public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Martin Bligh <mbligh@mbligh.org>
To: Christoph Lameter <clameter@sgi.com>
Cc: Andrew Morton <akpm@osdl.org>,
	linux-kernel@vger.kernel.org, Rohit Seth <rohitseth@google.com>
Subject: Re: ZONE_DMA
Date: Thu, 21 Sep 2006 16:15:57 -0700	[thread overview]
Message-ID: <45131D2D.8020403@mbligh.org> (raw)
In-Reply-To: <Pine.LNX.4.64.0609211045400.5959@schroedinger.engr.sgi.com>

>>If a device driver wants "DMAable" memory, and thus does a ZONE_DMA
>>allocation, and we've moved all its memory from ZONE_DMA to ZONE_NORMAL
>>(as I think you're proposing doing for PPC64 (and ia64?)), then the
>>allocation will fail.
> 
> I would not presume proposing anything for PPC64 and left everything the 
> way it is. The arch people can control if they want ZONE DMA or not and 
> the default is to leave things as is. If PPC64 wants to go ZONE_DMAless 
> then the arch code needs to be modified not refer to ZONE_DMA anymore and 
> if that works then we can switch CONFIG_ZONE_DMA off for PPC64. See the 
> patches in mm for examples how other arches have done it.

Sorry, there was some confusion then, maybe I misunderstood one of your
earlier emails.

>>So are we saying that no driver code should be calling with GFP_DMA
>>(a quick grep turns up 148 instances under driver/), that if they do
>>they should only work on specific architectures (some instances were
>>s390-only drivers)? If so, should we not be removing the definiton of
>>GFP_DMA itself if ZONE_DMA is config'ed out, so that it fails at
>>compile time, rather than runtime?
> 
> 
> This was covered at length before. Removing all GFP_DMA references would 
> require extensive #ifdefs. The limited patch in mm is only neutering 
> GFP_DMA for arches that do not need it. If an arch has removed its definition of 
> CONFIG_ZONE_DMA then __GFP_DMA will be ignored in the page allocator.

Just ignoring GFP_DMA in the allocator seems like a horrible violation
to me. GFP_DMA means "give me memory from ZONE_DMA". We're both well
aware that the whole concept of ZONE_DMA doesn't make much sense, but
still, that's what it does.

So if you just put all of memory in ZONE_DMA for your particular
machine, and bumped the DMA limit up to infinity, we wouldn't need
any of these patches, right? Which would also match what the other
arches do for this (eg PPC64).

Seems like a much simpler way of solving the problem to me. Or do you
think that won't work for some reason?

M.

  reply	other threads:[~2006-09-21 23:16 UTC|newest]

Thread overview: 128+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-09-20 20:54 2.6.19 -mm merge plans Andrew Morton
2006-09-20 21:08 ` Jeff Garzik
2006-09-20 21:42 ` Badari Pulavarty
2006-09-20 21:53 ` USB: fix autosuspend-autoresume with CONFIGRe: " Jiri Kosina
2006-09-20 22:09   ` Andrew Morton
2006-09-20 22:41     ` Greg KH
2006-09-20 21:55 ` Autofs4 breakage (was 2.6.19 -mm merge plans) Trond Myklebust
2006-09-20 22:13   ` Andrew Morton
2006-09-21  3:39   ` Andrew Morton
2006-09-21 23:25     ` Trond Myklebust
2006-09-22 15:55       ` Andrew Morton
2006-09-26  4:10   ` Ian Kent
2006-09-21  0:09 ` 2.6.19 -mm merge plans Martin J. Bligh
2006-09-21  0:22   ` ZONE_DMA (was: Re: 2.6.19 -mm merge plans) Andrew Morton
2006-09-21  0:31     ` Christoph Lameter
2006-09-21  1:03       ` Alan Cox
2006-09-21 15:59         ` James Bottomley
2006-09-21  0:50     ` ZONE_DMA Martin J. Bligh
2006-09-21  1:10       ` ZONE_DMA Christoph Lameter
2006-09-21  1:23         ` ZONE_DMA Martin J. Bligh
2006-09-21 15:58           ` ZONE_DMA Christoph Lameter
2006-09-21 16:57             ` ZONE_DMA Martin Bligh
2006-09-21 17:54               ` ZONE_DMA Christoph Lameter
2006-09-21 23:15                 ` Martin Bligh [this message]
2006-09-22  2:59                   ` ZONE_DMA Christoph Lameter
2006-09-22 17:21                     ` ZONE_DMA Jesse Barnes
2006-09-22 17:39                       ` ZONE_DMA Jesse Barnes
2006-09-22 18:08                         ` ZONE_DMA Christoph Lameter
2006-09-22 18:26                           ` ZONE_DMA Jesse Barnes
2006-09-22 18:32                             ` ZONE_DMA Christoph Lameter
2006-09-22 18:39                               ` ZONE_DMA Jesse Barnes
2006-09-22 18:40                                 ` ZONE_DMA Christoph Lameter
2006-09-22 19:06                                   ` ZONE_DMA Jesse Barnes
2006-09-22 19:07                                     ` ZONE_DMA Christoph Lameter
2006-09-22 17:35                     ` ZONE_DMA Martin Bligh
2006-09-22 17:37                       ` ZONE_DMA Christoph Lameter
2006-09-21  0:40   ` 2.6.19 -mm merge plans Christoph Lameter
2006-09-21  2:28 ` 2.6.19 -mm merge plans (NTP changes) john stultz
2006-09-21  2:40   ` Andrew Morton
2006-09-21 10:24   ` Roman Zippel
2006-09-25 16:50     ` john stultz
2006-09-25 17:04       ` Ray Lee
2006-09-25 17:38         ` Roman Zippel
2006-09-21  4:22 ` 2.6.19 -mm merge plans Jeff Garzik
2006-09-21  5:07   ` Andrew Morton
2006-09-21  5:23     ` Linus Torvalds
2006-09-21  9:12       ` Alan Cox
2006-09-21  6:36     ` Jeff Garzik
2006-09-21  6:48       ` Andrew Morton
2006-09-21 11:10         ` Rafael J. Wysocki
2006-09-22 13:16         ` release cycle (Re: 2.6.19 -mm merge plans) Pavel Machek
2006-09-21  9:16     ` 2.6.19 -mm merge plans Alan Cox
2006-09-21 10:55       ` Jan Engelhardt
2006-09-21 15:25       ` Linus Torvalds
2006-09-21 15:50         ` Jan Engelhardt
2006-09-21 17:59         ` Andrew Morton
2006-09-21 18:20           ` Jeff Garzik
2006-09-21 18:22           ` Linus Torvalds
2006-09-21 18:33             ` Jeff Garzik
2006-09-21 18:55               ` Andrew Morton
2006-09-21 19:46               ` Adrian Bunk
2006-09-21 20:37                 ` Diego Calleja
2006-09-21 21:35                   ` Randy.Dunlap
2006-09-21 20:38                 ` Jeff Garzik
2006-09-21 20:49                   ` Adrian Bunk
2006-09-21 21:33             ` Bill Davidsen
2006-09-21 21:33               ` Jeff Garzik
2006-09-21 21:52                 ` David Miller
2006-09-21 22:05                   ` Dave Jones
2006-09-21 22:44                     ` David Miller
2006-09-22  8:35                     ` Russell King
2006-09-22 15:48                       ` Dave Jones
2006-09-22 16:21                         ` Linus Torvalds
2006-09-22 20:08                           ` Roland Dreier
2006-09-22 20:22                             ` Jeff Garzik
2006-09-23  8:29                               ` Ryan Anderson
2006-09-24  7:48                           ` Lennert Buytenhek
2006-09-24  9:20                             ` Russell King
     [not found]                               ` <20060924142353.6c725128.seanlkml@sympatico.ca>
2006-09-24 18:23                                 ` Sean
2006-09-24 22:34                                   ` Stefan Richter
     [not found]                                     ` <20060924190758.132c0008.seanlkml@sympatico.ca>
2006-09-24 23:07                                       ` Sean
2006-09-24 23:09                                 ` Russell King
     [not found]                                   ` <20060924192308.ef60880a.seanlkml@sympatico.ca>
2006-09-24 23:23                                     ` Sean
2006-09-25  0:48                                       ` Linus Torvalds
2006-10-11 14:09                                         ` Pierre Ossman
2006-09-24 19:38                               ` Mike Galbraith
2006-09-22 16:24                         ` Alan Cox
2006-09-22 16:08                           ` Dave Jones
2006-09-22 16:29                         ` Jeff Garzik
2006-09-22 17:09                           ` Dave Jones
2006-09-22 17:11                           ` Dave Jones
2006-09-22 18:26                             ` Jeff Garzik
2006-09-22 18:26                         ` Andrew Morton
2006-09-24  6:33                           ` Lennert Buytenhek
2006-09-22  1:03                 ` Bill Davidsen
2006-09-24 18:51             ` Ingo Molnar
2006-09-25  3:07               ` Nick Piggin
2006-09-25 11:53                 ` Ingo Molnar
2006-09-25 18:57                   ` Rafael J. Wysocki
2006-09-22  6:33           ` Andi Kleen
2006-09-23  8:06   ` Andrew Morton
2006-09-23  9:52     ` David Woodhouse
2006-09-21 13:14 ` Ingo Molnar
2006-09-21 17:35   ` Andrew Morton
2006-09-22 13:06   ` Pavel Machek
2006-09-22 19:01     ` Ingo Molnar
2006-09-22 20:29       ` hires timer patchset [was Re: 2.6.19 -mm merge plans] Bill Rugolsky Jr.
2006-09-23 11:07       ` 2.6.19 -mm merge plans Pavel Machek
2006-09-22 20:12     ` Dave Jones
2006-09-21 14:04 ` Apple Motion Sensor (was: 2.6.19 -mm merge plans) Michael Hanselmann
     [not found] ` <20060921234637.GA9742@mail.ustc.edu.cn>
2006-09-21 23:46   ` 2.6.19 -mm merge plans Fengguang Wu
2006-09-21 23:59     ` Andrew Morton
     [not found]       ` <20060922003223.GA9952@mail.ustc.edu.cn>
2006-09-22  0:32         ` Fengguang Wu
2006-09-23 14:48           ` Gene Heskett
     [not found]             ` <20060924024239.GA11671@mail.ustc.edu.cn>
2006-09-24  2:42               ` Fengguang Wu
2006-09-24 14:25           ` Diego Calleja
2006-09-22  9:24 ` David Woodhouse
2006-09-22 10:10   ` David Woodhouse
2006-09-22 10:42     ` Jan Engelhardt
2006-09-22 11:20       ` David Woodhouse
2006-09-22 14:48     ` Randy.Dunlap
2006-09-23 10:44   ` David Woodhouse
2006-09-22 12:32 ` 2.6.19 -mm merge plans: AVR32 Haavard Skinnemoen
2006-09-22 14:42 ` 2.6.19 -mm merge plans Christoph Hellwig
2006-09-22 16:03   ` 2.6.19 -mm merge plans (ecryptfs) Andrew Morton
2006-09-22 16:53     ` Michael Halcrow
2006-09-25 14:59 ` [PATCH -mm updated] PCMCIA: Add few IDs into ide-cs Marcin Juszkiewicz
2006-09-25 15:48   ` Alan Cox

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=45131D2D.8020403@mbligh.org \
    --to=mbligh@mbligh.org \
    --cc=akpm@osdl.org \
    --cc=clameter@sgi.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=rohitseth@google.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