Re: [PATCH] failslab - failmalloc for slab allocator

[Akinobu Mita <mita@miraclelinux.com>]

On Sun, Aug 13, 2006 at 12:17:02PM -0700, Andrew Morton wrote:
> We would benefit from having some faul-injection capabilities in the
> mainline kernel.
> 
> - kmalloc failures
> 
> - alloc_pages() failures
> 
> - disk IO errors (there are rumours of a DM module for this, but I
>   haven't seen it).

What shold I do for this?
Should we have faul-injection capability in generic_make_request()
like this patch?

Index: work-failmalloc/block/ll_rw_blk.c
===================================================================
--- work-failmalloc.orig/block/ll_rw_blk.c
+++ work-failmalloc/block/ll_rw_blk.c
@@ -3077,6 +3077,9 @@ end_io:
 		if (unlikely(test_bit(QUEUE_FLAG_DEAD, &q->queue_flags)))
 			goto end_io;
 
+		if (should_fail(&fail_make_request, bio->bi_size))
+			goto end_io;
+
 		/*
 		 * If this device has partitions, remap block n
 		 * of partition p to block n+start(p) of the disk.


> They would need to be lightweight, clean and enabled/configured at runtime,
> not at boot time.

I'll make it configurable at runtime with using debugfs.