linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Andi Kleen <andi@firstfloor.org>
To: "Luiz Fernando N. Capitulino" <lcapitulino@mandriva.com.br>
Cc: Andi Kleen <andi@firstfloor.org>,
	linux-kernel@vger.kernel.org, linux-mm@kvack.org
Subject: Re: [PATCH] [0/13] General DMA zone rework
Date: Tue, 11 Mar 2008 18:35:40 +0100	[thread overview]
Message-ID: <20080311173540.GG27593@one.firstfloor.org> (raw)
In-Reply-To: <20080311142624.1dbd3af5@mandriva.com.br>

On Tue, Mar 11, 2008 at 02:26:24PM -0300, Luiz Fernando N. Capitulino wrote:
> ===================================================================
> --- linux-2.6.24.orig/sound/core/memalloc.c
> +++ linux-2.6.24/sound/core/memalloc.c
> @@ -218,7 +218,6 @@ static void *snd_malloc_dev_pages(struct
>  	snd_assert(dma != NULL, return NULL);
>  	pg = get_order(size);
>  	gfp_flags = GFP_KERNEL
> -		| __GFP_COMP	/* compound page lets parts be mapped */

Oops. Thanks. I'll double check that. mask allocator indeed doesn't
handle __GFP_COMP and nobody should be passing that into dma_alloc_coherent
anyways. But the bug you got was for the small size wasn't it?

>  		| __GFP_NORETRY /* don't trigger OOM-killer */
>  		| __GFP_NOWARN; /* no stack trace print - this call is non-critical */
>  	res = dma_alloc_coherent(dev, PAGE_SIZE << pg, dma, gfp_flags);
> """
> 
>  Also, I think you forgot to protect the __free_pages_mask()
> call with "#ifdef CONFIG_MASK_ALLOC" in patch mask-compat.

PageMaskAlloc() returns constant 0 in this case so the compiler should e
liminate it.

> 
>  Question: let's say that the devices I have consumes only
> 5MB of the whole reserved pool (16MB), of course that it's

It's likely less than 16MB because there are already some users in there
like the kernel text. The default sizing without swiotlb just grabs anything 
left over up to the 16Mb boundary.

> ok to reduce the pool to 5MB, right?

Yes, but not leaving any free over is a little risky, someone might need
more later. But e.g. going down to 8-9MB would be likely possible.

The long term plan (but that is some time off and a little vague still) 
would be to let the various subsystem size the mask zone dynamically for 
their need.

Then especially on 64bit swiotlb machines who allocate 64MB by default
for swiotlb that would save significant memory.

> 
>  I have more machines to test this stuff...

Thanks. Keep me posted.

-Andi

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

  reply	other threads:[~2008-03-11 17:35 UTC|newest]

Thread overview: 56+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-03-07  9:07 Andi Kleen
2008-03-07  9:07 ` [PATCH] [2/13] Make get_order(0) return 0 Andi Kleen
2008-03-07  9:07 ` [PATCH] [3/13] Make kvm bad_page symbol static Andi Kleen
2008-03-07  9:07 ` [PATCH] [4/13] Prepare page_alloc for the maskable allocator Andi Kleen
2008-03-07 18:19   ` Sam Ravnborg
2008-03-07 18:36     ` Cyrill Gorcunov
2008-03-07 19:02     ` Andi Kleen
2008-03-07  9:07 ` [PATCH] [5/13] Add mask allocator statistics to vmstat.[ch] Andi Kleen
2008-03-08  2:24   ` Christoph Lameter
2008-03-07  9:07 ` [PATCH] [6/13] Core maskable allocator Andi Kleen
2008-03-07 10:53   ` Johannes Weiner
2008-03-07 11:14     ` Andi Kleen
2008-03-07 17:05   ` Randy Dunlap
2008-03-07 17:31     ` Andi Kleen
2008-03-07 17:33       ` Randy Dunlap
2008-03-07 17:43         ` Andi Kleen
2008-03-07 17:51           ` Randy Dunlap
2008-03-07 21:13   ` Cyrill Gorcunov
2008-03-07 23:28     ` Andi Kleen
2008-03-08  5:03   ` KAMEZAWA Hiroyuki
2008-03-08  5:41     ` KAMEZAWA Hiroyuki
2008-03-08 11:41     ` Andi Kleen
2008-03-11 15:34   ` Jonathan Corbet
2008-03-11 15:54     ` Andi Kleen
2008-03-07  9:07 ` [PATCH] [7/13] Implement compat hooks for GFP_DMA Andi Kleen
2008-03-07  9:07 ` [PATCH] [8/13] Enable the mask allocator for x86 Andi Kleen
2008-03-07 18:32   ` Sam Ravnborg
2008-03-07 19:03     ` Andi Kleen
2008-03-07 19:09       ` Sam Ravnborg
2008-03-08  2:37   ` Christoph Lameter
2008-03-08  6:35     ` Yinghai Lu
2008-03-08  7:31       ` Christoph Lameter
2008-03-08 11:54     ` Andi Kleen
2008-03-10 17:13       ` Christoph Lameter
2008-03-07  9:07 ` [PATCH] [9/13] Remove set_dma_reserve Andi Kleen
2008-03-07  9:07 ` [PATCH] [10/13] Switch the 32bit dma_alloc_coherent functions over to use the maskable allocator Andi Kleen
2008-03-07  9:07 ` [PATCH] [11/13] Switch x86-64 dma_alloc_coherent over to " Andi Kleen
2008-03-07  9:07 ` [PATCH] [12/13] Add vmstat statistics for new swiotlb code Andi Kleen
2008-03-08  2:38   ` Christoph Lameter
2008-03-07  9:07 ` [PATCH] [13/13] Convert x86-64 swiotlb to use the mask allocator directly Andi Kleen
2008-03-07 15:18 ` [PATCH] [0/13] General DMA zone rework Rene Herman
2008-03-07 15:22   ` Rene Herman
2008-03-07 15:31     ` Andi Kleen
2008-03-07 15:34   ` Andi Kleen
2008-03-07 20:51 ` Luiz Fernando N. Capitulino
2008-03-08  0:46   ` Andi Kleen
2008-03-10 18:03     ` Luiz Fernando N. Capitulino
2008-03-10 18:08       ` Andi Kleen
2008-03-11 17:26         ` Luiz Fernando N. Capitulino
2008-03-11 17:35           ` Andi Kleen [this message]
2008-03-11 18:00             ` Luiz Fernando N. Capitulino
2008-03-11 18:49               ` Andi Kleen
2008-03-11 19:36                 ` Luiz Fernando N. Capitulino
2008-03-08  2:42 ` Christoph Lameter
2008-03-08 11:57   ` Andi Kleen
2008-03-10 17:14     ` Christoph Lameter

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=20080311173540.GG27593@one.firstfloor.org \
    --to=andi@firstfloor.org \
    --cc=lcapitulino@mandriva.com.br \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.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