Great! :) On 07/04/16 20:33, Nico Bouthoorn wrote: > I've pulled the latest master, in a fresh install. It is working! > > thanks, > > Nico > > Johan Ström wrote: >> Since this HW is out there in unknown number of deployments, it makes more >> sense to add a SW fix. I added a verify_mask which ignores the button >> pull-ups on read-back. >> >> Nico: Can you pull latest master and try that? >> If 'git show HEAD' shows your revert at the top, you must first do 'git >> reset --hard HEAD^' (assuming you have no other uncommited changes). Then >> you can do 'git pull'. >> >> Johan >> >> On 03/04/16 20:30, Johan Ström wrote: >>> Hello again, >>> >>> I've gotten some helpful off-list response from Der Tiger, and to sum it up. >>> >>> The hardware Tiger uses (Hobbyboard) is electrically identical to the on >>> Nico uses (https://fstab.nl/en/lcd_owfs), except that the Hobbyboard >>> schematic has 3 pushbuttons between P0..P3 and GND, with 10nF cap's in >>> parallel, and 3 10k pull-ups between VCC and P0..P3. The schematic Nico >>> uses leaves P0..P3 floating. >>> >>> One theory put forward by Der Tiger is that, since the patch disables the >>> PIO output transistors for P0..P3 and thus leaving the outputs in an >>> undefined state, the chip/owfs could be confused. >>> >>> Also, with the help of Jan's explanation on the command sequence, I cannot >>> really see any apparent issues with the patch. >>> >>> After studying the logs (which says ' Error sending initial attention'), >>> and the corresponding code, I'm quite sure the lacking pull-ups are the >>> source of the problem. >>> OW_Hinit->OW_w_pios writes a byte and then read backs the PIO status, >>> expecting them to be the same. If no pull-ups are present, it will i.e >>> write 0x37, but read back 0x30, and fail. >>> This is the case in the log (see below, [1]). >>> >>> >>> Nico: Can you try to add some 10k resistors between P0-VCC, P1-VCC and >>> P3-VCC? Remember to test with the original code (without the extra revert). >>> Hopefully that will help! >>> Thanks to Tiger and Jan for helping to (hopefully) fix this issue :) >>> >>> Johan >>> >>> >>> >>> [1] >>> TRAFFIC IN <NETREAD> bus=0 (/dev/ttyUSB0) >>> Byte buffer DS9097U, length=1 >>> --000: 37 (Here we write 0x37) >>> <7> >>> TRAFFIC IN <NETREAD> bus=0 (/dev/ttyUSB0) >>> Byte buffer DS9097U, length=1 >>> --000: C8 >>> <.> >>> TRAFFIC IN <NETREAD> bus=0 (/dev/ttyUSB0) >>> Byte buffer DS9097U, length=1 >>> --000: AA >>> <.> >>> TRAFFIC IN <NETREAD> bus=0 (/dev/ttyUSB0) >>> Byte buffer DS9097U, length=1 >>> --000: 30 (Here we read back 0x30, bit 0-3 is not pulled up.) >>> <0> >>> DEBUG: ow_tcp_read.c:(113) read: 4 - 0 = 4 >>> DEBUG: ow_transaction.c:(145) modify = 0 >>> DEBUG: ow_transaction.c:(208) end = 0 >>> DEBUG: ow_2408.c:(442) Error sending initial attention >>> DEBUG: ow_2408.c:(406) Screen initialization error >>> >>> >> >> >> ------------------------------------------------------------------------------ >> >> >> >> _______________________________________________ >> Owfs-developers mailing list >> Owfs-developers@lists.sourceforge.net >> https://lists.sourceforge.net/lists/listinfo/owfs-developers >>
------------------------------------------------------------------------------ _______________________________________________ Owfs-developers mailing list Owfs-developers@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/owfs-developers