gpu: nvgpu: update device management framework to remove unusable engines

On certain platforms, not all copy engine instances are usable. The user
shouldn't submit any work to these engines. To enforce this, remove
these engines from active/host_engine list, this should ensure that these
engines do not get advertised to userspace. In order to accomplish this
introduce the following functions:
- nvgpu_engine_remove_one_dev: This function removes the specified device
  entry from following device lists: fifo->host_engines, fifo->active_engines,
  runlist->rl_dev_list, runlist->eng_bitmask.

Replace iteration over LCE device type entries using
nvgpu_device_for_each(g, dev, NVGPU_DEVTYPE_LCE), along with this introduce
macro nvgpu_device_for_each_safe.

Introduce gpu_dbg_ce flag for CE debugging.

Bug 3370462

Change-Id: I2e21f18363c6e53630d129da241c8fece106cd33
Signed-off-by: Antony Clince Alex <aalex@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2616711
Reviewed-by: Seema Khowala <seemaj@nvidia.com>
Reviewed-by: Vaibhav Kachore <vkachore@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: Gerrit_Virtual_Submit
This commit is contained in:
Antony Clince Alex
2021-10-26 10:09:20 +00:00
committed by mobile promotions
parent 79ee724b04
commit cce1d7ad84
7 changed files with 82 additions and 32 deletions

View File

@@ -37,6 +37,16 @@ int nvgpu_ce_init_support(struct gk20a *g)
g->ops.ce.set_pce2lce_mapping(g);
}
/*
* Bug 1895019
* Each time PCE2LCE config is updated and if it happens to
* map a LCE which was previously unmapped, then ELCG would have turned
* off the clock to the unmapped LCE and when the LCE config is updated,
* a race occurs between the config update and ELCG turning on the clock
* to that LCE, this might result in LCE dropping the config update.
* To avoid such a race, each time PCE2LCE config is updated toggle
* resets for all LCEs.
*/
err = nvgpu_mc_reset_devtype(g, NVGPU_DEVTYPE_LCE);
if (err != 0) {
nvgpu_err(g, "NVGPU_DEVTYPE_LCE reset failed");