Re: 2.6.test11 bug

[Linus Torvalds]

On Tue, 9 Dec 2003, Rafal Skoczylas wrote:
>
> Anyway, while compiling with frame pointers I got another oops ;/
> This one is different those we have seen so far:

Actually, it's very similar to the ones you had earlier - the backtrace is
different, but it's the same thing again:

	ecx: 5944f160   edx: d944f160

that's a doubly-linked list, and the two values _should_ be the same, I
bet.

Except one of them has the high bit clear, and when following the bogus
pointer, you get a page fault.

I think the second one (the "kernel bug in mmap.c" one) is just a result
of the first oops one leaving the mm lists in a bad state and map_count
being off as a result, or something like that.

			Linus
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/