| Message ID | 1394542976-20095-3-git-send-email-ogerlitz@mellanox.com |
|---|---|
| State | Changes Requested, archived |
| Delegated to: | David Miller |
| Headers | show |
Hello. On 03/11/2014 04:02 PM, Or Gerlitz wrote: > From: Jack Morgenstein <jackm@dev.mellanox.co.il> > For IB transport, the host determines the slave GIDs. For ETH (RoCE), > however, the slave's GID is determined by the IP address that the slave > itself assigns to the ETH device used by RoCE. > In this case, the slave must be able to write its GIDs to the HCA gid table > (at the GID indices that slave "owns"). > This commit adds processing for the SET_PORT_GID_TABLE opcode modifier > for the SET_PORT command wrapper (so that slaves may modify their GIDS > for RoCE). > Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> > Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> > --- > drivers/net/ethernet/mellanox/mlx4/port.c | 27 ++++++++++++++++++++++++++- > 1 files changed, 26 insertions(+), 1 deletions(-) > diff --git a/drivers/net/ethernet/mellanox/mlx4/port.c b/drivers/net/ethernet/mellanox/mlx4/port.c > index 9c063d6..095f74d 100644 > --- a/drivers/net/ethernet/mellanox/mlx4/port.c > +++ b/drivers/net/ethernet/mellanox/mlx4/port.c [...] > @@ -581,6 +584,28 @@ static int mlx4_common_set_port(struct mlx4_dev *dev, int slave, u32 in_mod, > > gen_context->mtu = cpu_to_be16(master->max_mtu[port]); > break; > + case MLX4_SET_PORT_GID_TABLE: > + gid_entry = (struct mlx4_roce_gid_entry *)(inbox->buf); > + /* check that do not have duplicates */ > + if (memcmp(gid_entry->raw, zgid_entry.raw, 16)) { > + for (i = 0; i < MLX4_ROCE_MAX_GIDS; i++) { > + if (slave != i && > + !memcmp(gid_entry->raw, priv->roce_gids[port - 1][i].raw, 16)) { This line is over 80 columns, as checkpatch.pl would have told. > + mlx4_warn(dev, "requested gid entry for slave:%d " > + "is a duplicate of slave %d\n", checkpatch.pl would have warned not to break up the string literal here -- it's an exception from the 80-column rule. > + slave, i); > + return -EEXIST; > + } > + } > + } > + /* insert slave GID at proper index */ > + memcpy(priv->roce_gids[port - 1][slave].raw, gid_entry->raw, 16); This line is over 80 columns. > + /* rewrite roce port gids table to FW */ > + for (i = 0; i < MLX4_ROCE_MAX_GIDS; i++) { > + memcpy(gid_entry->raw, priv->roce_gids[port - 1][i].raw, 16); This one too. > + gid_entry++; > + } > + break; WBR, Sergei -- To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Tue, Mar 11, 2014 at 9:12 PM, Sergei Shtylyov <sergei.shtylyov@cogentembedded.com> wrote: > Hello. > > > On 03/11/2014 04:02 PM, Or Gerlitz wrote: > >> From: Jack Morgenstein <jackm@dev.mellanox.co.il> > > >> For IB transport, the host determines the slave GIDs. For ETH (RoCE), >> however, the slave's GID is determined by the IP address that the slave >> itself assigns to the ETH device used by RoCE. > > >> In this case, the slave must be able to write its GIDs to the HCA gid >> table >> (at the GID indices that slave "owns"). > > >> This commit adds processing for the SET_PORT_GID_TABLE opcode modifier >> for the SET_PORT command wrapper (so that slaves may modify their GIDS >> for RoCE). > > >> Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> >> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> >> --- >> drivers/net/ethernet/mellanox/mlx4/port.c | 27 >> ++++++++++++++++++++++++++- >> 1 files changed, 26 insertions(+), 1 deletions(-) > > >> diff --git a/drivers/net/ethernet/mellanox/mlx4/port.c >> b/drivers/net/ethernet/mellanox/mlx4/port.c >> index 9c063d6..095f74d 100644 >> --- a/drivers/net/ethernet/mellanox/mlx4/port.c >> +++ b/drivers/net/ethernet/mellanox/mlx4/port.c > > [...] > >> @@ -581,6 +584,28 @@ static int mlx4_common_set_port(struct mlx4_dev *dev, >> int slave, u32 in_mod, >> >> gen_context->mtu = >> cpu_to_be16(master->max_mtu[port]); >> break; >> + case MLX4_SET_PORT_GID_TABLE: >> + gid_entry = (struct mlx4_roce_gid_entry >> *)(inbox->buf); >> + /* check that do not have duplicates */ >> + if (memcmp(gid_entry->raw, zgid_entry.raw, 16)) { >> + for (i = 0; i < MLX4_ROCE_MAX_GIDS; i++) { >> + if (slave != i && >> + !memcmp(gid_entry->raw, >> priv->roce_gids[port - 1][i].raw, 16)) { > > > This line is over 80 columns, as checkpatch.pl would have told. Indeed, I don't think the 80 columns warning of checkpatch is considered a gating factor for accepting patches, so I would prefer not to re-spin the series just to fix this sort of warnings. > > >> + mlx4_warn(dev, "requested >> gid entry for slave:%d " >> + "is a duplicate >> of slave %d\n", > > > checkpatch.pl would have warned not to break up the string literal here > -- it's an exception from the 80-column rule. I see, yep, I assume we can fix that if needed. -- To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
diff --git a/drivers/net/ethernet/mellanox/mlx4/port.c b/drivers/net/ethernet/mellanox/mlx4/port.c index 9c063d6..095f74d 100644 --- a/drivers/net/ethernet/mellanox/mlx4/port.c +++ b/drivers/net/ethernet/mellanox/mlx4/port.c @@ -505,6 +505,7 @@ int mlx4_get_port_ib_caps(struct mlx4_dev *dev, u8 port, __be32 *caps) mlx4_free_cmd_mailbox(dev, outmailbox); return err; } +static struct mlx4_roce_gid_entry zgid_entry; static int mlx4_common_set_port(struct mlx4_dev *dev, int slave, u32 in_mod, u8 op_mod, struct mlx4_cmd_mailbox *inbox) @@ -515,6 +516,7 @@ static int mlx4_common_set_port(struct mlx4_dev *dev, int slave, u32 in_mod, struct mlx4_slave_state *slave_st = &master->slave_state[slave]; struct mlx4_set_port_rqp_calc_context *qpn_context; struct mlx4_set_port_general_context *gen_context; + struct mlx4_roce_gid_entry *gid_entry; int reset_qkey_viols; int port; int is_eth; @@ -535,7 +537,8 @@ static int mlx4_common_set_port(struct mlx4_dev *dev, int slave, u32 in_mod, /* Slaves cannot perform SET_PORT operations except changing MTU */ if (is_eth) { if (slave != dev->caps.function && - in_modifier != MLX4_SET_PORT_GENERAL) { + in_modifier != MLX4_SET_PORT_GENERAL && + in_modifier != MLX4_SET_PORT_GID_TABLE) { mlx4_warn(dev, "denying SET_PORT for slave:%d\n", slave); return -EINVAL; @@ -581,6 +584,28 @@ static int mlx4_common_set_port(struct mlx4_dev *dev, int slave, u32 in_mod, gen_context->mtu = cpu_to_be16(master->max_mtu[port]); break; + case MLX4_SET_PORT_GID_TABLE: + gid_entry = (struct mlx4_roce_gid_entry *)(inbox->buf); + /* check that do not have duplicates */ + if (memcmp(gid_entry->raw, zgid_entry.raw, 16)) { + for (i = 0; i < MLX4_ROCE_MAX_GIDS; i++) { + if (slave != i && + !memcmp(gid_entry->raw, priv->roce_gids[port - 1][i].raw, 16)) { + mlx4_warn(dev, "requested gid entry for slave:%d " + "is a duplicate of slave %d\n", + slave, i); + return -EEXIST; + } + } + } + /* insert slave GID at proper index */ + memcpy(priv->roce_gids[port - 1][slave].raw, gid_entry->raw, 16); + /* rewrite roce port gids table to FW */ + for (i = 0; i < MLX4_ROCE_MAX_GIDS; i++) { + memcpy(gid_entry->raw, priv->roce_gids[port - 1][i].raw, 16); + gid_entry++; + } + break; } return mlx4_cmd(dev, inbox->dma, in_mod, op_mod, MLX4_CMD_SET_PORT, MLX4_CMD_TIME_CLASS_B,