Re: [Xen-devel] x86 swiotlb questions - Herbert Xu - com.xensource.lists.xen-devel

29 messages in com.xensource.lists.xen-develRe: [Xen-devel] x86 swiotlb questions

From	Sent On	Attachments
Jan Beulich	15 Dec 2006 04:50
Keir Fraser	15 Dec 2006 05:35
Jan Beulich	15 Dec 2006 05:52
Keir Fraser	15 Dec 2006 06:03
Jan Beulich	15 Dec 2006 06:16
Keir Fraser	15 Dec 2006 06:18
Jan Beulich	15 Dec 2006 06:46
Alan	15 Dec 2006 08:19
Keir Fraser	15 Dec 2006 08:47
Jan Beulich	17 Dec 2006 23:44
Keir Fraser	18 Dec 2006 01:38
Jan Beulich	19 Dec 2006 04:47
Keir Fraser	19 Dec 2006 06:14
Jan Beulich	19 Dec 2006 06:39
Keir Fraser	19 Dec 2006 06:46
Muli Ben-Yehuda	19 Dec 2006 09:06
Jan Beulich	20 Dec 2006 08:39	.patch, .patch, .patch, 2 more
Jan Beulich	22 Dec 2006 06:49	.patch, .patch, .patch, 2 more
Jan Beulich	22 Dec 2006 08:20
Herbert Xu	22 Dec 2006 13:00
Keir Fraser	23 Dec 2006 01:48
Muli Ben-Yehuda	24 Dec 2006 20:49
Keir Fraser	25 Dec 2006 02:20
Jan Beulich	30 Dec 2006 09:32
Keir Fraser	30 Dec 2006 09:46
Jan Beulich	02 Jan 2007 00:39
Jan Beulich	02 Jan 2007 23:09
Keir Fraser	03 Jan 2007 01:32
Jan Beulich	03 Jan 2007 03:01

Subject:	Re: [Xen-devel] x86 swiotlb questions
From:	Herbert Xu (herb...@gondor.apana.org.au)
Date:	12/22/2006 01:00:12 PM
List:	com.xensource.lists.xen-devel

On Fri, Dec 22, 2006 at 04:20:28PM +0000, Jan Beulich wrote:

One more thing: Is it really necessary to restrict dma_alloc_coherent() to
dma_bits? I.e., couldn't we, once the bit-level page allocator is merged, use the real bit
width needed for the requesting device here? If not, this would then permit using the

Yes I think that should work. If the width is less than what the hypervisor supports it'll simply fail which is better than returning the wrong memory silently.

Cheers,