On Mon, 30 Jan 2012 09:38:52 +0100 "Feldschmid, Ingo" ifel@msc-ge.com wrote:
Hi,
thanks for all the input. I have created a new patch file, which addresses the problems mentioned in the last few replies:
- print.c: added entry in boards_known array- board_enable.c: phase 2
function name changed to "p2_not_a_laptop"
- board_enable.c: new entry in board_matches array sorted by vendor name
- board_enable.c: board detection mechanism changed to use first pci
identifier for tunnel creek and second pci identifier for EG20T and dmi information for vendor name
The detection mechanism should now work for all boards with Tunnel Creek CPU with EG20T southbridge from vendor "MSC Vertriebs GmbH". I tried to narrow it down with the CPU/chipset/DMI combination, since we don't have any company-specific pci devices on this board.
Signed-off-by: Ingo Feldschmid ifel@msc-ge.com
thanks for the rework!
Acked-by: Stefan Tauner stefan.tauner@student.tuwien.ac.at
i have changed a few things additionally: most noteworthy i have changed the board enable to return 0 instead of 1. this is not checked atm, but if it ever will, then 0 will be the right choice. I have also added a debug print so that "DMI chassis-type is not specific enough." is followed by "Laptop detection overridden by P2 board enable." in verbose output.
committed in r1487.