On Wed, Mar 22, 2023 at 11:22:56PM +0200, Ville Syrjälä wrote: > On Wed, Mar 22, 2023 at 04:36:09PM -0400, Rodrigo Vivi wrote: > > On Wed, Mar 22, 2023 at 08:12:19PM +0200, Ville Syrjala wrote: > > > From: Ville Syrjälä <ville.syrjala@xxxxxxxxxxxxxxx> > > > > > > Currently we have no sane way to forcibly disable SAGV, which > > > makes debugging things a PITA. Manually poking at the pcode > > > mailbox with it's various SAGV/QGV/PSF formats is no fun, > > > and likely to be clobbered by the driver anyway. > > > > > > Let's add a modparam for this. > > > > > > Signed-off-by: Ville Syrjälä <ville.syrjala@xxxxxxxxxxxxxxx> > > > --- > > > drivers/gpu/drm/i915/display/skl_watermark.c | 4 ++++ > > > drivers/gpu/drm/i915/i915_params.c | 3 +++ > > > drivers/gpu/drm/i915/i915_params.h | 1 + > > > 3 files changed, 8 insertions(+) > > > > > > diff --git a/drivers/gpu/drm/i915/display/skl_watermark.c b/drivers/gpu/drm/i915/display/skl_watermark.c > > > index 50a9a6adbe32..ff70225c0263 100644 > > > --- a/drivers/gpu/drm/i915/display/skl_watermark.c > > > +++ b/drivers/gpu/drm/i915/display/skl_watermark.c > > > @@ -411,6 +411,9 @@ static bool intel_crtc_can_enable_sagv(const struct intel_crtc_state *crtc_state > > > struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); > > > struct drm_i915_private *i915 = to_i915(crtc->base.dev); > > > > > > + if (!i915->params.enable_sagv) > > > + return false; > > > > would this be one more ifdef for xe? > > Why? xe needs it just as much. yes, I know... but the param will be different... open like this we need to have ifdef, or we need to hide in a macro, or have it in a xe/ext/display/ duplicated file... it was a comment more to raise the thoughts around items like this that impacts xe... definitely not a blocker or anything like that... > > > should we hide this on a HAS_SAGV? > > If we have no sagv then we're not going to enable sagv anyway. > Doesn't matter if it's blocked a few lines earlier by the > modparam rather than falling through to checking the other > stuff. > > > > > anyway, let's move on: > > > > Reviewed-by: Rodrigo Vivi <rodrigo.vivi@xxxxxxxxx> > > Thanks > > > > > > > > + > > > if (DISPLAY_VER(i915) >= 12) > > > return tgl_crtc_can_enable_sagv(crtc_state); > > > else > > > @@ -3696,6 +3699,7 @@ static int intel_sagv_status_show(struct seq_file *m, void *unused) > > > }; > > > > > > seq_printf(m, "SAGV available: %s\n", str_yes_no(intel_has_sagv(i915))); > > > + seq_printf(m, "SAGV modparam: %s\n", str_enabled_disabled(i915->params.enable_sagv)); > > > seq_printf(m, "SAGV status: %s\n", sagv_status[i915->display.sagv.status]); > > > seq_printf(m, "SAGV block time: %d usec\n", i915->display.sagv.block_time_us); > > > > > > diff --git a/drivers/gpu/drm/i915/i915_params.c b/drivers/gpu/drm/i915/i915_params.c > > > index ade744cccfea..66141906ea05 100644 > > > --- a/drivers/gpu/drm/i915/i915_params.c > > > +++ b/drivers/gpu/drm/i915/i915_params.c > > > @@ -121,6 +121,9 @@ i915_param_named_unsafe(enable_psr2_sel_fetch, bool, 0400, > > > "(0=disabled, 1=enabled) " > > > "Default: 0"); > > > > > > +i915_param_named_unsafe(enable_sagv, bool, 0600, > > > + "Enable system agent voltage/frequency scaling (SAGV) (default: true)"); > > > + > > > i915_param_named_unsafe(force_probe, charp, 0400, > > > "Force probe options for specified supported devices. " > > > "See CONFIG_DRM_I915_FORCE_PROBE for details."); > > > diff --git a/drivers/gpu/drm/i915/i915_params.h b/drivers/gpu/drm/i915/i915_params.h > > > index 3f51f90145b6..6798b5c2363d 100644 > > > --- a/drivers/gpu/drm/i915/i915_params.h > > > +++ b/drivers/gpu/drm/i915/i915_params.h > > > @@ -56,6 +56,7 @@ struct drm_printer; > > > param(int, enable_psr, -1, 0600) \ > > > param(bool, psr_safest_params, false, 0400) \ > > > param(bool, enable_psr2_sel_fetch, true, 0400) \ > > > + param(bool, enable_sagv, true, 0600) \ > > > param(int, disable_power_well, -1, 0400) \ > > > param(int, enable_ips, 1, 0600) \ > > > param(int, invert_brightness, 0, 0600) \ > > > -- > > > 2.39.2 > > > > > -- > Ville Syrjälä > Intel
