Message ID | 20171211183659.21524.88045.stgit@localhost6.localdomain6 |
---|---|
State | Accepted |
Delegated to: | Jeff Kirsher |
Headers | show |
Series | ixgbevf: update Rx/Tx code path for build_skb | expand |
> -----Original Message----- > From: Intel-wired-lan [mailto:intel-wired-lan-bounces@osuosl.org] On Behalf > Of Emil Tantilov > Sent: Monday, December 11, 2017 10:37 AM > To: intel-wired-lan@lists.osuosl.org > Subject: [Intel-wired-lan] [PATCH 3/9] ixgbevf: use length to determine if > descriptor is done > > Based on: > commit 7ec0116c9131 ("igb: Use length to determine if descriptor is done") > > This change makes it so that we use the length of the packet instead of the > DD status bit to determine if a new descriptor is ready to be processed. > The obvious advantage is that it cuts down on reads as we don't really even > need the DD bit if going from a 0 to a non-zero value on size is enough to > inform us that the packet has been completed. > > In addition we only reset the Rx descriptor length for descriptor zero when > resetting a ring instead of having to do a memset with 0 over the entire > ring. By doing this we can save some time on initialization. > > Signed-off-by: Emil Tantilov <emil.s.tantilov@intel.com> > --- Tested-by: Krishneil Singh <krishneil.k.singh@intel.com>
> -----Original Message----- > From: Intel-wired-lan [mailto:intel-wired-lan-bounces@osuosl.org] On Behalf Of > Emil Tantilov > Sent: Monday, December 11, 2017 10:37 AM > To: intel-wired-lan@lists.osuosl.org > Subject: [Intel-wired-lan] [PATCH 3/9] ixgbevf: use length to determine if > descriptor is done > > Based on: > commit 7ec0116c9131 ("igb: Use length to determine if descriptor is done") > > This change makes it so that we use the length of the packet instead of the > DD status bit to determine if a new descriptor is ready to be processed. > The obvious advantage is that it cuts down on reads as we don't really even > need the DD bit if going from a 0 to a non-zero value on size is enough to > inform us that the packet has been completed. > > In addition we only reset the Rx descriptor length for descriptor zero when > resetting a ring instead of having to do a memset with 0 over the entire > ring. By doing this we can save some time on initialization. > > Signed-off-by: Emil Tantilov <emil.s.tantilov@intel.com> > --- > drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c | 14 ++++++++------ > 1 file changed, 8 insertions(+), 6 deletions(-) > > diff --git a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c > b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c > index 0cc2688..725fe2d 100644 > --- a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c > +++ b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c > @@ -653,8 +653,8 @@ static void ixgbevf_alloc_rx_buffers(struct ixgbevf_ring > *rx_ring, > i -= rx_ring->count; > } > > - /* clear the hdr_addr for the next_to_use descriptor */ > - rx_desc->read.hdr_addr = 0; > + /* clear the length for the next_to_use descriptor */ > + rx_desc->wb.upper.length = 0; > > cleaned_count--; > } while (cleaned_count); > @@ -938,7 +938,7 @@ static int ixgbevf_clean_rx_irq(struct ixgbevf_q_vector > *q_vector, > > rx_desc = IXGBEVF_RX_DESC(rx_ring, rx_ring->next_to_clean); > > - if (!ixgbevf_test_staterr(rx_desc, IXGBE_RXD_STAT_DD)) > + if (!rx_desc->wb.upper.length) > break; > > /* This memory barrier is needed to keep us from reading > @@ -1729,6 +1729,7 @@ static void ixgbevf_configure_rx_ring(struct > ixgbevf_adapter *adapter, > struct ixgbevf_ring *ring) > { > struct ixgbe_hw *hw = &adapter->hw; > + union ixgbe_adv_rx_desc *rx_desc; > u64 rdba = ring->dma; > u32 rxdctl; > u8 reg_idx = ring->reg_idx; > @@ -1757,6 +1758,10 @@ static void ixgbevf_configure_rx_ring(struct > ixgbevf_adapter *adapter, > IXGBE_WRITE_REG(hw, IXGBE_VFRDT(reg_idx), 0); > ring->tail = adapter->io_addr + IXGBE_VFRDT(reg_idx); > > + /* initialize Rx descriptor 0 */ > + rx_desc = IXGBEVF_RX_DESC(ring, 0); > + rx_desc->wb.upper.length = 0; > + > /* reset ntu and ntc to place SW in sync with hardwdare */ > ring->next_to_clean = 0; > ring->next_to_use = 0; > @@ -2141,9 +2146,6 @@ static void ixgbevf_clean_rx_ring(struct ixgbevf_ring > *rx_ring) > > size = sizeof(struct ixgbevf_rx_buffer) * rx_ring->count; > memset(rx_ring->rx_buffer_info, 0, size); > - > - /* Zero out the descriptor ring */ > - memset(rx_ring->desc, 0, rx_ring->size); > } > > /** > > _______________________________________________ > Intel-wired-lan mailing list > Intel-wired-lan@osuosl.org > https://lists.osuosl.org/mailman/listinfo/intel-wired-lan Tested-by: Krishneil Singh <krishneil.k.singh@intel.com>
diff --git a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c index 0cc2688..725fe2d 100644 --- a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c +++ b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c @@ -653,8 +653,8 @@ static void ixgbevf_alloc_rx_buffers(struct ixgbevf_ring *rx_ring, i -= rx_ring->count; } - /* clear the hdr_addr for the next_to_use descriptor */ - rx_desc->read.hdr_addr = 0; + /* clear the length for the next_to_use descriptor */ + rx_desc->wb.upper.length = 0; cleaned_count--; } while (cleaned_count); @@ -938,7 +938,7 @@ static int ixgbevf_clean_rx_irq(struct ixgbevf_q_vector *q_vector, rx_desc = IXGBEVF_RX_DESC(rx_ring, rx_ring->next_to_clean); - if (!ixgbevf_test_staterr(rx_desc, IXGBE_RXD_STAT_DD)) + if (!rx_desc->wb.upper.length) break; /* This memory barrier is needed to keep us from reading @@ -1729,6 +1729,7 @@ static void ixgbevf_configure_rx_ring(struct ixgbevf_adapter *adapter, struct ixgbevf_ring *ring) { struct ixgbe_hw *hw = &adapter->hw; + union ixgbe_adv_rx_desc *rx_desc; u64 rdba = ring->dma; u32 rxdctl; u8 reg_idx = ring->reg_idx; @@ -1757,6 +1758,10 @@ static void ixgbevf_configure_rx_ring(struct ixgbevf_adapter *adapter, IXGBE_WRITE_REG(hw, IXGBE_VFRDT(reg_idx), 0); ring->tail = adapter->io_addr + IXGBE_VFRDT(reg_idx); + /* initialize Rx descriptor 0 */ + rx_desc = IXGBEVF_RX_DESC(ring, 0); + rx_desc->wb.upper.length = 0; + /* reset ntu and ntc to place SW in sync with hardwdare */ ring->next_to_clean = 0; ring->next_to_use = 0; @@ -2141,9 +2146,6 @@ static void ixgbevf_clean_rx_ring(struct ixgbevf_ring *rx_ring) size = sizeof(struct ixgbevf_rx_buffer) * rx_ring->count; memset(rx_ring->rx_buffer_info, 0, size); - - /* Zero out the descriptor ring */ - memset(rx_ring->desc, 0, rx_ring->size); } /**
Based on: commit 7ec0116c9131 ("igb: Use length to determine if descriptor is done") This change makes it so that we use the length of the packet instead of the DD status bit to determine if a new descriptor is ready to be processed. The obvious advantage is that it cuts down on reads as we don't really even need the DD bit if going from a 0 to a non-zero value on size is enough to inform us that the packet has been completed. In addition we only reset the Rx descriptor length for descriptor zero when resetting a ring instead of having to do a memset with 0 over the entire ring. By doing this we can save some time on initialization. Signed-off-by: Emil Tantilov <emil.s.tantilov@intel.com> --- drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-)