nv-tegra-mirror/linux-nvgpu
1
0
Fork 0
mirror of git://nv-tegra.nvidia.com/linux-nvgpu.git synced 2025-12-23 01:50:07 +03:00
Code Issues Packages Projects Releases Wiki Activity

gpu: nvgpu: Add nvgpu_early_poweron() support

Browse Source 
1) NvGpu dev node needs to be created in gpu power on
early stage to avoid latency introduced by udevd.
For creating dev node, device and grmgr init
needs to move to early stage of GPU power on.
After grmgr init, NvGpu can identify the number of MIG
instance required for each physical GPU.
For that, added a new API nvgpu_early_poweron() to handle
early init which is required for before dev node creation.

2) Removed fifo dependency in nvgpu_init_gr_manager()

3) Used get_max_subctx_count() directly to query
the veid/subctx count.

JIRA NVGPU-6633

Change-Id: Ib9d7c3e184c71237b0da9305515ccd8ceda1d5ad
Signed-off-by: Lakshmanan M <lm@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2517173
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
This commit is contained in:
Lakshmanan M
2021-04-19 22:56:18 +05:30
committed by mobile promotions
parent 3648fe4655
commit 3f8c562004
4 changed files with 63 additions and 15 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
drivers/gpu/nvgpu/os/linux/module.c
View File

@@ -41,7 +41,6 @@
#include <soc/tegra/fuse.h>
#endif /* CONFIG_NVGPU_TEGRA_FUSE */
#include <nvgpu/hal_init.h>
#include <nvgpu/dma.h>
#include <nvgpu/kmem.h>
#include <nvgpu/nvgpu_common.h>
@@ -434,14 +433,12 @@ int gk20a_pm_finalize_poweron(struct device *dev)
nvgpu_restore_usermode_for_poweron(g);
err = nvgpu_detect_chip(g);
if (err)
err = nvgpu_early_poweron(g);
if (err != 0) {
nvgpu_err(g, "nvgpu_early_poweron failed[%d]", err);
goto done;
}
/**
* TODO: Need to add nvgpu_early_poweron() sequence before
* creating device nodes.
*/
if (!l->dev_nodes_created) {
err = gk20a_user_init(dev);
if (err) {