Re: [PATCH] swiotlb-0.8.2^H3: seperation of physical/virtual address translation and core memory allocation.

[Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>]

> > in your -mm tree.
> 
> I don't put individual git trees into -mm any more - I get them from
> linux-next.

Ahh. OK.
> 
> I could grab the individual patches but I don't see that this gains
> anyone anything.  I'd suggest that you ask Stephen to include this tree
> in linux-next then ask Linus to pull it into 2.6.36-rc1.

Stephen,

I was wondering if you wouldn't mind including the following git tree
in your linux-next?

git://git.kernel.org/pub/scm/linux/kernel/git/konrad/swiotlb-2.6.git

The patches are in the MAIN branch (and also in swiotlb-0.8.3).

They have the Tested-by, Signed-off-by, and Acked-by signatures.

Andrew,
> > [nice description]

Thank you.
> 
> The sad thing is that this description will get lost because it isn't
.. snip..
> and then follow it with the specific description for [patch 1/n].  I
> suggest that you make such change in this tree.

Done!
> 
> It's not a very good solution because anyone who is looking at [patch
.. snip ..
> "see <patch name> for a description of this work" or similar.

Done!
> 
> Even better would be if git understood this workflow - if it was able
> to take commentary-only commits and to bidirectionally link those
> commentaries to the invividual commits which they describe.
> 
..snip.
> mixture of "enum dma_data_direction dir" and "enum dma_data_direction
> direction", which could be cleaned up.

Oh, good catch. Fixed that.