Hi Marek,
On Tue, 21 Aug 2012 17:01:08 +0200
Marek Szyprowski wrote:
> > > > -__iommu_alloc_remap(struct page **pages, size_t size, gfp_t gfp,
> > > > pgprot_t prot)
> > > > +__iommu_alloc_remap(struct page **pages, size_t size, gfp_t gfp,
> > > > pgprot_t prot,
> > > > +
Hi Marek,
On Tue, 21 Aug 2012 17:01:08 +0200
Marek Szyprowski m.szyprow...@samsung.com wrote:
-__iommu_alloc_remap(struct page **pages, size_t size, gfp_t gfp,
pgprot_t prot)
+__iommu_alloc_remap(struct page **pages, size_t size, gfp_t gfp,
pgprot_t prot,
+
Hi Hiroshi,
On Tuesday, August 21, 2012 2:15 PM Hiroshi Doyu wrote:
> Hiroshi Doyu wrote @ Tue, 21 Aug 2012 13:22:35 +0200:
>
> > Hi,
> >
> > On Mon, 30 Jul 2012 10:28:19 +0200
> > Marek Szyprowski wrote:
> >
> > > This patch changes dma-mapping subsystem to use generic vmalloc areas
> > >
Russell King - ARM Linux wrote @ Tue, 21 Aug 2012
14:34:51 +0200:
> On Tue, Aug 21, 2012 at 02:22:35PM +0300, Hiroshi Doyu wrote:
> > The following "__get_vm_area_node()" can take gfp_mask, it means that
> > this function is expected to be called from atomic context, but why
> > it's _NOT_
On Tue, Aug 21, 2012 at 02:22:35PM +0300, Hiroshi Doyu wrote:
> The following "__get_vm_area_node()" can take gfp_mask, it means that
> this function is expected to be called from atomic context, but why
> it's _NOT_ allowed _ONLY_ from interrupt context?
One reason is it takes read/write locks
Hiroshi Doyu wrote @ Tue, 21 Aug 2012 13:22:35 +0200:
> Hi,
>
> On Mon, 30 Jul 2012 10:28:19 +0200
> Marek Szyprowski wrote:
>
> > This patch changes dma-mapping subsystem to use generic vmalloc areas
> > for all consistent dma allocations. This increases the total size limit
> > of the
Hi,
On Mon, 30 Jul 2012 10:28:19 +0200
Marek Szyprowski wrote:
> This patch changes dma-mapping subsystem to use generic vmalloc areas
> for all consistent dma allocations. This increases the total size limit
> of the consistent allocations and removes platform hacks and a lot of
> duplicated
Hi,
On Mon, 30 Jul 2012 10:28:19 +0200
Marek Szyprowski m.szyprow...@samsung.com wrote:
This patch changes dma-mapping subsystem to use generic vmalloc areas
for all consistent dma allocations. This increases the total size limit
of the consistent allocations and removes platform hacks and a
Hiroshi Doyu hd...@nvidia.com wrote @ Tue, 21 Aug 2012 13:22:35 +0200:
Hi,
On Mon, 30 Jul 2012 10:28:19 +0200
Marek Szyprowski m.szyprow...@samsung.com wrote:
This patch changes dma-mapping subsystem to use generic vmalloc areas
for all consistent dma allocations. This increases the
On Tue, Aug 21, 2012 at 02:22:35PM +0300, Hiroshi Doyu wrote:
The following __get_vm_area_node() can take gfp_mask, it means that
this function is expected to be called from atomic context, but why
it's _NOT_ allowed _ONLY_ from interrupt context?
One reason is it takes read/write locks
Russell King - ARM Linux li...@arm.linux.org.uk wrote @ Tue, 21 Aug 2012
14:34:51 +0200:
On Tue, Aug 21, 2012 at 02:22:35PM +0300, Hiroshi Doyu wrote:
The following __get_vm_area_node() can take gfp_mask, it means that
this function is expected to be called from atomic context, but why
Hi Hiroshi,
On Tuesday, August 21, 2012 2:15 PM Hiroshi Doyu wrote:
Hiroshi Doyu hd...@nvidia.com wrote @ Tue, 21 Aug 2012 13:22:35 +0200:
Hi,
On Mon, 30 Jul 2012 10:28:19 +0200
Marek Szyprowski m.szyprow...@samsung.com wrote:
This patch changes dma-mapping subsystem to use
This patch changes dma-mapping subsystem to use generic vmalloc areas
for all consistent dma allocations. This increases the total size limit
of the consistent allocations and removes platform hacks and a lot of
duplicated code.
Atomic allocations are served from special pool preallocated on
This patch changes dma-mapping subsystem to use generic vmalloc areas
for all consistent dma allocations. This increases the total size limit
of the consistent allocations and removes platform hacks and a lot of
duplicated code.
Atomic allocations are served from special pool preallocated on
14 matches
Mail list logo