Displaying 6 results from an estimated 6 matches for "page_walk".
2019 Aug 16
2
[PATCH 04/15] mm: remove the pgmap field from struct hmm_vma_walk
...operations was one of the motivations for plumbing
> get_dev_pagemap() with a percpu-ref.
hmm_range_fault() doesn't use page pins at all, so if a ZONE_DEVICE
page comes out of it then it needs to use another locking pattern.
If I follow it all right:
We can do a get_dev_pagemap inside the page_walk and touch the pgmap,
or we can do the 'device mutex && retry' pattern and touch the pgmap
in the driver, under that lock.
However in all cases the current get_dev_pagemap()'s in the page walk
are not necessary, and we can delete them.
?
Jason
2019 Aug 16
1
[PATCH 04/15] mm: remove the pgmap field from struct hmm_vma_walk
On Fri, Aug 16, 2019 at 10:21:41AM -0700, Dan Williams wrote:
> > We can do a get_dev_pagemap inside the page_walk and touch the pgmap,
> > or we can do the 'device mutex && retry' pattern and touch the pgmap
> > in the driver, under that lock.
> >
> > However in all cases the current get_dev_pagemap()'s in the page walk
> > are not necessary, and we can delete...
2008 Jan 18
0
[PATCH] minios: support COW for a zero page
...d7ed41be2 extras/mini-os/arch/x86/mm.c
--- a/extras/mini-os/arch/x86/mm.c Fri Jan 18 15:55:13 2008 +0000
+++ b/extras/mini-os/arch/x86/mm.c Fri Jan 18 16:02:32 2008 +0000
@@ -50,6 +50,7 @@
#endif
unsigned long *phys_to_machine_mapping;
+unsigned long mfn_zero;
extern char stack[];
extern void page_walk(unsigned long virt_addr);
@@ -492,10 +493,13 @@ static void clear_bootstrap(void)
static void clear_bootstrap(void)
{
struct xen_memory_reservation reservation;
- xen_pfn_t mfns[] = { virt_to_mfn(0), virt_to_mfn(&shared_info) };
+ xen_pfn_t mfns[] = { virt_to_mfn(&shared_info...
2019 Aug 16
0
[PATCH 04/15] mm: remove the pgmap field from struct hmm_vma_walk
...for plumbing
> > get_dev_pagemap() with a percpu-ref.
>
> hmm_range_fault() doesn't use page pins at all, so if a ZONE_DEVICE
> page comes out of it then it needs to use another locking pattern.
>
> If I follow it all right:
>
> We can do a get_dev_pagemap inside the page_walk and touch the pgmap,
> or we can do the 'device mutex && retry' pattern and touch the pgmap
> in the driver, under that lock.
>
> However in all cases the current get_dev_pagemap()'s in the page walk
> are not necessary, and we can delete them.
Yes, as long as ...
2019 Aug 16
2
[PATCH 04/15] mm: remove the pgmap field from struct hmm_vma_walk
On Thu, Aug 15, 2019 at 01:47:12PM -0700, Dan Williams wrote:
> On Thu, Aug 15, 2019 at 1:41 PM Jason Gunthorpe <jgg at mellanox.com> wrote:
> >
> > On Thu, Aug 15, 2019 at 04:33:06PM -0400, Jerome Glisse wrote:
> >
> > > So nor HMM nor driver should dereference the struct page (i do not
> > > think any iommu driver would either),
> >
> >
2012 Nov 26
13
[PATCH 0 of 4] Minios improvements for app development
This patch series contains a set of patches making minios rather easier
to use, from an application development point of view.
Overview of patches:
1 Command line argument parsing support, from Xen.
2 Weak console handler function.
3 Build system tweaks for application directories.
4 Trailing whitespace cleanup. (because it is very messy)
Patch 4 is likely to be more controversial than