gpu: nvgpu: Add ELPG_MS protected call for L2 flush

- if L2 flush is done when ELPG_MS feature is engaged
  then it can cause some of the signals to go non-idle.
  This can cause idle snap in ELPG_MS.
- To avoid the idle snap, add elpg_ms protected call before
  L2 flush operation

Bug 200763448

Change-Id: I651875bc051c3b7d26d2bb0b593083512a5765b2
Signed-off-by: Divya <dsinghatwari@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2599459
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
Reviewed-by: Mahantesh Kumbar <mkumbar@nvidia.com>
Reviewed-by: Vijayakumar Subbu <vsubbu@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:
Divya
2021-08-14 07:58:39 +00:00
committed by mobile promotions
parent 727a2573dc
commit d538737ba1
11 changed files with 39 additions and 17 deletions

View File

@@ -1952,7 +1952,7 @@ static int nvgpu_dbg_gpu_ioctl_get_gr_context(struct dbg_session_gk20a *dbg_s,
/* Channel gr_ctx buffer is gpu cacheable.
Flush and invalidate before cpu update. */
err = g->ops.mm.cache.l2_flush(g, true);
err = nvgpu_pg_elpg_ms_protected_call(g, g->ops.mm.cache.l2_flush(g, true));
if (err != 0) {
nvgpu_err(g, "l2_flush failed");
goto done;