This patch started life as as533, and has been re-diffed against the current tree.
Disconnect processing in usb-storage naturally divides into two parts: one
to quiesce the driver (make sure no commands are executing or queued) and
remove the host, and the other to deallocate all the USB and non-USB
resources. This patch creates two subroutines to handle those two parts.
Mostly it's just code movement, but there is one significant change. If
the scsi-scanning thread fails to initialize but the host has successfully
been added, we need to quiesce the driver before removing the host.
After all, it's possible that scanning could have been initiated from
somewhere else, such as userspace -- very low probability, but it's easily
handled by calling the new subroutine.
Greg, please apply.
Matt
Signed-off-by: Alan Stern <[EMAIL PROTECTED]>
Signed-off-by: Matthew Dharm <[EMAIL PROTECTED]>
Index: linux-2.6.13-rc3/drivers/usb/storage/usb.c
===================================================================
--- linux-2.6.13-rc3.orig/drivers/usb/storage/usb.c 2005-07-12
21:46:46.000000000 -0700
+++ linux-2.6.13-rc3/drivers/usb/storage/usb.c 2005-07-24 21:40:46.000000000
-0700
@@ -786,6 +786,7 @@
* any more commands.
*/
US_DEBUGP("-- sending exit command to thread\n");
+ set_bit(US_FLIDX_DISCONNECTING, &us->flags);
up(&us->sema);
/* Call the destructor routine, if it exists */
@@ -816,6 +817,36 @@
usb_set_intfdata(us->pusb_intf, NULL);
}
+/* First stage of disconnect processing: stop all commands and remove
+ * the host */
+static void quiesce_and_remove_host(struct us_data *us)
+{
+ /* Prevent new USB transfers, stop the current command, and
+ * interrupt a SCSI-scan or device-reset delay */
+ set_bit(US_FLIDX_DISCONNECTING, &us->flags);
+ usb_stor_stop_transport(us);
+ wake_up(&us->delay_wait);
+
+ /* It doesn't matter if the SCSI-scanning thread is still running.
+ * The thread will exit when it sees the DISCONNECTING flag. */
+
+ /* Wait for the current command to finish, then remove the host */
+ down(&us->dev_semaphore);
+ up(&us->dev_semaphore);
+ scsi_remove_host(us_to_host(us));
+}
+
+/* Second stage of disconnect processing: deallocate all resources */
+static void release_everything(struct us_data *us)
+{
+ usb_stor_release_resources(us);
+ dissociate_dev(us);
+
+ /* Drop our reference to the host; the SCSI core will free it
+ * (and "us" along with it) when the refcount becomes 0. */
+ scsi_host_put(us_to_host(us));
+}
+
/* Thread to carry out delayed SCSI-device scanning */
static int usb_stor_scan_thread(void * __us)
{
@@ -956,7 +987,7 @@
if (result < 0) {
printk(KERN_WARNING USB_STORAGE
"Unable to start the device-scanning thread\n");
- scsi_remove_host(host);
+ quiesce_and_remove_host(us);
goto BadDevice;
}
atomic_inc(&total_threads);
@@ -969,10 +1000,7 @@
/* We come here if there are any problems */
BadDevice:
US_DEBUGP("storage_probe() failed\n");
- set_bit(US_FLIDX_DISCONNECTING, &us->flags);
- usb_stor_release_resources(us);
- dissociate_dev(us);
- scsi_host_put(host);
+ release_everything(us);
return result;
}
@@ -982,28 +1010,8 @@
struct us_data *us = usb_get_intfdata(intf);
US_DEBUGP("storage_disconnect() called\n");
-
- /* Prevent new USB transfers, stop the current command, and
- * interrupt a SCSI-scan or device-reset delay */
- set_bit(US_FLIDX_DISCONNECTING, &us->flags);
- usb_stor_stop_transport(us);
- wake_up(&us->delay_wait);
-
- /* It doesn't matter if the SCSI-scanning thread is still running.
- * The thread will exit when it sees the DISCONNECTING flag. */
-
- /* Wait for the current command to finish, then remove the host */
- down(&us->dev_semaphore);
- up(&us->dev_semaphore);
- scsi_remove_host(us_to_host(us));
-
- /* Wait for everything to become idle and release all our resources */
- usb_stor_release_resources(us);
- dissociate_dev(us);
-
- /* Drop our reference to the host; the SCSI core will free it
- * (and "us" along with it) when the refcount becomes 0. */
- scsi_host_put(us_to_host(us));
+ quiesce_and_remove_host(us);
+ release_everything(us);
}
/***********************************************************************
--
Matthew Dharm Home: [EMAIL PROTECTED]
Maintainer, Linux USB Mass Storage Driver
We can customize our colonels.
-- Tux
User Friendly, 12/1/1998
pgpV8H0B3y5gk.pgp
Description: PGP signature
