| Message ID | 20190812104827.5935-3-yangbo.lu@nxp.com |
|---|---|
| State | Changes Requested |
| Delegated to: | David Miller |
| Headers | show |
| Series | ocelot_ace: fix and improve the driver | expand |
The 08/12/2019 18:48, Yangbo Lu wrote: > The ingress ports setting of rule should support covering all ports. > This patch is to use u16 ingress_port for ingress port mask setting > for ace rule. One bit corresponds one port. That is how the HW is working, and it would be nice if we could operate on a port masks/lists instead. But how can this be used? Can you please explain how/when this will make a difference? > Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com> > --- > drivers/net/ethernet/mscc/ocelot_ace.c | 2 +- > drivers/net/ethernet/mscc/ocelot_ace.h | 2 +- > drivers/net/ethernet/mscc/ocelot_flower.c | 2 +- > 3 files changed, 3 insertions(+), 3 deletions(-) > > diff --git a/drivers/net/ethernet/mscc/ocelot_ace.c b/drivers/net/ethernet/mscc/ocelot_ace.c > index 5580a58..91250f3 100644 > --- a/drivers/net/ethernet/mscc/ocelot_ace.c > +++ b/drivers/net/ethernet/mscc/ocelot_ace.c > @@ -352,7 +352,7 @@ static void is2_entry_set(struct ocelot *ocelot, int ix, > data.type = IS2_ACTION_TYPE_NORMAL; > > VCAP_KEY_ANY_SET(PAG); > - VCAP_KEY_SET(IGR_PORT_MASK, 0, ~BIT(ace->chip_port)); > + VCAP_KEY_SET(IGR_PORT_MASK, 0, ~ace->ingress_port); > VCAP_KEY_BIT_SET(FIRST, OCELOT_VCAP_BIT_1); > VCAP_KEY_BIT_SET(HOST_MATCH, OCELOT_VCAP_BIT_ANY); > VCAP_KEY_BIT_SET(L2_MC, ace->dmac_mc); > diff --git a/drivers/net/ethernet/mscc/ocelot_ace.h b/drivers/net/ethernet/mscc/ocelot_ace.h > index ce72f02..0fe23e0 100644 > --- a/drivers/net/ethernet/mscc/ocelot_ace.h > +++ b/drivers/net/ethernet/mscc/ocelot_ace.h > @@ -193,7 +193,7 @@ struct ocelot_ace_rule { > > enum ocelot_ace_action action; > struct ocelot_ace_stats stats; > - int chip_port; > + u16 ingress_port; > > enum ocelot_vcap_bit dmac_mc; > enum ocelot_vcap_bit dmac_bc; > diff --git a/drivers/net/ethernet/mscc/ocelot_flower.c b/drivers/net/ethernet/mscc/ocelot_flower.c > index 7c60e8c..bfddc50 100644 > --- a/drivers/net/ethernet/mscc/ocelot_flower.c > +++ b/drivers/net/ethernet/mscc/ocelot_flower.c > @@ -184,7 +184,7 @@ struct ocelot_ace_rule *ocelot_ace_rule_create(struct flow_cls_offload *f, > return NULL; > > rule->ocelot = block->port->ocelot; > - rule->chip_port = block->port->chip_port; > + rule->ingress_port = BIT(block->port->chip_port); > return rule; > } -- Allan
Hi Allan, > -----Original Message----- > From: Allan W. Nielsen <allan.nielsen@microchip.com> > Sent: Monday, August 12, 2019 8:38 PM > To: Y.b. Lu <yangbo.lu@nxp.com> > Cc: netdev@vger.kernel.org; David S . Miller <davem@davemloft.net>; > Alexandre Belloni <alexandre.belloni@bootlin.com>; Microchip Linux Driver > Support <UNGLinuxDriver@microchip.com> > Subject: Re: [PATCH 2/3] ocelot_ace: fix ingress ports setting for rule > > The 08/12/2019 18:48, Yangbo Lu wrote: > > The ingress ports setting of rule should support covering all ports. > > This patch is to use u16 ingress_port for ingress port mask setting > > for ace rule. One bit corresponds one port. > That is how the HW is working, and it would be nice if we could operate on a > port masks/lists instead. But how can this be used? [Y.b. Lu] Will the changes affect anything? Current usage in ocelot_flower.c will be converted as below. - rule->chip_port = block->port->chip_port; + rule->ingress_port = BIT(block->port->chip_port); > > Can you please explain how/when this will make a difference? [Y.b. Lu] Actually I have another internal patch based on this patch-set for setting rule of trapping IEEE 1588 PTP Ethernet frames. For such rule which should be applied on several ingress ports, we can set it once when ocelot initialization I think. The internal patch I mentioned is for felix which had different ports number (VCAP_PORT_CNT). So I hadn't sent it out. Let me just send v2 patch-set with the patch dropping VCAP_PORT_CNT changes for your reviewing. Please feel free to provide suggestion. Thanks a lot:) > > > Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com> > > --- > > drivers/net/ethernet/mscc/ocelot_ace.c | 2 +- > > drivers/net/ethernet/mscc/ocelot_ace.h | 2 +- > > drivers/net/ethernet/mscc/ocelot_flower.c | 2 +- > > 3 files changed, 3 insertions(+), 3 deletions(-) > > > > diff --git a/drivers/net/ethernet/mscc/ocelot_ace.c > > b/drivers/net/ethernet/mscc/ocelot_ace.c > > index 5580a58..91250f3 100644 > > --- a/drivers/net/ethernet/mscc/ocelot_ace.c > > +++ b/drivers/net/ethernet/mscc/ocelot_ace.c > > @@ -352,7 +352,7 @@ static void is2_entry_set(struct ocelot *ocelot, int ix, > > data.type = IS2_ACTION_TYPE_NORMAL; > > > > VCAP_KEY_ANY_SET(PAG); > > - VCAP_KEY_SET(IGR_PORT_MASK, 0, ~BIT(ace->chip_port)); > > + VCAP_KEY_SET(IGR_PORT_MASK, 0, ~ace->ingress_port); > > VCAP_KEY_BIT_SET(FIRST, OCELOT_VCAP_BIT_1); > > VCAP_KEY_BIT_SET(HOST_MATCH, OCELOT_VCAP_BIT_ANY); > > VCAP_KEY_BIT_SET(L2_MC, ace->dmac_mc); diff --git > > a/drivers/net/ethernet/mscc/ocelot_ace.h > > b/drivers/net/ethernet/mscc/ocelot_ace.h > > index ce72f02..0fe23e0 100644 > > --- a/drivers/net/ethernet/mscc/ocelot_ace.h > > +++ b/drivers/net/ethernet/mscc/ocelot_ace.h > > @@ -193,7 +193,7 @@ struct ocelot_ace_rule { > > > > enum ocelot_ace_action action; > > struct ocelot_ace_stats stats; > > - int chip_port; > > + u16 ingress_port; > > > > enum ocelot_vcap_bit dmac_mc; > > enum ocelot_vcap_bit dmac_bc; > > diff --git a/drivers/net/ethernet/mscc/ocelot_flower.c > > b/drivers/net/ethernet/mscc/ocelot_flower.c > > index 7c60e8c..bfddc50 100644 > > --- a/drivers/net/ethernet/mscc/ocelot_flower.c > > +++ b/drivers/net/ethernet/mscc/ocelot_flower.c > > @@ -184,7 +184,7 @@ struct ocelot_ace_rule > *ocelot_ace_rule_create(struct flow_cls_offload *f, > > return NULL; > > > > rule->ocelot = block->port->ocelot; > > - rule->chip_port = block->port->chip_port; > > + rule->ingress_port = BIT(block->port->chip_port); > > return rule; > > } > > -- Allan
diff --git a/drivers/net/ethernet/mscc/ocelot_ace.c b/drivers/net/ethernet/mscc/ocelot_ace.c index 5580a58..91250f3 100644 --- a/drivers/net/ethernet/mscc/ocelot_ace.c +++ b/drivers/net/ethernet/mscc/ocelot_ace.c @@ -352,7 +352,7 @@ static void is2_entry_set(struct ocelot *ocelot, int ix, data.type = IS2_ACTION_TYPE_NORMAL; VCAP_KEY_ANY_SET(PAG); - VCAP_KEY_SET(IGR_PORT_MASK, 0, ~BIT(ace->chip_port)); + VCAP_KEY_SET(IGR_PORT_MASK, 0, ~ace->ingress_port); VCAP_KEY_BIT_SET(FIRST, OCELOT_VCAP_BIT_1); VCAP_KEY_BIT_SET(HOST_MATCH, OCELOT_VCAP_BIT_ANY); VCAP_KEY_BIT_SET(L2_MC, ace->dmac_mc); diff --git a/drivers/net/ethernet/mscc/ocelot_ace.h b/drivers/net/ethernet/mscc/ocelot_ace.h index ce72f02..0fe23e0 100644 --- a/drivers/net/ethernet/mscc/ocelot_ace.h +++ b/drivers/net/ethernet/mscc/ocelot_ace.h @@ -193,7 +193,7 @@ struct ocelot_ace_rule { enum ocelot_ace_action action; struct ocelot_ace_stats stats; - int chip_port; + u16 ingress_port; enum ocelot_vcap_bit dmac_mc; enum ocelot_vcap_bit dmac_bc; diff --git a/drivers/net/ethernet/mscc/ocelot_flower.c b/drivers/net/ethernet/mscc/ocelot_flower.c index 7c60e8c..bfddc50 100644 --- a/drivers/net/ethernet/mscc/ocelot_flower.c +++ b/drivers/net/ethernet/mscc/ocelot_flower.c @@ -184,7 +184,7 @@ struct ocelot_ace_rule *ocelot_ace_rule_create(struct flow_cls_offload *f, return NULL; rule->ocelot = block->port->ocelot; - rule->chip_port = block->port->chip_port; + rule->ingress_port = BIT(block->port->chip_port); return rule; }
The ingress ports setting of rule should support covering all ports. This patch is to use u16 ingress_port for ingress port mask setting for ace rule. One bit corresponds one port. Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com> --- drivers/net/ethernet/mscc/ocelot_ace.c | 2 +- drivers/net/ethernet/mscc/ocelot_ace.h | 2 +- drivers/net/ethernet/mscc/ocelot_flower.c | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-)