Hi Robert,
thanks for your report!
There are a few issues with teh DC395 driver and that's the reason I did not
send it to Linus and/or Alan and asked for inclusion into the kernel.
The problem seems to be related with clearing or not clearing the SCSI FIFO
before disconnection, arbitration, ... and after reconnection, and the
little docu about the chip I have is unfortunately not very clear on this.
Your problem seems to be different, and I wonder why all your SCSI requests
have pid 0. Probably every device starts with pid 0 now? (Yes, I know it's
high time to have a deep look into the new 2.4 SCSI queueing algos.)
The messages about the Unexpected disconnection are probably harmless. Just
the scanning for devices in 2.4 has slightly changed, and the driver now
produces these messages instead of simply reporting back that the devices do
not exist.
On Fri, Jul 28, 2000 at 09:14:31PM -0400, Robert Dale wrote:
> I have experienced some problems with this card/driver. Although it's not
> yet in the kernel, I thought I would discuss the problems here and hope we
> can work together to find a solution. (It gets a little lengthy here..)
>
> Using the dc395-127 patch with 2.4.0-test2, I get this:
>
> # dmesg
>
> DC395x (TRM-S1040) SCSI driver 1.27, 2000-05-23
> DC395x: Target 00: Sync: 48ns Offset 15 (20.8 MB/s)
> Vendor: PIONEER Model: CD-ROM DR-966 Rev: 1.00
> Type: CD-ROM ANSI SCSI revision: 02
> Detected scsi CD-ROM sr0 at scsi0, channel 0, id 0, lun 0
> DC395x: Unexpected Disconnection (pid 0)!
> pid 0(592):12 00..(1-0) Start 00 S b8 DISC UnExpD *
[...]
> DC395x: Target 06: Sync: 100ns Offset 15 (10.0 MB/s)
> Vendor: FUJITSU Model: M2513EL Rev: 0020
> Type: Optical Device ANSI SCSI revision: 02
> Detected scsi removable disk sda at scsi0, channel 0, id 6, lun 0
> DC395x: Unexpected Disconnection (pid 0)!
> pid 0(1190):12 00..(8-0) Start 00 S b8 DISC UnExpD *
[...]
> scsi : detected 1 SCSI cdrom 1 SCSI disk total.
> sr0: scsi3-mmc drive: 40x/40x cd/rw xa/form2 cdda tray
> Uniform CD-ROM driver Revision: 3.11
> SCSI device sda: hdwr sector= 2048 bytes. Sectors= 310352 [606 MB] [0.6 GB]
> sda: Write Protect is off
> sda: sda1
>
> # mount /dev/sda1 /mnt
> # ls -R /mnt
>
> scsi : aborting command due to timeout : pid 0, scsi0, channel 0, id 6, lun 0
> Read (10) 00 00 04 82 a7 00 00 01 00
> DC395x_abort: pid=0, target=06-0
> DC395x: abort: Lower SyncFreq to for dev 06-0!
> pid 0(36998):28 00..(6-0) Start 00 60 WtG MIP1 #MIP0(04).DISC +abort (SN)!*
> TmpSRB:
It looks like the trouble really starts right away. The command trace looks
a little bit strange, as I normally don't expect a device to first send a
message, before even the command is transfered.
I hope, I can find a similar problem on my machine ...
[...]
> Needless to say, it didn't work too well. I haven't tried any other
> recent dev kernels.
>
> Then on 2.2.16:
>
> DC395x (TRM-S1040) SCSI driver 1.27, 2000-05-23
> DC395x: Used settings: AdaptID=7
> DC395 : Connectors: int50 Termination: Auto Low High
> DC395x (TRM-S1040): 1 adapters found
> scsi0 : Tekram DC395U/UW/F DC315/U V1.27, 2000-05-23
> scsi : 1 host.
> DC395x: Target 00: Sync: 48ns Offset 15 (20.8 MB/s)
> Vendor: PIONEER Model: CD-ROM DR-966 Rev: 1.00
> Type: CD-ROM ANSI SCSI revision: 02
> Detected scsi CD-ROM sr0 at scsi0, channel 0, id 0, lun 0
> DC395x: Target 06: Sync: 100ns Offset 15 (10.0 MB/s)
> Vendor: FUJITSU Model: M2513EL Rev: 0020
> Type: Optical Device ANSI SCSI revision: 02
> Detected scsi removable disk sda at scsi0, channel 0, id 6, lun 0
> scsi : detected 2 SCSI generics 1 SCSI cdrom 1 SCSI disk total.
> sr0: scsi3-mmc drive: 40x/40x cd/rw xa/form2 cdda tray
> Uniform CD-ROM driver Revision: 3.09
> SCSI device sda: hdwr sector= 2048 bytes. Sectors= 310352 [606 MB] [0.6 GB]
> sda: Write Protect is off
> Partition check:
> sda: sda1
>
> So far so good, but...
>
> Using a Pioneer 40x CD-ROM (DR-966), kernel 2.2.16 goes up to a load of 2.0.
> The system becomes 0% idle. And that's simply installing some games from the
> CDROM to an IDE harddrive. However, I have had no problems reading/writing
> to/from the MO drive (which is probably much slower than the CDROM).
> This is pretty much described in the README. It's here just incase.
>
> Other hardware: athlon, asus k7v, ibm hd.
>
> I would be interested in any ideas. I hope to take a look at it within
> the next couple of weeks. Thanks.
I can't tell why your load goes to two. Maybe the CPU actually is busy
PIOing the data to your IDE harddisk.
In that case, hdparm -d1, would both boost your hd performance and lowering
your CPU load. Maybe you need some of Andre's IDE patches in order to make
your IDE chipset (VIA) work with DMA.
Regards,
--
Kurt Garloff <[EMAIL PROTECTED]> Eindhoven, NL
GPG key: See mail header, key servers Linux kernel development
SuSE GmbH, Nuernberg, FRG SCSI, Security
PGP signature
