* Gordon Farquharson <[EMAIL PROTECTED]> [2008-02-10 14:08]: > # Let's pad the kernel to 131072 * 10 + 1 so it will be rounded up > # by slugimage to 131072 * 11, i.e. 11 blocks. > util/arm/nslu2/pad $(TEMP)/$(KERNELNAME).nslu2 1310724
BTW, 1310724 seems to be 131072 * 10 + 4 rather than +1 as the comment claims. I wonder why I made that mistake. > Shouldn't 1310724 be 1441792 otherwise the script will fail with a > kernel size of 1337692? No, if the kernel is larger than 1310724 it won't get padded but this is not important since slugimage needs to round up to a full block anyway, and 1310724 is more than 10 blocks so it will take 11 blocks, which is what we want Apparently slugimage allocates 11 blocks when you give it a file with 1310705 bytes. This makes sense since 131072 * 10 - 16 = 1310704 (- 16 because of the header). In summary, I believe any value between 1310705 and 1441776 will work. Right, I just tested this. Maybe we should pad to 1310705 then. -- Martin Michlmayr http://www.cyrius.com/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
