Re: [ewg] [PATCH] mlx4: Fix bug in mlx4_ib_mcg_attach

2009-12-10 Thread Eli Cohen
On Wed, Dec 09, 2009 at 02:49:47PM -0800, Roland Dreier wrote:
> 
> I don't see anything touching this code there.  The patch that
> introduced this code upstream, 521e575b ("IB/mlx4: Add support for
> blocking multicast loopback packets") doesn't have this bug and I don't
> see anything else that changed that area of the code.
> 

Your branch, my bug :-) I put it in one of my commits.
___
ewg mailing list
ewg@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg


Re: [ewg] [PATCH] mlx4: Fix bug in mlx4_ib_mcg_attach

2009-12-09 Thread Roland Dreier

 > > This bug doesn't seem to ever have been present in the upstream
 > > kernel -- what are you generating this patch against?

 > I think it came from your for-next branch.

I don't see anything touching this code there.  The patch that
introduced this code upstream, 521e575b ("IB/mlx4: Add support for
blocking multicast loopback packets") doesn't have this bug and I don't
see anything else that changed that area of the code.

 - R.
___
ewg mailing list
ewg@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg


Re: [ewg] [PATCH] mlx4: Fix bug in mlx4_ib_mcg_attach

2009-12-09 Thread Eli Cohen
On Wed, Dec 09, 2009 at 02:33:31PM -0800, Roland Dreier wrote:
> This bug doesn't seem to ever have been present in the upstream
> kernel -- what are you generating this patch against?
> 

I think it came from your for-next branch.
___
ewg mailing list
ewg@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg


Re: [ewg] [PATCH] mlx4: Fix bug in mlx4_ib_mcg_attach

2009-12-09 Thread Roland Dreier
This bug doesn't seem to ever have been present in the upstream
kernel -- what are you generating this patch against?

 - R.
___
ewg mailing list
ewg@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg


[ewg] [PATCH] mlx4: Fix bug in mlx4_ib_mcg_attach

2009-12-08 Thread Eli Cohen
The "!" operator has precedence over the "&" operator so parenthesis are
required to properly evaluate the user's loopback requirement.

Signed-off-by: Eli Cohen 
---
 drivers/infiniband/hw/mlx4/main.c |4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/infiniband/hw/mlx4/main.c 
b/drivers/infiniband/hw/mlx4/main.c
index 0f6ef38..736eea0 100644
--- a/drivers/infiniband/hw/mlx4/main.c
+++ b/drivers/infiniband/hw/mlx4/main.c
@@ -586,8 +586,8 @@ static int mlx4_ib_mcg_attach(struct ib_qp *ibqp, union 
ib_gid *gid, u16 lid)
struct mlx4_ib_dev *mdev = to_mdev(ibqp->device);
struct mlx4_ib_qp *mqp = to_mqp(ibqp);
 
-   err = mlx4_multicast_attach(mdev->dev, &mqp->mqp, gid->raw, 
!!mqp->flags &
-   MLX4_IB_QP_BLOCK_MULTICAST_LOOPBACK);
+   err = mlx4_multicast_attach(mdev->dev, &mqp->mqp, gid->raw, 
!!(mqp->flags &
+   MLX4_IB_QP_BLOCK_MULTICAST_LOOPBACK));
if (err)
return err;
 
-- 
1.6.5.5

___
ewg mailing list
ewg@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg