On Tue, 6 May 2003, Jens Axboe wrote:
> You can play with the c code, you've demonstrated that much so far. So
> play some more, find out which commands are aborted and why. The log
> messages even tell you which ones. Now find out if these are necessary
> for proper MO functionality or not. Or maybe some vital commands are
> even missing, lots of fun there :). But it really should not be very
> hard.
Ok, I got a little bit further with the patch below. I can mount to disk
read-write and do a few things:
$ mount -t ext2 /dev/hde /mnt/mo
$ echo "foo" > /mnt/mo/bar
$ cat /mnt/mo/bar
foo
$ sync
No problem so far. But then
$ umount /mnt/mo
gets me:
------------[ cut here ]------------
kernel BUG at fs/buffer.c:2607!
invalid operand: 0000 [#1]
CPU: 0
EIP: 0060:[<c01466f6>] Not tainted
EFLAGS: 00010202
EIP is at submit_bh+0x26/0x110
eax: 00000010 ebx: da449bd0 ecx: dfbb4c34 edx: c1494e98
esi: 00000001 edi: c0394900 ebp: de77d200 esp: de04bea8
ds: 007b es: 007b ss: 0068
Process umount (pid: 1161, threadinfo=de04a000 task=de210c60)
Stack: dd52a400 da449bd0 c01468e0 00000001 da449bd0 dfbb4b84 00000004 da449bd0
de77d200 dd52a400 de77d200 c0188342 da449bd0 00000000 df7c0498 c0186fb9
de77d200 dd52a400 de04a000 00000000 de04bf08 c0157e44 de04bf08 dfbb4c18
Call Trace:
[<c01468e0>] sync_dirty_buffer+0x80/0xb0
[<c0188342>] ext2_sync_super+0x42/0x50
[<c0186fb9>] ext2_put_super+0x29/0x90
[<c0157e44>] invalidate_inodes+0xa4/0xc0
[<c0147d70>] generic_shutdown_super+0xa0/0x140
[<c0148682>] kill_block_super+0x12/0x30
[<c0147bb7>] deactivate_super+0x67/0xa0
[<c015a467>] __mntput+0x17/0x30
[<c014d837>] path_release+0x27/0x30
[<c015abff>] sys_umount+0x7f/0x90
[<c0138783>] sys_munmap+0x43/0x60
[<c015ac1b>] sys_oldumount+0xb/0x10
[<c0108dbf>] syscall_call+0x7/0xb
Code: 0f 0b 2f 0a a6 3a 33 c0 8b 4b 20 85 c9 75 08 0f 0b 30 0a a6
umount then segfaults. The BUG is from:
int submit_bh(int rw, struct buffer_head * bh)
{
struct bio *bio;
BUG_ON(!buffer_locked(bh));
BUG_ON(!buffer_mapped(bh));
^^^^^^^^^^^^^^^^^^^^^^^^^^^ this one
After this, the MO drive is dead until I reboot. Trying to mount the disk
read-only after that reboot works without problems. Trying to mount
read-write gives the same BUG_ON again on umount.
Running e2fsck from a 2.4 kernel shows that the directory entry for
/mnt/mo/bar did in fact get written to the disk, but fsck reports it
as using an unused inode, so I guess the inode and data writes did
not happen due to the BUG_ON.
Something seems to missing still.
My patch:
--- linux-2.5/include/linux/cdrom.h Thu Apr 24 16:39:43 2003
+++ build25/include/linux/cdrom.h Tue May 6 18:39:41 2003
@@ -387,6 +387,7 @@ struct cdrom_generic_command
#define CDC_DVD 0x8000 /* drive is a DVD */
#define CDC_DVD_R 0x10000 /* drive can write DVD-R */
#define CDC_DVD_RAM 0x20000 /* drive can write DVD-RAM */
+#define CDC_MO 0x40000 /* drive is an MO */
/* drive status possibilities returned by CDROM_DRIVE_STATUS ioctl */
#define CDS_NO_INFO 0 /* if not implemented */
--- linux-2.5/drivers/cdrom/cdrom.c Fri Mar 21 07:01:46 2003
+++ build25/drivers/cdrom/cdrom.c Tue May 6 18:40:51 2003
@@ -425,7 +425,8 @@ int cdrom_open(struct cdrom_device_info
if ((fp->f_flags & O_NONBLOCK) && (cdi->options & CDO_USE_FFLAGS))
ret = cdi->ops->open(cdi, 1);
else {
- if ((fp->f_mode & FMODE_WRITE) && !CDROM_CAN(CDC_DVD_RAM))
+ if ((fp->f_mode & FMODE_WRITE)
+ && !(CDROM_CAN(CDC_DVD_RAM) || CDROM_CAN(CDC_MO)))
return -EROFS;
ret = open_for_data(cdi);
--- linux-2.5/drivers/ide/ide.c Thu May 1 15:01:12 2003
+++ build25/drivers/ide/ide.c Tue May 6 17:01:34 2003
@@ -1518,6 +1518,7 @@ int generic_ide_ioctl(struct block_devic
return ide_taskfile_ioctl(drive, cmd, arg);
#ifdef CONFIG_PKT_TASK_IOCTL
case ide_cdrom:
+ case ide_optical:
case ide_tape:
case ide_floppy:
return pkt_taskfile_ioctl(drive, cmd, arg);
--- linux-2.5/drivers/ide/ide-probe.c Sat Apr 26 18:49:57 2003
+++ build25/drivers/ide/ide-probe.c Tue May 6 19:07:37 2003
@@ -1235,7 +1235,7 @@ struct gendisk *ata_probe(dev_t dev, int
(void) request_module("ide-disk");
if (drive->scsi)
(void) request_module("ide-scsi");
- if (drive->media == ide_cdrom)
+ if ((drive->media == ide_cdrom) || (drive->media == ide_optical))
(void) request_module("ide-cd");
if (drive->media == ide_tape)
(void) request_module("ide-tape");
--- linux-2.5/drivers/ide/ide-cd.h Sat Nov 2 22:12:01 2002
+++ build25/drivers/ide/ide-cd.h Tue May 6 18:33:28 2003
@@ -85,7 +85,8 @@ struct ide_cd_config_flags {
__u8 audio_play : 1; /* can do audio related commands */
__u8 close_tray : 1; /* can close the tray */
__u8 writing : 1; /* pseudo write in progress */
- __u8 reserved : 3;
+ __u8 mo : 1; /* Drive is an MO */
+ __u8 reserved : 2;
byte max_speed; /* Max speed of the drive */
};
#define CDROM_CONFIG_FLAGS(drive) (&(((struct cdrom_info *)(drive->driver_data))->config_flags))
--- linux-2.5/drivers/ide/ide-cd.c Sun Apr 20 23:52:14 2003
+++ build25/drivers/ide/ide-cd.c Tue May 6 19:26:24 2003
@@ -2793,7 +2793,7 @@ static struct cdrom_device_ops ide_cdrom
CDC_MEDIA_CHANGED | CDC_PLAY_AUDIO | CDC_RESET |
CDC_IOCTLS | CDC_DRIVE_STATUS | CDC_CD_R |
CDC_CD_RW | CDC_DVD | CDC_DVD_R| CDC_DVD_RAM |
- CDC_GENERIC_PACKET,
+ CDC_GENERIC_PACKET | CDC_MO,
.generic_packet = ide_cdrom_packet,
};
@@ -2826,6 +2826,8 @@ static int ide_cdrom_register (ide_drive
devinfo->mask |= CDC_PLAY_AUDIO;
if (!CDROM_CONFIG_FLAGS(drive)->close_tray)
devinfo->mask |= CDC_CLOSE_TRAY;
+ if (!CDROM_CONFIG_FLAGS(drive)->mo)
+ devinfo->mask |= CDC_MO;
return register_cdrom(devinfo);
}
@@ -2871,12 +2873,17 @@ int ide_cdrom_probe_capabilities (ide_dr
struct atapi_capabilities_page cap;
int nslots = 1;
+ if (drive->media == ide_optical) {
+ CDROM_CONFIG_FLAGS(drive)->mo = 1;
+ return nslots;
+ }
+
if (CDROM_CONFIG_FLAGS(drive)->nec260) {
CDROM_CONFIG_FLAGS(drive)->no_eject = 0;
CDROM_CONFIG_FLAGS(drive)->audio_play = 1;
return nslots;
}
-
+
if (ide_cdrom_get_capabilities(drive, &cap))
return 0;
@@ -3104,6 +3111,7 @@ int ide_cdrom_setup (ide_drive_t *drive)
CDROM_CONFIG_FLAGS(drive)->supp_disc_present = 0;
CDROM_CONFIG_FLAGS(drive)->audio_play = 0;
CDROM_CONFIG_FLAGS(drive)->close_tray = 1;
+ CDROM_CONFIG_FLAGS(drive)->mo = 0;
/* limit transfer size per interrupt. */
CDROM_CONFIG_FLAGS(drive)->limit_nframes = 0;
@@ -3180,7 +3188,8 @@ int ide_cdrom_setup (ide_drive_t *drive)
nslots = ide_cdrom_probe_capabilities (drive);
- if (CDROM_CONFIG_FLAGS(drive)->dvd_ram)
+ if (CDROM_CONFIG_FLAGS(drive)->dvd_ram
+ || CDROM_CONFIG_FLAGS(drive)->mo)
set_disk_ro(drive->disk, 0);
#if 0
@@ -3337,7 +3346,7 @@ static int ide_cdrom_attach (ide_drive_t
goto failed;
if (!drive->present)
goto failed;
- if (drive->media != ide_cdrom)
+ if ((drive->media != ide_cdrom) && (drive->media != ide_optical))
goto failed;
/* skip drives that we were told to ignore */
if (ignore != NULL) {
-- Ciao, Pascal- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Wed May 07 2003 - 22:00:27 EST