On Wed, Jul 30, 2003 at 10:08:27AM -0400, Alan Stern wrote:
> Matt:
> 
> Daniel reports that the patch to change the length of the 
> cache-determination MODE-SENSE worked (patch included below).  Here is the 
> relevant part of the system log.  It turns out that only the first 11 
> bytes of the cache page are needed...
[...]
> Do you think I should submit this to James Bottomley?  The way I see it,
> the code should be fine either way, for conformant devices.  At least one
> device (Daniel's) works with the patch but not without it.  I'm worried
> that other devices might turn out to do exactly the opposite.
> 
> Maybe a better approach would be:
> 
>       len = data.length;
>       if (len > 512)
>               len = data.header_length + data.block_descriptor_length + 3;

My guess is that the vast majority of devices will be fine with any length
over the header size -- some of James' code tries to guarantee that.

Do we have any idea how large that page is supposed to be?

Whatever length you request should also be a multiple of 4, just to make
sure devices stay happy.

Other than that, I say send it to James.

Matt

-- 
Matthew Dharm                              Home: [EMAIL PROTECTED] 
Maintainer, Linux USB Mass Storage Driver

Now payink attention, please.  This is mouse.  Click-click. Easy to 
use, da? Now you try...
                                        -- Pitr to Miranda
User Friendly, 10/11/1998

Attachment: pgp00000.pgp
Description: PGP signature

Reply via email to