mbox series

[0/8] drm/i915: dbuf cleanups

Message ID 20210416171011.19012-1-ville.syrjala@linux.intel.com (mailing list archive)
Headers show
Series drm/i915: dbuf cleanups | expand

Message

Ville Syrjälä April 16, 2021, 5:10 p.m. UTC
From: Ville Syrjälä <ville.syrjala@linux.intel.com>

A bunch of drive-by-cleanup While I was reading through
the dbuf code.

Ville Syrjälä (8):
  drm/i915: Collect dbuf device info into a sub-struct
  drm/i915: Handle dbuf bypass path allocation earlier
  drm/i915: Store dbuf slice mask in device info
  drm/i915: Use intel_dbuf_slice_size()
  drm/i915: Use intel_de_rmw() for DBUF_POWER_REQUEST
  drm/i915: Polish for_each_dbuf_slice()
  drm/i915: Add enabledisable()
  drm/i915: Say "enable foo" instead of "set foo to enabled"

 drivers/gpu/drm/i915/display/intel_bw.c       | 11 ++---
 drivers/gpu/drm/i915/display/intel_ddi.c      |  4 +-
 drivers/gpu/drm/i915/display/intel_display.h  |  9 ++--
 .../drm/i915/display/intel_display_power.c    | 24 ++++------
 drivers/gpu/drm/i915/display/intel_dp.c       | 14 +++---
 .../drm/i915/display/intel_dp_aux_backlight.c |  2 +-
 drivers/gpu/drm/i915/display/intel_tc.c       |  4 +-
 drivers/gpu/drm/i915/i915_pci.c               | 16 +++----
 drivers/gpu/drm/i915/i915_utils.h             |  5 +++
 drivers/gpu/drm/i915/intel_device_info.h      |  6 ++-
 drivers/gpu/drm/i915/intel_pm.c               | 45 +++++++------------
 drivers/gpu/drm/i915/intel_pm.h               |  1 +
 12 files changed, 65 insertions(+), 76 deletions(-)

Comments

Jani Nikula April 20, 2021, 12:57 p.m. UTC | #1
On Fri, 16 Apr 2021, Ville Syrjala <ville.syrjala@linux.intel.com> wrote:
> From: Ville Syrjälä <ville.syrjala@linux.intel.com>
>
> A bunch of drive-by-cleanup While I was reading through
> the dbuf code.

Good stuff. On the series,

Reviewed-by: Jani Nikula <jani.nikula@intel.com>

Some thoughts provoked by the series:

* Time to start splitting up dbuf/ddb/wm stuff from intel_pm.[ch] to
  separate file(s) under display/?

* Time to start moving more display related struct intel_device_info
  fields under the display substruct?

BR,
Jani.


>
> Ville Syrjälä (8):
>   drm/i915: Collect dbuf device info into a sub-struct
>   drm/i915: Handle dbuf bypass path allocation earlier
>   drm/i915: Store dbuf slice mask in device info
>   drm/i915: Use intel_dbuf_slice_size()
>   drm/i915: Use intel_de_rmw() for DBUF_POWER_REQUEST
>   drm/i915: Polish for_each_dbuf_slice()
>   drm/i915: Add enabledisable()
>   drm/i915: Say "enable foo" instead of "set foo to enabled"
>
>  drivers/gpu/drm/i915/display/intel_bw.c       | 11 ++---
>  drivers/gpu/drm/i915/display/intel_ddi.c      |  4 +-
>  drivers/gpu/drm/i915/display/intel_display.h  |  9 ++--
>  .../drm/i915/display/intel_display_power.c    | 24 ++++------
>  drivers/gpu/drm/i915/display/intel_dp.c       | 14 +++---
>  .../drm/i915/display/intel_dp_aux_backlight.c |  2 +-
>  drivers/gpu/drm/i915/display/intel_tc.c       |  4 +-
>  drivers/gpu/drm/i915/i915_pci.c               | 16 +++----
>  drivers/gpu/drm/i915/i915_utils.h             |  5 +++
>  drivers/gpu/drm/i915/intel_device_info.h      |  6 ++-
>  drivers/gpu/drm/i915/intel_pm.c               | 45 +++++++------------
>  drivers/gpu/drm/i915/intel_pm.h               |  1 +
>  12 files changed, 65 insertions(+), 76 deletions(-)
Ville Syrjälä April 20, 2021, 1:22 p.m. UTC | #2
On Tue, Apr 20, 2021 at 03:57:45PM +0300, Jani Nikula wrote:
> On Fri, 16 Apr 2021, Ville Syrjala <ville.syrjala@linux.intel.com> wrote:
> > From: Ville Syrjälä <ville.syrjala@linux.intel.com>
> >
> > A bunch of drive-by-cleanup While I was reading through
> > the dbuf code.
> 
> Good stuff. On the series,
> 
> Reviewed-by: Jani Nikula <jani.nikula@intel.com>
> 
> Some thoughts provoked by the series:
> 
> * Time to start splitting up dbuf/ddb/wm stuff from intel_pm.[ch] to
>   separate file(s) under display/?

I guess we could start shoveling. I've been putting it off due to
having bunch of in-flight stuff (wm latency cleanups, s/intel/ilk/
rename, and some wip skl+ dbuf patches come to mind). But I suppose
those aren't going to make any progress without a reposting anyway.
I could always respin them after the code has settled in its new
home.

> 
> * Time to start moving more display related struct intel_device_info
>   fields under the display substruct?

Perhaps. I was a bit torn between .dbuf and .display.dbuf for this
series. In the end I decided that .display.dbuf is a bit too long
for my liking. Not sure if s/.display/.de/ might make it more
palatable...