From: Bartlomiej Zolnierkiewicz <B.Zolnierkiewicz@elka.pw.edu.pl>
To: Jeff Garzik <jgarzik@pobox.com>
Cc: Linus Torvalds <torvalds@transmeta.com>, <linux-kernel@vger.kernel.org>
Subject: Re: [BK PATCHES] add ata scsi driver
Date: Mon, 26 May 2003 13:37:01 +0200 (MET DST) [thread overview]
Message-ID: <Pine.SOL.4.30.0305261328300.26499-100000@mion.elka.pw.edu.pl> (raw)
In-Reply-To: <3ED1F6C6.3050400@pobox.com>
On Mon, 26 May 2003, Jeff Garzik wrote:
> Bartlomiej Zolnierkiewicz wrote:
> > On Mon, 26 May 2003, Jeff Garzik wrote:
> >
> >
> >>Linus Torvalds wrote:
> >>
> >>>On Mon, 26 May 2003, Jeff Garzik wrote:
> >>>
> >>>
> >>>>Just to echo some comments I said in private, this driver is _not_
> >>>>a replacement for drivers/ide. This is not, and has never been,
> >>>>the intention. In fact, I need drivers/ide's continued existence,
> >>>>so that I may have fewer boundaries on future development.
> >>>
> >>>
> >>>Just out of interest, is there any _point_ to this driver? I can
> >>>appreciate the approach, but I'd like to know if it does anything (at all)
> >>>better than the native IDE driver? Faster? Anything?
> >>
> >>
> >>Direction: SATA is much more suited to SCSI, because otherwise you wind
> >>up re-creating all the queueing and error handling mess that SCSI
> >>already does for you. The SATA2 host controllers coming out soon do
> >>full host-side TCQ, not the dain-bramaged ATA TCQ bus-release stuff.
> >>Doing SATA2 devel in drivers/ide will essentially be re-creating the
> >>SCSI mid-layer.
> >
> >
> > And now you are recreating ATA in SCSI ;-).
> >
> > Don't get me wrong: I like idea very much, but why you can't
> > share common code between drivers/ide and your ATA-SCSI.
>
> SATA2 is really 100% different from PATA in all respects except for the
> ATA commands themselves being sent down the pipe. Doing that inside
> drivers/ide really means two driver cores at that point.
>
> Sharing code is definitely an option, though!
> Why do you think I named it "libata"? :)
>
>
> Another SATA wrinkle: the phy layer.
>
> SATA, like network cards, has an underlying connection layer. The host
> connects to the device. Your link state may be up or down. This is
> easily mapped to SCSI semantics (initiator connection to target). Much
> more code if writing from scratch.
>
>
> >>Legacy-free: Because I don't have to worry about legacy host
> >>controllers, I can ignore limitations drivers/ide cannot. In
> >>drivers/ide, each host IO (PIO/MMIO) is done via function pointer. If
> >>your arch has a mach_vec, more function pointers. Mine does direct
> >>calls to the asm/io.h functions in faster. So, ATA command submission
> >>is measureably faster.
> >
> >
> > I think it is simply wrong, you should use function pointers.
> > You can have ie. two PCI hosts, one using PIO and one using MMIO.
> >
> > "measureably faster", I doubt.
> > IO operations are REALLY slow when compared to CPU cycles.
>
> You misunderstand.
>
> drivers/ide -- uses function pointers HWIF->outb, etc. inside a static
> taskfile-submit function. The TF submit functions statically order each
> HWIF->outb() call, and you cannot change that order from the low-level
> driver.
>
> my driver -- uses separate taskfile-submit functions for different hardware.
>
> My method directly uses outb(), writel(), or whatever is necessary. It
> allows one to use __writel() and barriers to optimize as needed, or to
> support special cache coherency needs. Further, for SATA and some
> advanced PATA controllers, my method allows one to use a completely
> alien and new taskfile submission method. For example, building SATA2
> FIS's as an array of u32's, and then queueing that FIS onto a DMA ring.
> The existing drivers/ide code is nowhere close to that.
Its not that hard, but nevermind.
> Putting it into drivers/ide language, my driver adds a
> HWIF->do_flagged_taskfile hook, and eliminates the HWIF->outb hooks.
> Turns ~13 hook calls into one, from looking at flagged_taskfile.
Okay, I misunderstood.
This is cool, but you are using ie. outb not only in taskfiles.
> And FWIW, I would not state "measureably faster" unless I actually
> benchmarked it. :) It's a pittance in the grand scheme of things,
> since you're inevitably limited by device speed, but lower CPU usage is
> something nobody complains about...
>
>
> >>sysfs: James and co are putting time into getting scsi sysfs right. I
> >>would rather ride their coattails, and have my driver Just Work with
> >>sysfs and the driver model.
> >
> >
> > No big deal here, ATA will get it too.
>
> Oh agreed. But scsi is pretty much there now (it's in testing in a BK
> tree), and we're darn close to 2.6.0.
>
>
> >>PIO data transfer is faster and more scheduler-friendly, since it polls
> >>from a kernel thread.
> >
> >
> > CPU polling is faster than IRQs?
>
> Yes, almost always.
>
> The problem with polling is that it chews up CPU if you're not careful,
> starving out other processes. I'm careful :)
Good.
> >>And for specifically Intel SATA, drivers/ide flat out doesn't work (even
> >>though it claims to).
> >
> >
> > So fix it ;-).
>
> As you can see from the discussion, I think a new driver best serves the
> future of SATA2 as well as today's SATA. So I'm not gonna bother... I
> have a driver that works.
>
> If you're interested in working the problem, the result of booting ICH5
> SATA in drivers/ide is a hang at
> hdX: attached to ide-disk driver
>
> Nothing happens after that. sysrq doesn't work, so my assumption was
> that the bus was locked up.
>
>
> >>So, I conclude: faster, smaller, and better future direction. IMO, of
> >>course :)
> >
> >
> > And right now ugly and incomplete.
> > IMO, of course ;-).
>
> Incomplete? Kinda sorta. SATAPI isn't here yet, so all it needs is
> more fine-grained error handling. I call it 100% stable now, though.
>
> Ugly? Compared to drivers/ide? I actually have a lot of admiration and
> respect for the PATA knowledge embedded in drivers/ide. But I would
> never call it pretty :)
Silly excuse ;-).
Thanks,
--
Bartlomiej
> Jeff
next prev parent reply other threads:[~2003-05-26 11:24 UTC|newest]
Thread overview: 68+ messages / expand[flat|nested] mbox.gz Atom feed top
2003-05-26 4:58 [BK PATCHES] add ata scsi driver Jeff Garzik
2003-05-26 5:15 ` Linus Torvalds
2003-05-26 5:30 ` Jeff Garzik
2003-05-26 5:36 ` Jeff Garzik
2003-05-26 5:42 ` Linus Torvalds
2003-05-26 6:01 ` Jeff Garzik
2003-05-26 16:56 ` Linus Torvalds
2003-05-26 17:47 ` Jeff Garzik
2003-05-26 20:09 ` Linus Torvalds
2003-05-27 0:29 ` Alan Cox
2003-05-27 6:07 ` Jeff Garzik
2003-05-27 6:30 ` Linus Torvalds
2003-05-27 6:51 ` Linus Torvalds
2003-05-27 7:29 ` Jeff Garzik
2003-05-26 5:40 ` Linus Torvalds
2003-05-26 5:53 ` Jeff Garzik
2003-05-26 6:21 ` Jeff Garzik
2003-05-26 16:57 ` Linus Torvalds
2003-05-26 17:24 ` Jens Axboe
2003-05-26 17:54 ` Jeff Garzik
2003-05-26 17:59 ` Jeff Garzik
2003-05-26 18:11 ` Jens Axboe
2003-05-27 0:22 ` Alan Cox
2003-05-27 4:15 ` Linus Torvalds
2003-05-26 10:32 ` Bartlomiej Zolnierkiewicz
2003-05-26 11:13 ` Jeff Garzik
2003-05-26 11:37 ` Bartlomiej Zolnierkiewicz [this message]
2003-05-26 5:59 ` Benjamin Herrenschmidt
2003-05-26 6:03 ` Jeff Garzik
2003-06-02 9:46 ` Andre Hedrick
2003-06-02 13:56 ` Alan Cox
2003-05-26 18:12 James Bottomley
2003-05-26 18:18 ` Jens Axboe
2003-05-26 18:47 ` James Bottomley
2003-05-26 19:07 ` Jens Axboe
2003-05-26 19:17 ` James Bottomley
2003-05-26 19:33 ` Jens Axboe
2003-05-27 12:39 ` Jens Axboe
2003-05-27 14:26 ` James Bottomley
2003-05-27 17:16 ` Jens Axboe
2003-05-27 18:09 ` James Bottomley
2003-05-27 18:21 ` Jens Axboe
2003-05-27 18:30 ` James Bottomley
2003-05-26 20:27 ` Linus Torvalds
2003-05-26 20:36 ` James Bottomley
2003-05-26 20:45 ` Linus Torvalds
2003-05-26 20:51 ` Jens Axboe
2003-05-26 20:56 ` James Bottomley
2003-05-26 20:38 ` Jens Axboe
2003-05-26 20:49 ` Linus Torvalds
2003-05-26 20:57 ` Jens Axboe
2003-05-26 21:34 ` Linus Torvalds
2003-05-26 23:58 ` Nick Piggin
2003-05-27 0:09 ` Linus Torvalds
2003-05-27 0:49 ` Nick Piggin
2003-05-27 0:16 ` Alan Cox
2003-05-27 6:54 ` Jens Axboe
2003-05-27 14:20 ` James Bottomley
2003-05-27 14:36 ` Linus Torvalds
2003-05-27 14:59 ` James Bottomley
2003-05-27 15:21 ` Jeff Garzik
2003-05-27 15:38 ` James Bottomley
2003-05-27 15:50 ` Jeff Garzik
2003-05-27 16:00 ` James Bottomley
2003-05-27 16:16 ` Jeff Garzik
2003-05-28 9:35 ` Christoph Hellwig
2003-05-28 10:50 ` Lincoln Dale
2003-05-27 19:43 ` Jens Axboe
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=Pine.SOL.4.30.0305261328300.26499-100000@mion.elka.pw.edu.pl \
--to=b.zolnierkiewicz@elka.pw.edu.pl \
--cc=jgarzik@pobox.com \
--cc=linux-kernel@vger.kernel.org \
--cc=torvalds@transmeta.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).