Hello.
I've just start with Accessing Pixels in a Frame on the DM643x in my evmDM6437.
I think is the best project to probe an sobel edge detection, but I have problems. I would like to create another "frameBuffPtr" (named "frameBuffPtrOut") where putting my proccessed pixels. For that, I define:
PSP_VPSSSurfaceParams *frameBuffPtrOut;
And in the first step I do is transfering pixels from original buffer to this second buffer:
for (i=0;i<ROWS_IMAGE_IN;i++)
{
for (j=0;j<COLS_IMAGE_IN*2;j++)
{
*((unsigned char*)frameBuffPtrOut->frame.frameBufferPtr+i*COLS_IMAGE_IN*2+j)=
*((unsigned char*)frameBuffPtr->frame.frameBufferPtr+i*COLS_IMAGE_IN*2+j);
}
}
where ROWS_IMAGE=480 and COLS_IMAGE_IN=720
After that, I change some parameters in function calls to represent this second buffer in output video signal
BCACHE_wbInv((void*)(frameBuffPtrOut->frame.frameBufferPtr), ROWS_IMAGE_IN*COLS_IMAGE_IN*2, 1);
/* display the video frame */
FVID_exchange(hGioVpbeVid0, &frameBuffPtrOut);
But I don't see anything. I put a breakpoint and I surprised because this second buffer is located at address 0!!
How anyone why this happens? I think this variable/buffer would be allocated after 0x80000000 where DDR2 is defined.
Do I need to do some changes in .tcf configuration?
Thanks a lot