| Message ID | 20200306235403.18751-1-andre.guedes@intel.com |
|---|---|
| State | Accepted |
| Delegated to: | Jeff Kirsher |
| Headers | show |
| Series | igc: Fix NFC queue redirection support | expand |
On 3/6/2020 15:54, Andre Guedes wrote: > The support for ethtool Network Flow Classification (NFC) queue > redirection based on destination MAC address is currently broken in IGC. > For instance, if we add the following rule, matching frames aren't > enqueued on the expected rx queue. > > $ ethtool -N IFNAME flow-type ether dst 3c:fd:fe:9e:7f:71 queue 2 > > The issue here is due to the fact that igc_rar_set_index() is missing > code to enable the queue selection feature from Receive Address High > (RAH) register. This patch adds the missing code and fixes the issue. > > Signed-off-by: Andre Guedes <andre.guedes@intel.com> > --- > drivers/net/ethernet/intel/igc/igc_defines.h | 5 ++++- > drivers/net/ethernet/intel/igc/igc_main.c | 11 ++++++++--- > 2 files changed, 12 insertions(+), 4 deletions(-) > > diff --git a/drivers/net/ethernet/intel/igc/igc_defines.h b/drivers/net/ethernet/intel/igc/igc_defines.h > index e5116337b68d..63d3d34763da 100644 > --- a/drivers/net/ethernet/intel/igc/igc_defines.h > +++ b/drivers/net/ethernet/intel/igc/igc_defines.h > @@ -63,8 +63,11 @@ > * (RAR[15]) for our directed address used by controllers with > * manageability enabled, allowing us room for 15 multicast addresses. > */ > +#define IGC_RAH_QSEL_MASK 0x000C0000 > +#define IGC_RAH_QSEL_SHIFT 18 > +#define IGC_RAH_QSEL_ENABLE BIT(28) > #define IGC_RAH_AV 0x80000000 /* Receive descriptor valid */ > -#define IGC_RAH_POOL_1 0x00040000 > + > #define IGC_RAL_MAC_ADDR_LEN 4 > #define IGC_RAH_MAC_ADDR_LEN 2 > > diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c > index 025f2e439aef..49a0d0bff727 100644 > --- a/drivers/net/ethernet/intel/igc/igc_main.c > +++ b/drivers/net/ethernet/intel/igc/igc_main.c > @@ -780,13 +780,18 @@ static void igc_rar_set_index(struct igc_adapter *adapter, u32 index) > rar_low = le32_to_cpup((__le32 *)(addr)); > rar_high = le16_to_cpup((__le16 *)(addr + 4)); > > + if (adapter->mac_table[index].state & IGC_MAC_STATE_QUEUE_STEERING) { > + u8 queue = adapter->mac_table[index].queue; > + u32 qsel = IGC_RAH_QSEL_MASK & (queue << IGC_RAH_QSEL_SHIFT); > + > + rar_high |= qsel; > + rar_high |= IGC_RAH_QSEL_ENABLE; > + } > + > /* Indicate to hardware the Address is Valid. */ > if (adapter->mac_table[index].state & IGC_MAC_STATE_IN_USE) { > if (is_valid_ether_addr(addr)) > rar_high |= IGC_RAH_AV; > - > - rar_high |= IGC_RAH_POOL_1 << > - adapter->mac_table[index].queue; > } > > wr32(IGC_RAL(index), rar_low); > Acked-by: Sasha Neftin <sasha.neftin@intel.com>
> From: Intel-wired-lan <intel-wired-lan-bounces@osuosl.org> On Behalf Of > Andre Guedes > Sent: Friday, March 6, 2020 3:54 PM > To: intel-wired-lan@lists.osuosl.org > Subject: [Intel-wired-lan] [PATCH] igc: Fix NFC queue redirection support > > The support for ethtool Network Flow Classification (NFC) queue > redirection based on destination MAC address is currently broken in IGC. > For instance, if we add the following rule, matching frames aren't > enqueued on the expected rx queue. > > $ ethtool -N IFNAME flow-type ether dst 3c:fd:fe:9e:7f:71 queue 2 > > The issue here is due to the fact that igc_rar_set_index() is missing > code to enable the queue selection feature from Receive Address High > (RAH) register. This patch adds the missing code and fixes the issue. > > Signed-off-by: Andre Guedes <andre.guedes@intel.com> > --- > drivers/net/ethernet/intel/igc/igc_defines.h | 5 ++++- > drivers/net/ethernet/intel/igc/igc_main.c | 11 ++++++++--- > 2 files changed, 12 insertions(+), 4 deletions(-) Tested-by: Aaron Brown <aaron.f.brown@intel.com>
diff --git a/drivers/net/ethernet/intel/igc/igc_defines.h b/drivers/net/ethernet/intel/igc/igc_defines.h index e5116337b68d..63d3d34763da 100644 --- a/drivers/net/ethernet/intel/igc/igc_defines.h +++ b/drivers/net/ethernet/intel/igc/igc_defines.h @@ -63,8 +63,11 @@ * (RAR[15]) for our directed address used by controllers with * manageability enabled, allowing us room for 15 multicast addresses. */ +#define IGC_RAH_QSEL_MASK 0x000C0000 +#define IGC_RAH_QSEL_SHIFT 18 +#define IGC_RAH_QSEL_ENABLE BIT(28) #define IGC_RAH_AV 0x80000000 /* Receive descriptor valid */ -#define IGC_RAH_POOL_1 0x00040000 + #define IGC_RAL_MAC_ADDR_LEN 4 #define IGC_RAH_MAC_ADDR_LEN 2 diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c index 025f2e439aef..49a0d0bff727 100644 --- a/drivers/net/ethernet/intel/igc/igc_main.c +++ b/drivers/net/ethernet/intel/igc/igc_main.c @@ -780,13 +780,18 @@ static void igc_rar_set_index(struct igc_adapter *adapter, u32 index) rar_low = le32_to_cpup((__le32 *)(addr)); rar_high = le16_to_cpup((__le16 *)(addr + 4)); + if (adapter->mac_table[index].state & IGC_MAC_STATE_QUEUE_STEERING) { + u8 queue = adapter->mac_table[index].queue; + u32 qsel = IGC_RAH_QSEL_MASK & (queue << IGC_RAH_QSEL_SHIFT); + + rar_high |= qsel; + rar_high |= IGC_RAH_QSEL_ENABLE; + } + /* Indicate to hardware the Address is Valid. */ if (adapter->mac_table[index].state & IGC_MAC_STATE_IN_USE) { if (is_valid_ether_addr(addr)) rar_high |= IGC_RAH_AV; - - rar_high |= IGC_RAH_POOL_1 << - adapter->mac_table[index].queue; } wr32(IGC_RAL(index), rar_low);
The support for ethtool Network Flow Classification (NFC) queue redirection based on destination MAC address is currently broken in IGC. For instance, if we add the following rule, matching frames aren't enqueued on the expected rx queue. $ ethtool -N IFNAME flow-type ether dst 3c:fd:fe:9e:7f:71 queue 2 The issue here is due to the fact that igc_rar_set_index() is missing code to enable the queue selection feature from Receive Address High (RAH) register. This patch adds the missing code and fixes the issue. Signed-off-by: Andre Guedes <andre.guedes@intel.com> --- drivers/net/ethernet/intel/igc/igc_defines.h | 5 ++++- drivers/net/ethernet/intel/igc/igc_main.c | 11 ++++++++--- 2 files changed, 12 insertions(+), 4 deletions(-)