Paven, Which release code are you referring to? The ccdc_config_raw() function has these undesirable code that was removed in LSP210. You could remove them as well. The VINT0 and VINT1 are set as desire in ccdc_setwin(). They are software programmable interrupt lines.
Murali Karicheri Software Design Engineer Texas Instruments Inc. Germantown, MD 20874 Phone : 301-515-3736 email: [email protected] ________________________________ From: [email protected] [mailto:[email protected]] On Behalf Of Pavel Han Sent: Tuesday, June 23, 2009 9:57 PM To: [email protected] Subject: When The VINT0 and VINT1 occurs on DM355? Hello EveryBody! Now I encountered a very strange problem on DM355 EVM.I use DM355 to capture 2592x1944 resolution picture from the MT9P031 Sensor,but sometimes I can only get a disrupt image,and In the capture thread,I find the DQBUF time only need 1ms in the most time(If I do not use IPIPE and only do the DQBUF/QBUF pairs,the DQBUF time will be 102ms,I think this time is correct).In addtion, from the ccdc_dm355 driver, I get the following code: val = params->win.height / 64; mdelay(5); regw(val, VDINT0); mdelay(5); regw(0, VDINT1); It is very stange!I think the VDINT0 interrupt should be configured to occured on the completion time to make the current video-buffer to be STATE_DONE,and the VDINT1 interrupt should be configured to occured on the current frame time to make the next frame to be STATE_ACTIVE.But the register setup is not the case,so I want to konw what is the real means of the VDINT register configure code section. Thank you Very much!
_______________________________________________ Davinci-linux-open-source mailing list [email protected] http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source
