diff options
author | 2017-01-16 17:29:57 +0100 | |
---|---|---|
committer | 2017-02-02 10:30:37 +0100 | |
commit | e66774dd6f6a3d44559599e4eeb785734c28d034 (patch) | |
tree | e87ff31d2685fae6e2a5009d426f63abe3c3af85 /drivers/gpu/drm/etnaviv/etnaviv_mmu.h | |
parent | drm/etnaviv: get cmdbuf physical address through the cmdbuf abstraction (diff) | |
download | linux-dev-e66774dd6f6a3d44559599e4eeb785734c28d034.tar.xz linux-dev-e66774dd6f6a3d44559599e4eeb785734c28d034.zip |
drm/etnaviv: add cmdbuf suballocator
There are 3 big benefits to suballocating a single big DMA buffer
for command submission:
1. Avoid hammering CMA. The old way of allocating and freeing a DMA
buffer for each submission was hitting some of the real slow
pathes in CMA, as this allocator was not designed for a concurrent
small buffers load.
2. Less TLB flushes on IOMMUv2. If a new command buffer is mapped into
the GPU address space the MMU TLBs need to be flushed. By having
one big buffer statically mapped to the GPU, a lot of those flushes
can be avoided.
3. No funky workarounds for GC3000. The FE TLB flush on GC3000 isn't
reliable. To work around that we tried to lay out the cmdbufs in
the GPU address space in a way to avoid this issue. This hasn't
always worked if the address space is crowded. A single statically
mapped buffer avoids the erratum completely.
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Diffstat (limited to 'drivers/gpu/drm/etnaviv/etnaviv_mmu.h')
-rw-r--r-- | drivers/gpu/drm/etnaviv/etnaviv_mmu.h | 10 |
1 files changed, 6 insertions, 4 deletions
diff --git a/drivers/gpu/drm/etnaviv/etnaviv_mmu.h b/drivers/gpu/drm/etnaviv/etnaviv_mmu.h index e787e49c9693..54be289e5981 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_mmu.h +++ b/drivers/gpu/drm/etnaviv/etnaviv_mmu.h @@ -62,10 +62,12 @@ void etnaviv_iommu_unmap_gem(struct etnaviv_iommu *mmu, struct etnaviv_vram_mapping *mapping); void etnaviv_iommu_destroy(struct etnaviv_iommu *iommu); -u32 etnaviv_iommu_get_cmdbuf_va(struct etnaviv_gpu *gpu, - struct etnaviv_cmdbuf *buf); -void etnaviv_iommu_put_cmdbuf_va(struct etnaviv_gpu *gpu, - struct etnaviv_cmdbuf *buf); +int etnaviv_iommu_get_suballoc_va(struct etnaviv_gpu *gpu, dma_addr_t paddr, + struct drm_mm_node *vram_node, size_t size, + u32 *iova); +void etnaviv_iommu_put_suballoc_va(struct etnaviv_gpu *gpu, + struct drm_mm_node *vram_node, size_t size, + u32 iova); size_t etnaviv_iommu_dump_size(struct etnaviv_iommu *iommu); void etnaviv_iommu_dump(struct etnaviv_iommu *iommu, void *buf); |