diff options
author | Alex Waterman <alexw@nvidia.com> | 2017-10-18 16:24:53 -0400 |
---|---|---|
committer | mobile promotions <svcmobile_promotions@nvidia.com> | 2017-11-10 18:46:54 -0500 |
commit | ee4970a33f41b56f2ada6a0b5ab6f9c400e39d88 (patch) | |
tree | 65d26ac7fd8667ac10cee8330a7647e9e72a745c /drivers/gpu/nvgpu/include | |
parent | 6911b4d48c414279731580f1212e29e4b691b04c (diff) |
gpu: nvgpu: Make buf alignment generic
Drastically simplify and move the aligment computation for buffers
getting mapped into the SGT code. An SGT is all that is needed for
computing the alignment.
However, this did require that a new SGT op was added:
nvgpu_sgt_iommuable()
This function returns true if the passed SGT is IOMMU'able and must
be implemented by an SGT implementation that has IOMMU'able buffers.
If this function is left as NULL then it is assumed that the buffer
is not IOMMU'able.
Also cleanup the parameter ordering convention among all nvgpu_sgt
functions. Previously there was a mishmash of different parameter
orderings. This patch now standardizes on the gk20a first approach
seen everywhere else in the driver.
JIRA NVGPU-30
JIRA NVGPU-246
JIRA NVGPU-71
Change-Id: Ic4ab7b752847cf795c7cfafed5a07818217bba86
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1583985
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Diffstat (limited to 'drivers/gpu/nvgpu/include')
-rw-r--r-- | drivers/gpu/nvgpu/include/nvgpu/nvgpu_mem.h | 12 |
1 files changed, 10 insertions, 2 deletions
diff --git a/drivers/gpu/nvgpu/include/nvgpu/nvgpu_mem.h b/drivers/gpu/nvgpu/include/nvgpu/nvgpu_mem.h index 6feacff7..2b8b7015 100644 --- a/drivers/gpu/nvgpu/include/nvgpu/nvgpu_mem.h +++ b/drivers/gpu/nvgpu/include/nvgpu/nvgpu_mem.h | |||
@@ -62,6 +62,11 @@ struct nvgpu_sgt_ops { | |||
62 | u64 (*sgl_gpu_addr)(struct gk20a *g, void *sgl, | 62 | u64 (*sgl_gpu_addr)(struct gk20a *g, void *sgl, |
63 | struct nvgpu_gmmu_attrs *attrs); | 63 | struct nvgpu_gmmu_attrs *attrs); |
64 | /* | 64 | /* |
65 | * If left NULL then iommuable is assumed to be false. | ||
66 | */ | ||
67 | bool (*sgt_iommuable)(struct gk20a *g, struct nvgpu_sgt *sgt); | ||
68 | |||
69 | /* | ||
65 | * Note: this operates on the whole SGT not a specific SGL entry. | 70 | * Note: this operates on the whole SGT not a specific SGL entry. |
66 | */ | 71 | */ |
67 | void (*sgt_free)(struct gk20a *g, struct nvgpu_sgt *sgt); | 72 | void (*sgt_free)(struct gk20a *g, struct nvgpu_sgt *sgt); |
@@ -232,9 +237,12 @@ void *nvgpu_sgt_get_next(struct nvgpu_sgt *sgt, void *sgl); | |||
232 | u64 nvgpu_sgt_get_phys(struct nvgpu_sgt *sgt, void *sgl); | 237 | u64 nvgpu_sgt_get_phys(struct nvgpu_sgt *sgt, void *sgl); |
233 | u64 nvgpu_sgt_get_dma(struct nvgpu_sgt *sgt, void *sgl); | 238 | u64 nvgpu_sgt_get_dma(struct nvgpu_sgt *sgt, void *sgl); |
234 | u64 nvgpu_sgt_get_length(struct nvgpu_sgt *sgt, void *sgl); | 239 | u64 nvgpu_sgt_get_length(struct nvgpu_sgt *sgt, void *sgl); |
235 | u64 nvgpu_sgt_get_gpu_addr(struct nvgpu_sgt *sgt, struct gk20a *g, void *sgl, | 240 | u64 nvgpu_sgt_get_gpu_addr(struct gk20a *g, struct nvgpu_sgt *sgt, void *sgl, |
236 | struct nvgpu_gmmu_attrs *attrs); | 241 | struct nvgpu_gmmu_attrs *attrs); |
237 | void nvgpu_sgt_free(struct nvgpu_sgt *sgt, struct gk20a *g); | 242 | void nvgpu_sgt_free(struct gk20a *g, struct nvgpu_sgt *sgt); |
243 | |||
244 | bool nvgpu_sgt_iommuable(struct gk20a *g, struct nvgpu_sgt *sgt); | ||
245 | u64 nvgpu_sgt_alignment(struct gk20a *g, struct nvgpu_sgt *sgt); | ||
238 | 246 | ||
239 | /** | 247 | /** |
240 | * nvgpu_mem_create_from_mem - Create a new nvgpu_mem struct from an old one. | 248 | * nvgpu_mem_create_from_mem - Create a new nvgpu_mem struct from an old one. |