Comment 4 for bug 724102

Currently, I am reviewing 3 different patches for this issue, but I am not able to get predictable results. My testing method is to fill an 8G SD with random data (sudo dd bs=4M if=/dev/random of=/dev/mmcblk0) and then read that in in both linux and win32-image-writer and compare the reads with md5sum. This is proving to take a very long time and not providing me with consistent results. I will push out v0.4 without this fix if I can not resolve this by Jan 2, 2012.