[PATCH] fix-glamofb-flicker-72hz-refresh.patch

Andy Green andy at openmoko.com
Mon Mar 10 16:34:12 CET 2008


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Somebody in the thread at some point said:

> Can we create a hmm... "inverse patch" or sth like that for GTA01 - just for 
> testing purposes, to provoke flicker there and make sure it's _absolutely_ 
> nothing in hw of GTA02, but really is nothing but a bug in LCM?
> Just an idea...
> And maybe manufacturer of LCM can tell what's the very nature of this bug, so 
> this became more of a real fix for a well understood problem, rather than a 
> workaround which seems to help for now but the bug may resurrect when nobody 
> expects.

It is a good idea... if someone else wants to take it up with TPO (I
don't have a GTA01).  I don't think it buys us much struggling to
understand the structure of an undocumented Toshiba ASIC (no longer
mentioned on their site) on an OEM panel when there are so many other
issues to take care of, eventually we will use different panels, etc.
Working around it is probably the correct "solution".  (If the mystery
is in our own stuff though, then it is worth more time to try to own it
properly.)

The issue of landscape refresh rate needs re-testing with the patches
too, it may also act differently with flicker with the blanking stuff
removed.  If it is still flickering badly we also need to either can
hardware landscape for now or mount an expedition into Glamo-land to see
what can be done about the artifacts at 24.5MHz PCLK in landscape.

> Also, you said the PCLK signal is very corrupted due to the R-C. Even while 
> apparently not causing actual trouble, this should be given a closer look, to 
> feel good regarding worst case scenarios. Someone has to have an explanation 
> _why_ this C is there and should be this value. Otherwise just replace with a 
> better value for MP?

I guess the idea is emissions LPF... I attach what it does to this
24.5Mhz clock.  This is on a 200MHz bandwidth 'scope, that's really what
the "square wave" actually looks like at the LCM connector :-)

If there is no objection maybe we could try 12pF or somesuch there
instead of 47pF.  But at this time near production I can imagine the
urge is to let all sleeping dogs lie.

- -Andy
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iD8DBQFH1VT0OjLpvpq7dMoRAhVqAJ9cYCeGIw3TOvY7FMHhOrD3yGYmBgCeNIxg
jqOKL9HdpZIEppGnDVaNdGc=
=qxI8
-----END PGP SIGNATURE-----
-------------- next part --------------
A non-text attachment was scrubbed...
Name: f0016tek.jpg
Type: image/jpeg
Size: 105287 bytes
Desc: not available
Url : http://lists.openmoko.org/pipermail/openmoko-kernel/attachments/20080310/f8f9f54c/attachment.jpg 


More information about the openmoko-kernel mailing list