Re: 2.5.46: ide-cd cdrecord (almost) success report

[Jens Axboe <axboe@suse.de>]

On Wed, Nov 06 2002, Adam Kropelin wrote:
> On Wed, Nov 06, 2002 at 08:44:57AM +0100, Jens Axboe wrote:
> > On Wed, Nov 06 2002, Jens Axboe wrote:
> > > > when the cdrecord buffer underran was surprising, though: oops below.
> > > > Very repeatable. Can supply copious hw details if it helps.
> > > 
> > > I'll try and reproduce that here, there's been a similar report (same
> > > oops) before. If you can just send me the dmesg output after a boot that
> > > should be fine.
> > 
> > Could you reproduce with this patch? I'd like to see the request state
> > when this happens.
> 
> Here you go... This was with max tags locked at 4. I've included some of
> the surrounding lines from cdrecord.

ok

> --Adam
> 
> Track 01:   85 of  437 MB written (fifo   0%) [buf  31%]   3.3x./opt/schily/bin/cdrecord: Input/output error. write_g1: scsi sendcmd: no error
> CDB:  2A 00 00 00 AA BE 00 00 1F 00
> status: 0x1 (GOOD STATUS)
> resid: 63488
> cmd finished after 0.011s timeout 40s
> 
> write track data: error after 89518080 bytes
> cdrom_newpc_intr: dev hdc: flags = REQ_RW REQ_NOMERGE REQ_STARTED REQ_BLOCK_PC REQ_FAILED REQ_QUIET
> sector 0, nr/cnr 124/8
> bio 00000000, biotail c1cd0860, buffer c968d000, data 00000000, len 63488
> cdb: 2a 00 00 00 aa be 00 00 1f 00 00 00 00 00 00 00
> hdc: padding 63488 bytes
> Sense Bytes: 70 00 00 00 00 00 00 12 00 00 00 00 00 00 00 00 00 00

>From the ide-cd POV, the best we can do here is just try to continue
safely in case of error. So a slight modification to last patch should
be fine, I think.

At least that is the atapi part of.

===== drivers/ide/ide-cd.c 1.32 vs edited =====
--- 1.32/drivers/ide/ide-cd.c	Sun Nov  3 19:57:35 2002
+++ edited/drivers/ide/ide-cd.c	Thu Nov  7 15:25:05 2002
@@ -1722,6 +1722,11 @@
 			blen = bio_iovec(rq->bio)->bv_len;
 		}
 
+		if (!ptr) {
+			printk("%s: confused, missing data\n", drive->name);
+			break;
+		}
+
 		if (blen > thislen)
 			blen = thislen;
 
@@ -1741,8 +1746,6 @@
 	 * pad, if necessary
 	 */
 	if (len) {
-		printk("%s: padding %u bytes\n", drive->name, len);
-
 		while (len) {
 			int pad = 0;
 

-- 
Jens Axboe