3 Commits

Author	SHA1	Message	Date
Sagar Kamble	ed16377983	gpu: nvgpu: allocate comptags and store metadata in REGISTER_BUFFER ioctl ... To enable userspace query about comptags allocation status of a buffer, comptags are to be allocated only during buffer registration done by nvrm_gpu. Earlier, they were allocated during map. nvrm_gpu will be sending metadata blob to be associated with the buffer. This will have to be stored in the dmabuf privdata for all the buffers registered by nvrm_gpu. This patch moves the privdata allocation to buffer registration ioctl. Remove g->mm.priv_lock as it is not needed now. This lock was added to protect dmabuf private data setup. That private data is now handled through dmabuf->ops and setup of dmabuf->ops is done under dmabuf->lock. To support legacy userspace, this patch still allocates comptags on demand on map calls for unregistered buffers. Bug 200586313 Change-Id: I88b2ca04c733dd02a84bcbf05060bddc00147790 Signed-off-by: Sagar Kamble <skamble@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2480761 Tested-by: mobile promotions <svcmobile_promotions@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>	2021-09-02 11:42:08 -07:00
scottl	cd3ad1ccc7	gpu: nvgpu: fix REMAP android build failure ... Rework nvgpu_vm_remap_os_buf structure initialization to avoid android/clang build issues with the use of a single pair of {} to initialize certain structures. The os-dependent nvgpu_vm_remap_os_buf_get() routine now does a memset of the structure prior to initializing its contents. Jira NVGPU-6804 Change-Id: I08682c6ab7b8324a605a56ed660dea5bea11d16b Signed-off-by: scottl <scottl@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2553193 Tested-by: mobile promotions <svcmobile_promotions@nvidia.com> Reviewed-by: svcacv <svcacv@nvidia.com> Reviewed-by: svc_kernel_abi <svc_kernel_abi@nvidia.com> Reviewed-by: Alex Waterman <alexw@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> GVS: Gerrit_Virtual_Submit	2021-07-03 02:05:25 -07:00
scottl	3cd256b344	gpu: nvgpu: add linux REMAP support ... Add REMAP ioctl and accompanying support to the linux nvgpu driver. REMAP support provides per-page control over sparse VM areas using the concept of a virtual memory pool. The REMAP ioctl accepts a list of operations (each a map or unmap) that modify the VM area pages tracked by the virtual mmemory pool. Inclusion of REMAP support in the nvgpu build is controlled by the new CONFIG_NVGPU_REMAP flag. This flag is enabled by default for linux builds. A new NVGPU_GPU_FLAGS_SUPPORT_REMAP characteristics flag is added for use in detecting when REMAP support is available. When a VM allocation tagged with NVGPU_VM_AREA_ALLOC_SPARSE is made the base virtual memory pool resources are allocated. Per-page resources are later allocated when the NVGPU_AS_IOCTL_REMAP ioctl is issued. All REMAP resources are released when the corresponding VM area is freed. Jira NVGPU-6804 Change-Id: I1f2cdc0c06c1698a62640c1c6fbcb2f9db24a0bc Signed-off-by: scottl <scottl@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2542178 Tested-by: mobile promotions <svcmobile_promotions@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>	2021-06-28 22:39:06 -07:00