2010/3/2 Thomas Hellström tho...@shipmail.org:
Michel Dänzer wrote:
On Tue, 2010-03-02 at 00:32 +0200, Pauli Nieminen wrote:
bo allocation is still too expensive operation for dma buffers in
classic mesa. It takes quite a lot cpu time in bind memory (agp
system) and ttm_mem_global
While this is almost surely a good idea, note that userspace caching
and suballocation substantially improves Mesa performance even on PCIe
systems.
This is mostly due to the unavoidable overhead of kernel calls and
pagetable modifications, as well as the avoidable linear search the
kernel
Luca Barbieri wrote:
While this is almost surely a good idea, note that userspace caching
and suballocation substantially improves Mesa performance even on PCIe
systems.
This is mostly due to the unavoidable overhead of kernel calls and
pagetable modifications,
as well as the avoidable
^^^ Luca, I've never seen this show up high on a profile (yet). Do you see
that with Nouveau? I used to have an rb-tree implementation of drm_mm_xxx
lying around, but I didn't use it because I didn't have a case where it
showed up?
Yes, before I did userspace allocation, in doom3 profiles,
Luca Barbieri wrote:
^^^ Luca, I've never seen this show up high on a profile (yet). Do you see
that with Nouveau? I used to have an rb-tree implementation of drm_mm_xxx
lying around, but I didn't use it because I didn't have a case where it
showed up?
Yes, before I did userspace
On Wed, Mar 3, 2010 at 4:50 PM, Jerome Glisse gli...@freedesktop.org wrote:
On Wed, Mar 03, 2010 at 04:23:08PM +0200, Pauli Nieminen wrote:
On Wed, Mar 3, 2010 at 3:33 PM, Jerome Glisse gli...@freedesktop.org wrote:
On Tue, Mar 02, 2010 at 12:32:54AM +0200, Pauli Nieminen wrote:
On Sun, Feb
On Wed, Mar 03, 2010 at 05:46:43PM +0200, Pauli Nieminen wrote:
On Wed, Mar 3, 2010 at 4:50 PM, Jerome Glisse gli...@freedesktop.org wrote:
On Wed, Mar 03, 2010 at 04:23:08PM +0200, Pauli Nieminen wrote:
On Wed, Mar 3, 2010 at 3:33 PM, Jerome Glisse gli...@freedesktop.org
wrote:
On Tue,
On Tue, 2010-03-02 at 00:32 +0200, Pauli Nieminen wrote:
bo allocation is still too expensive operation for dma buffers in
classic mesa. It takes quite a lot cpu time in bind memory (agp
system) and ttm_mem_global functions. Would it be possible to move
some parts those expensive operations
On AGP system we might allocate/free routinely uncached or wc memory,
changing page from cached (wb) to uc or wc is very expensive and involves
a lot of flushing. To improve performance this allocator use a pool
of uc,wc pages.
Pools are linked lists of pages. Ordered so that first is the latest
