Messages in this thread | | | Date | Tue, 01 Mar 2011 11:44:29 -0800 | From | "H. Peter Anvin" <> | Subject | Re: questions about init_memory_mapping_high() |
| |
On 03/01/2011 12:29 AM, Tejun Heo wrote: > Hey, > > (sorry about the earlier empty reply, fat finger on my phone) > > On Mon, Feb 28, 2011 at 10:14:44AM -0800, H. Peter Anvin wrote: >>> 1. The only rationale given in the commit description is that a >>> RED-PEN is killed, which was the following. >>> >>> /* >>> * RED-PEN putting page tables only on node 0 could >>> * cause a hotspot and fill up ZONE_DMA. The page tables >>> * need roughly 0.5KB per GB. >>> */ >>> >>> This already wasn't true with top-down memblock allocation. >>> >>> The 0.5KB per GiB comment is for 32bit w/ 3 level mapping. On >>> 64bit, it's ~4KiB per GiB when using 2MiB mappings and, well, very >>> small per GiB if 1GiB mapping is used. Even with 2MiB mapping, >>> 1TiB mapping would only be 4MiB. Under ZONE_DMA, this could be >>> problematic but with top-down this can't be a problem in any >>> realistic way in foreseeable future. >>> >> >> It's true on 64 bits too when PAE is not available (e.g. with Xen.) > > Hmm... I don't follow. Can you elaborate? If PAE is not available > for whatever reason, the physical memory is limited to 4GiB but I > don't follow what that has to do with the above. >
Sorry, PSE, not PAE.
-hpa
| |