Re: [PATCH] SCSI: limit mode sense usage

[Pat LaVarre <p.lavarre@ieee.org>]

> >   There now in -test9 I think I see: ...
> > We guess writable if need be, but we prefer to settle for: ...
> > sdkp->write_prot = ((data.device_specific & 0x80) != 0); ...
> > after the first command that works in the sequence: ...
> > sd_do_mode_sense(..., 0x3F, ..., 4, ...);
> > sd_do_mode_sense(..., 0, buffer, 4, ...);
> > sd_do_mode_sense(..., 0x3F, buffer, 255, ...);
> > 
> > I mean to suggest, ... more closely ... Talk Like Windows ...
> > e.g.: ...
> > sd_do_mode_sense(..., 0x08, ..., 4, ...);
> 
> There's an excellent reason for _not_ doing this, at least as far as 
> usb-storage is concerned.  Namely, many USB Mass Storage devices die when 
> they receive MODE-SENSE(10) for page x08 but handle page x3F perfectly 
> well.  Ronald's device is a rather unusual exception.

Do these devices die for all lengths of op x5A MODE_SENSE_10 of page
x08, or only for the rude, header-only, 4 or 8 byte length that we get
by default if we patch drivers/scsi/ minimally?

That is, do these devices die only when assaulted by:

sd_do_mode_sense(..., 0x08, ..., 4, ...);

in place of the more Windows-like:

sd_do_mode_sense(..., 0x08, ..., 0x1C, ...); // if x5A MODE_SENSE_10
sd_do_mode_sense(..., 0x08, ..., 0x18, ...); // if x1A MODE_SENSE

?

Pat LaVarre