Re: [PATCH 04/10] x86, mm: Don't clear page table if next range is ram

2012-10-09 Thread Yinghai Lu
On Tue, Oct 9, 2012 at 9:04 AM, Konrad Rzeszutek Wilk wrote: > How do we clean it wrongly? >> And it only happens when we are trying to map range one by one range >> separately. >> >> After we add checking before clearing the related page table, that panic will >> not happen anymore. > > So we do

Re: [PATCH 04/10] x86, mm: Don't clear page table if next range is ram

2012-10-09 Thread Konrad Rzeszutek Wilk
On Mon, Oct 08, 2012 at 09:39:12PM -0700, Yinghai Lu wrote: > After we add code use BRK to map buffer for final page table, .. mention the name of the patch that adds this. What is 'final page table'? Isn't this just the existing bootup tables modified to cover more memory. > > It should be safe

[PATCH 04/10] x86, mm: Don't clear page table if next range is ram

2012-10-08 Thread Yinghai Lu
After we add code use BRK to map buffer for final page table, It should be safe to remove early_memmap for page table accessing. But we get panic with that. It turns out we clear the initial page table wrongly for next range that is separated by holes. And it only happens when we are trying to m