gpu: nvgpu: add sched domain bind ioctl

Support binding TSGs to some other scheduling domain than the default
one. Binding happens by name until a more robust interface appears in
the future, as the name is a natural identifier for users.

No other domains are actually created anywhere yet; that will happen in
next patches.

Jira NVGPU-6788

Change-Id: I5abcdea869b525b0a0e9937302f106f7eee94ec2
Signed-off-by: Konsta Hölttä <kholtta@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2628047
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
Reviewed-by: Alex Waterman <alexw@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: Gerrit_Virtual_Submit
This commit is contained in:
Konsta Hölttä
2021-10-12 16:09:05 +03:00
committed by mobile promotions
parent e692cd9913
commit fe7ae02f5f
4 changed files with 70 additions and 3 deletions

View File

@@ -373,6 +373,8 @@ void nvgpu_tsg_disable(struct nvgpu_tsg *tsg);
int nvgpu_tsg_bind_channel(struct nvgpu_tsg *tsg,
struct nvgpu_channel *ch);
int nvgpu_tsg_bind_domain(struct nvgpu_tsg *tsg, const char *domain_name);
/**
* @brief Get pointer to #nvgpu_tsg for the tsgid.
*