| Message ID | 20210923140813.13541-13-biju.das.jz@bp.renesas.com (mailing list archive) |
|---|---|
| State | Superseded |
| Delegated to: | Geert Uytterhoeven |
| Headers | show |
| Series | Add Gigabit Ethernet driver support | expand |
On 9/23/21 5:08 PM, Biju Das wrote: > R-Car AVB-DMAC supports timestamp feature. > Add a timestamp hw feature bit to struct ravb_hw_info > to add this feature only for R-Car. > > Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com> > --- > drivers/net/ethernet/renesas/ravb.h | 2 + > drivers/net/ethernet/renesas/ravb_main.c | 68 +++++++++++++++--------- > 2 files changed, 45 insertions(+), 25 deletions(-) > > diff --git a/drivers/net/ethernet/renesas/ravb.h b/drivers/net/ethernet/renesas/ravb.h > index ab4909244276..2505de5d4a28 100644 > --- a/drivers/net/ethernet/renesas/ravb.h > +++ b/drivers/net/ethernet/renesas/ravb.h > @@ -1034,6 +1034,7 @@ struct ravb_hw_info { > unsigned mii_rgmii_selection:1; /* E-MAC supports mii/rgmii selection */ > unsigned half_duplex:1; /* E-MAC supports half duplex mode */ > unsigned rx_2k_buffers:1; /* AVB-DMAC has Max 2K buf size on RX */ > + unsigned timestamp:1; /* AVB-DMAC has timestamp */ Isn't this a matter of the gPTP support as well, i.e. no separate flag needed? [...] > @@ -1089,6 +1090,7 @@ struct ravb_private { > unsigned int num_tx_desc; /* TX descriptors per packet */ > > int duplex; > + struct ravb_rx_desc *rgeth_rx_ring[NUM_RX_QUEUE]; Strange place to declare this... > > const struct ravb_hw_info *info; > struct reset_control *rstc; > diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c > index 9c0d35f4b221..2c375002ebcb 100644 > --- a/drivers/net/ethernet/renesas/ravb_main.c > +++ b/drivers/net/ethernet/renesas/ravb_main.c > @@ -949,11 +949,14 @@ static bool ravb_queue_interrupt(struct net_device *ndev, int q) > > static bool ravb_timestamp_interrupt(struct net_device *ndev) > { > + struct ravb_private *priv = netdev_priv(ndev); > + const struct ravb_hw_info *info = priv->info; > u32 tis = ravb_read(ndev, TIS); > > if (tis & TIS_TFUF) { > ravb_write(ndev, ~(TIS_TFUF | TIS_RESERVED), TIS); > - ravb_get_tx_tstamp(ndev); > + if (info->timestamp) > + ravb_get_tx_tstamp(ndev); Shouldn't we just disable TIS.TFUF permanently instead for the non-gPTP case? [...] MBR, Sergey
HI Sergei, Thanks for the feedback. > -----Original Message----- > Subject: Re: [RFC/PATCH 12/18] ravb: Add timestamp to struct ravb_hw_info > > On 9/23/21 5:08 PM, Biju Das wrote: > > > R-Car AVB-DMAC supports timestamp feature. > > Add a timestamp hw feature bit to struct ravb_hw_info to add this > > feature only for R-Car. > > > > Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com> > > --- > > drivers/net/ethernet/renesas/ravb.h | 2 + > > drivers/net/ethernet/renesas/ravb_main.c | 68 > > +++++++++++++++--------- > > 2 files changed, 45 insertions(+), 25 deletions(-) > > > > diff --git a/drivers/net/ethernet/renesas/ravb.h > > b/drivers/net/ethernet/renesas/ravb.h > > index ab4909244276..2505de5d4a28 100644 > > --- a/drivers/net/ethernet/renesas/ravb.h > > +++ b/drivers/net/ethernet/renesas/ravb.h > > @@ -1034,6 +1034,7 @@ struct ravb_hw_info { > > unsigned mii_rgmii_selection:1; /* E-MAC supports mii/rgmii > selection */ > > unsigned half_duplex:1; /* E-MAC supports half duplex mode */ > > unsigned rx_2k_buffers:1; /* AVB-DMAC has Max 2K buf size on RX > */ > > + unsigned timestamp:1; /* AVB-DMAC has timestamp */ > > Isn't this a matter of the gPTP support as well, i.e. no separate flag > needed? Agreed. Previously it is suggested to use timestamp. I will change it to as part of gPTP support cases. > > [...] > > @@ -1089,6 +1090,7 @@ struct ravb_private { > > unsigned int num_tx_desc; /* TX descriptors per packet */ > > > > int duplex; > > + struct ravb_rx_desc *rgeth_rx_ring[NUM_RX_QUEUE]; > > Strange place to declare this... Agreed. This has to be on later patch. Will move it. > > > > > const struct ravb_hw_info *info; > > struct reset_control *rstc; > > diff --git a/drivers/net/ethernet/renesas/ravb_main.c > > b/drivers/net/ethernet/renesas/ravb_main.c > > index 9c0d35f4b221..2c375002ebcb 100644 > > --- a/drivers/net/ethernet/renesas/ravb_main.c > > +++ b/drivers/net/ethernet/renesas/ravb_main.c > > @@ -949,11 +949,14 @@ static bool ravb_queue_interrupt(struct > > net_device *ndev, int q) > > > > static bool ravb_timestamp_interrupt(struct net_device *ndev) { > > + struct ravb_private *priv = netdev_priv(ndev); > > + const struct ravb_hw_info *info = priv->info; > > u32 tis = ravb_read(ndev, TIS); > > > > if (tis & TIS_TFUF) { > > ravb_write(ndev, ~(TIS_TFUF | TIS_RESERVED), TIS); > > - ravb_get_tx_tstamp(ndev); > > + if (info->timestamp) > > + ravb_get_tx_tstamp(ndev); > > Shouldn't we just disable TIS.TFUF permanently instead for the non-gPTP > case? Good catch. As ravb_dmac_init_rgeth(will be renamed to "ravb_dmac_init_gbeth") is not enabling this interrupt as it is not documented in RZ/G2L hardware manual. So this function never gets called for non-gPTP case. I will remove this check. Regards, Biju > > [...] > > MBR, Sergey
Hi Sergei, > Subject: RE: [RFC/PATCH 12/18] ravb: Add timestamp to struct ravb_hw_info > > HI Sergei, > > Thanks for the feedback. > > > -----Original Message----- > > Subject: Re: [RFC/PATCH 12/18] ravb: Add timestamp to struct > > ravb_hw_info > > > > On 9/23/21 5:08 PM, Biju Das wrote: > > > > > R-Car AVB-DMAC supports timestamp feature. > > > Add a timestamp hw feature bit to struct ravb_hw_info to add this > > > feature only for R-Car. > > > > > > Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com> > > > --- > > > drivers/net/ethernet/renesas/ravb.h | 2 + > > > drivers/net/ethernet/renesas/ravb_main.c | 68 > > > +++++++++++++++--------- > > > 2 files changed, 45 insertions(+), 25 deletions(-) > > > > > > diff --git a/drivers/net/ethernet/renesas/ravb.h > > > b/drivers/net/ethernet/renesas/ravb.h > > > index ab4909244276..2505de5d4a28 100644 > > > --- a/drivers/net/ethernet/renesas/ravb.h > > > +++ b/drivers/net/ethernet/renesas/ravb.h > > > @@ -1034,6 +1034,7 @@ struct ravb_hw_info { > > > unsigned mii_rgmii_selection:1; /* E-MAC supports mii/rgmii > > selection */ > > > unsigned half_duplex:1; /* E-MAC supports half duplex mode */ > > > unsigned rx_2k_buffers:1; /* AVB-DMAC has Max 2K buf size on RX > > */ > > > + unsigned timestamp:1; /* AVB-DMAC has timestamp */ > > > > Isn't this a matter of the gPTP support as well, i.e. no separate > > flag needed? > > Agreed. Previously it is suggested to use timestamp. I will change it to > as part of gPTP support cases. > > > > > [...] > > > @@ -1089,6 +1090,7 @@ struct ravb_private { > > > unsigned int num_tx_desc; /* TX descriptors per packet */ > > > > > > int duplex; > > > + struct ravb_rx_desc *rgeth_rx_ring[NUM_RX_QUEUE]; > > > > Strange place to declare this... > > Agreed. This has to be on later patch. Will move it. > > > > > > > > > const struct ravb_hw_info *info; > > > struct reset_control *rstc; > > > diff --git a/drivers/net/ethernet/renesas/ravb_main.c > > > b/drivers/net/ethernet/renesas/ravb_main.c > > > index 9c0d35f4b221..2c375002ebcb 100644 > > > --- a/drivers/net/ethernet/renesas/ravb_main.c > > > +++ b/drivers/net/ethernet/renesas/ravb_main.c > > > @@ -949,11 +949,14 @@ static bool ravb_queue_interrupt(struct > > > net_device *ndev, int q) > > > > > > static bool ravb_timestamp_interrupt(struct net_device *ndev) { > > > + struct ravb_private *priv = netdev_priv(ndev); > > > + const struct ravb_hw_info *info = priv->info; > > > u32 tis = ravb_read(ndev, TIS); > > > > > > if (tis & TIS_TFUF) { > > > ravb_write(ndev, ~(TIS_TFUF | TIS_RESERVED), TIS); > > > - ravb_get_tx_tstamp(ndev); > > > + if (info->timestamp) > > > + ravb_get_tx_tstamp(ndev); > > > > Shouldn't we just disable TIS.TFUF permanently instead for the > > non-gPTP case? > > Good catch. As ravb_dmac_init_rgeth(will be renamed to > "ravb_dmac_init_gbeth") is not enabling this interrupt as it is not > documented in RZ/G2L hardware manual. > So this function never gets called for non-gPTP case. > > I will remove this check. As discussed, I will drop this patch in the next version and contents of this patch will be added to gPTP support cases. Regards, Biju
On 9/26/21 9:34 AM, Biju Das wrote: >> -----Original Message----- >> Subject: Re: [RFC/PATCH 12/18] ravb: Add timestamp to struct ravb_hw_info >> >> On 9/23/21 5:08 PM, Biju Das wrote: >> >>> R-Car AVB-DMAC supports timestamp feature. >>> Add a timestamp hw feature bit to struct ravb_hw_info to add this >>> feature only for R-Car. >>> >>> Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com> >>> --- >>> drivers/net/ethernet/renesas/ravb.h | 2 + >>> drivers/net/ethernet/renesas/ravb_main.c | 68 >>> +++++++++++++++--------- >>> 2 files changed, 45 insertions(+), 25 deletions(-) >>> >>> diff --git a/drivers/net/ethernet/renesas/ravb.h >>> b/drivers/net/ethernet/renesas/ravb.h >>> index ab4909244276..2505de5d4a28 100644 >>> --- a/drivers/net/ethernet/renesas/ravb.h >>> +++ b/drivers/net/ethernet/renesas/ravb.h >>> @@ -1034,6 +1034,7 @@ struct ravb_hw_info { >>> unsigned mii_rgmii_selection:1; /* E-MAC supports mii/rgmii >> selection */ >>> unsigned half_duplex:1; /* E-MAC supports half duplex mode */ >>> unsigned rx_2k_buffers:1; /* AVB-DMAC has Max 2K buf size on RX >> */ >>> + unsigned timestamp:1; /* AVB-DMAC has timestamp */ >> >> Isn't this a matter of the gPTP support as well, i.e. no separate flag >> needed? > > Agreed. Previously it is suggested to use timestamp. I will change it to as part of gPTP support cases. TIA. :-) >> >> [...] >>> @@ -1089,6 +1090,7 @@ struct ravb_private { >>> unsigned int num_tx_desc; /* TX descriptors per packet */ >>> >>> int duplex; >>> + struct ravb_rx_desc *rgeth_rx_ring[NUM_RX_QUEUE]; >> >> Strange place to declare this... > > Agreed. This has to be on later patch. Will move it. I only meant that these fields should go together with rx_ring[]. Apparently we have a case of wrong patch ordering here (as this patch needs this field declared)... >>> const struct ravb_hw_info *info; >>> struct reset_control *rstc; >>> diff --git a/drivers/net/ethernet/renesas/ravb_main.c >>> b/drivers/net/ethernet/renesas/ravb_main.c >>> index 9c0d35f4b221..2c375002ebcb 100644 >>> --- a/drivers/net/ethernet/renesas/ravb_main.c >>> +++ b/drivers/net/ethernet/renesas/ravb_main.c >>> @@ -949,11 +949,14 @@ static bool ravb_queue_interrupt(struct >>> net_device *ndev, int q) >>> >>> static bool ravb_timestamp_interrupt(struct net_device *ndev) { >>> + struct ravb_private *priv = netdev_priv(ndev); >>> + const struct ravb_hw_info *info = priv->info; >>> u32 tis = ravb_read(ndev, TIS); >>> >>> if (tis & TIS_TFUF) { >>> ravb_write(ndev, ~(TIS_TFUF | TIS_RESERVED), TIS); >>> - ravb_get_tx_tstamp(ndev); >>> + if (info->timestamp) >>> + ravb_get_tx_tstamp(ndev); >> >> Shouldn't we just disable TIS.TFUF permanently instead for the non-gPTP >> case? > > Good catch. As ravb_dmac_init_rgeth(will be renamed to "ravb_dmac_init_gbeth") is not enabling this interrupt as it is not documented in RZ/G2L hardware manual. > So this function never gets called for non-gPTP case. > > I will remove this check. TIA! > Regards, > Biju MBR, Sergey
On 9/26/21 11:45 PM, Sergey Shtylyov wrote: >>> -----Original Message----- >>> Subject: Re: [RFC/PATCH 12/18] ravb: Add timestamp to struct ravb_hw_info >>> >>> On 9/23/21 5:08 PM, Biju Das wrote: >>> >>>> R-Car AVB-DMAC supports timestamp feature. >>>> Add a timestamp hw feature bit to struct ravb_hw_info to add this >>>> feature only for R-Car. >>>> >>>> Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com> >>>> --- >>>> drivers/net/ethernet/renesas/ravb.h | 2 + >>>> drivers/net/ethernet/renesas/ravb_main.c | 68 >>>> +++++++++++++++--------- >>>> 2 files changed, 45 insertions(+), 25 deletions(-) >>>> >>>> diff --git a/drivers/net/ethernet/renesas/ravb.h >>>> b/drivers/net/ethernet/renesas/ravb.h >>>> index ab4909244276..2505de5d4a28 100644 >>>> --- a/drivers/net/ethernet/renesas/ravb.h >>>> +++ b/drivers/net/ethernet/renesas/ravb.h >>>> @@ -1034,6 +1034,7 @@ struct ravb_hw_info { >>>> unsigned mii_rgmii_selection:1; /* E-MAC supports mii/rgmii >>> selection */ >>>> unsigned half_duplex:1; /* E-MAC supports half duplex mode */ >>>> unsigned rx_2k_buffers:1; /* AVB-DMAC has Max 2K buf size on RX >>> */ >>>> + unsigned timestamp:1; /* AVB-DMAC has timestamp */ >>> >>> Isn't this a matter of the gPTP support as well, i.e. no separate flag >>> needed? >> >> Agreed. Previously it is suggested to use timestamp. I will change it to as part of gPTP support cases. > > TIA. :-) > >>> >>> [...] >>>> @@ -1089,6 +1090,7 @@ struct ravb_private { >>>> unsigned int num_tx_desc; /* TX descriptors per packet */ >>>> >>>> int duplex; >>>> + struct ravb_rx_desc *rgeth_rx_ring[NUM_RX_QUEUE]; >>> >>> Strange place to declare this... >> >> Agreed. This has to be on later patch. Will move it. > > I only meant that these fields should go together with rx_ring[]. Apparently Sorry, this field. > we have a case of wrong patch ordering here (as this patch needs this field declared)... [...] >> Regards, >> Biju MBR, Sergey
Hi Sergei, > Subject: Re: [RFC/PATCH 12/18] ravb: Add timestamp to struct ravb_hw_info > > On 9/26/21 9:34 AM, Biju Das wrote: > > >> -----Original Message----- > >> Subject: Re: [RFC/PATCH 12/18] ravb: Add timestamp to struct > >> ravb_hw_info > >> > >> On 9/23/21 5:08 PM, Biju Das wrote: > >> > >>> R-Car AVB-DMAC supports timestamp feature. > >>> Add a timestamp hw feature bit to struct ravb_hw_info to add this > >>> feature only for R-Car. > >>> > >>> Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com> > >>> --- > >>> drivers/net/ethernet/renesas/ravb.h | 2 + > >>> drivers/net/ethernet/renesas/ravb_main.c | 68 > >>> +++++++++++++++--------- > >>> 2 files changed, 45 insertions(+), 25 deletions(-) > >>> > >>> diff --git a/drivers/net/ethernet/renesas/ravb.h > >>> b/drivers/net/ethernet/renesas/ravb.h > >>> index ab4909244276..2505de5d4a28 100644 > >>> --- a/drivers/net/ethernet/renesas/ravb.h > >>> +++ b/drivers/net/ethernet/renesas/ravb.h > >>> @@ -1034,6 +1034,7 @@ struct ravb_hw_info { > >>> unsigned mii_rgmii_selection:1; /* E-MAC supports mii/rgmii > >> selection */ > >>> unsigned half_duplex:1; /* E-MAC supports half duplex mode */ > >>> unsigned rx_2k_buffers:1; /* AVB-DMAC has Max 2K buf size on RX > >> */ > >>> + unsigned timestamp:1; /* AVB-DMAC has timestamp */ > >> > >> Isn't this a matter of the gPTP support as well, i.e. no separate > >> flag needed? > > > > Agreed. Previously it is suggested to use timestamp. I will change it to > as part of gPTP support cases. > > TIA. :-) > > >> > >> [...] > >>> @@ -1089,6 +1090,7 @@ struct ravb_private { > >>> unsigned int num_tx_desc; /* TX descriptors per packet */ > >>> > >>> int duplex; > >>> + struct ravb_rx_desc *rgeth_rx_ring[NUM_RX_QUEUE]; > >> > >> Strange place to declare this... > > > > Agreed. This has to be on later patch. Will move it. > > I only meant that these fields should go together with rx_ring[]. > Apparently we have a case of wrong patch ordering here (as this patch > needs this field declared)... Exactly that is the case, it is moved to later patch (https://patchwork.kernel.org/project/linux-renesas-soc/patch/20210923140813.13541-14-biju.das.jz@bp.renesas.com/) which is the first user of This variable. Regards, Biju > > >>> const struct ravb_hw_info *info; > >>> struct reset_control *rstc; > >>> diff --git a/drivers/net/ethernet/renesas/ravb_main.c > >>> b/drivers/net/ethernet/renesas/ravb_main.c > >>> index 9c0d35f4b221..2c375002ebcb 100644 > >>> --- a/drivers/net/ethernet/renesas/ravb_main.c > >>> +++ b/drivers/net/ethernet/renesas/ravb_main.c > >>> @@ -949,11 +949,14 @@ static bool ravb_queue_interrupt(struct > >>> net_device *ndev, int q) > >>> > >>> static bool ravb_timestamp_interrupt(struct net_device *ndev) { > >>> + struct ravb_private *priv = netdev_priv(ndev); > >>> + const struct ravb_hw_info *info = priv->info; > >>> u32 tis = ravb_read(ndev, TIS); > >>> > >>> if (tis & TIS_TFUF) { > >>> ravb_write(ndev, ~(TIS_TFUF | TIS_RESERVED), TIS); > >>> - ravb_get_tx_tstamp(ndev); > >>> + if (info->timestamp) > >>> + ravb_get_tx_tstamp(ndev); > >> > >> Shouldn't we just disable TIS.TFUF permanently instead for the > >> non-gPTP case? > > > > Good catch. As ravb_dmac_init_rgeth(will be renamed to > "ravb_dmac_init_gbeth") is not enabling this interrupt as it is not > documented in RZ/G2L hardware manual. > > So this function never gets called for non-gPTP case. > > > > I will remove this check. > > TIA! > > > Regards, > > Biju > > MBR, Sergey
Hi Sergei, > Subject: Re: [RFC/PATCH 12/18] ravb: Add timestamp to struct ravb_hw_info > > On 9/26/21 11:45 PM, Sergey Shtylyov wrote: > > >>> -----Original Message----- > >>> Subject: Re: [RFC/PATCH 12/18] ravb: Add timestamp to struct > >>> ravb_hw_info > >>> > >>> On 9/23/21 5:08 PM, Biju Das wrote: > >>> > >>>> R-Car AVB-DMAC supports timestamp feature. > >>>> Add a timestamp hw feature bit to struct ravb_hw_info to add this > >>>> feature only for R-Car. > >>>> > >>>> Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com> > >>>> --- > >>>> drivers/net/ethernet/renesas/ravb.h | 2 + > >>>> drivers/net/ethernet/renesas/ravb_main.c | 68 > >>>> +++++++++++++++--------- > >>>> 2 files changed, 45 insertions(+), 25 deletions(-) > >>>> > >>>> diff --git a/drivers/net/ethernet/renesas/ravb.h > >>>> b/drivers/net/ethernet/renesas/ravb.h > >>>> index ab4909244276..2505de5d4a28 100644 > >>>> --- a/drivers/net/ethernet/renesas/ravb.h > >>>> +++ b/drivers/net/ethernet/renesas/ravb.h > >>>> @@ -1034,6 +1034,7 @@ struct ravb_hw_info { > >>>> unsigned mii_rgmii_selection:1; /* E-MAC supports > mii/rgmii > >>> selection */ > >>>> unsigned half_duplex:1; /* E-MAC supports half duplex > mode */ > >>>> unsigned rx_2k_buffers:1; /* AVB-DMAC has Max 2K buf size > on RX > >>> */ > >>>> + unsigned timestamp:1; /* AVB-DMAC has timestamp */ > >>> > >>> Isn't this a matter of the gPTP support as well, i.e. no separate > >>> flag needed? > >> > >> Agreed. Previously it is suggested to use timestamp. I will change it > to as part of gPTP support cases. > > > > TIA. :-) > > > >>> > >>> [...] > >>>> @@ -1089,6 +1090,7 @@ struct ravb_private { > >>>> unsigned int num_tx_desc; /* TX descriptors per packet */ > >>>> > >>>> int duplex; > >>>> + struct ravb_rx_desc *rgeth_rx_ring[NUM_RX_QUEUE]; > >>> > >>> Strange place to declare this... > >> > >> Agreed. This has to be on later patch. Will move it. > > > > I only meant that these fields should go together with rx_ring[]. > > Apparently > > Sorry, this field. It is a mistake from my side which ended up this variable in this patch. Function pointers are introduced to avoid more checks and only rx functions pf RZ/G2L are using this variable. It will be clear you to once you finish reviewing the remaining patches(patch 13 - patch 18). Regards, Biju Regards, Biju > > > we have a case of wrong patch ordering here (as this patch needs this > field declared)... > [...] > > >> Regards, > >> Biju > > MBR, Sergey
diff --git a/drivers/net/ethernet/renesas/ravb.h b/drivers/net/ethernet/renesas/ravb.h index ab4909244276..2505de5d4a28 100644 --- a/drivers/net/ethernet/renesas/ravb.h +++ b/drivers/net/ethernet/renesas/ravb.h @@ -1034,6 +1034,7 @@ struct ravb_hw_info { unsigned mii_rgmii_selection:1; /* E-MAC supports mii/rgmii selection */ unsigned half_duplex:1; /* E-MAC supports half duplex mode */ unsigned rx_2k_buffers:1; /* AVB-DMAC has Max 2K buf size on RX */ + unsigned timestamp:1; /* AVB-DMAC has timestamp */ }; struct ravb_private { @@ -1089,6 +1090,7 @@ struct ravb_private { unsigned int num_tx_desc; /* TX descriptors per packet */ int duplex; + struct ravb_rx_desc *rgeth_rx_ring[NUM_RX_QUEUE]; const struct ravb_hw_info *info; struct reset_control *rstc; diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c index 9c0d35f4b221..2c375002ebcb 100644 --- a/drivers/net/ethernet/renesas/ravb_main.c +++ b/drivers/net/ethernet/renesas/ravb_main.c @@ -949,11 +949,14 @@ static bool ravb_queue_interrupt(struct net_device *ndev, int q) static bool ravb_timestamp_interrupt(struct net_device *ndev) { + struct ravb_private *priv = netdev_priv(ndev); + const struct ravb_hw_info *info = priv->info; u32 tis = ravb_read(ndev, TIS); if (tis & TIS_TFUF) { ravb_write(ndev, ~(TIS_TFUF | TIS_RESERVED), TIS); - ravb_get_tx_tstamp(ndev); + if (info->timestamp) + ravb_get_tx_tstamp(ndev); return true; } return false; @@ -1071,16 +1074,24 @@ static int ravb_poll(struct napi_struct *napi, int budget) struct net_device *ndev = napi->dev; struct ravb_private *priv = netdev_priv(ndev); const struct ravb_hw_info *info = priv->info; + struct ravb_rx_desc *desc; unsigned long flags; int q = napi - priv->napi; int mask = BIT(q); int quota = budget; + unsigned int entry; + if (!info->timestamp) { + entry = priv->cur_rx[q] % priv->num_rx_ring[q]; + desc = &priv->rgeth_rx_ring[q][entry]; + } /* Processing RX Descriptor Ring */ /* Clear RX interrupt */ ravb_write(ndev, ~(mask | RIS0_RESERVED), RIS0); - if (ravb_rx(ndev, "a, q)) - goto out; + if (info->timestamp || desc->die_dt != DT_FEMPTY) { + if (ravb_rx(ndev, "a, q)) + goto out; + } /* Processing TX Descriptor Ring */ spin_lock_irqsave(&priv->lock, flags); @@ -1689,6 +1700,7 @@ static void ravb_tx_timeout_work(struct work_struct *work) static netdev_tx_t ravb_start_xmit(struct sk_buff *skb, struct net_device *ndev) { struct ravb_private *priv = netdev_priv(ndev); + const struct ravb_hw_info *info = priv->info; unsigned int num_tx_desc = priv->num_tx_desc; u16 q = skb_get_queue_mapping(skb); struct ravb_tstamp_skb *ts_skb; @@ -1765,28 +1777,30 @@ static netdev_tx_t ravb_start_xmit(struct sk_buff *skb, struct net_device *ndev) desc->dptr = cpu_to_le32(dma_addr); /* TX timestamp required */ - if (q == RAVB_NC) { - ts_skb = kmalloc(sizeof(*ts_skb), GFP_ATOMIC); - if (!ts_skb) { - if (num_tx_desc > 1) { - desc--; - dma_unmap_single(ndev->dev.parent, dma_addr, - len, DMA_TO_DEVICE); + if (info->timestamp) { + if (q == RAVB_NC) { + ts_skb = kmalloc(sizeof(*ts_skb), GFP_ATOMIC); + if (!ts_skb) { + if (num_tx_desc > 1) { + desc--; + dma_unmap_single(ndev->dev.parent, dma_addr, + len, DMA_TO_DEVICE); + } + goto unmap; } - goto unmap; + ts_skb->skb = skb_get(skb); + ts_skb->tag = priv->ts_skb_tag++; + priv->ts_skb_tag &= 0x3ff; + list_add_tail(&ts_skb->list, &priv->ts_skb_list); + + /* TAG and timestamp required flag */ + skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS; + desc->tagh_tsr = (ts_skb->tag >> 4) | TX_TSR; + desc->ds_tagl |= cpu_to_le16(ts_skb->tag << 12); } - ts_skb->skb = skb_get(skb); - ts_skb->tag = priv->ts_skb_tag++; - priv->ts_skb_tag &= 0x3ff; - list_add_tail(&ts_skb->list, &priv->ts_skb_list); - /* TAG and timestamp required flag */ - skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS; - desc->tagh_tsr = (ts_skb->tag >> 4) | TX_TSR; - desc->ds_tagl |= cpu_to_le16(ts_skb->tag << 12); + skb_tx_timestamp(skb); } - - skb_tx_timestamp(skb); /* Descriptor type must be set after all the above writes */ dma_wmb(); if (num_tx_desc > 1) { @@ -1897,10 +1911,12 @@ static int ravb_close(struct net_device *ndev) "device will be stopped after h/w processes are done.\n"); /* Clear the timestamp list */ - list_for_each_entry_safe(ts_skb, ts_skb2, &priv->ts_skb_list, list) { - list_del(&ts_skb->list); - kfree_skb(ts_skb->skb); - kfree(ts_skb); + if (info->timestamp) { + list_for_each_entry_safe(ts_skb, ts_skb2, &priv->ts_skb_list, list) { + list_del(&ts_skb->list); + kfree_skb(ts_skb->skb); + kfree(ts_skb); + } } /* PHY disconnect */ @@ -2165,6 +2181,7 @@ static const struct ravb_hw_info ravb_gen3_hw_info = { .multi_tsrq = 1, .magic_pkt = 1, .rx_2k_buffers = 1, + .timestamp = 1, }; static const struct ravb_hw_info ravb_gen2_hw_info = { @@ -2186,6 +2203,7 @@ static const struct ravb_hw_info ravb_gen2_hw_info = { .multi_tsrq = 1, .magic_pkt = 1, .rx_2k_buffers = 1, + .timestamp = 1, }; static const struct ravb_hw_info rgeth_hw_info = {
R-Car AVB-DMAC supports timestamp feature. Add a timestamp hw feature bit to struct ravb_hw_info to add this feature only for R-Car. Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com> --- drivers/net/ethernet/renesas/ravb.h | 2 + drivers/net/ethernet/renesas/ravb_main.c | 68 +++++++++++++++--------- 2 files changed, 45 insertions(+), 25 deletions(-)