When doing memcpy/memset of EQEs, we should use sizeof struct
mlx4_eqe as the base size and not caps.eqe_size which could be bigger.
If caps.eqe_size is bigger than the struct mlx4_eqe then we corrupt
data in the master context.
When using a 64 byte stride, the memcpy copied over 63 bytes to the
slave_eq structure. This resulted in copying over the entire eqe of
interest, including its ownership bit -- and also 31 bytes of garbage
into the next WQE in the slave EQ -- which did NOT include the ownership
bit (and therefore had no impact).
However, once the stride is increased to 128, we are overwriting the
ownership bits of *three* eqes in the slave_eq struct. This results
in an incorrect ownership bit for those eqes, which causes the eq to
seem to be full. The issue therefore surfaced only once 128-byte EQEs
started being used in SRIOV and (overarchitectures that have 128/256
byte cache-lines such as PPC) - e.g after commit
77507aa249ae
"net/mlx4_core: Enable CQE/EQE stride support".
Fixes:
08ff32352d6f ('mlx4: 64-byte CQE/EQE support')
Signed-off-by: Carol L Soto <clsoto@linux.vnet.ibm.com>
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
}
}
- memset(&priv->mfunc.master.cmd_eqe, 0, dev->caps.eqe_size);
+ memset(&priv->mfunc.master.cmd_eqe, 0, sizeof(struct mlx4_eqe));
priv->mfunc.master.cmd_eqe.type = MLX4_EVENT_TYPE_CMD;
INIT_WORK(&priv->mfunc.master.comm_work,
mlx4_master_comm_channel);
return;
}
- memcpy(s_eqe, eqe, dev->caps.eqe_size - 1);
+ memcpy(s_eqe, eqe, sizeof(struct mlx4_eqe) - 1);
s_eqe->slave_id = slave;
/* ensure all information is written before setting the ownersip bit */
dma_wmb();