[sane-devel] HP ScanJet IIcx /T --- any hope?

abel deuring a.deuring@satzbau-gmbh.de
Sat, 04 Jan 2003 16:17:34 +0100

Peter Kirchgessner wrote:
> Hi,
> from the logfile is see that everything is working well up to that point 
> when the scan-data is read from the scanner. It immediately fails when 
> trying to read 32768 bytes of data.
> You are using the option dumb-read of the hp-backend. What happens 
> without that option ? If scanning fails earlier, I would say the problem 
> is the kernel driver of the card. This option could be used for Buslogic 
> cards to work around a problem of the corresponding kernel driver.
> It simply reads data from the scanner bytewise until an error occurs. 
> This makes sense for small amount of data when setting up options in the 
> scanner. But it makes no sense when reading the image data.
> Just for a test please drop the option dumb-read and rerun scanimage
> to create a new logfile.

The host status 7 for the failing command:

dev_max(currently)=8 max_active_device=3 (origin 1)
  scsi_dma_free_sectors=176 sg_pool_secs_aval=320 def_reserved_size=32768
  >>> device=sg2 scsi1 chan=0 id=2 lun=0   em=0 sg_tablesize=128 excl=1
    FD(1): timeout=120000ms bufflen=131072 (res)sgat=4 low_dma=0
    cmd_q=1 f_packid=0 k_orphan=0 closed=0
      rb>> rcv: id=3058 blen=32768 dur=126010ms sgat=0 op=0x08
      rcv: id=3059 blen=32768 dur=126000ms sgat=0 op=0x08
[sanei_scsi] sanei_scsi.issue: 0x401d0008
[sanei_scsi] sanei_scsi_req_wait: read 64 bytes
[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: 0000
[hp] do_read: Error from scsi_req_wait: Error during device I/O

indicates an error somewhere in the Linux SCSI system, most probably in 
the adapter driver. Most 2.4.x kernel have two drivers for the Adaptec 
7xxx chips: aic7xxx and aic7xxx_old. Moritz, you could try to use the 
other driver. Another question: Do you see anything related to this 
problem in /var/log/messages?