Patch 1 of 1 (9th in a series, if that makes sense)
This patch fixes the problem Bjorn reported. The busy_initializing flag
should have cleared before going into the for loop.
I didn't see this because I wasn't booting cciss on my test system.
The other patches I submitted should be applied first, or reapplied
whatever the case may be...
Please include this patch.
Signed-off-by: Mike Miller <mike.miller@xxxxxx>
cciss.c | 5 +----
1 files changed, 1 insertion(+), 4 deletions(-)
--------------------------------------------------------------------------------
diff -burNp lx2613-mm3/drivers/block/cciss.c lx2613-mm3-busy-bug/drivers/block/cciss.c
--- lx2613-mm3/drivers/block/cciss.c 2005-09-13 12:10:52.000000000 -0500
+++ lx2613-mm3-busy-bug/drivers/block/cciss.c 2005-09-17 15:13:05.596434712 -0500
@@ -483,9 +483,6 @@ static int cciss_open(struct inode *inod
printk(KERN_DEBUG "cciss_open %s\n", inode->i_bdev->bd_disk->disk_name);
#endif /* CCISS_DEBUG */
- if (host->busy_initializing)
- return -EBUSY;
-
if (host->busy_initializing || drv->busy_configuring)
return -EBUSY;
/*
@@ -2991,6 +2988,7 @@ static int __devinit cciss_init_one(stru
hba[i]->access.set_intr_mask(hba[i], CCISS_INTR_ON);
cciss_procinit(i);
+ hba[i]->busy_initializing = 0;
for(j=0; j < NWD; j++) { /* mfm */
drive_info_struct *drv = &(hba[i]->drv[j]);
@@ -3033,7 +3031,6 @@ static int __devinit cciss_init_one(stru
add_disk(disk);
}
- hba[i]->busy_initializing = 0;
return(1);
clean4:
-
: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
