Re: [hatari-devel] Wrong MODE SENSE data length for page code 0x3f

[Eero Tamminen <oak@xxxxxxxxxxxxxx>]

Hi Uwe,

On 26.10.2024 1.04, Uwe Seimet wrote:
> will definitely cause issues with an OS that expects SCSI, especially
> SCSI-2, drives. Sending a different number of bytes than the initiator
> requested can easily cause issues. If 0 bytes are requested, for instance,
> this means 0 bytes have to be transferred, not 4 bytes.
> It looks to me as if this code tries to simulate the behavior of old drives
> like SH/Megafile or SASI drives, but for Linux this will probably not work.

Maybe it still makes sense to log INFO message?
"INFO: REQUEST SENSE max size (%d) differs from what old SCSI-1 drives 
use (4-22)"


> The bus reset you can see in the log is very likely caused by a mismatch of
> the number of bytes expected/requested and the number actually returned by
> Hatari.
> I guess that (at least for a test) removing this code and returning the
> proper number of bytes will resolve some of the Linux issues with SCSI
> drives.

The only change from my patch is log for each of those SENSE requests 
changing like this:
-WARN : HDC: *** Strange REQUEST SENSE ***!
-DEBUG: raw_scsi: data in 22 bytes waiting
-DEBUG: raw_scsi: data in finished, 22 bytes: status phase
+DEBUG: raw_scsi: data in 96 bytes waiting
+DEBUG: raw_scsi: data in finished, 96 bytes: status phase

There are still as many of them, and boot ends to same failure.

Maybe it's related more to the non-existing partition table?


	- Eero