[Intel-wired-lan] [PATCH] [iwl-net] Revert "igc: set TP bit in 'supported' and 'advertising' fields of ethtool_link_ksettings"

Neftin, Sasha sasha.neftin at intel.com
Tue Sep 26 11:00:06 UTC 2023


On 26/09/2023 8:23, Neftin, Sasha wrote:
> On 25/09/2023 22:40, Prasad Koya wrote:
>> Hi,
>>
>> Here is the ethtool output before and after changing the speed with 
>> the commit 9ac3fc2f42e5ffa1e927dcbffb71b15fa81459e2:
>>
>> -bash-4.2# ethtool ma1
>> Settings for ma1:
>>          Supported ports: [ TP ]
>>          Supported link modes:   10baseT/Half 10baseT/Full
>>                                  100baseT/Half 100baseT/Full
>>                                  1000baseT/Full
>>                                  2500baseT/Full
>>          Supported pause frame use: Symmetric
>>          Supports auto-negotiation: Yes
>>          Supported FEC modes: Not reported
>>          Advertised link modes:  10baseT/Half 10baseT/Full
>>                                  100baseT/Half 100baseT/Full
>>                                  1000baseT/Full
>>                                  2500baseT/Full
>>          Advertised pause frame use: Symmetric
>>          Advertised auto-negotiation: Yes
>>          Advertised FEC modes: Not reported
>>          Speed: 1000Mb/s
>>          Duplex: Full
>>          Auto-negotiation: on
>>          Port: Twisted Pair
>>          PHYAD: 0
>>          Transceiver: internal
>>          MDI-X: off (auto)
>>          Supports Wake-on: pumbg
>>          Wake-on: d
>>          Current message level: 0x00000007 (7)
>>                                 drv probe link
>>          Link detected: yes
>> -bash-4.2#
>> -bash-4.2# ethtool  -s ma1 speed 100 duplex full autoneg on
>> -bash-4.2#
>> -bash-4.2# ethtool ma1
>> Settings for ma1:
>>          Supported ports: [ TP ]
>>          Supported link modes:   10baseT/Half 10baseT/Full
>>                                  100baseT/Half 100baseT/Full
>>                                  1000baseT/Full
>>                                  2500baseT/Full
>>          Supported pause frame use: Symmetric
>>          Supports auto-negotiation: Yes
>>          Supported FEC modes: Not reported
>>          Advertised link modes:  100baseT/Full
>>                                  2500baseT/Full
>>          Advertised pause frame use: Symmetric
>>          Advertised auto-negotiation: Yes
>>          Advertised FEC modes: Not reported
>>          Speed: 100Mb/s
>>          Duplex: Full
>>          Auto-negotiation: on
>>          Port: Twisted Pair
>>          PHYAD: 0
>>          Transceiver: internal
>>          MDI-X: off (auto)
>>          Supports Wake-on: pumbg
>>          Wake-on: d
>>          Current message level: 0x00000007 (7)
>>                                 drv probe link
>>          Link detected: yes
>> -bash-4.2#
>>
>> With the patch reverted:
>>
>> -bash-4.2# ethtool -s ma1 speed 100 duplex full autoneg on
>> -bash-4.2#
>> -bash-4.2# ethtool ma1
>> Settings for ma1:
>>          Supported ports: [ TP ]
>>          Supported link modes:   10baseT/Half 10baseT/Full
>>                                  100baseT/Half 100baseT/Full
>>                                  1000baseT/Full
>>                                  2500baseT/Full
>>          Supported pause frame use: Symmetric
>>          Supports auto-negotiation: Yes
>>          Supported FEC modes: Not reported
>>          Advertised link modes:  100baseT/Full
>>          Advertised pause frame use: Symmetric
>>          Advertised auto-negotiation: Yes
>>          Advertised FEC modes: Not reported
>>          Speed: 100Mb/s
>>          Duplex: Full
>>          Port: Twisted Pair
>>          PHYAD: 0
>>          Transceiver: internal
>>          Auto-negotiation: on
>>          MDI-X: off (auto)
>>          Supports Wake-on: pumbg
>>          Wake-on: d
>>          Current message level: 0x00000007 (7)
>>                                 drv probe link
>>          Link detected: yes
>> -bash-4.2#
>>
>> with the patch enabled:
>> ==================
>>
>> Default 'advertising' field is: 0x8000000020ef
>> ie., 10Mbps_half, 10Mbps_full, 100Mbps_half, 100Mbps_full, 
>> 1000Mbps_full, Autoneg, TP, Pause and 2500Mbps_full bits set.
>>
>> and 'hw->phy.autoneg_advertised' is 0xaf
>>
>> During "ethtool -s ma1 speed 100 duplex full autoneg on"
>>
>> ethtool sends 'advertising' as 0x20c8 ie., 100Mbps_full, Autoneg, TP, 
>> Pause bits set which are correct.
>>
>> However, to reset the link with new 'advertising' bits, code takes 
>> this path:
>>
>> [  255.073847]  igc_setup_copper_link+0x73c/0x750
>> [  255.073851]  igc_setup_link+0x4a/0x170
>> [  255.073852]  igc_init_hw_base+0x98/0x100
>> [  255.073855]  igc_reset+0x69/0xe0
>> [  255.073857]  igc_down+0x22b/0x230
>> [  255.073859]  igc_ethtool_set_link_ksettings+0x25f/0x270
>> [  255.073863]  ethtool_set_link_ksettings+0xa9/0x140
>> [  255.073866]  dev_ethtool+0x1236/0x2570
>>
>> igc_setup_copper_link() calls igc_copper_link_autoneg(). 
>> igc_copper_link_autoneg() changes phy->autoneg_advertised
>>
>>      phy->autoneg_advertised &= phy->autoneg_mask;
>>
>> and autoneg_mask is IGC_ALL_SPEED_DUPLEX_2500 which is 0xaf:
>>
>> /* 1Gbps and 2.5Gbps half duplex is not supported, nor spec-compliant. */
>> #define ADVERTISE_10_HALF       0x0001
>> #define ADVERTISE_10_FULL       0x0002
>> #define ADVERTISE_100_HALF      0x0004
>> #define ADVERTISE_100_FULL      0x0008
>> #define ADVERTISE_1000_HALF     0x0010 /* Not used, just FYI */
>> #define ADVERTISE_1000_FULL     0x0020
>> #define ADVERTISE_2500_HALF     0x0040 /* Not used, just FYI */
>> #define ADVERTISE_2500_FULL     0x0080
>>
>> #define IGC_ALL_SPEED_DUPLEX_2500 ( \
>>      ADVERTISE_10_HALF | ADVERTISE_10_FULL | ADVERTISE_100_HALF | \
>>      ADVERTISE_100_FULL | ADVERTISE_1000_FULL | ADVERTISE_2500_FULL)
>>
>> so 0x20c8 & 0xaf becomes 0x88 ie., the TP bit (bit 7 
>> of ethtool_link_mode_bit_indices) in 0x20c8 got interpreted as 
>> ADVERTISE_2500_FULL. so after igc_reset(), hw->phy.autoneg_advertised 
>> is 0x88. Post that, 'ethtool <interface>' reports 2500Mbps can also be 
>> advertised.
>>
>> @@ -445,9 +451,19 @@ static s32 igc_copper_link_autoneg(struct igc_hw 
>> *hw)
>>          u16 phy_ctrl;
>>          s32 ret_val;
>>
>>          /* Perform some bounds checking on the autoneg advertisement
>>           * parameter.
>>           */
>> +       if (!(phy->autoneg_advertised & ADVERTISED_2500baseX_Full))
>> +               phy->autoneg_advertised &= ~ADVERTISE_2500_FULL;
>> +       if ((phy->autoneg_advertised & ADVERTISED_2500baseX_Full))
>> +               phy->autoneg_advertised |= ADVERTISE_2500_FULL;
>> +
> 
> It will introduce more ambiguity. ADVERTISED_2500baseX_Full (is bit 15), 
> 2500 Base-X is a different type not supported by i225/6 parts. i225/6 
> parts support 2500baseT_Full_BIT (bit 47 in new structure).
> Look, ethtool used (same as igb) ethtool_convert_link_mode_to_legacy_u32 
> method, but there is no option for 2500baseT_Full_BIT. (since i225 only 
> copper mode, the TP advertisement was omitted intentionally in an 
> original code, I thought).
> 
>>          phy->autoneg_advertised &= phy->autoneg_mask;
>>
>> I see phy->autoneg_advertised modified similarly 
>> in igc_phy_setup_autoneg() as well.
>>
>> Above diff works for:
>>
>> ethtool -s <intf> speed 10/100/1000 duplex full autoneg on
>> or
>> ethtool -s <intf> advertise 0x3f (0x03 or 0x0f etc)
>>
>> but I haven't tested on a 2500 Mbps link. ADVERTISE_2500_FULL is there 
>> only for igc.
>>
>> Thanks
>> Prasad
>>
>>
>> On Sun, Sep 24, 2023 at 7:51 AM Andrew Lunn <andrew at lunn.ch 
>> <mailto:andrew at lunn.ch>> wrote:
>>
>>     On Sun, Sep 24, 2023 at 10:09:17AM +0300, Neftin, Sasha wrote:
>>      > On 22/09/2023 19:38, Prasad Koya wrote:
>>      > > This reverts commit 9ac3fc2f42e5ffa1e927dcbffb71b15fa81459e2.
>>      > >
>>      > > After the command "ethtool -s enps0 speed 100 duplex full
>>     autoneg on",
>>      > > i.e., advertise only 100Mbps speed to the peer, "ethtool enps0"
>>     shows
>>      > > advertised speeds as 100Mbps and 2500Mbps. Same behavior is seen
>>      > > when changing the speed to 10Mbps or 1000Mbps.
>>      > >
>>      > > This applies to I225/226 parts, which only supports copper mode.
>>      > > Reverting to original till the ambiguity is resolved.
>>      > >
>>      > > Fixes: 9ac3fc2f42e5 ("igc: set TP bit in 'supported' and
>>      > > 'advertising' fields of ethtool_link_ksettings")
>>      > > Signed-off-by: Prasad Koya <prasad at arista.com
>>     <mailto:prasad at arista.com>>
>>      >
>>      > Acked-by: Sasha Neftin <sasha.neftin at intel.com
>>     <mailto:sasha.neftin at intel.com>>
>>      >
>>      > > ---
>>      > >   drivers/net/ethernet/intel/igc/igc_ethtool.c | 2 --
>>      > >   1 file changed, 2 deletions(-)
>>      > >
>>      > > diff --git a/drivers/net/ethernet/intel/igc/igc_ethtool.c
>>     b/drivers/net/ethernet/intel/igc/igc_ethtool.c
>>      > > index 93bce729be76..0e2cb00622d1 100644
>>      > > --- a/drivers/net/ethernet/intel/igc/igc_ethtool.c
>>      > > +++ b/drivers/net/ethernet/intel/igc/igc_ethtool.c
>>      > > @@ -1708,8 +1708,6 @@ static int
>>     igc_ethtool_get_link_ksettings(struct net_device *netdev,
>>      > >     /* twisted pair */
>>      > >     cmd->base.port = PORT_TP;
>>      > >     cmd->base.phy_address = hw->phy.addr;
>>      > > -   ethtool_link_ksettings_add_link_mode(cmd, supported, TP);
>>      > > -   ethtool_link_ksettings_add_link_mode(cmd, advertising, TP);
>>
>>     This looks very odd. Please can you confirm this revert really does
>>     make ethtool report the correct advertisement when it has been 
>> limited
>>     to 100Mbps. Because looking at this patch, i have no idea how this is
>>     going wrong.
> 
> Andrew, yes, I can confirm. (revert works).
> We need a process fix, but it will be a different patch. Also, I prefer 
> not to leave upstream code with a regression.
> 
> ethtool_convert_link_mode_to_legacy_u32 have no option to work with 
> 2500baseT_Full_BIT (47 > 32). Need to use a new structure for 
> auto-negotiation advertisement, mask... (not u16). We need to think how 
> to fix it right.
> 
Colleagues, let us check the option not to use the 
'ethtool_convert_link_mode_to_legacy_u32' method. Looks like we can 
suggest another solution and not process the revert.

>>
>>              Andrew
>>
> 
> _______________________________________________
> Intel-wired-lan mailing list
> Intel-wired-lan at osuosl.org
> https://lists.osuosl.org/mailman/listinfo/intel-wired-lan



More information about the Intel-wired-lan mailing list