Re: [PATCH] async_tx: fix kmap_atomic usage in async_memcpy

[Andrew Morton <akpm@linux-foundation.org>]

On Thu, 19 Jul 2007 18:12:52 -0700
Dan Williams <dan.j.williams@intel.com> wrote:

> Andrew Morton:
> 	[async_memcpy] is very wrong if both ASYNC_TX_KMAP_DST and
> 	ASYNC_TX_KMAP_SRC can ever be set.  We'll end up using the same kmap
> 	slot for both src add dest and we get either corrupted data or a BUG.
> 
> Evgeniy Polyakov:
> 	Btw, shouldn't it always be kmap_atomic() even if flag is not set.
> 	That pages are usual one returned by alloc_page().
> 
> So fix the usage of kmap_atomic and kill the ASYNC_TX_KMAP_DST and
> ASYNC_TX_KMAP_SRC flags.
> 
> Cc: Andrew Morton <akpm@linux-foundation.org>
> Cc: Evgeniy Polyakov <johnpol@2ka.mipt.ru>
> Signed-off-by: Dan Williams <dan.j.williams@intel.com>
> ---
> 
>  crypto/async_tx/async_memcpy.c |   19 ++++---------------
>  drivers/md/raid5.c             |    4 ++--
>  include/linux/async_tx.h       |    6 ------
>  3 files changed, 6 insertions(+), 23 deletions(-)

Neato.

You (or at least, Shannon) run a git tree.  I always get confused when
git-tree-owners send me patches, because I expect them to put the
patches into their git trees.

So if you, a git-tree-owner, wish me to merge-test-and-forward a patch,
please explicitly tell me.