Messages in this thread | | | Date | Thu, 18 Jul 2002 00:44:21 -0700 | From | William Lee Irwin III <> | Subject | Re: [PATCH] Initcall depends automagic |
| |
On Thu, Jul 18, 2002 at 05:36:09PM +1000, Rusty Russell wrote: > This patch generates initcall ordering based on the theory that if A.o > references a symbol in B.o, then B.o's initcall must preceed A.o's. > Kai and I came up with the approach, and Stephen Rothwell did all the > implementation brain sweat. I did the typing. > Unfortunately, there are dependency loops in the kernel which make > this approach fail, but if someone wants to play with it and try to > untangle them, they are more than welcome. One option would be to > look at more course-grained .o files, rather than *all* .o files > (eg. fs/ext2/ext2.o rather than fs/ext2/*.o). > I will be persuing my previous explicit initcall depends patch in the > meantime, as it offers an elegent solution that works now.
Any chance you can post the strongly connected components of the dependency graph?
Thanks, Bill - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |