nv-tegra-mirror/linux-nvgpu
1
0
Fork 0
mirror of git://nv-tegra.nvidia.com/linux-nvgpu.git synced 2025-12-22 17:36:20 +03:00
Code Issues Packages Projects Releases Wiki Activity

gpu: nvgpu: gr: ignore second zcull request to ctx

Browse Source 
All channels in TSG will share same zcull context. Any
attempt to add a second zcull buffer will be ignored.

Bug 3364302

Change-Id: I04e18dfe8e5fac4ca131c3b625755aa90a23180d
Signed-off-by: Seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2616677
Reviewed-by: svcacv <svcacv@nvidia.com>
Reviewed-by: Seema Khowala <seemaj@nvidia.com>
Reviewed-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@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:
Seshendra Gadagottu
2021-10-26 02:42:23 -07:00
committed by mobile promotions
parent 1c053a75af
commit 03b1a81ab1
1 changed files with 18 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

19
drivers/gpu/nvgpu/common/gr/gr_setup.c
View File

@@ -1,5 +1,5 @@
/*
* Copyright (c) 2019-2021, NVIDIA CORPORATION. All rights reserved.
* Copyright (c) 2019-2022, NVIDIA CORPORATION. All rights reserved.
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
@@ -94,6 +94,23 @@ int nvgpu_gr_setup_bind_ctxsw_zcull(struct gk20a *g, struct nvgpu_channel *c,
}
gr_ctx = tsg->gr_ctx;
/*
* Each TSG shares same context with all the channels in the tsg
* and zcull cannot be set per channel. If any channel tries
* to add a second zcull buffer, it will be ignored.
* See Bug 3364302.
* TODO - https://jirasw.nvidia.com/browse/NVGPU-451
* When full subcontext(multiple VA) is supported by TSG
* then each channel can have separate VA address for same
* physical zcull buffer but then zcull va ptr cannot be stored
* at gr_ctx level and current design needs to be re-worked.
*/
if (nvgpu_gr_ctx_get_zcull_ctx_va(gr_ctx) != 0ULL) {
nvgpu_log(g, gpu_dbg_info,
"zcull bind is ignored for already bound ctx");
return 0;
}
nvgpu_gr_ctx_set_zcull_ctx(g, gr_ctx, mode, zcull_va);
return nvgpu_gr_setup_zcull(g, c, gr_ctx);