[sane-devel] Mustek backend in 1.0.7 and 1.0.8

[Henning Meier-Geinitz]

Hi,

On Sat, Jun 01, 2002 at 05:13:53PM +0100, James Tappin wrote:
> I've been having a problem with the Mustek backend with the last 2 releases 
> of SANE. Basically 
> scanimage -d mustek:/dev/scanner > tmp2.pnm
> stops part way through the scan with the error (after setting the debug 
> level to 255):

[...]

> The scanner is a ScanExpress 12000SP scsi scanner. This problem occurs 
> consistently with both 1.0.8 and 1.0.7 but it works fine with 1.0.6 and 
> earlier versions (there may have been intermittent problems with 1.0.5 I 
> don't recall for certain that far back).

Just a followup on this for other readers. I got detailed logfiles from
James for 1.0.6 and 1.0.8. He is using the BusLogic Mylex BT-930 card.

1.0.6:
[sanei_scsi] sanei_scsi_req_wait: SCSI command complained: Success
[sanei_scsi] sense buffer: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
[sanei_scsi] target status: 00 host status: 0007 driver status: 0027
[mustek] sense_handler: got sense code 00 for fd 3 (arg = null)

1.0.8
[sanei_scsi] sanei_scsi_req_wait: SCSI command complained: Success
[sanei_scsi] sense buffer: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
[sanei_scsi] target status: 00 host status: 0007 driver status: 0027
[mustek] reader_process: failed to read data, status: Error during device I/O, buffer: 2

So the error occurs with both versions. However, sanei_scsi from
1.0.8 actually reports the error while 1.0.6 calls the sense_handler
(which ignores the error because sense is clean).

From several messages about BusLogic problems I came to the conclusion
Status: O

that it's probably a problem with the Linux BusLogic kernel driver.
Anyone has any other ideas?

Bye,
  Henning