| Message ID | E1l1t3B-0005Vn-2N@rmk-PC.armlinux.org.uk (mailing list archive) |
|---|---|
| State | Changes Requested |
| Delegated to: | Netdev Maintainers |
| Headers | show |
| Series | dpaa2: add 1000base-X support | expand |
| Context | Check | Description |
|---|---|---|
| netdev/cover_letter | success | Link |
| netdev/fixes_present | success | Link |
| netdev/patch_count | success | Link |
| netdev/tree_selection | success | Clearly marked for net-next |
| netdev/subject_prefix | success | Link |
| netdev/cc_maintainers | success | CCed 5 of 5 maintainers |
| netdev/source_inline | success | Was 0 now: 0 |
| netdev/verify_signedoff | success | Link |
| netdev/module_param | success | Was 0 now: 0 |
| netdev/build_32bit | success | Errors and warnings before: 0 this patch: 0 |
| netdev/kdoc | success | Errors and warnings before: 0 this patch: 0 |
| netdev/verify_fixes | success | Link |
| netdev/checkpatch | success | total: 0 errors, 0 warnings, 0 checks, 11 lines checked |
| netdev/build_allmodconfig_warn | success | Errors and warnings before: 0 this patch: 0 |
| netdev/header_inline | success | Link |
| netdev/stable | success | Stable not CCed |
On Tue, Jan 19, 2021 at 03:36:09PM +0000, Russell King wrote: > Add support for backplane link mode, which is, according to discussions > with NXP earlier in the year, is a mode where the OS (Linux) is able to > manage the PCS and Serdes itself. Indeed, DPMACs in TYPE_BACKPLANE can have both their PCS and SerDes managed by Linux (since the firmware is not touching these). That being said, DPMACs in TYPE_PHY (the type that is already supported in dpaa2-mac) can also have their PCS managed by Linux (no interraction from the firmware's part with the PCS, just the SerDes). All in all, this patch is not needed for this particular usecase, where the switch between 1000Base-X and SGMII is done by just a minor reconfiguration in the PCS, without the need for SerDes changes. Also, with just the changes from this patch, a interface connected to a DPMAC in TYPE_BACKPLANE is not even creating a phylink instance. It's mainly because of this check from dpaa2-eth: if (dpaa2_eth_is_type_phy(priv)) { err = dpaa2_mac_connect(mac); I would suggest just dropping this patch. Ioana > > This commit prepares the ground work for allowing 1G fiber connections > to be used with DPAA2 on the SolidRun CEX7 platforms. > > Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk> > --- > drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.c | 5 +++-- > 1 file changed, 3 insertions(+), 2 deletions(-) > > diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.c > index 3ddfb40eb5e4..ccaf7e35abeb 100644 > --- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.c > +++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.c > @@ -315,8 +315,9 @@ int dpaa2_mac_connect(struct dpaa2_mac *mac) > goto err_put_node; > } > > - if (mac->attr.link_type == DPMAC_LINK_TYPE_PHY && > - mac->attr.eth_if != DPMAC_ETH_IF_RGMII) { > + if ((mac->attr.link_type == DPMAC_LINK_TYPE_PHY && > + mac->attr.eth_if != DPMAC_ETH_IF_RGMII) || > + mac->attr.link_type == DPMAC_LINK_TYPE_BACKPLANE) { > err = dpaa2_pcs_create(mac, dpmac_node, mac->attr.id); > if (err) > goto err_put_node; > -- > 2.20.1 >
On Wed, Jan 20, 2021 at 10:19:01PM +0000, Ioana Ciornei wrote: > On Tue, Jan 19, 2021 at 03:36:09PM +0000, Russell King wrote: > > Add support for backplane link mode, which is, according to discussions > > with NXP earlier in the year, is a mode where the OS (Linux) is able to > > manage the PCS and Serdes itself. > > Indeed, DPMACs in TYPE_BACKPLANE can have both their PCS and SerDes managed > by Linux (since the firmware is not touching these). > That being said, DPMACs in TYPE_PHY (the type that is already supported > in dpaa2-mac) can also have their PCS managed by Linux (no interraction > from the firmware's part with the PCS, just the SerDes). This is not what was discussed last year. It clearly shows in the slides that Bogdan sent in April 2020 "PCS Representation in Linux.pptx" page 4 that the firmware manages the PCS in TYPE_PHY and TYPE_FIXED mode. It was explained during the call that Linux must not access the internal MDIO nor touch the PCS _except_ when using TYPE_BACKPLANE mode. Touching the internal MDIO was stated as unsupported in other modes as the MC firmware will be performing MDIO accesses. > Also, with just the changes from this patch, a interface connected to a > DPMAC in TYPE_BACKPLANE is not even creating a phylink instance. It's > mainly because of this check from dpaa2-eth: > > if (dpaa2_eth_is_type_phy(priv)) { > err = dpaa2_mac_connect(mac); > > > I would suggest just dropping this patch. That is a recent change in net-next, but is not in my tree... In any case, if NXP have changed it such that, when in TYPE_PHY, the MC firmware no longer accesses the PCS and it is now safe to perform MDIO accesses, then we _still_ need this patch for older firmwares. In short, what you've put in your email does not tie up with the position that was discussed last year, and seems to me to be a total U-turn over what was being said.
On Wed, Jan 20, 2021 at 10:19:01PM +0000, Ioana Ciornei wrote: > On Tue, Jan 19, 2021 at 03:36:09PM +0000, Russell King wrote: > > Add support for backplane link mode, which is, according to discussions > > with NXP earlier in the year, is a mode where the OS (Linux) is able to > > manage the PCS and Serdes itself. > > Indeed, DPMACs in TYPE_BACKPLANE can have both their PCS and SerDes managed > by Linux (since the firmware is not touching these). > That being said, DPMACs in TYPE_PHY (the type that is already supported > in dpaa2-mac) can also have their PCS managed by Linux (no interraction > from the firmware's part with the PCS, just the SerDes). > > All in all, this patch is not needed for this particular usecase, where > the switch between 1000Base-X and SGMII is done by just a minor > reconfiguration in the PCS, without the need for SerDes changes. > > Also, with just the changes from this patch, a interface connected to a > DPMAC in TYPE_BACKPLANE is not even creating a phylink instance. It's > mainly because of this check from dpaa2-eth: > > if (dpaa2_eth_is_type_phy(priv)) { > err = dpaa2_mac_connect(mac); > > > I would suggest just dropping this patch. Hi Ioana, So what is happening with this series given our discussions off-list? Do I resend it as-is? Thanks.
On Thu, Feb 04, 2021 at 02:51:24PM +0000, Russell King - ARM Linux admin wrote: > On Wed, Jan 20, 2021 at 10:19:01PM +0000, Ioana Ciornei wrote: > > On Tue, Jan 19, 2021 at 03:36:09PM +0000, Russell King wrote: > > > Add support for backplane link mode, which is, according to discussions > > > with NXP earlier in the year, is a mode where the OS (Linux) is able to > > > manage the PCS and Serdes itself. > > > > Indeed, DPMACs in TYPE_BACKPLANE can have both their PCS and SerDes managed > > by Linux (since the firmware is not touching these). > > That being said, DPMACs in TYPE_PHY (the type that is already supported > > in dpaa2-mac) can also have their PCS managed by Linux (no interraction > > from the firmware's part with the PCS, just the SerDes). > > > > All in all, this patch is not needed for this particular usecase, where > > the switch between 1000Base-X and SGMII is done by just a minor > > reconfiguration in the PCS, without the need for SerDes changes. > > > > Also, with just the changes from this patch, a interface connected to a > > DPMAC in TYPE_BACKPLANE is not even creating a phylink instance. It's > > mainly because of this check from dpaa2-eth: > > > > if (dpaa2_eth_is_type_phy(priv)) { > > err = dpaa2_mac_connect(mac); > > > > > > I would suggest just dropping this patch. > > Hi Ioana, > > So what is happening with this series given our discussions off-list? > Let's also accept TYPE_BACKPLANE as you suggested. > Do I resend it as-is? > For the net-next, you would also need the following diff on top of your changes in patch 3/3: --- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.h +++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.h @@ -695,7 +695,9 @@ static inline unsigned int dpaa2_eth_rx_head_room(struct dpaa2_eth_priv *priv) static inline bool dpaa2_eth_is_type_phy(struct dpaa2_eth_priv *priv) { - if (priv->mac && priv->mac->attr.link_type == DPMAC_LINK_TYPE_PHY) + if (priv->mac && + (priv->mac->attr.link_type == DPMAC_LINK_TYPE_PHY || + priv->mac->attr.link_type == DPMAC_LINK_TYPE_BACKPLANE)) return true; return false; Would you mind amending your commit with this and resending the series? Thanks, Ioana
diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.c index 3ddfb40eb5e4..ccaf7e35abeb 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.c +++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.c @@ -315,8 +315,9 @@ int dpaa2_mac_connect(struct dpaa2_mac *mac) goto err_put_node; } - if (mac->attr.link_type == DPMAC_LINK_TYPE_PHY && - mac->attr.eth_if != DPMAC_ETH_IF_RGMII) { + if ((mac->attr.link_type == DPMAC_LINK_TYPE_PHY && + mac->attr.eth_if != DPMAC_ETH_IF_RGMII) || + mac->attr.link_type == DPMAC_LINK_TYPE_BACKPLANE) { err = dpaa2_pcs_create(mac, dpmac_node, mac->attr.id); if (err) goto err_put_node;
Add support for backplane link mode, which is, according to discussions with NXP earlier in the year, is a mode where the OS (Linux) is able to manage the PCS and Serdes itself. This commit prepares the ground work for allowing 1G fiber connections to be used with DPAA2 on the SolidRun CEX7 platforms. Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk> --- drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)