Re: [PATCH 0/5] Fixes and additions to EDID generation

Re: [PATCH 0/5] Fixes and additions to EDID generation

[Jani Nikula]

On Tue, 13 Nov 2018, Ben Hutchings <ben.hutchings@codethink.co.uk> wrote:
> This series adds two more EDIDs that I needed for a project some time ago,
> and fixes some problems that I found along the way.

See also [1]. The patches there have been applied via the linux-doc
tree.

> There isn't a listed maintainer for Documentation/EDID but it looks
> like it falls under the DRM umbrella.  Please let me know if this the
> wrong place.

Cc: linux-doc, Jon and Dave.

The Documentation/EDID directory was merged by Dave along with the EDID
firmware loader mechanism in da0df92b5731 ("drm: allow loading an EDID
as firmware to override broken monitor").

I think it would be reasonable to handle the EDID stuff via dri-devel
and the drm trees, and modify MAINTAINERS to reflect that. Thoughts?

(And in the long run we should use or write a tool to generate EDIDs
instead of using assembly with macros. And like Jon already mentioned in
the other thread, even the assembly probably shouldn't be under
Documentation. But I digress.)


BR,
Jani.

[1] http://mid.mail-archive.com/1541407715-5417-1-git-send-email-cniedermaier@dh-electronics.de

>
> Ben.
>
> Ben Hutchings (5):
>   drm: EDID: Remove a mess involving the number 63
>   drm: EDID: Fix bit masking of {X,Y}{OFFSET,PULSE}
>   drm: EDID: Don't force make to be silent
>   drm: EDID: Add a 1280x720 (720p) EDID
>   drm: EDID: Add a 1280x768 ("WXGA") EDID
>
>  Documentation/EDID/1024x768.S  |  4 ++--
>  Documentation/EDID/1280x1024.S |  4 ++--
>  Documentation/EDID/1280x720.S  | 45 ++++++++++++++++++++++++++++++++++++++++++
>  Documentation/EDID/1280x768.S  | 45 ++++++++++++++++++++++++++++++++++++++++++
>  Documentation/EDID/1600x1200.S |  4 ++--
>  Documentation/EDID/1680x1050.S |  4 ++--
>  Documentation/EDID/1920x1080.S |  4 ++--
>  Documentation/EDID/800x600.S   |  4 ++--
>  Documentation/EDID/HOWTO.txt   |  4 ++--
>  Documentation/EDID/Makefile    | 12 +++++------
>  Documentation/EDID/edid.S      | 10 ++++------
>  11 files changed, 114 insertions(+), 26 deletions(-)
>  create mode 100644 Documentation/EDID/1280x720.S
>  create mode 100644 Documentation/EDID/1280x768.S

-- 
Jani Nikula, Intel Open Source Graphics Center

Re: [PATCH 0/5] Fixes and additions to EDID generation

[Daniel Vetter]

On Wed, Nov 14, 2018 at 12:49:11PM +0200, Jani Nikula wrote:
> On Tue, 13 Nov 2018, Ben Hutchings <ben.hutchings@codethink.co.uk> wrote:
> > This series adds two more EDIDs that I needed for a project some time ago,
> > and fixes some problems that I found along the way.
> 
> See also [1]. The patches there have been applied via the linux-doc
> tree.
> 
> > There isn't a listed maintainer for Documentation/EDID but it looks
> > like it falls under the DRM umbrella.  Please let me know if this the
> > wrong place.
> 
> Cc: linux-doc, Jon and Dave.
> 
> The Documentation/EDID directory was merged by Dave along with the EDID
> firmware loader mechanism in da0df92b5731 ("drm: allow loading an EDID
> as firmware to override broken monitor").
> 
> I think it would be reasonable to handle the EDID stuff via dri-devel
> and the drm trees, and modify MAINTAINERS to reflect that. Thoughts?
> 
> (And in the long run we should use or write a tool to generate EDIDs
> instead of using assembly with macros. And like Jon already mentioned in
> the other thread, even the assembly probably shouldn't be under
> Documentation. But I digress.)

If we do that, maybe stuff the EDID things into Documentation/gpu? Would
need an ack from media folks I think, but imo would make sense.
-Daniel

> 
> 
> BR,
> Jani.
> 
> [1] http://mid.mail-archive.com/1541407715-5417-1-git-send-email-cniedermaier@dh-electronics.de
> 
> >
> > Ben.
> >
> > Ben Hutchings (5):
> >   drm: EDID: Remove a mess involving the number 63
> >   drm: EDID: Fix bit masking of {X,Y}{OFFSET,PULSE}
> >   drm: EDID: Don't force make to be silent
> >   drm: EDID: Add a 1280x720 (720p) EDID
> >   drm: EDID: Add a 1280x768 ("WXGA") EDID
> >
> >  Documentation/EDID/1024x768.S  |  4 ++--
> >  Documentation/EDID/1280x1024.S |  4 ++--
> >  Documentation/EDID/1280x720.S  | 45 ++++++++++++++++++++++++++++++++++++++++++
> >  Documentation/EDID/1280x768.S  | 45 ++++++++++++++++++++++++++++++++++++++++++
> >  Documentation/EDID/1600x1200.S |  4 ++--
> >  Documentation/EDID/1680x1050.S |  4 ++--
> >  Documentation/EDID/1920x1080.S |  4 ++--
> >  Documentation/EDID/800x600.S   |  4 ++--
> >  Documentation/EDID/HOWTO.txt   |  4 ++--
> >  Documentation/EDID/Makefile    | 12 +++++------
> >  Documentation/EDID/edid.S      | 10 ++++------
> >  11 files changed, 114 insertions(+), 26 deletions(-)
> >  create mode 100644 Documentation/EDID/1280x720.S
> >  create mode 100644 Documentation/EDID/1280x768.S
> 
> -- 
> Jani Nikula, Intel Open Source Graphics Center

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch

Re: [PATCH 0/5] Fixes and additions to EDID generation

[Gerd Hoffmann]

  Hi,

> > (And in the long run we should use or write a tool to generate EDIDs
> > instead of using assembly with macros. And like Jon already mentioned in
> > the other thread, even the assembly probably shouldn't be under
> > Documentation. But I digress.)
> 
> If we do that, maybe stuff the EDID things into Documentation/gpu? Would
> need an ack from media folks I think, but imo would make sense.

Oh, btw:  I've written a edid generator for qemu for the virtual display
devices, and I've also added a small test tool so you don't have to boot
a virtual machine to test the thing.  Will ship with the upcoming qemu
3.1 release.  You might find this useful ...

cheers,
  Gerd

============================== cut here ==============================
# qemu-edid -h

This is a test tool for the qemu edid generator.

Typically you'll pipe the output into edid-decode
to check if the generator works correctly.

usage: qemu-edid <options>
options:
    -h             print this text
    -o <file>      set output file (stdout by default)
    -v <vendor>    set monitor vendor (three letters)
    -n <name>      set monitor name
    -s <serial>    set monitor serial
    -d <dpi>       set display resolution
    -x <prefx>     set preferred width
    -y <prefy>     set preferred height
    -X <maxx>      set maximum width
    -Y <maxy>      set maximum height

# qemu-edid | edid-decode 
Extracted contents:
header:          00 ff ff ff ff ff ff 00
serial number:   49 14 34 12 00 00 00 00 2a 18
version:         01 04
basic params:    a5 28 1e 78 06
chroma info:     ee 91 a3 54 4c 99 26 0f 50 54
established:     21 08 00
standard:        e1 c0 d1 c0 01 01 01 01 01 01 01 01 01 01 01 01
descriptor 1:    25 20 00 66 41 00 1a 30 00 1e 33 40 93 2e 11 00 00 18
descriptor 2:    00 00 00 fd 00 32 7d 1e a0 78 01 0a 20 20 20 20 20 20
descriptor 3:    00 00 00 fc 00 51 45 4d 55 20 4d 6f 6e 69 74 6f 72 0a
descriptor 4:    00 00 00 f7 00 0a 00 4a a2 24 29 20 00 00 00 00 00 00
extensions:      01
checksum:        19

Manufacturer: RHT Model 1234 Serial Number 0
Made week 42 of 2014
EDID version: 1.4
Digital display
8 bits per primary color channel
DisplayPort interface
Maximum image size: 40 cm x 30 cm
Gamma: 2.20
Supported color formats: RGB 4:4:4
Default (sRGB) color space is primary color space
First detailed timing is preferred timing
Established timings supported:
  640x480@60Hz
  800x600@60Hz
  1024x768@60Hz
Standard timings supported:
  2048x1152@60Hz
  1920x1080@60Hz
Detailed mode: Clock 82.290 MHz, 403 mm x 302 mm
               1024 1280 1310 1382 hborder 0
                768  771  774  794 vborder 0
               -hsync -vsync 
Monitor ranges (bare limits): 50-125Hz V, 30-160kHz H, max dotclock 1200MHz
Monitor name: QEMU Monitor
Established timings III
Has 1 extension blocks
Checksum: 0x19 (valid)

CEA extension block
Extension version: 3
6 bytes of CEA data
  Video data block
    VIC 125 Unknown mode 
    VIC 101 4096x2160@50Hz 
    VIC  96 3840x2160@50Hz 
    VIC  89 2560x1080@50Hz 
    VIC  31 1920x1080@50Hz 
0 native detailed modes
Checksum: 0xf2 (valid)