RE: [Patch] Allocate sparse vmemmap block above 4G

!MAILaRCHIVE_VOTE_RePLACE
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
To: Zou, Nanhai <nanhai.zou@...>
Cc: Mel Gorman <mel@...>, LKML <linux-kernel@...>, Linus Torvalds <torvalds@...>, Greg KH <greg@...>, Dave Jones <davej@...>, Martin Ebourne <fedora@...>, Siddha, Suresh B <suresh.b.siddha@...>, Andi Kleen <ak@...>, Andrew Morton <akpm@...>, Andy Whitcroft <apw@...>
Date: Thursday, November 8, 2007 - 9:54 pm

On Fri, 9 Nov 2007, Zou, Nanhai wrote:


The 4GB boundary here is MAX_DMA32_ADDRESS I guess? We are only having 
this problem because of the two DMA zones on x86_64. I thought Andi was 
getting rid of the first one at 16MB. If he would do so then ZONE_DMA 
could be used instead of DMA32 and everything will be fine.

For now you may want to put

#ifdef CONFIG_ZONE_DMA32

around this code since it depends on DMA32.
-
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

Messages in current thread:
[Patch] Allocate sparse vmemmap block above 4G, Zou Nan hai, (Wed Nov 7, 8:52 pm)
Re: [Patch] Allocate sparse vmemmap block above 4G, Mel Gorman, (Thu Nov 8, 10:07 am)
RE: [Patch] Allocate sparse vmemmap block above 4G, Zou, Nanhai, (Thu Nov 8, 9:28 pm)
Re: [Patch] Allocate sparse vmemmap block above 4G, Mel Gorman, (Fri Nov 9, 10:42 am)
RE: [Patch] Allocate sparse vmemmap block above 4G, Christoph Lameter, (Thu Nov 8, 9:54 pm)