2003-03-18 22:59:40

by Eric Benson

[permalink] [raw]
Subject: ide-scsi failure on 2.5.65

I installed Red Hat 8.0 on an IBM NetVista 2283-55U,
an all-in-one desktop 1.8ghz P4. I downloaded and
compiled the 2.5.65 kernel with ide-scsi emulation and
kernel debugging enabled.

These are probably the relevant settings from the
config:
CONFIG_BLK_DEV_IDESCSI=y
CONFIG_SCSI=y
CONFIG_DEBUG_KERNEL=y
CONFIG_KALLSYMS=y
CONFIG_DEBUG_SPINLOCK_SLEEP=y
CONFIG_FRAME_POINTER=y

And these are extracted from the bootup messages:
PCI: Using IRQ router PIIX [8086/2440] at 00:1f.0
ICH2: IDE controller at PCI slot 00:1f.1

hdb: SAMSUNG CDRW/DVD SM-308B, ATAPI CD/DVD-ROM drive
scsi0 : SCSI host adapter emulation for IDE ATAPI
devices
Vendor: SAMSUNG Model: CDRW/DVD SM-308B Rev: T103

The application was "cdrdao read-cd", a recent
version that worked without difficulty on Red Hat's
2.4.18 kernels. Running the same command on 2.5.65
causes a failure that prints the following into
/var/log/messages.

14:18:02 hdb: irq timeout: status=0xd0 { Busy }
14:18:02 hdb: DMA disabled
14:18:02 hdb: ATAPI reset complete
14:18:02 hdb: status error: status=0x08 { DataRequest
}
14:18:02 hdb: drive not ready for command
14:18:02 hdb: irq timeout: status=0xd0 { Busy }
14:18:02 hdb: ATAPI reset complete
14:18:02 hdb: irq timeout: status=0x80 { Busy }
14:18:02 hdb: status timeout: status=0x80 { Busy }
14:18:02 hdb: drive not ready for command
14:18:02 hdb: ATAPI reset complete
14:18:02 hdb: status error: status=0x08 { DataRequest
}
14:18:02 hdb: drive not ready for command
14:18:02 hdb: status error: status=0x08 { DataRequest
}
14:18:02 hdb: drive not ready for command
14:18:03 hdb: status error: status=0x08 { DataRequest
}
14:18:33 hdb: drive not ready for command
14:18:33 ide-scsi: reset called for 0
14:18:33 bad: scheduling while atomic!
14:18:33 Call Trace:
14:18:33 [<c028c366>] ide_dma_intr+0x0/0xba
14:18:33 [<c01160fb>] schedule+0x3a3/0x3a8
14:18:33 [<c028c366>] ide_dma_intr+0x0/0xba
14:18:33 [<c0120f36>] schedule_timeout+0x5a/0xac
14:18:33 [<c0119d31>] printk+0x11d/0x17c
14:18:33 [<c0120ed0>] process_timeout+0x0/0xc
14:18:33 [<c0299641>] idescsi_reset+0x105/0x11c
14:18:33 [<c0293245>]
scsi_try_bus_device_reset+0x57/0x8e
14:18:33 [<c02932e1>]
scsi_eh_bus_device_reset+0x65/0x108
14:18:33 [<c0293ab0>] scsi_eh_ready_devs+0x28/0x74
14:18:33 [<c0293c2c>] scsi_unjam_host+0xc2/0xcc
14:18:33 [<c0293d16>] scsi_error_handler+0xe0/0x124
14:18:33 [<c0293c36>] scsi_error_handler+0x0/0x124
14:18:33 [<c0107289>] kernel_thread_helper+0x5/0xc
14:18:33
14:18:33 ------------[ cut here ]------------
14:18:33 kernel BUG at kernel/timer.c:155!
14:18:33 invalid operand: 0000
14:18:33 CPU: 0
14:18:33 EIP: 0060:[<c0120286>] Not tainted
14:18:33 EFLAGS: 00010002
14:18:33 EIP is at add_timer+0x124/0x132
14:18:33 eax: 00000001 ebx: f7daa4a4 ecx: c03d3360
edx: c048b5c4
14:18:33 esi: f7d06000 edi: 00000092 ebp: f7d07e90
esp: f7d07e88
14:18:33 ds: 007b es: 007b ss: 0068
14:18:33 Process scsi_eh_0 (pid: 9,
threadinfo=f7d06000 task=f7db6700)
14:18:33 Stack: f7d07ed0 f7daa480 f7d07ec0 c02817d1
f7daa4a4 c011602e 00000060 00000246
14:18:33 00000000 00000032 c02818c4 00000000
c048b5c4 c048b5c4 f7d07ef0 c0281eb2
14:18:33 c048b5c4 c02818c4 00000032 00000000
f7daa480 c048b240 00000092 00000000
14:18:33 Call Trace:
14:18:33 [<c02817d1>] ide_set_handler+0x5f/0xa0
14:18:33 [<c011602e>] schedule+0x2d6/0x3a8
14:18:33 [<c02818c4>] atapi_reset_pollfunc+0x0/0x11a
14:18:33 [<c0281eb2>] do_reset1+0x214/0x232
14:18:33 [<c02818c4>] atapi_reset_pollfunc+0x0/0x11a
14:18:33 [<c0281f01>] ide_do_reset+0x31/0x84
14:18:33 [<c029961a>] idescsi_reset+0xde/0x11c
14:18:33 [<c0293245>]
scsi_try_bus_device_reset+0x57/0x8e
14:18:33 [<c02932e1>]
scsi_eh_bus_device_reset+0x65/0x108
14:18:33 [<c0293ab0>] scsi_eh_ready_devs+0x28/0x74
14:18:33 [<c0293c2c>] scsi_unjam_host+0xc2/0xcc
14:18:33 [<c0293d16>] scsi_error_handler+0xe0/0x124
14:18:33 [<c0293c36>] scsi_error_handler+0x0/0x124
14:18:33 [<c0107289>] kernel_thread_helper+0x5/0xc
14:18:33
14:18:33 Code: 0f 0b 9b 00 8b 8a 37 c0 e9 01 ff ff ff
90 55 31 c0 89 e5 53
14:18:33 <6>note: scsi_eh_0[9] exited with
preempt_count 3
14:18:33 hda: ATAPI reset


__________________________________________________
Do you Yahoo!?
Yahoo! Platinum - Watch CBS' NCAA March Madness, live on your desktop!
http://platinum.yahoo.com


2003-03-19 06:53:45

by dan carpenter

[permalink] [raw]
Subject: Re: ide-scsi failure on 2.5.65

On Wednesday 19 March 2003 12:10 am, Eric Benson wrote:
> I installed Red Hat 8.0 on an IBM NetVista 2283-55U,
> an all-in-one desktop 1.8ghz P4. I downloaded and
> compiled the 2.5.65 kernel with ide-scsi emulation and
> kernel debugging enabled.
>
...
> 14:18:33 hdb: drive not ready for command
> 14:18:33 ide-scsi: reset called for 0
> 14:18:33 bad: scheduling while atomic!

This problem is known. Try the driver from the -ac kernel.

thanks,
dan carpenter

2003-03-19 13:09:03

by Alan

[permalink] [raw]
Subject: Re: ide-scsi failure on 2.5.65

On Tue, 2003-03-18 at 13:44, dan carpenter wrote:
> On Wednesday 19 March 2003 12:10 am, Eric Benson wrote:
> > I installed Red Hat 8.0 on an IBM NetVista 2283-55U,
> > an all-in-one desktop 1.8ghz P4. I downloaded and
> > compiled the 2.5.65 kernel with ide-scsi emulation and
> > kernel debugging enabled.
> >
> ...
> > 14:18:33 hdb: drive not ready for command
> > 14:18:33 ide-scsi: reset called for 0
> > 14:18:33 bad: scheduling while atomic!
>
> This problem is known. Try the driver from the -ac kernel.

You need 2.4.21pre-ac for that, the 2.5 code is wildly different
and I've not tackled it yet. Its about six weeks down the todo list
at the moment