Re: [2.6.8-rc2-mm2] oops report (crashes on booting)

[Hugh Dickins]

On Wed, 4 Aug 2004, Bernd Schubert wrote:
> 
> we just tried to boot one of our systems using 2.6.8-rc2-mm2,
> unfortunately it oopses after running pivot_root and init:
> 
> (written down manually)
> 
> Debug: sleeping function called from invalid context at 
> include/asm/uaccess.h:471
> 
> in_atomic():1, irqs_disabled():0
> dump_stack
> __might_sleep
> shmem_file_write

It shouldn't be an oops, but it'll certainly be a very disturbing
splurge of messages.  Sorry, I should have caught it sooner.  Please
apply patch below (which not only sidesteps the warnings, but appears
to be the _right_ thing to do, from a look at filemap.c)...

tmpfs must use __copy_from_user_inatomic now, to avoid might_sleep
warning, when knowingly using __copy_from_user with an atomic kmap.

Signed-off-by: Hugh Dickins <hugh@xxxxxxxxxxx>

--- 2.6.8-rc2-mm2/mm/shmem.c	2004-08-02 13:03:33.000000000 +0100
+++ linux/mm/shmem.c	2004-08-04 18:05:07.917110256 +0100
@@ -1323,7 +1323,8 @@ shmem_file_write(struct file *file, cons
 			__get_user(dummy, buf + bytes - 1);
 
 			kaddr = kmap_atomic(page, KM_USER0);
-			left = __copy_from_user(kaddr + offset, buf, bytes);
+			left = __copy_from_user_inatomic(kaddr + offset,
+							buf, bytes);
 			kunmap_atomic(kaddr, KM_USER0);
 		}
 		if (left) {

-
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/