[PATCH libdrm 1/2] amdgpu: Add wrappers for AMDGPU_VM IOCTL.
Christian König
ckoenig.leichtzumerken at gmail.com
Fri Oct 27 07:44:56 UTC 2017
Am 27.10.2017 um 02:15 schrieb Andrey Grodzovsky:
> Change-Id: I7eafb85c1ca96d6d255f0183bed0ce4129746fe0
> Signed-off-by: Andrey Grodzovsky <Andrey.Grodzovsky at amd.com>
> ---
> amdgpu/Makefile.sources | 1 +
> amdgpu/amdgpu.h | 20 +++++++++++++++++++
> amdgpu/amdgpu_vm.c | 52 +++++++++++++++++++++++++++++++++++++++++++++++++
> 3 files changed, 73 insertions(+)
> create mode 100644 amdgpu/amdgpu_vm.c
>
> diff --git a/amdgpu/Makefile.sources b/amdgpu/Makefile.sources
> index bc3abaa..498b64c 100644
> --- a/amdgpu/Makefile.sources
> +++ b/amdgpu/Makefile.sources
> @@ -6,6 +6,7 @@ LIBDRM_AMDGPU_FILES := \
> amdgpu_gpu_info.c \
> amdgpu_internal.h \
> amdgpu_vamgr.c \
> + amdgpu_vm.c \
> util_hash.c \
> util_hash.h \
> util_hash_table.c \
> diff --git a/amdgpu/amdgpu.h b/amdgpu/amdgpu.h
> index ecc975f..07f2851 100644
> --- a/amdgpu/amdgpu.h
> +++ b/amdgpu/amdgpu.h
> @@ -1489,6 +1489,26 @@ void amdgpu_cs_chunk_fence_to_dep(struct amdgpu_cs_fence *fence,
> void amdgpu_cs_chunk_fence_info_to_data(struct amdgpu_cs_fence_info *fence_info,
> struct drm_amdgpu_cs_chunk_data *data);
>
> +/**
> + * Reserve VMID
> + * \param context - \c [in] GPU Context
> + * \param flags - \c [in] TBD
> + *
> + * \return 0 on success otherwise POSIX Error code
> +*/
> +int amdgpu_vm_alloc_reserved_vmid(amdgpu_context_handle context,
> + uint32_t flags);
> +
> +/**
> + * Free reserved VMID
> + * \param context - \c [in] GPU Context
> + * \param flags - \c [in] TBD
> + *
> + * \return 0 on success otherwise POSIX Error code
> +*/
> +int amdgpu_vm_free_reserved_vmid(amdgpu_context_handle context,
> + uint32_t flags);
> +
> #ifdef __cplusplus
> }
> #endif
> diff --git a/amdgpu/amdgpu_vm.c b/amdgpu/amdgpu_vm.c
> new file mode 100644
> index 0000000..1664b7b
> --- /dev/null
> +++ b/amdgpu/amdgpu_vm.c
> @@ -0,0 +1,52 @@
> +/*
> + * Copyright 2017 Advanced Micro Devices, Inc.
> + *
> + * Permission is hereby granted, free of charge, to any person obtaining a
> + * copy of this software and associated documentation files (the "Software"),
> + * to deal in the Software without restriction, including without limitation
> + * the rights to use, copy, modify, merge, publish, distribute, sublicense,
> + * and/or sell copies of the Software, and to permit persons to whom the
> + * Software is furnished to do so, subject to the following conditions:
> + *
> + * The above copyright notice and this permission notice shall be included in
> + * all copies or substantial portions of the Software.
> + *
> + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
> + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
> + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
> + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
> + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
> + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
> + * OTHER DEALINGS IN THE SOFTWARE.
> + *
> + */
> +
> +
> +#include "amdgpu.h"
> +#include "amdgpu_drm.h"
> +#include "xf86drm.h"
> +#include "amdgpu_internal.h"
> +
> +int amdgpu_vm_alloc_reserved_vmid(amdgpu_context_handle context,
> + uint32_t flags)
Could just be the mail client, but flags looks way to much indented to
the right.
Additional to that I would keep the name of the kernel IOCTL for the
function name.
E.g. amdgpu_vm_reserve_vmid() and amdgpu_vm_unreserve_vmid().
Apart from that both patches look good to me.
Christian.
> +{
> + union drm_amdgpu_vm vm;
> +
> + vm.in.op = AMDGPU_VM_OP_RESERVE_VMID;
> + vm.in.flags = flags;
> +
> + return drmCommandWriteRead(context->dev->fd, DRM_AMDGPU_VM,
> + &vm, sizeof(vm));
> +}
> +
> +int amdgpu_vm_free_reserved_vmid(amdgpu_context_handle context,
> + uint32_t flags)
> +{
> + union drm_amdgpu_vm vm;
> +
> + vm.in.op = AMDGPU_VM_OP_UNRESERVE_VMID;
> + vm.in.flags = flags;
> +
> + return drmCommandWriteRead(context->dev->fd, DRM_AMDGPU_VM,
> + &vm, sizeof(vm));
> +}
More information about the amd-gfx
mailing list