gpu: nvgpu: add/remove l2 cache flush when updating the ctx buffers

gr ctx buffer in non-cacheable hence there is no need to do L2 cache
flush when updating the buffer. Remove the flushes.

pm ctx buffer is cacheable hence add l2 flush in the function
nvgpu_profiler_quiesce_hwpm_streamout_non_resident since it
updates the buffer.

Bug 3677982

Change-Id: I0c15ec7a7f8fa250af1d25891122acc24443a872
Signed-off-by: Sagar Kamble <skamble@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2713916
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
This commit is contained in:
Sagar Kamble
2022-05-17 18:28:46 +05:30
committed by mobile promotions
parent 65e7baf856
commit bfa20f62c6
5 changed files with 12 additions and 59 deletions

View File

@@ -84,12 +84,6 @@ void nvgpu_gr_subctx_load_ctx_header(struct gk20a *g,
struct nvgpu_gr_ctx *gr_ctx, u64 gpu_va)
{
struct nvgpu_mem *ctxheader = &subctx->ctx_header;
int err = 0;
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");
}
#ifdef CONFIG_NVGPU_SET_FALCON_ACCESS_MAP
/* set priv access map */