On Wed, 2014-02-19 at 22:05 -0800, Ben Widawsky wrote:
> The previous allocation mechanism would get 2 contiguous allocations,
> one for the page directories, and one for the page tables. As each page
> table is 1 page, and there are 512 of these per page directory, this
> goes to 2MB. An unfriendl
The previous allocation mechanism would get 2 contiguous allocations,
one for the page directories, and one for the page tables. As each page
table is 1 page, and there are 512 of these per page directory, this
goes to 2MB. An unfriendly request at best. Worse still, our HW now
supports 4 page dire
On Wed, Feb 19, 2014 at 11:20:51PM +0200, Imre Deak wrote:
> On Wed, 2014-02-19 at 13:06 -0800, Ben Widawsky wrote:
> > On Wed, Feb 19, 2014 at 09:11:46PM +0200, Imre Deak wrote:
> > > On Wed, 2014-02-12 at 14:28 -0800, Ben Widawsky wrote:
> > > > The previous allocation mechanism would get 2 conti
On Wed, 2014-02-19 at 13:06 -0800, Ben Widawsky wrote:
> On Wed, Feb 19, 2014 at 09:11:46PM +0200, Imre Deak wrote:
> > On Wed, 2014-02-12 at 14:28 -0800, Ben Widawsky wrote:
> > > The previous allocation mechanism would get 2 contiguous allocations,
> > > one for the page directories, and one for
On Wed, Feb 19, 2014 at 09:11:46PM +0200, Imre Deak wrote:
> On Wed, 2014-02-12 at 14:28 -0800, Ben Widawsky wrote:
> > The previous allocation mechanism would get 2 contiguous allocations,
> > one for the page directories, and one for the page tables. As each page
> > table is 1 page, and there ar
On Wed, 2014-02-19 at 21:11 +0200, Imre Deak wrote:
> On Wed, 2014-02-12 at 14:28 -0800, Ben Widawsky wrote:
> > The previous allocation mechanism would get 2 contiguous allocations,
> > one for the page directories, and one for the page tables. As each page
> > table is 1 page, and there are 512 o
On Wed, 2014-02-12 at 14:28 -0800, Ben Widawsky wrote:
> The previous allocation mechanism would get 2 contiguous allocations,
> one for the page directories, and one for the page tables. As each page
> table is 1 page, and there are 512 of these per page directory, this
> goes to 1MB. An unfriendl
On Wed, Feb 12, 2014 at 11:45:59PM +, Chris Wilson wrote:
> On Wed, Feb 12, 2014 at 02:28:48PM -0800, Ben Widawsky wrote:
> > - for (i = first_pte; i < last_pte; i++)
> > + for (i = which_pte; i < last_pte; i++) {
> > pt_vaddr[i] = scratch_pte;
> > +
On Wed, Feb 12, 2014 at 02:28:48PM -0800, Ben Widawsky wrote:
> - for (i = first_pte; i < last_pte; i++)
> + for (i = which_pte; i < last_pte; i++) {
> pt_vaddr[i] = scratch_pte;
> + num_entries--;
> + BUG_ON(num_
The previous allocation mechanism would get 2 contiguous allocations,
one for the page directories, and one for the page tables. As each page
table is 1 page, and there are 512 of these per page directory, this
goes to 1MB. An unfriendly request at best. Worse still, our HW now
supports 4 page dire
10 matches
Mail list logo