From: "David S. Miller" <davem@davemloft.net>
To: Jesse Barnes <jbarnes@engr.sgi.com>
Cc: torvalds@osdl.org, willy@debian.org,
James.Bottomley@steeleye.com, geert@linux-m68k.org,
linux-arch@vger.kernel.org,
viro@parcelfarce.linux.theplanet.co.uk, akpm@osdl.org,
alan@lxorguk.ukuu.org.uk, davem@redhat.com, jgarzik@pobox.com
Subject: Re: RFC: being more anal about iospace accesses..
Date: Thu, 16 Sep 2004 13:04:50 -0700 [thread overview]
Message-ID: <20040916130450.40b3f8a1.davem@davemloft.net> (raw)
In-Reply-To: <200409161152.54691.jbarnes@engr.sgi.com>
On Thu, 16 Sep 2004 11:52:54 -0700
Jesse Barnes <jbarnes@engr.sgi.com> wrote:
> On Thursday, September 16, 2004 11:21 am, Linus Torvalds wrote:
> > It's literally only the sn2 _platform_ that does something stupid. In
> > other words, it's not ia64 that is broken (surprise surprise - you never
> > expected me to have said anything good about ia64), it's literally the SN2
> > platform that seems broken.
>
> No, it's just the only smart platform. Apparently we're the only ones who
> realized that most PIO reads do not rely on prior DMA being complete (and
> that it's expensive to make sure that they do), so we optimized that out of
> our hardware. Only interrupts will guarantee DMA completion on SGI
> platforms.
Jesse, I think you might not need that hack in the I/O accessor
routines to handle this. pci_dma_unmap(), pci_dma_sync_*() and
friends should be the place to perform this "prior DMA sync" thing.
Drivers aren't allowed to touch active DMA data areas until they
call one of those interfaces (see Documentation/DMA-mapping.txt)
so that should just work.
next prev parent reply other threads:[~2004-09-16 20:07 UTC|newest]
Thread overview: 63+ messages / expand[flat|nested] mbox.gz Atom feed top
2004-09-08 22:57 RFC: being more anal about iospace accesses Linus Torvalds
2004-09-08 23:07 ` David S. Miller
2004-09-08 23:25 ` Linus Torvalds
2004-09-09 1:19 ` Linus Torvalds
2004-09-09 4:36 ` David S. Miller
2004-09-09 5:56 ` Richard Henderson
2004-09-09 5:04 ` viro
2004-09-09 5:05 ` David S. Miller
2004-09-09 5:13 ` Linus Torvalds
2004-09-09 6:08 ` viro
2004-09-09 8:27 ` Geert Uytterhoeven
2004-09-09 6:23 ` David Woodhouse
2004-09-09 13:14 ` Alan Cox
2004-09-11 6:09 ` Linus Torvalds
2004-09-11 6:42 ` Anton Blanchard
2004-09-11 7:26 ` Benjamin Herrenschmidt
2004-09-11 7:29 ` Geert Uytterhoeven
2004-09-11 7:23 ` Benjamin Herrenschmidt
2004-09-11 14:42 ` Alan Cox
2004-09-15 15:03 ` Linus Torvalds
2004-09-15 19:02 ` Geert Uytterhoeven
2004-09-15 19:16 ` Linus Torvalds
2004-09-15 19:40 ` Matthew Wilcox
2004-09-15 20:10 ` Linus Torvalds
2004-09-15 20:17 ` Linus Torvalds
2004-09-16 12:17 ` David Woodhouse
2004-09-16 13:52 ` Linus Torvalds
2004-09-15 20:20 ` Russell King
2004-09-15 20:34 ` Linus Torvalds
2004-09-15 20:51 ` Linus Torvalds
2004-09-15 22:38 ` James Bottomley
2004-09-16 2:33 ` Matthew Wilcox
2004-09-16 4:28 ` Linus Torvalds
2004-09-16 4:57 ` Benjamin Herrenschmidt
2004-09-16 4:58 ` Benjamin Herrenschmidt
2004-09-16 13:41 ` Matthew Wilcox
2004-09-16 18:21 ` Linus Torvalds
2004-09-16 18:52 ` Jesse Barnes
2004-09-16 19:09 ` Linus Torvalds
2004-09-16 20:02 ` Jesse Barnes
2004-09-16 20:37 ` James Bottomley
2004-09-16 20:42 ` Jesse Barnes
2004-09-16 21:37 ` Grant Grundler
2004-09-16 20:04 ` David S. Miller [this message]
2004-09-16 20:13 ` Jeff Garzik
2004-09-16 20:45 ` David S. Miller
2004-09-16 20:20 ` Jesse Barnes
2004-09-17 5:17 ` Benjamin Herrenschmidt
2004-09-17 15:30 ` Jesse Barnes
2004-09-16 19:01 ` Linus Torvalds
2004-09-16 19:13 ` Jeff Garzik
2004-09-16 19:50 ` Linus Torvalds
2004-09-16 20:07 ` Alan Cox
2004-09-17 5:44 ` Benjamin Herrenschmidt
2004-09-17 5:20 ` Benjamin Herrenschmidt
2004-09-17 15:03 ` Linus Torvalds
2004-09-16 22:30 ` Matthew Wilcox
2004-09-16 22:42 ` Linus Torvalds
2004-09-16 22:46 ` Jeff Garzik
2004-09-16 23:15 ` Linus Torvalds
2004-09-16 23:30 ` Jeff Garzik
2004-09-16 23:43 ` Linus Torvalds
2004-09-17 12:44 ` Matthew Wilcox
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=20040916130450.40b3f8a1.davem@davemloft.net \
--to=davem@davemloft.net \
--cc=James.Bottomley@steeleye.com \
--cc=akpm@osdl.org \
--cc=alan@lxorguk.ukuu.org.uk \
--cc=davem@redhat.com \
--cc=geert@linux-m68k.org \
--cc=jbarnes@engr.sgi.com \
--cc=jgarzik@pobox.com \
--cc=linux-arch@vger.kernel.org \
--cc=torvalds@osdl.org \
--cc=viro@parcelfarce.linux.theplanet.co.uk \
--cc=willy@debian.org \
/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