On Mon, Dec 09, 2024 at 09:54:54PM +0530, Arun R Murthy wrote: > Display histogram is a hardware functionality where a statistics for 'x' > number of frames is generated to form a histogram data. This is notified > to the user via histogram event. Compositor will then upon sensing the > histogram event will read the histogram data from KMD via crtc property. > A library can be developed to take this generated histogram as an > input and apply some algorithm to generate an Image EnhancemenT(IET). > This is further fed back to the KMD via crtc property. KMD will use this > IET as a multiplicand factor to multiply with the incoming pixels at the > end of the pipe which is then pushed onto the display. > > One such library Global Histogram Enhancement(GHE) will take the histogram > as input and applied the algorithm to enhance the density and then > return the enhanced factor. This library can be located @ > https://github.com/intel/ghe > > The corresponding mutter changes to enable/disable histogram, read the > histogram data, communicate with the library and write the enhanced data > back to the KMD is also pushed for review at https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3873 > and https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3873/diffs?commit_id=270808ca7c8be48513553d95b4a47541f5d40206 > The IGT changes for validating the histogram event and reading the > histogram is also pushed for review at https://patchwork.freedesktop.org/series/135789/ I think other people have already asked this on previous postings of these patches, but please don't try to manually hack the version numbers within a series. What you just posted has "PATCHv10" on the cover letter, "PATCHv2" on one patch, "PATCHv3" on three patches, and the rest are unversioned "PATCH." The general expectation these days is that versioning in the subject applies to the series as a whole, not the individual patches, so this causes a lot of confusion. Posting like you did here also wrecks havoc on a lot of the tools people use to manage and compare series like the "b4" tool. When generating and sending a new series, you should just do something like "git format-patch -v10 ..." so that the proper "v10" numbering is automatically applied to all the patches and we don't wind up with this strange jumble. Matt > > Test-with: 20240705091333.328322-1-mohammed.thasleem@xxxxxxxxx > > Arun R Murthy (10): > drm/crtc: Add histogram properties > drm/crtc: Expose API to create drm crtc property for histogram > drm/i915/histogram: Define registers for histogram > drm/i915/histogram: Add support for histogram > drm/xe: Add histogram support to Xe builds > drm/i915/histogram: histogram interrupt handling > drm/i915/display: handle drm-crtc histogram property updates > drm/i915/histogram: histogram delay counter doesnt reset > drm/i915/histogram: Histogram changes for Display 20+ > drm/i915/histogram: Enable pipe dithering > > drivers/gpu/drm/drm_atomic_state_helper.c | 6 + > drivers/gpu/drm/drm_atomic_uapi.c | 17 + > drivers/gpu/drm/drm_crtc.c | 30 ++ > drivers/gpu/drm/i915/Makefile | 1 + > drivers/gpu/drm/i915/display/intel_atomic.c | 1 + > drivers/gpu/drm/i915/display/intel_crtc.c | 7 + > drivers/gpu/drm/i915/display/intel_display.c | 17 + > .../gpu/drm/i915/display/intel_display_irq.c | 6 +- > .../drm/i915/display/intel_display_types.h | 4 + > .../gpu/drm/i915/display/intel_histogram.c | 380 ++++++++++++++++++ > .../gpu/drm/i915/display/intel_histogram.h | 40 ++ > .../drm/i915/display/intel_histogram_regs.h | 75 ++++ > drivers/gpu/drm/i915/i915_reg.h | 5 +- > drivers/gpu/drm/xe/Makefile | 1 + > include/drm/drm_crtc.h | 49 +++ > include/uapi/drm/drm_mode.h | 11 + > 16 files changed, 647 insertions(+), 3 deletions(-) > create mode 100644 drivers/gpu/drm/i915/display/intel_histogram.c > create mode 100644 drivers/gpu/drm/i915/display/intel_histogram.h > create mode 100644 drivers/gpu/drm/i915/display/intel_histogram_regs.h > > -- > 2.25.1 > -- Matt Roper Graphics Software Engineer Linux GPU Platform Enablement Intel Corporation
