On Thu, Jun 12, 2008 at 5:22 AM, Justin Mattock <justinmattock@xxxxxxxxx> wrote: > On Thu, Jun 12, 2008 at 4:15 AM, Tejun Heo <htejun@xxxxxxxxx> wrote: >> Hello, >> >> Please do the followings. >> >> 1. apply the attached patch >> 2. turn on device drivers -> SCSI -> SCSI verbose logging >> 3. rebuild the kernel and boot the new kernel. >> 4. run "sysctl -w dev.scsi.logging_level=1" >> 5. run the suspend/resume test w/ dd and report the result. >> >> Thanks. >> >> -- >> tejun >> >> --- >> drivers/ata/libata-eh.c | 6 ++++++ >> 1 file changed, 6 insertions(+) >> >> Index: work/drivers/ata/libata-eh.c >> =================================================================== >> --- work.orig/drivers/ata/libata-eh.c >> +++ work/drivers/ata/libata-eh.c >> @@ -2507,6 +2507,9 @@ static int atapi_eh_clear_ua(struct ata_ >> return -EIO; >> } >> >> + ata_dev_printk(dev, KERN_WARNING, "XXX: TUR err_mask=0x%x sense_key=0x%x\n", >> + err_mask, sense_key); >> + >> if (!err_mask || sense_key != UNIT_ATTENTION) >> return 0; >> >> @@ -2516,6 +2519,9 @@ static int atapi_eh_clear_ua(struct ata_ >> "UNIT ATTENTION (err_mask=0x%x)\n", err_mask); >> return -EIO; >> } >> + >> + ata_dev_printk(dev, KERN_WARNING, "XXX: REQUEST SENSE err_mask=0x%x ASC=0x%x ASCQ=0x%x\n", >> + err_mask, sense_buffer[12], sense_buffer[13]); >> } >> >> ata_dev_printk(dev, KERN_WARNING, >> >> > > Alright; I'll let you know what happens. > regards; > > -- > Justin P. Mattock > O.K. applied the patch and did what you have described,(except for sysctl -w / ended up inserting dev.scsi.logging_level=1 in /etc/sysctl.conf) This is what dd reported:(same as before) dd: reading `/dev/sr0':Input/output error 41+1 records in 41+1 records out 43122688 bytes (43 MB) copied, 41.0967 seconds, 1.0 MB/s Attached is dmesg of using dd on the first suspend, and playing a dvd on the second. regards; -- Justin P. Mattock
Attachment:
dmesg
Description: Binary data
