Hey Folks, I know roughly what the issue is with the kallsyms problem in the kernel now, or I think I do. It's late, but, quickly, the kernel does the following during build: 1). Generate a tmp vmlinux that isn't the final one with empty kallsyms 2). Run the kallsyms generation to make a compressed table 3). Dump out temporary System.map using this 4). Rerun kallsyms and generate a second System.map 5). Diff the two System.map files The system.map files should be identical, but in this case they are shifted after kallsyms_names (from kallsyms_markers) onwards. I believe the problem is that the linker is doing something screwy the second time around and we're exactly 4 words off with our addresses. Anyway. Now I think I'm looking in the right area and understand what we're trying to do it is of course too late to finish tracking this down. But I'll aim to have it figured out before the weekend. Jon. _______________________________________________ arm mailing list arm@xxxxxxxxxxxxxxxxxxxxxxx https://admin.fedoraproject.org/mailman/listinfo/arm
