Hrm... I guess I was mis-reading your messages... The translation should still be left there, of course... you never know when a userspace application will come along and do something stupid. Just having it there doesn't hurt anything -- worst case, it becomes an obsolete and unused code path.
But, you should just be able to change the higher layers and it should Just Work(tm). Matt On Sun, Jan 13, 2002 at 02:25:56AM +0100, Peter Osterlund wrote: > Matthew Dharm <[EMAIL PROTECTED]> writes: > > > On Sat, Jan 12, 2002 at 11:25:57PM +0100, Peter Osterlund wrote: > > > On Sat, 12 Jan 2002, Matthew Dharm wrote: > > > > > > I think my device *is* the Traveller. Would it be feasible to detect this > > > at run-time? Send a MODE_SENSE(6) and if it fails, switch to the > > > MODE_SENSE-conversion protocol. > > > > That's really a job for the higher-level drivers to do. Having low-level > > drivers second-guess high-level drivers is a sure path to trouble. > > > > > Anyway, what would break if you remove the MODE_SENSE/SELECT conversion > > > from the usb_stor_ATAPI_command function? It sounds like nobody should be > > > using the 6-byte commands anyway. > > > > OnStream tape drives would simply stop working. So would a couple of other > > tape drive units that use this dongle. > > Why would OnStream break? Note that I'm now talking about removing the > conversion from the 8020 protocol, not adding it to the QIC protocol. > >From a previous mail you sent, I thought OnStream is using the QIC > protocol because it is connected to a Freecom cable, so it shouldn't > be affected by such a change. > > My reasoning was this: > > 1. You say we can't do MODE_SENSE 6 -> 10 translation for my CDRW, > because it would break other stuff. > 2. My CDRW doesn't support MODE_SENSE 6. > 3. To make my CDRW work, we have to fix higher layers so that they > don't use MODE_SENSE 6. > 4. Once 3 has been done, the MODE_SENSE 6 -> 10 translation in the > 8020 protocol is no longer needed and can be removed. This will > simplify the code. > > Is there a flaw in that reasoning? > > -- > Peter Osterlund - [EMAIL PROTECTED] > http://w1.894.telia.com/~u89404340 -- Matthew Dharm Home: [EMAIL PROTECTED] Maintainer, Linux USB Mass Storage Driver Hi. I have my back hairs caught in my computer fan. -- Customer User Friendly, 8/20/1998
msg03912/pgp00000.pgp
Description: PGP signature
