| Message ID | 20190911014801.26821-1-kenneth@whitecape.org (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | drm/i915: Whitelist COMMON_SLICE_CHICKEN2 | expand |
Quoting Kenneth Graunke (2019-09-11 02:48:01) > This allows userspace to use "legacy" mode for push constants, where > they are committed at 3DPRIMITIVE or flush time, rather than being > committed at 3DSTATE_BINDING_TABLE_POINTERS_XS time. Gen6-8 and Gen11 > both use the "legacy" behavior - only Gen9 works in the "new" way. > > Conflating push constants with binding tables is painful for userspace, > we would like to be able to avoid doing so. > > Signed-off-by: Kenneth Graunke <kenneth@whitecape.org> > Cc: stable@vger.kernel.org Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> -Chris
Quoting Chris Wilson (2019-09-11 08:42:22) > Quoting Kenneth Graunke (2019-09-11 02:48:01) > > This allows userspace to use "legacy" mode for push constants, where > > they are committed at 3DPRIMITIVE or flush time, rather than being > > committed at 3DSTATE_BINDING_TABLE_POINTERS_XS time. Gen6-8 and Gen11 > > both use the "legacy" behavior - only Gen9 works in the "new" way. > > > > Conflating push constants with binding tables is painful for userspace, > > we would like to be able to avoid doing so. > > > > Signed-off-by: Kenneth Graunke <kenneth@whitecape.org> > > Cc: stable@vger.kernel.org > Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Pushed. Do you also want to do this for icl? -Chris
On Wednesday, September 11, 2019 1:00:51 AM PDT Chris Wilson wrote: > Quoting Chris Wilson (2019-09-11 08:42:22) > > Quoting Kenneth Graunke (2019-09-11 02:48:01) > > > This allows userspace to use "legacy" mode for push constants, where > > > they are committed at 3DPRIMITIVE or flush time, rather than being > > > committed at 3DSTATE_BINDING_TABLE_POINTERS_XS time. Gen6-8 and Gen11 > > > both use the "legacy" behavior - only Gen9 works in the "new" way. > > > > > > Conflating push constants with binding tables is painful for userspace, > > > we would like to be able to avoid doing so. > > > > > > Signed-off-by: Kenneth Graunke <kenneth@whitecape.org> > > > Cc: stable@vger.kernel.org > > Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> > > Pushed. Do you also want to do this for icl? > -Chris Thanks! I don't think it's necessary for ICL, the bit seems to be gone and it appears to perform the legacy behavior all the time. --Ken
diff --git a/drivers/gpu/drm/i915/gt/intel_workarounds.c b/drivers/gpu/drm/i915/gt/intel_workarounds.c index 243d3f77be13..41d0f786e06d 100644 --- a/drivers/gpu/drm/i915/gt/intel_workarounds.c +++ b/drivers/gpu/drm/i915/gt/intel_workarounds.c @@ -1062,6 +1062,9 @@ static void gen9_whitelist_build(struct i915_wa_list *w) /* WaAllowUMDToModifyHDCChicken1:skl,bxt,kbl,glk,cfl */ whitelist_reg(w, GEN8_HDC_CHICKEN1); + + /* WaSendPushConstantsFromMMIO:skl,bxt */ + whitelist_reg(w, COMMON_SLICE_CHICKEN2); } static void skl_whitelist_build(struct intel_engine_cs *engine)
This allows userspace to use "legacy" mode for push constants, where they are committed at 3DPRIMITIVE or flush time, rather than being committed at 3DSTATE_BINDING_TABLE_POINTERS_XS time. Gen6-8 and Gen11 both use the "legacy" behavior - only Gen9 works in the "new" way. Conflating push constants with binding tables is painful for userspace, we would like to be able to avoid doing so. Signed-off-by: Kenneth Graunke <kenneth@whitecape.org> Cc: stable@vger.kernel.org --- drivers/gpu/drm/i915/gt/intel_workarounds.c | 3 +++ 1 file changed, 3 insertions(+)