* [PATCH 1/2] bnx2x: resurrect RX hashing
@ 2011-08-23 16:15 Michal Schmidt
2011-08-23 16:15 ` [PATCH 2/2] bnx2x: expose HW RX VLAN stripping toggle Michal Schmidt
` (2 more replies)
0 siblings, 3 replies; 5+ messages in thread
From: Michal Schmidt @ 2011-08-23 16:15 UTC (permalink / raw)
To: netdev
Cc: Michal Schmidt, Vladislav Zolotarov, Eilon Greenstein,
Dmitry Kravkov, MichałMirosław
bnx2x used to be able to set rxhash, but this was lost in the conversion
to hw_features (commit 66371c441).
Restore it and enable it by default.
Signed-off-by: Michal Schmidt <mschmidt@redhat.com>
CC: Vladislav Zolotarov <vladz@broadcom.com>
CC: Eilon Greenstein <eilong@broadcom.com>
CC: Dmitry Kravkov <dmitry@broadcom.com>
CC: Michał Mirosław <mirq-linux@rere.qmqm.pl>
---
drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c | 4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
index 7204789..85dd294 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
@@ -10285,8 +10285,8 @@ static int __devinit bnx2x_init_dev(struct pci_dev *pdev,
dev->priv_flags |= IFF_UNICAST_FLT;
dev->hw_features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
- NETIF_F_TSO | NETIF_F_TSO_ECN | NETIF_F_TSO6 |
- NETIF_F_RXCSUM | NETIF_F_LRO | NETIF_F_HW_VLAN_TX;
+ NETIF_F_TSO | NETIF_F_TSO_ECN | NETIF_F_TSO6 | NETIF_F_LRO |
+ NETIF_F_RXCSUM | NETIF_F_RXHASH | NETIF_F_HW_VLAN_TX;
dev->vlan_features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
NETIF_F_TSO | NETIF_F_TSO_ECN | NETIF_F_TSO6 | NETIF_F_HIGHDMA;
^ permalink raw reply related [flat|nested] 5+ messages in thread* [PATCH 2/2] bnx2x: expose HW RX VLAN stripping toggle 2011-08-23 16:15 [PATCH 1/2] bnx2x: resurrect RX hashing Michal Schmidt @ 2011-08-23 16:15 ` Michal Schmidt 2011-08-24 11:16 ` Vlad Zolotarov 2011-08-24 11:18 ` [PATCH 1/2] bnx2x: resurrect RX hashing Vlad Zolotarov 2011-08-26 16:51 ` David Miller 2 siblings, 1 reply; 5+ messages in thread From: Michal Schmidt @ 2011-08-23 16:15 UTC (permalink / raw) To: netdev Cc: Michal Schmidt, Vladislav Zolotarov, Eilon Greenstein, Dmitry Kravkov Allow disabling of HW RX VLAN stripping with ethtool. Signed-off-by: Michal Schmidt <mschmidt@redhat.com> CC: Vladislav Zolotarov <vladz@broadcom.com> CC: Eilon Greenstein <eilong@broadcom.com> CC: Dmitry Kravkov <dmitry@broadcom.com> --- drivers/net/ethernet/broadcom/bnx2x/bnx2x.h | 1 + drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c | 18 ++++++++++++++---- drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c | 10 +++++----- 3 files changed, 20 insertions(+), 9 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h index f127768..dd43634 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h @@ -1178,6 +1178,7 @@ struct bnx2x { #define NO_MCP_FLAG (1 << 9) #define BP_NOMCP(bp) (bp->flags & NO_MCP_FLAG) +#define RX_VLAN_STRIP_FLAG (1 << 10) #define MF_FUNC_DIS (1 << 11) #define OWN_CNIC_IRQ (1 << 12) #define NO_ISCSI_OOO_FLAG (1 << 13) diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c index 93bff08..340bff9 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c @@ -385,6 +385,10 @@ static inline u16 bnx2x_set_lro_mss(struct bnx2x *bp, u16 parsing_flags, else /* IPv4 */ hdrs_len += sizeof(struct iphdr); + /* VLAN header present and not stripped by HW */ + if ((parsing_flags & PARSING_FLAGS_VLAN) && + !(bp->flags & RX_VLAN_STRIP_FLAG)) + hdrs_len += VLAN_HLEN; /* Check if there was a TCP timestamp, if there is it's will * always be 12 bytes length: nop nop kind length echo val. @@ -412,7 +416,7 @@ static int bnx2x_fill_frag_skb(struct bnx2x *bp, struct bnx2x_fastpath *fp, frag_size = le16_to_cpu(cqe->pkt_len) - len_on_bd; pages = SGE_PAGE_ALIGN(frag_size) >> SGE_PAGE_SHIFT; - /* This is needed in order to enable forwarding support */ + /* Doing LRO, let TCP know the receive MSS */ if (frag_size) skb_shinfo(skb)->gso_size = bnx2x_set_lro_mss(bp, tpa_info->parsing_flags, len_on_bd); @@ -514,7 +518,8 @@ static void bnx2x_tpa_stop(struct bnx2x *bp, struct bnx2x_fastpath *fp, skb->ip_summed = CHECKSUM_UNNECESSARY; if (!bnx2x_fill_frag_skb(bp, fp, queue, skb, cqe, cqe_idx)) { - if (tpa_info->parsing_flags & PARSING_FLAGS_VLAN) + if ((tpa_info->parsing_flags & PARSING_FLAGS_VLAN) && + (bp->flags & RX_VLAN_STRIP_FLAG)) __vlan_hwaccel_put_tag(skb, tpa_info->vlan_tag); napi_gro_receive(&fp->napi, skb); } else { @@ -757,8 +762,8 @@ reuse_rx: skb_record_rx_queue(skb, fp->index); - if (le16_to_cpu(cqe_fp->pars_flags.flags) & - PARSING_FLAGS_VLAN) + if ((le16_to_cpu(cqe_fp->pars_flags.flags) & + PARSING_FLAGS_VLAN) && (bp->flags & RX_VLAN_STRIP_FLAG)) __vlan_hwaccel_put_tag(skb, le16_to_cpu(cqe_fp->vlan_tag)); napi_gro_receive(&fp->napi, skb); @@ -3431,6 +3436,11 @@ int bnx2x_set_features(struct net_device *dev, u32 features) else flags &= ~TPA_ENABLE_FLAG; + if (features & NETIF_F_HW_VLAN_RX) + flags |= RX_VLAN_STRIP_FLAG; + else + flags &= ~RX_VLAN_STRIP_FLAG; + if (features & NETIF_F_LOOPBACK) { if (bp->link_params.loopback_mode != LOOPBACK_BMAC) { bp->link_params.loopback_mode = LOOPBACK_BMAC; diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c index 85dd294..e444e7d 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c @@ -2722,9 +2722,8 @@ static inline unsigned long bnx2x_get_q_flags(struct bnx2x *bp, __set_bit(BNX2X_Q_FLG_MCAST, &flags); } - /* Always set HW VLAN stripping */ - __set_bit(BNX2X_Q_FLG_VLAN, &flags); - + if (bp->flags & RX_VLAN_STRIP_FLAG) + __set_bit(BNX2X_Q_FLG_VLAN, &flags); return flags | bnx2x_get_common_flags(bp, fp, true); } @@ -10286,12 +10285,13 @@ static int __devinit bnx2x_init_dev(struct pci_dev *pdev, dev->hw_features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | NETIF_F_TSO | NETIF_F_TSO_ECN | NETIF_F_TSO6 | NETIF_F_LRO | - NETIF_F_RXCSUM | NETIF_F_RXHASH | NETIF_F_HW_VLAN_TX; + NETIF_F_RXCSUM | NETIF_F_RXHASH | + NETIF_F_HW_VLAN_TX | NETIF_F_HW_VLAN_RX; dev->vlan_features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | NETIF_F_TSO | NETIF_F_TSO_ECN | NETIF_F_TSO6 | NETIF_F_HIGHDMA; - dev->features |= dev->hw_features | NETIF_F_HW_VLAN_RX; + dev->features |= dev->hw_features; if (bp->flags & USING_DAC_FLAG) dev->features |= NETIF_F_HIGHDMA; ^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH 2/2] bnx2x: expose HW RX VLAN stripping toggle 2011-08-23 16:15 ` [PATCH 2/2] bnx2x: expose HW RX VLAN stripping toggle Michal Schmidt @ 2011-08-24 11:16 ` Vlad Zolotarov 0 siblings, 0 replies; 5+ messages in thread From: Vlad Zolotarov @ 2011-08-24 11:16 UTC (permalink / raw) To: Michal Schmidt; +Cc: netdev@vger.kernel.org, Eilon Greenstein, Dmitry Kravkov On Tuesday 23 August 2011 19:15:39 Michal Schmidt wrote: > Allow disabling of HW RX VLAN stripping with ethtool. > > Signed-off-by: Michal Schmidt <mschmidt@redhat.com> > CC: Vladislav Zolotarov <vladz@broadcom.com> > CC: Eilon Greenstein <eilong@broadcom.com> > CC: Dmitry Kravkov <dmitry@broadcom.com> > --- > > drivers/net/ethernet/broadcom/bnx2x/bnx2x.h | 1 + > drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c | 18 ++++++++++++++---- > drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c | 10 +++++----- > 3 files changed, 20 insertions(+), 9 deletions(-) > > diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h > b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h index f127768..dd43634 > 100644 > --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h > +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h > @@ -1178,6 +1178,7 @@ struct bnx2x { > #define NO_MCP_FLAG (1 << 9) > > #define BP_NOMCP(bp) (bp->flags & NO_MCP_FLAG) > +#define RX_VLAN_STRIP_FLAG (1 << 10) > #define MF_FUNC_DIS (1 << 11) > #define OWN_CNIC_IRQ (1 << 12) > #define NO_ISCSI_OOO_FLAG (1 << 13) > diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c > b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c index 93bff08..340bff9 > 100644 > --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c > +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c > @@ -385,6 +385,10 @@ static inline u16 bnx2x_set_lro_mss(struct bnx2x *bp, > u16 parsing_flags, else /* IPv4 */ > hdrs_len += sizeof(struct iphdr); > > + /* VLAN header present and not stripped by HW */ > + if ((parsing_flags & PARSING_FLAGS_VLAN) && > + !(bp->flags & RX_VLAN_STRIP_FLAG)) > + hdrs_len += VLAN_HLEN; > > /* Check if there was a TCP timestamp, if there is it's will > * always be 12 bytes length: nop nop kind length echo val. > @@ -412,7 +416,7 @@ static int bnx2x_fill_frag_skb(struct bnx2x *bp, struct > bnx2x_fastpath *fp, frag_size = le16_to_cpu(cqe->pkt_len) - len_on_bd; > pages = SGE_PAGE_ALIGN(frag_size) >> SGE_PAGE_SHIFT; > > - /* This is needed in order to enable forwarding support */ > + /* Doing LRO, let TCP know the receive MSS */ > if (frag_size) > skb_shinfo(skb)->gso_size = bnx2x_set_lro_mss(bp, > tpa_info->parsing_flags, len_on_bd); > @@ -514,7 +518,8 @@ static void bnx2x_tpa_stop(struct bnx2x *bp, struct > bnx2x_fastpath *fp, skb->ip_summed = CHECKSUM_UNNECESSARY; > > if (!bnx2x_fill_frag_skb(bp, fp, queue, skb, cqe, cqe_idx)) { > - if (tpa_info->parsing_flags & PARSING_FLAGS_VLAN) > + if ((tpa_info->parsing_flags & PARSING_FLAGS_VLAN) && > + (bp->flags & RX_VLAN_STRIP_FLAG)) > __vlan_hwaccel_put_tag(skb, tpa_info- >vlan_tag); > napi_gro_receive(&fp->napi, skb); > } else { > @@ -757,8 +762,8 @@ reuse_rx: > > skb_record_rx_queue(skb, fp->index); > > - if (le16_to_cpu(cqe_fp->pars_flags.flags) & > - PARSING_FLAGS_VLAN) > + if ((le16_to_cpu(cqe_fp->pars_flags.flags) & > + PARSING_FLAGS_VLAN) && (bp->flags & RX_VLAN_STRIP_FLAG)) > __vlan_hwaccel_put_tag(skb, > le16_to_cpu(cqe_fp->vlan_tag)); > napi_gro_receive(&fp->napi, skb); > @@ -3431,6 +3436,11 @@ int bnx2x_set_features(struct net_device *dev, u32 > features) else > flags &= ~TPA_ENABLE_FLAG; > > + if (features & NETIF_F_HW_VLAN_RX) > + flags |= RX_VLAN_STRIP_FLAG; > + else > + flags &= ~RX_VLAN_STRIP_FLAG; > + > if (features & NETIF_F_LOOPBACK) { > if (bp->link_params.loopback_mode != LOOPBACK_BMAC) { > bp->link_params.loopback_mode = LOOPBACK_BMAC; > diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c > b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c index 85dd294..e444e7d > 100644 > --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c > +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c > @@ -2722,9 +2722,8 @@ static inline unsigned long bnx2x_get_q_flags(struct > bnx2x *bp, __set_bit(BNX2X_Q_FLG_MCAST, &flags); > } > > - /* Always set HW VLAN stripping */ > - __set_bit(BNX2X_Q_FLG_VLAN, &flags); > - > + if (bp->flags & RX_VLAN_STRIP_FLAG) > + __set_bit(BNX2X_Q_FLG_VLAN, &flags); > > return flags | bnx2x_get_common_flags(bp, fp, true); > } > @@ -10286,12 +10285,13 @@ static int __devinit bnx2x_init_dev(struct > pci_dev *pdev, > > dev->hw_features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | > NETIF_F_TSO | NETIF_F_TSO_ECN | NETIF_F_TSO6 | NETIF_F_LRO | > - NETIF_F_RXCSUM | NETIF_F_RXHASH | NETIF_F_HW_VLAN_TX; > + NETIF_F_RXCSUM | NETIF_F_RXHASH | > + NETIF_F_HW_VLAN_TX | NETIF_F_HW_VLAN_RX; > > dev->vlan_features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | > NETIF_F_TSO | NETIF_F_TSO_ECN | NETIF_F_TSO6 | NETIF_F_HIGHDMA; > > - dev->features |= dev->hw_features | NETIF_F_HW_VLAN_RX; > + dev->features |= dev->hw_features; > if (bp->flags & USING_DAC_FLAG) > dev->features |= NETIF_F_HIGHDMA; NACK as u set the RX_VLAN_STRIP_FLAG in bp->flags before u reset the function, thus there is a real posibility that u would handle the pending packets (during unload) under the new stripping policy while it's not yet configured in the HW. thanks, vlad ^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH 1/2] bnx2x: resurrect RX hashing 2011-08-23 16:15 [PATCH 1/2] bnx2x: resurrect RX hashing Michal Schmidt 2011-08-23 16:15 ` [PATCH 2/2] bnx2x: expose HW RX VLAN stripping toggle Michal Schmidt @ 2011-08-24 11:18 ` Vlad Zolotarov 2011-08-26 16:51 ` David Miller 2 siblings, 0 replies; 5+ messages in thread From: Vlad Zolotarov @ 2011-08-24 11:18 UTC (permalink / raw) To: Michal Schmidt Cc: netdev@vger.kernel.org, Eilon Greenstein, Dmitry Kravkov, MichałMirosław On Tuesday 23 August 2011 19:15:32 Michal Schmidt wrote: > bnx2x used to be able to set rxhash, but this was lost in the conversion > to hw_features (commit 66371c441). > Restore it and enable it by default. > > Signed-off-by: Michal Schmidt <mschmidt@redhat.com> > CC: Vladislav Zolotarov <vladz@broadcom.com> > CC: Eilon Greenstein <eilong@broadcom.com> > CC: Dmitry Kravkov <dmitry@broadcom.com> > CC: Michał Mirosław <mirq-linux@rere.qmqm.pl> > --- > > drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c | 4 ++-- > 1 files changed, 2 insertions(+), 2 deletions(-) > > diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c > b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c index 7204789..85dd294 > 100644 > --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c > +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c > @@ -10285,8 +10285,8 @@ static int __devinit bnx2x_init_dev(struct pci_dev > *pdev, dev->priv_flags |= IFF_UNICAST_FLT; > > dev->hw_features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | > - NETIF_F_TSO | NETIF_F_TSO_ECN | NETIF_F_TSO6 | > - NETIF_F_RXCSUM | NETIF_F_LRO | NETIF_F_HW_VLAN_TX; > + NETIF_F_TSO | NETIF_F_TSO_ECN | NETIF_F_TSO6 | NETIF_F_LRO | > + NETIF_F_RXCSUM | NETIF_F_RXHASH | NETIF_F_HW_VLAN_TX; > > dev->vlan_features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | > NETIF_F_TSO | NETIF_F_TSO_ECN | NETIF_F_TSO6 | NETIF_F_HIGHDMA; ACK Thanks, Michal. vlad ^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH 1/2] bnx2x: resurrect RX hashing 2011-08-23 16:15 [PATCH 1/2] bnx2x: resurrect RX hashing Michal Schmidt 2011-08-23 16:15 ` [PATCH 2/2] bnx2x: expose HW RX VLAN stripping toggle Michal Schmidt 2011-08-24 11:18 ` [PATCH 1/2] bnx2x: resurrect RX hashing Vlad Zolotarov @ 2011-08-26 16:51 ` David Miller 2 siblings, 0 replies; 5+ messages in thread From: David Miller @ 2011-08-26 16:51 UTC (permalink / raw) To: mschmidt; +Cc: netdev, vladz, eilong, dmitry, mirq-linux From: Michal Schmidt <mschmidt@redhat.com> Date: Tue, 23 Aug 2011 18:15:32 +0200 > bnx2x used to be able to set rxhash, but this was lost in the conversion > to hw_features (commit 66371c441). > Restore it and enable it by default. > > Signed-off-by: Michal Schmidt <mschmidt@redhat.com> Applied. ^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2011-08-26 16:51 UTC | newest] Thread overview: 5+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2011-08-23 16:15 [PATCH 1/2] bnx2x: resurrect RX hashing Michal Schmidt 2011-08-23 16:15 ` [PATCH 2/2] bnx2x: expose HW RX VLAN stripping toggle Michal Schmidt 2011-08-24 11:16 ` Vlad Zolotarov 2011-08-24 11:18 ` [PATCH 1/2] bnx2x: resurrect RX hashing Vlad Zolotarov 2011-08-26 16:51 ` David Miller
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).