Messages in this thread | | | From | "Eric Youngdale" <> | Subject | Re: [Patch] Cleanup struct gendisk registration, 2.3.40-pre1 | Date | Sun, 16 Jan 2000 12:09:05 -0500 |
| |
----- Original Message ----- From: "Alan Cox" <alan@lxorguk.ukuu.org.uk> To: "Andre Hedrick" <andre@suse.com> Cc: "Linus Torvalds" <torvalds@transmeta.com>; "Alan Cox" <alan@lxorguk.ukuu.org.uk>; "Alexander Viro" <aviro@redhat.com>; "Guest section DW" <dwguest@win.tue.nl>; "Stephen C. Tweedie" <sct@redhat.com>; <linux-kernel@vger.rutgers.edu>; "Ben LaHaise" <bcrl@redhat.com>; "Alan Cox" <number6@the-village.bc.nu>; "Eric Youngdale" <eric@andante.org> Sent: Sunday, January 16, 2000 11:27 AM Subject: Re: [Patch] Cleanup struct gendisk registration, 2.3.40-pre1
> > I suggested Linux consider the move to a default packet global block > > device layer and both SCSI and IDE plug into it. > > Not all devices are remotely packet based. Later IDE (ATAPI) is, current > SCSI is, i2o had totally different needs. > > Eric is building a nice library, maybe the library for merging/ISA bouncing > should be shared but don't jam the policy into the block layer. For some high > end cards we already have far too much policy in the block layer.
It should be possible to migrate all of the remaining policy in the assorted block devices down to the individual drivers. I added enough hooks for what I needed to do, but some of the policy questions (i.e. the question as to whether the head entry on the queue is active or not) can now be set in the actual driver. Those big switch statements in ll_rw_blk.c can all be made to go away (or at least I think they can, with a little bit of work). Any remaining policy in the block layer (at least that which is directly related to queueing) can probably be moved to their respective drivers by extending request_queue_t if needed.
The only reason I didn't do it was that since I had changed so many files for the SCSI queueing, the constant merging as new kernel releases were coming out was becoming a significant burden. Finishing some of this stuff is buried down on my list of things to do - if no one else gets to it, I would probably do it the same time that I went in to fix the locking on those damned tables in the block layer.
-Eric
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |