 	if (!list_empty(&scanning_hosts)) {
 		struct async_scan_data *next = list_entry(scanning_hosts.next,


Index: kernel.spec
===================================================================
RCS file: /cvs/pkgs/rpms/kernel/devel/kernel.spec,v
retrieving revision 1.120
retrieving revision 1.121
diff -u -r1.120 -r1.121
--- kernel.spec	26 Aug 2007 22:57:18 -0000	1.120
+++ kernel.spec	26 Aug 2007 23:00:23 -0000	1.121
@@ -605,6 +605,7 @@
 Patch350: linux-2.6-devmem.patch
 Patch370: linux-2.6-crash-driver.patch
 Patch400: linux-2.6-scsi-cpqarray-set-master.patch
+Patch401: linux-2.6-scsi-async-double-add.patch
 Patch420: linux-2.6-squashfs.patch
 Patch423: linux-2.6-gfs-locking-exports.patch
 Patch430: linux-2.6-net-silence-noisy-printks.patch
@@ -1061,6 +1062,8 @@
 #
 # fix cpqarray pci enable
 ApplyPatch linux-2.6-scsi-cpqarray-set-master.patch
+# Fix async scanning double-add problems
+ApplyPatch linux-2.6-scsi-async-double-add.patch
 
 # Filesystem patches.
 # Squashfs
@@ -1764,6 +1767,9 @@
 
 %changelog
 * Sun Aug 26 2007 Dave Jones <davej at redhat.com>
+- Fix async scanning double-add problems.
+
+* Sun Aug 26 2007 Dave Jones <davej at redhat.com>
 - 2.6.23-rc3-git10
 
 * Fri Aug 24 2007 Kristian Høgsberg <krh at redhat.com>