On 2/12/20 4:49 PM, Brian Welty wrote:
On 2/12/2020 4:34 PM, Chris Wilson wrote:
Quoting Brian Welty (2020-02-13 00:14:18)
For DGFX devices, the MOCS control value is not initialized or used.
Then why is the table populated?
-Chris
The format has changed (been reduced?) for DGFX. drm_i915_mocs_entry.l3cc_value is what is still initialized/used.
Probably first needed is the patch that defines the table entries for DGFX.
Ugh, I didn't notice this wasn't applied yet. Let me ask about this.
We do have:
commit e6e2ac07118b15f25683fcbd59ea1be73ec9465d
Author: Lucas De Marchi <lucas.demarchi@xxxxxxxxx>
Date: Thu Oct 24 12:51:21 2019 -0700
drm/i915: do not set MOCS control values on dgfx
So I see no reason not to add this change to the test side to match
that. Maybe we can add an additional check in the test to validate that
all the control_entries are set to 0 in the table on DGFX?
Daniele
-Brian
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/intel-gfx