On Mon, May 11, 2009 at 10:32 AM, ron minnich rminnich@gmail.com wrote:
attached.
This came up when cbfs (correctly) diagnosed a rom as corrupt ... it was a bug in cbfs, but that's how it's supposed to work -- catch a bad rom at build time, not boot time.
+ csize = headersize(name); + + strcpy(c->magic, COMPONENT_MAGIC); + + c->offset = htonl(csize);
I think this code would be clearer without csize.
+ c->offset = htonl(headersize(name));
The only other comment I have is that rom_add is pretty trivial now. It could disappear since it is only a wrapper for rom_alloc. Is there any time we want to do rom_alloc when we don't want to copy in the data?
Acked-by: Myles Watsonmylesgw@gmail.com
BTW, you can simplify the existing coreboot code a bit if you add a "char filename[0];" entry to 'struct cbfs_file' - as SeaBIOS does.
I like that idea.
Thanks, Myles