diff options
| author | Arthur Hanson Rasmusson <arthur@vgpu.io> | 2026-10-05 13:40:15 -0400 |
|---|---|---|
| committer | Arthur Hanson Rasmusson <arthur@vgpu.io> | 2026-10-05 13:40:15 -0400 |
| commit | d07555097d5f46d0f51c98572a220235ea5fcc0d (patch) | |
| tree | 0e90f80fc2ce3e9d7f72236ed8ddb82b0ce6a944 /kernel/gsp_phys_gpu.h | |
| parent | 97e7d0136fe63b08823d8b66d5bab84cfecf435b (diff) | |
| download | open-iov-d07555097d5f46d0f51c98572a220235ea5fcc0d.tar.gz open-iov-d07555097d5f46d0f51c98572a220235ea5fcc0d.zip | |
vgsp: add gsp_phys_rmcall_channel to force one GR channel variant
Lets a bisection pair any promote shape with any channel shape, in
particular V6 (SKIP_CTXBUFFER_ALLOC, KMD-managed) with the server-managed
VA-only promote: with skip-ctxbuffer the server's kgrctxMapCtxBuffers does
not pre-book the engine context vaList, so the VA branch's
vaListSetManaged(NV_FALSE) gate should pass and the per-buffer commit can
program the engine context VA. Default 0 keeps the mode-selected walk
byte-identical. Not hardware-tested (GPU wedged; see Handoff ADDENDUM 51).
Diffstat (limited to 'kernel/gsp_phys_gpu.h')
| -rw-r--r-- | kernel/gsp_phys_gpu.h | 7 |
1 files changed, 7 insertions, 0 deletions
diff --git a/kernel/gsp_phys_gpu.h b/kernel/gsp_phys_gpu.h index a270df8..925bfb2 100644 --- a/kernel/gsp_phys_gpu.h +++ b/kernel/gsp_phys_gpu.h @@ -90,6 +90,13 @@ bool gsp_phys_srv_ctxbuf_value(void); bool gsp_phys_srv_ctxcommit_value(void); /* + * gsp_phys_rmcall_channel: force one channel-alloc variant id on the GR + * ladder (0 = the promote-mode-selected default). 5 = V5 (client memdescs, + * no skip-ctxbuffer), 6 = V6 (SKIP_CTXBUFFER_ALLOC, KMD-managed ctx). + */ +int gsp_phys_rmcall_channel_value(void); + +/* * True when the gsp_phys_work_forward module parameter is set (default 0: the * guest's CE pushes stay software-emulated exactly as before). When set, the * sema service shadows each decoded CE push as one copy on the debugfs ladder's |