• State
  • Locked Locked
  • Replies 1 reply
  • Subscribers 98 subscribers
  • Views 230 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

EVM DM365 - CPLD wrong addressing bug in MV - fix

Juraj
Prodigy 245 points
Other Parts Discussed in Thread: TVP5146

Hi,

While working with the CPLD I found wrong behaviour after the function davinci_cfg_reg is called. After hours of experimenting I found the reason. It was wrong condition used in the arch/arm/mach-davinci/mux.c

/* Check the PinMux register in question */

if (cfg->mux_reg)

should be replaced with

if (cfg->mask)

This is fixed in GIT since March 2009 (!), but not in Montavista kernel.

See patch:

http://www.mail-archive.com/davinci-linux-open-source@linux.davincidsp.com/msg10667.html

 

Thank You

Juraj

  • 0
    Expert 2315 points

    I am busy sorting out an issue in the encodedecode application where it calls the function davinci_cfg_reg() in which this bug listed is found and would like to know if I should make the change like given by Juraj as my board was "working" before and also "after" I actually implemented it to see what it is doing?

    The probelem I have is that I get my interrupt set up and working on gpio0 during inserting module irqk but when I run encodedecode demo where the interrupt will be used, the evm on initialisation needs to change the state of gpio40 to prevent clashes between i2c of the tvp5146 and the imager interface and then it buggers up my gpio settings and my interrupt doesn't want to work during my encodedecode running.

    Help appreciated. Thanks, Jinh T.