On 10/12/2012 06:55 AM, Yinghai Lu wrote:
> On Thu, Oct 11, 2012 at 7:41 AM, Konrad Rzeszutek Wilk
> wrote:
>> On Wed, Oct 10, 2012 at 08:49:05AM -0700, Yinghai Lu wrote:
>>>
>>> yes that is some extreme case:
>>> assume that 2M range is [2T-2M, 2T),
>>
>> What is T in here? Terabyte? Is the '['
On Thu, Oct 11, 2012 at 7:41 AM, Konrad Rzeszutek Wilk
wrote:
> On Wed, Oct 10, 2012 at 08:49:05AM -0700, Yinghai Lu wrote:
>>
>> yes that is some extreme case:
>> assume that 2M range is [2T-2M, 2T),
>
> What is T in here? Terabyte? Is the '[' vs ')' a significance in your
> explanation? Should
On Wed, Oct 10, 2012 at 08:49:05AM -0700, Yinghai Lu wrote:
> On Wed, Oct 10, 2012 at 7:17 AM, Konrad Rzeszutek Wilk
> wrote:
> > On Tue, Oct 09, 2012 at 06:03:33PM -0700, Yinghai Lu wrote:
>
> >> good point, i add one line comment to the revised patch.
> >>
> >> +/* need 3 4k for initial
On Wed, Oct 10, 2012 at 08:49:05AM -0700, Yinghai Lu wrote:
On Wed, Oct 10, 2012 at 7:17 AM, Konrad Rzeszutek Wilk
kon...@kernel.org wrote:
On Tue, Oct 09, 2012 at 06:03:33PM -0700, Yinghai Lu wrote:
good point, i add one line comment to the revised patch.
+/* need 3 4k for initial
On Thu, Oct 11, 2012 at 7:41 AM, Konrad Rzeszutek Wilk
kon...@kernel.org wrote:
On Wed, Oct 10, 2012 at 08:49:05AM -0700, Yinghai Lu wrote:
yes that is some extreme case:
assume that 2M range is [2T-2M, 2T),
What is T in here? Terabyte? Is the '[' vs ')' a significance in your
explanation?
On 10/12/2012 06:55 AM, Yinghai Lu wrote:
On Thu, Oct 11, 2012 at 7:41 AM, Konrad Rzeszutek Wilk
kon...@kernel.org wrote:
On Wed, Oct 10, 2012 at 08:49:05AM -0700, Yinghai Lu wrote:
yes that is some extreme case:
assume that 2M range is [2T-2M, 2T),
What is T in here? Terabyte? Is the '['
On Wed, Oct 10, 2012 at 7:17 AM, Konrad Rzeszutek Wilk
wrote:
> On Tue, Oct 09, 2012 at 06:03:33PM -0700, Yinghai Lu wrote:
>> good point, i add one line comment to the revised patch.
>>
>> +/* need 3 4k for initial PMD_SIZE, 4k for 0-ISA_END_ADDRESS */
>
> Can you explain the math please? It
On Tue, Oct 09, 2012 at 06:03:33PM -0700, Yinghai Lu wrote:
> On Tue, Oct 9, 2012 at 9:01 AM, Konrad Rzeszutek Wilk
> wrote:
> >> +RESERVE_BRK(early_pgt_alloc, 16384);
> >
> > How did you come up with 16KB being the right size? What is this
> > based on? Can you provide a comment explaining why
On Tue, Oct 09, 2012 at 06:03:33PM -0700, Yinghai Lu wrote:
On Tue, Oct 9, 2012 at 9:01 AM, Konrad Rzeszutek Wilk kon...@kernel.org
wrote:
+RESERVE_BRK(early_pgt_alloc, 16384);
How did you come up with 16KB being the right size? What is this
based on? Can you provide a comment
On Wed, Oct 10, 2012 at 7:17 AM, Konrad Rzeszutek Wilk
kon...@kernel.org wrote:
On Tue, Oct 09, 2012 at 06:03:33PM -0700, Yinghai Lu wrote:
good point, i add one line comment to the revised patch.
+/* need 3 4k for initial PMD_SIZE, 4k for 0-ISA_END_ADDRESS */
Can you explain the math
On Tue, Oct 9, 2012 at 9:01 AM, Konrad Rzeszutek Wilk wrote:
>> +RESERVE_BRK(early_pgt_alloc, 16384);
>
> How did you come up with 16KB being the right size? What is this
> based on? Can you provide a comment explaining why 16KB is the
> right value on 32-bit and 64-bit machines?
good point, i
On Mon, Oct 08, 2012 at 09:39:11PM -0700, Yinghai Lu wrote:
> Get pgt_buf early from BRK, and use it to map page table at first.
>
> also use the left at first, then use new one.
Left?
>
> -v2: extra xen call back for that new range.
> -v3: fix compiling about #llx in print out that is reported
On Mon, Oct 08, 2012 at 09:39:11PM -0700, Yinghai Lu wrote:
Get pgt_buf early from BRK, and use it to map page table at first.
also use the left at first, then use new one.
Left?
-v2: extra xen call back for that new range.
-v3: fix compiling about #llx in print out that is reported by
On Tue, Oct 9, 2012 at 9:01 AM, Konrad Rzeszutek Wilk kon...@kernel.org wrote:
+RESERVE_BRK(early_pgt_alloc, 16384);
How did you come up with 16KB being the right size? What is this
based on? Can you provide a comment explaining why 16KB is the
right value on 32-bit and 64-bit machines?
good
Get pgt_buf early from BRK, and use it to map page table at first.
also use the left at first, then use new one.
-v2: extra xen call back for that new range.
-v3: fix compiling about #llx in print out that is reported by Fengguang
-v4: remove the early_pgt_buf_* stuff, because xen interface is
Get pgt_buf early from BRK, and use it to map page table at first.
also use the left at first, then use new one.
-v2: extra xen call back for that new range.
-v3: fix compiling about #llx in print out that is reported by Fengguang
-v4: remove the early_pgt_buf_* stuff, because xen interface is
16 matches
Mail list logo