Re: [V9fs-developer] [PATCH] 9P: Add memory barriers to protect request fields over cb/rpc threads handoff

[Dominique Martinet <dominique.martinet@cea.fr>]

Hi,

Dominique Martinet wrote on Fri, Jan 17, 2014 :
> We need barriers to guarantee this pattern works as intended:
> [w] req->rc, 1		[r] req->status, 1
> wmb			rmb
> [w] req->status, 1	[r] req->rc
> 
> Where the wmb ensures that rc gets written before status,
> and the rmb ensures that if you observe status == 1, rc is the new value.
> 
> Signed-off-by: Dominique Martinet <dominique.martinet@cea.fr>
> ---
>  include/net/9p/client.h |  2 +-
>  net/9p/client.c         | 16 +++++++++++++++-
>  net/9p/trans_fd.c       | 15 ++++++---------
>  net/9p/trans_rdma.c     |  3 +--
>  net/9p/trans_virtio.c   |  3 +--
>  5 files changed, 24 insertions(+), 15 deletions(-)

Reminder to get some attention now that the pull request for merge
window has passed :)
Please ask if this needs a resend, but should still apply as is.


Would be particularily nice to get confirmation that this doesn't slow
everything down for tcp/virtio folk or raises another kind of blocker
(although I have nothing better short of another transport-specific
function call after the p9_client_rpc wait is over, which is worse than
this to me, but suggestions are always welcome!)

Cheers,
-- 
Dominique Martinet