Re: kern/42314: IC Plus IP100x PHY support

[SAITOH Masanobu <msaitoh%execsw.org@localhost>]

The following reply was made to PR kern/42314; it has been noted by GNATS.

From: SAITOH Masanobu <msaitoh%execsw.org@localhost>
To: Andrius V <vezhlys%gmail.com@localhost>, gnats-bugs%netbsd.org@localhost
Cc: msaitoh%execsw.org@localhost, msaitoh%netbsd.org@localhost, gnats-admin%netbsd.org@localhost,
 netbsd-bugs%netbsd.org@localhost, tharada%oucrc.org@localhost
Subject: Re: kern/42314: IC Plus IP100x PHY support
Date: Tue, 19 Nov 2019 00:04:49 +0900

 On 2019/11/18 23:46, SAITOH Masanobu wrote:
 > On 2019/11/15 2:42, Andrius V wrote:
 >> Hi,
 >>
 >> Tested, didn't notice any regressions, auto-negotiation configures
 >> 1000baseT full-duplex as expected (doesn't matter if on boot or using
 >> "ifconfig vge0 media auto" manually). Setting media type explicitly to
 >> 1000baseT selects 100baseTX though (which is an improvement, since it
 >> was completely  failing with the original code).
 > 
 > On my machine, forcing 1000baseT-FDX usually got a link as it is, but
 > sometimes got a link as 100baseTX-FDX. I suspect it depend on your
 > cable and/or the link partner. Some PHYs include IP1000A and IP1001
 > have a feature called downshift. If the PHY detect it can't link 1000baseT,
 > it downgrade to 100baseT. It can be disabled by clearing IPGPHY_SCSR_DOWNSHIFT_ENB
 > bit.
 > 
 >>  From dmesg:
 >>
 >> [     1.050249] vge0 at pci4 dev 0 function 0: VIA VT612X Gigabit
 >> Ethernet (rev. 0x82)
 >> [     1.050249] vge0: interrupting at ioapic1 pin 12
 >> [     1.050249] vge0: Ethernet address xx:xx:xx:xx:xx:xx
 >> [     1.050249] ipgphy0 at vge0 phy 22: IP1001 10/100/1000 PHY, rev. 0
 >> [     1.050249] 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX,
 >> 1000baseT, 1000baseT-FDX, auto
 > 
 > I have the same one (I have both stge+IP1000A and vge+IP1001).
 > 
 >> ifconfig vge0 (auto)
 >> vge0: flags=0x8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
 >>         capabilities=3f00<IP4CSUM_Rx,IP4CSUM_Tx,TCP4CSUM_Rx,TCP4CSUM_Tx>
 >>         capabilities=3f00<UDP4CSUM_Rx,UDP4CSUM_Tx>
 >>         enabled=0
 >>         ec_capabilities=7<VLAN_MTU,VLAN_HWTAGGING,JUMBO_MTU>
 >>         ec_enabled=2<VLAN_HWTAGGING>
 >>         address: xx:xx:xx:xx:xx:xx
 >>         media: Ethernet autoselect (1000baseT
 >> full-duplex,flowcontrol,master,rxpause,txpause)
 >>         status: active
 >>         inet 192.168.1.10/24 broadcast 192.168.1.255 flags 0x0
 >>         inet6 fe80::8c9e:c8a0:eb4b:731f%vge0/64 flags 0x0 scopeid 0x1
 >> ifconfig vge0 (1000baseT):
 >> vge0: flags=0x8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
 >>         capabilities=3f00<IP4CSUM_Rx,IP4CSUM_Tx,TCP4CSUM_Rx,TCP4CSUM_Tx>
 >>         capabilities=3f00<UDP4CSUM_Rx,UDP4CSUM_Tx>
 >>         enabled=0
 >>         ec_capabilities=7<VLAN_MTU,VLAN_HWTAGGING,JUMBO_MTU>
 >>         ec_enabled=2<VLAN_HWTAGGING>
 >>         address: xx:xx:xx:xx:xx:xx
 >>         media: Ethernet 1000baseT (100baseTX
 >> full-duplex,flowcontrol,rxpause,txpause)
 >>         status: active
 > 
 > It seems you  set "ifconfig vge0 media 1000baseT" and the resulted link
 > is 100baseT. It would be good to try ifconfig vge0 media "1000baseT-FDX"
 > 
 > On my environment:
 >> # ifconfig vge0
 >> vge0: flags=0x8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
 >>         capabilities=3f00<IP4CSUM_Rx,IP4CSUM_Tx,TCP4CSUM_Rx,TCP4CSUM_Tx>
 >>         capabilities=3f00<UDP4CSUM_Rx,UDP4CSUM_Tx>
 >>         enabled=0
 >>         ec_capabilities=7<VLAN_MTU,VLAN_HWTAGGING,JUMBO_MTU>
 >>         ec_enabled=2<VLAN_HWTAGGING>
 >>         address: 00:02:2a:e0:4b:a9
 >>         media: Ethernet 1000baseT full-duplex (1000baseT full-duplex,flowcontrol,rxpause,txpause)
 >>         status: active
 
 Even though the link was established as 1000baseTX-FDX,
 vge0 got watchdog timeout as:
 > vge0: soft reset timed outvge0: watchdog timeout
 
 and repeat linkdown -> linkup -> watchdog. I suspect some addition
 might be required into if_vge.c. stge0 doesn't have such problem.
 
 
 > Thanks.
 > 
 >>         inet 192.168.1.10/24 broadcast 192.168.1.255 flags 0x0
 >>         inet6 fe80::8c9e:c8a0:eb4b:731f%vge0/64 flags 0x0 scopeid 0x1
 >>
 >> Regards,
 >> Andrius V
 >>
 >>
 >> On Thu, Nov 14, 2019 at 11:48 AM Andrius V <vezhlys%gmail.com@localhost> wrote:
 >>>
 >>> Hi,
 >>>
 >>> Sure, will do sometime today, once the changes will be synced with
 >>> fossil/git repos.
 >>>
 >>> On Thu, Nov 14, 2019 at 11:10 AM Masanobu SAITOH <msaitoh%execsw.org@localhost> wrote:
 >>>>
 >>>> The following reply was made to PR kern/42314; it has been noted by GNATS.
 >>>>
 >>>> From: Masanobu SAITOH <msaitoh%execsw.org@localhost>
 >>>> To: Andrius V <vezhlys%gmail.com@localhost>
 >>>> Cc: msaitoh%execsw.org@localhost, msaitoh%netbsd.org@localhost, gnats-admin%netbsd.org@localhost,
 >>>>  netbsd-bugs%netbsd.org@localhost, tharada%oucrc.org@localhost, gnats-bugs%netbsd.org@localhost
 >>>> Subject: Re: kern/42314: IC Plus IP100x PHY support
 >>>> Date: Thu, 14 Nov 2019 18:09:20 +0900
 >>>>
 >>>>  On 2019/10/09 0:38, Andrius V wrote:
 >>>>  > Hi,
 >>>>  >
 >>>>  >> It's merged!
 >>>>  >
 >>>>  > Thank you! Tested, working as expected.
 >>>>  >
 >>>>  >> Additional patch are also welcomed.
 >>>>  >
 >>>>  > Sure, if I will have something.
 >>>>  >
 >>>>  > Regards,
 >>>>  > Andrius V
 >>>>
 >>>>  Could you test with the latest ipgphy.c(rev. 1.4)?
 >>>>
 >>>>  Also please show me the output of the dmesg (both MAC and PHY).
 >>>>
 >>>>  Thanks in advance.
 >>>>
 >>>>  --
 >>>>  -----------------------------------------------
 >>>>                  SAITOH Masanobu (msaitoh%execsw.org@localhost
 >>>>                                   msaitoh%netbsd.org@localhost)
 >>>>
 > 
 > 
 
 
 -- 
 -----------------------------------------------
                 SAITOH Masanobu (msaitoh%execsw.org@localhost
                                  msaitoh%netbsd.org@localhost)