Re: [PATCH] drm/amd/display: do not allocate display_mode_lib unnecessarily

[Dor Askayo]

On Fri, Jan 17, 2020 at 12:59 PM Dor Askayo <dor.askayo@xxxxxxxxx> wrote:
>
> On Sat, Jan 4, 2020 at 2:23 PM Dor Askayo <dor.askayo@xxxxxxxxx> wrote:
> >
> > This allocation isn't required and can fail when resuming from suspend.
> >
> > Bug: https://gitlab.freedesktop.org/drm/amd/issues/1009
> > Signed-off-by: Dor Askayo <dor.askayo@xxxxxxxxx>
> > ---
> >  drivers/gpu/drm/amd/display/dc/core/dc.c | 17 +++++++++--------
> >  1 file changed, 9 insertions(+), 8 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
> > index dd4731ab935c..83ebb716166b 100644
> > --- a/drivers/gpu/drm/amd/display/dc/core/dc.c
> > +++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
> > @@ -2179,12 +2179,7 @@ void dc_set_power_state(
> >         enum dc_acpi_cm_power_state power_state)
> >  {
> >         struct kref refcount;
> > -       struct display_mode_lib *dml = kzalloc(sizeof(struct display_mode_lib),
> > -                                               GFP_KERNEL);
> > -
> > -       ASSERT(dml);
> > -       if (!dml)
> > -               return;
> > +       struct display_mode_lib *dml;
> >
> >         switch (power_state) {
> >         case DC_ACPI_CM_POWER_STATE_D0:
> > @@ -2206,6 +2201,12 @@ void dc_set_power_state(
> >                  * clean state, and dc hw programming optimizations will not
> >                  * cause any trouble.
> >                  */
> > +               dml = kzalloc(sizeof(struct display_mode_lib),
> > +                               GFP_KERNEL);
> > +
> > +               ASSERT(dml);
> > +               if (!dml)
> > +                       return;
> >
> >                 /* Preserve refcount */
> >                 refcount = dc->current_state->refcount;
> > @@ -2219,10 +2220,10 @@ void dc_set_power_state(
> >                 dc->current_state->refcount = refcount;
> >                 dc->current_state->bw_ctx.dml = *dml;
> >
> > +               kfree(dml);
> > +
> >                 break;
> >         }
> > -
> > -       kfree(dml);
> >  }
> >
> >  void dc_resume(struct dc *dc)
> > --
> > 2.24.1
> >
>
> I've been running with this fix applied on top of Fedora's
> 5.3.16-300.fc31.x86_64 kernel for the past two weeks, suspending
> and resuming often. This the first time since I bought my RX 580 8GB
> more than a year ago that I can suspend and resume reliably.
>
> I'd appreciate a quick review for the above, it really is a trivial change.
>
> Thanks,
> Dor

Bumping this up again. I've been running with this change for the past
20 days without issues.

Thanks,
Dor

On Fri, Jan 17, 2020 at 12:59 PM Dor Askayo <dor.askayo@xxxxxxxxx> wrote:
>
> On Sat, Jan 4, 2020 at 2:23 PM Dor Askayo <dor.askayo@xxxxxxxxx> wrote:
> >
> > This allocation isn't required and can fail when resuming from suspend.
> >
> > Bug: https://gitlab.freedesktop.org/drm/amd/issues/1009
> > Signed-off-by: Dor Askayo <dor.askayo@xxxxxxxxx>
> > ---
> >  drivers/gpu/drm/amd/display/dc/core/dc.c | 17 +++++++++--------
> >  1 file changed, 9 insertions(+), 8 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
> > index dd4731ab935c..83ebb716166b 100644
> > --- a/drivers/gpu/drm/amd/display/dc/core/dc.c
> > +++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
> > @@ -2179,12 +2179,7 @@ void dc_set_power_state(
> >         enum dc_acpi_cm_power_state power_state)
> >  {
> >         struct kref refcount;
> > -       struct display_mode_lib *dml = kzalloc(sizeof(struct display_mode_lib),
> > -                                               GFP_KERNEL);
> > -
> > -       ASSERT(dml);
> > -       if (!dml)
> > -               return;
> > +       struct display_mode_lib *dml;
> >
> >         switch (power_state) {
> >         case DC_ACPI_CM_POWER_STATE_D0:
> > @@ -2206,6 +2201,12 @@ void dc_set_power_state(
> >                  * clean state, and dc hw programming optimizations will not
> >                  * cause any trouble.
> >                  */
> > +               dml = kzalloc(sizeof(struct display_mode_lib),
> > +                               GFP_KERNEL);
> > +
> > +               ASSERT(dml);
> > +               if (!dml)
> > +                       return;
> >
> >                 /* Preserve refcount */
> >                 refcount = dc->current_state->refcount;
> > @@ -2219,10 +2220,10 @@ void dc_set_power_state(
> >                 dc->current_state->refcount = refcount;
> >                 dc->current_state->bw_ctx.dml = *dml;
> >
> > +               kfree(dml);
> > +
> >                 break;
> >         }
> > -
> > -       kfree(dml);
> >  }
> >
> >  void dc_resume(struct dc *dc)
> > --
> > 2.24.1
> >
>
> I've been running with this fix applied on top of Fedora's
> 5.3.16-300.fc31.x86_64 kernel for
> the past two weeks, suspending and resuming often. This the first time
> since I bought my
> RX 580 8GB more than a year ago that I can suspend and resume reliably.
>
> I'd appreciate a quick review for the above, it really is a trivial change.
>
> Thanks,
> Dor