Hi Heiner,
There's an 8168fp/8117 chip has SFP+ port instead of RJ45, the phy device ID matches "Generic FE-GE Realtek PHY" nevertheless.
The problems is that, since it uses SFP+, both BMCR and BMSR read are always zero, so Realtek phylib never knows if the link is up.
However, the old method to read through MMIO correctly shows the link is up:
static unsigned int rtl8169_xmii_link_ok(struct rtl8169_private *tp)
{
return RTL_R8(tp, PHYstatus) & LinkStatus;
}
Few ideas here:
- Add a link state callback for phylib like phylink's phylink_fixed_state_cb(). However there's no guarantee that other parts of this chip works.
- Add SFP+ support for this chip. However the phy device matches to "Generic FE-GE Realtek PHY" which may complicate things.
Any advice will be welcome.
Kai-Heng
On Thu, Jan 02, 2020 at 02:59:42PM +0800, Kai Heng Feng wrote:
> Hi Heiner,
>
> There's an 8168fp/8117 chip has SFP+ port instead of RJ45, the phy device ID matches "Generic FE-GE Realtek PHY" nevertheless.
> The problems is that, since it uses SFP+, both BMCR and BMSR read are always zero, so Realtek phylib never knows if the link is up.
>
> However, the old method to read through MMIO correctly shows the link is up:
> static unsigned int rtl8169_xmii_link_ok(struct rtl8169_private *tp)
> {
> return RTL_R8(tp, PHYstatus) & LinkStatus;
> }
>
> Few ideas here:
> - Add a link state callback for phylib like phylink's phylink_fixed_state_cb(). However there's no guarantee that other parts of this chip works.
> - Add SFP+ support for this chip. However the phy device matches to "Generic FE-GE Realtek PHY" which may complicate things.
>
> Any advice will be welcome.
Hi Kai
Is the i2c bus accessible? Is there any documentation or example code?
In order to correctly support SFP+ cages, we need access to the i2c
bus to determine what sort of module has been inserted. It would also
be good to have access to LOS, transmitter disable, etc, from the SFP
cage.
Andrew
Hi Andrew,
> On Jan 2, 2020, at 23:21, Andrew Lunn <[email protected]> wrote:
>
> On Thu, Jan 02, 2020 at 02:59:42PM +0800, Kai Heng Feng wrote:
>> Hi Heiner,
>>
>> There's an 8168fp/8117 chip has SFP+ port instead of RJ45, the phy device ID matches "Generic FE-GE Realtek PHY" nevertheless.
>> The problems is that, since it uses SFP+, both BMCR and BMSR read are always zero, so Realtek phylib never knows if the link is up.
>>
>> However, the old method to read through MMIO correctly shows the link is up:
>> static unsigned int rtl8169_xmii_link_ok(struct rtl8169_private *tp)
>> {
>> return RTL_R8(tp, PHYstatus) & LinkStatus;
>> }
>>
>> Few ideas here:
>> - Add a link state callback for phylib like phylink's phylink_fixed_state_cb(). However there's no guarantee that other parts of this chip works.
>> - Add SFP+ support for this chip. However the phy device matches to "Generic FE-GE Realtek PHY" which may complicate things.
>>
>> Any advice will be welcome.
>
> Hi Kai
>
> Is the i2c bus accessible?
I don't think so. It seems to be a regular Realtek 8168 device with generic PCI ID [10ec:8168].
> Is there any documentation or example code?
Unfortunately no.
>
> In order to correctly support SFP+ cages, we need access to the i2c
> bus to determine what sort of module has been inserted. It would also
> be good to have access to LOS, transmitter disable, etc, from the SFP
> cage.
Seems like we need Realtek to provide more information to support this chip with SFP+.
Kai-Heng
>
> Andrew
On 02.01.2020 17:46, Kai-Heng Feng wrote:
> Hi Andrew,
>
>> On Jan 2, 2020, at 23:21, Andrew Lunn <[email protected]> wrote:
>>
>> On Thu, Jan 02, 2020 at 02:59:42PM +0800, Kai Heng Feng wrote:
>>> Hi Heiner,
>>>
>>> There's an 8168fp/8117 chip has SFP+ port instead of RJ45, the phy device ID matches "Generic FE-GE Realtek PHY" nevertheless.
>>> The problems is that, since it uses SFP+, both BMCR and BMSR read are always zero, so Realtek phylib never knows if the link is up.
>>>
>>> However, the old method to read through MMIO correctly shows the link is up:
>>> static unsigned int rtl8169_xmii_link_ok(struct rtl8169_private *tp)
>>> {
>>> return RTL_R8(tp, PHYstatus) & LinkStatus;
>>> }
>>>
>>> Few ideas here:
>>> - Add a link state callback for phylib like phylink's phylink_fixed_state_cb(). However there's no guarantee that other parts of this chip works.
>>> - Add SFP+ support for this chip. However the phy device matches to "Generic FE-GE Realtek PHY" which may complicate things.
>>>
>>> Any advice will be welcome.
>>
>> Hi Kai
>>
>> Is the i2c bus accessible?
>
> I don't think so. It seems to be a regular Realtek 8168 device with generic PCI ID [10ec:8168].
>
>> Is there any documentation or example code?
>
> Unfortunately no.
>
>>
>> In order to correctly support SFP+ cages, we need access to the i2c
>> bus to determine what sort of module has been inserted. It would also
>> be good to have access to LOS, transmitter disable, etc, from the SFP
>> cage.
>
> Seems like we need Realtek to provide more information to support this chip with SFP+.
>
Indeed it would be good to have some more details how this chip handles SFP+,
therefore I add Hau to the discussion.
As I see it the PHY registers are simply dummies on this chip. Or does this chip
support both, PHY and SFP+? Hopefully SFP presence can be autodetected, we could
skip the complete PHY handling in this case. Interesting would be which parts of
the SFP interface are exposed how via (proprietary) registers.
Recently the STMMAC driver was converted from phylib to phylink, maybe we have
to do the same with r8169 one fine day. But w/o more details this is just
speculation, much appreciated would be documentation from Realtek about the
SFP+ interface.
Kai, which hardware/board are we talking about?
> Kai-Heng
>
>>
>> Andrew
>
Heiner
> On Jan 3, 2020, at 05:24, Heiner Kallweit <[email protected]> wrote:
>
> On 02.01.2020 17:46, Kai-Heng Feng wrote:
>> Hi Andrew,
>>
>>> On Jan 2, 2020, at 23:21, Andrew Lunn <[email protected]> wrote:
>>>
>>> On Thu, Jan 02, 2020 at 02:59:42PM +0800, Kai Heng Feng wrote:
>>>> Hi Heiner,
>>>>
>>>> There's an 8168fp/8117 chip has SFP+ port instead of RJ45, the phy device ID matches "Generic FE-GE Realtek PHY" nevertheless.
>>>> The problems is that, since it uses SFP+, both BMCR and BMSR read are always zero, so Realtek phylib never knows if the link is up.
>>>>
>>>> However, the old method to read through MMIO correctly shows the link is up:
>>>> static unsigned int rtl8169_xmii_link_ok(struct rtl8169_private *tp)
>>>> {
>>>> return RTL_R8(tp, PHYstatus) & LinkStatus;
>>>> }
>>>>
>>>> Few ideas here:
>>>> - Add a link state callback for phylib like phylink's phylink_fixed_state_cb(). However there's no guarantee that other parts of this chip works.
>>>> - Add SFP+ support for this chip. However the phy device matches to "Generic FE-GE Realtek PHY" which may complicate things.
>>>>
>>>> Any advice will be welcome.
>>>
>>> Hi Kai
>>>
>>> Is the i2c bus accessible?
>>
>> I don't think so. It seems to be a regular Realtek 8168 device with generic PCI ID [10ec:8168].
>>
>>> Is there any documentation or example code?
>>
>> Unfortunately no.
>>
>>>
>>> In order to correctly support SFP+ cages, we need access to the i2c
>>> bus to determine what sort of module has been inserted. It would also
>>> be good to have access to LOS, transmitter disable, etc, from the SFP
>>> cage.
>>
>> Seems like we need Realtek to provide more information to support this chip with SFP+.
>>
> Indeed it would be good to have some more details how this chip handles SFP+,
> therefore I add Hau to the discussion.
>
> As I see it the PHY registers are simply dummies on this chip. Or does this chip
> support both, PHY and SFP+? Hopefully SFP presence can be autodetected, we could
> skip the complete PHY handling in this case. Interesting would be which parts of
> the SFP interface are exposed how via (proprietary) registers.
> Recently the STMMAC driver was converted from phylib to phylink, maybe we have
> to do the same with r8169 one fine day. But w/o more details this is just
> speculation, much appreciated would be documentation from Realtek about the
> SFP+ interface.
>
> Kai, which hardware/board are we talking about?
It's a regular Intel PC.
The ethernet is function 1 of the PCI device, function 0 isn't bound to any driver:
02:00.0 Unassigned class [ff00]: Realtek Semiconductor Co., Ltd. Device [10ec:816e] (rev 1a)
02:00.1 Ethernet controller [0200]: Realtek Semiconductor Co., Ltd. RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller [10ec:8168] (rev 22)
Kai-Heng
>
>> Kai-Heng
>>
>>>
>>> Andrew
>>
> Heiner
Chun-Hao,
> On Jan 3, 2020, at 12:53, Kai-Heng Feng <[email protected]> wrote:
>
>
>
>> On Jan 3, 2020, at 05:24, Heiner Kallweit <[email protected]> wrote:
>>
>> On 02.01.2020 17:46, Kai-Heng Feng wrote:
>>> Hi Andrew,
>>>
>>>> On Jan 2, 2020, at 23:21, Andrew Lunn <[email protected]> wrote:
>>>>
>>>> On Thu, Jan 02, 2020 at 02:59:42PM +0800, Kai Heng Feng wrote:
>>>>> Hi Heiner,
>>>>>
>>>>> There's an 8168fp/8117 chip has SFP+ port instead of RJ45, the phy device ID matches "Generic FE-GE Realtek PHY" nevertheless.
>>>>> The problems is that, since it uses SFP+, both BMCR and BMSR read are always zero, so Realtek phylib never knows if the link is up.
>>>>>
>>>>> However, the old method to read through MMIO correctly shows the link is up:
>>>>> static unsigned int rtl8169_xmii_link_ok(struct rtl8169_private *tp)
>>>>> {
>>>>> return RTL_R8(tp, PHYstatus) & LinkStatus;
>>>>> }
>>>>>
>>>>> Few ideas here:
>>>>> - Add a link state callback for phylib like phylink's phylink_fixed_state_cb(). However there's no guarantee that other parts of this chip works.
>>>>> - Add SFP+ support for this chip. However the phy device matches to "Generic FE-GE Realtek PHY" which may complicate things.
>>>>>
>>>>> Any advice will be welcome.
>>>>
>>>> Hi Kai
>>>>
>>>> Is the i2c bus accessible?
>>>
>>> I don't think so. It seems to be a regular Realtek 8168 device with generic PCI ID [10ec:8168].
>>>
>>>> Is there any documentation or example code?
>>>
>>> Unfortunately no.
>>>
>>>>
>>>> In order to correctly support SFP+ cages, we need access to the i2c
>>>> bus to determine what sort of module has been inserted. It would also
>>>> be good to have access to LOS, transmitter disable, etc, from the SFP
>>>> cage.
>>>
>>> Seems like we need Realtek to provide more information to support this chip with SFP+.
>>>
>> Indeed it would be good to have some more details how this chip handles SFP+,
>> therefore I add Hau to the discussion.
>>
>> As I see it the PHY registers are simply dummies on this chip. Or does this chip
>> support both, PHY and SFP+? Hopefully SFP presence can be autodetected, we could
>> skip the complete PHY handling in this case. Interesting would be which parts of
>> the SFP interface are exposed how via (proprietary) registers.
>> Recently the STMMAC driver was converted from phylib to phylink, maybe we have
>> to do the same with r8169 one fine day. But w/o more details this is just
>> speculation, much appreciated would be documentation from Realtek about the
>> SFP+ interface.
>>
>> Kai, which hardware/board are we talking about?
>
> It's a regular Intel PC.
>
> The ethernet is function 1 of the PCI device, function 0 isn't bound to any driver:
> 02:00.0 Unassigned class [ff00]: Realtek Semiconductor Co., Ltd. Device [10ec:816e] (rev 1a)
> 02:00.1 Ethernet controller [0200]: Realtek Semiconductor Co., Ltd. RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller [10ec:8168] (rev 22)
Would it be possible to share some info on SFP support?
Kai-Heng
>
> Kai-Heng
>
>>
>>> Kai-Heng
>>>
>>>>
>>>> Andrew
>>>
>> Heiner
>
> Chun-Hao,
>
> > On Jan 3, 2020, at 12:53, Kai-Heng Feng <[email protected]>
> wrote:
> >
> >
> >
> >> On Jan 3, 2020, at 05:24, Heiner Kallweit <[email protected]> wrote:
> >>
> >> On 02.01.2020 17:46, Kai-Heng Feng wrote:
> >>> Hi Andrew,
> >>>
> >>>> On Jan 2, 2020, at 23:21, Andrew Lunn <[email protected]> wrote:
> >>>>
> >>>> On Thu, Jan 02, 2020 at 02:59:42PM +0800, Kai Heng Feng wrote:
> >>>>> Hi Heiner,
> >>>>>
> >>>>> There's an 8168fp/8117 chip has SFP+ port instead of RJ45, the phy
> device ID matches "Generic FE-GE Realtek PHY" nevertheless.
> >>>>> The problems is that, since it uses SFP+, both BMCR and BMSR read
> are always zero, so Realtek phylib never knows if the link is up.
> >>>>>
> >>>>> However, the old method to read through MMIO correctly shows the
> link is up:
> >>>>> static unsigned int rtl8169_xmii_link_ok(struct rtl8169_private
> >>>>> *tp) {
> >>>>> return RTL_R8(tp, PHYstatus) & LinkStatus; }
> >>>>>
> >>>>> Few ideas here:
> >>>>> - Add a link state callback for phylib like phylink's
> phylink_fixed_state_cb(). However there's no guarantee that other parts of
> this chip works.
> >>>>> - Add SFP+ support for this chip. However the phy device matches to
> "Generic FE-GE Realtek PHY" which may complicate things.
> >>>>>
> >>>>> Any advice will be welcome.
> >>>>
> >>>> Hi Kai
> >>>>
> >>>> Is the i2c bus accessible?
> >>>
> >>> I don't think so. It seems to be a regular Realtek 8168 device with generic
> PCI ID [10ec:8168].
> >>>
> >>>> Is there any documentation or example code?
> >>>
> >>> Unfortunately no.
> >>>
> >>>>
> >>>> In order to correctly support SFP+ cages, we need access to the i2c
> >>>> bus to determine what sort of module has been inserted. It would
> >>>> also be good to have access to LOS, transmitter disable, etc, from
> >>>> the SFP cage.
> >>>
> >>> Seems like we need Realtek to provide more information to support this
> chip with SFP+.
> >>>
> >> Indeed it would be good to have some more details how this chip
> >> handles SFP+, therefore I add Hau to the discussion.
> >>
> >> As I see it the PHY registers are simply dummies on this chip. Or
> >> does this chip support both, PHY and SFP+? Hopefully SFP presence can
> >> be autodetected, we could skip the complete PHY handling in this
> >> case. Interesting would be which parts of the SFP interface are exposed
> how via (proprietary) registers.
> >> Recently the STMMAC driver was converted from phylib to phylink,
> >> maybe we have to do the same with r8169 one fine day. But w/o more
> >> details this is just speculation, much appreciated would be
> >> documentation from Realtek about the
> >> SFP+ interface.
> >>
> >> Kai, which hardware/board are we talking about?
> >
> > It's a regular Intel PC.
> >
> > The ethernet is function 1 of the PCI device, function 0 isn't bound to any
> driver:
> > 02:00.0 Unassigned class [ff00]: Realtek Semiconductor Co., Ltd.
> > Device [10ec:816e] (rev 1a)
> > 02:00.1 Ethernet controller [0200]: Realtek Semiconductor Co., Ltd.
> > RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller [10ec:8168]
> > (rev 22)
>
> Would it be possible to share some info on SFP support?
Hi Kai-Heng,
Could you use r8168 to dump hardware info with following command.
cat /proc/net/r8168/ethx/*
I want to make sure which chip you use and try to add support it in r8168/r8169.
Hau
>
> Kai-Heng
>
> >
> > Kai-Heng
> >
> >>
> >>> Kai-Heng
> >>>
> >>>>
> >>>> Andrew
> >>>
> >> Heiner
> >
>
>
> ------Please consider the environment before printing this e-mail.
> On Feb 14, 2020, at 5:07 PM, Hau <[email protected]> wrote:
>
>> Chun-Hao,
>>
>>> On Jan 3, 2020, at 12:53, Kai-Heng Feng <[email protected]>
>> wrote:
>>>
>>>
>>>
>>>> On Jan 3, 2020, at 05:24, Heiner Kallweit <[email protected]> wrote:
>>>>
>>>> On 02.01.2020 17:46, Kai-Heng Feng wrote:
>>>>> Hi Andrew,
>>>>>
>>>>>> On Jan 2, 2020, at 23:21, Andrew Lunn <[email protected]> wrote:
>>>>>>
>>>>>> On Thu, Jan 02, 2020 at 02:59:42PM +0800, Kai Heng Feng wrote:
>>>>>>> Hi Heiner,
>>>>>>>
>>>>>>> There's an 8168fp/8117 chip has SFP+ port instead of RJ45, the phy
>> device ID matches "Generic FE-GE Realtek PHY" nevertheless.
>>>>>>> The problems is that, since it uses SFP+, both BMCR and BMSR read
>> are always zero, so Realtek phylib never knows if the link is up.
>>>>>>>
>>>>>>> However, the old method to read through MMIO correctly shows the
>> link is up:
>>>>>>> static unsigned int rtl8169_xmii_link_ok(struct rtl8169_private
>>>>>>> *tp) {
>>>>>>> return RTL_R8(tp, PHYstatus) & LinkStatus; }
>>>>>>>
>>>>>>> Few ideas here:
>>>>>>> - Add a link state callback for phylib like phylink's
>> phylink_fixed_state_cb(). However there's no guarantee that other parts of
>> this chip works.
>>>>>>> - Add SFP+ support for this chip. However the phy device matches to
>> "Generic FE-GE Realtek PHY" which may complicate things.
>>>>>>>
>>>>>>> Any advice will be welcome.
>>>>>>
>>>>>> Hi Kai
>>>>>>
>>>>>> Is the i2c bus accessible?
>>>>>
>>>>> I don't think so. It seems to be a regular Realtek 8168 device with generic
>> PCI ID [10ec:8168].
>>>>>
>>>>>> Is there any documentation or example code?
>>>>>
>>>>> Unfortunately no.
>>>>>
>>>>>>
>>>>>> In order to correctly support SFP+ cages, we need access to the i2c
>>>>>> bus to determine what sort of module has been inserted. It would
>>>>>> also be good to have access to LOS, transmitter disable, etc, from
>>>>>> the SFP cage.
>>>>>
>>>>> Seems like we need Realtek to provide more information to support this
>> chip with SFP+.
>>>>>
>>>> Indeed it would be good to have some more details how this chip
>>>> handles SFP+, therefore I add Hau to the discussion.
>>>>
>>>> As I see it the PHY registers are simply dummies on this chip. Or
>>>> does this chip support both, PHY and SFP+? Hopefully SFP presence can
>>>> be autodetected, we could skip the complete PHY handling in this
>>>> case. Interesting would be which parts of the SFP interface are exposed
>> how via (proprietary) registers.
>>>> Recently the STMMAC driver was converted from phylib to phylink,
>>>> maybe we have to do the same with r8169 one fine day. But w/o more
>>>> details this is just speculation, much appreciated would be
>>>> documentation from Realtek about the
>>>> SFP+ interface.
>>>>
>>>> Kai, which hardware/board are we talking about?
>>>
>>> It's a regular Intel PC.
>>>
>>> The ethernet is function 1 of the PCI device, function 0 isn't bound to any
>> driver:
>>> 02:00.0 Unassigned class [ff00]: Realtek Semiconductor Co., Ltd.
>>> Device [10ec:816e] (rev 1a)
>>> 02:00.1 Ethernet controller [0200]: Realtek Semiconductor Co., Ltd.
>>> RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller [10ec:8168]
>>> (rev 22)
>>
>> Would it be possible to share some info on SFP support?
> Hi Kai-Heng,
>
> Could you use r8168 to dump hardware info with following command.
> cat /proc/net/r8168/ethx/*
>
> I want to make sure which chip you use and try to add support it in r8168/r8169.
Dump Driver Variable
Variable Value
---------- -----
MODULENAME r8168
driver version 8.048.00-NAPI
chipset 31
chipset_name RTL8168FP/8111FP
mtu 1500
NUM_RX_DESC 0x400
cur_rx 0x0
dirty_rx 0x0
NUM_TX_DESC 0x400
cur_tx 0x0
dirty_tx 0x0
rx_buf_sz 0x5f3
esd_flag 0x0
pci_cfg_is_read 0x1
rtl8168_rx_config 0xcf00
cp_cmd 0x20e1
intr_mask 0x115
timer_intr_mask 0x4000
wol_enabled 0x1
wol_opts 0x20
efuse_ver 0x3
eeprom_type 0x0
autoneg 0x1
duplex 0x1
speed 1000
advertising 0x3f
eeprom_len 0x0
cur_page 0x0
bios_setting 0x0
features 0x2
org_pci_offset_99 0x0
org_pci_offset_180 0xf
issue_offset_99_event 0x0
org_pci_offset_80 0x42
org_pci_offset_81 0x1
use_timer_interrrupt 0x1
HwIcVerUnknown 0x0
NotWrRamCodeToMicroP 0x0
NotWrMcuPatchCode 0x0
HwHasWrRamCodeToMicroP 0x0
sw_ram_code_ver 0x3
hw_ram_code_ver 0x0
rtk_enable_diag 0x0
ShortPacketSwChecksum 0x0
UseSwPaddingShortPkt 0x0
RequireAdcBiasPatch 0x0
AdcBiasPatchIoffset 0x0
RequireAdjustUpsTxLinkPulseTiming 0x0
SwrCnt1msIni 0x0
HwSuppNowIsOobVer 0x1
HwFiberModeVer 0x0
HwFiberStat 0x0
HwSwitchMdiToFiber 0x0
NicCustLedValue 0xca9
RequiredSecLanDonglePatch 0x0
HwSuppDashVer 0x3
DASH 0x1
dash_printer_enabled 0x0
HwSuppKCPOffloadVer 0x0
speed_mode 0x3e8
duplex_mode 0x1
autoneg_mode 0x1
advertising_mode 0x3f
aspm 0x1
s5wol 0x1
s5_keep_curr_mac 0x0
eee_enable 0x0
hwoptimize 0x0
proc_init_num 0x2
s0_magic_packet 0x0
HwSuppMagicPktVer 0x2
HwSuppCheckPhyDisableModeVer 0x3
HwPkgDet 0x6
HwSuppGigaForceMode 0x1
random_mac 0x0
org_mac_addr 54:b2:03:86:1b:40
perm_addr 54:b2:03:86:1b:40
dev_addr 54:b2:03:86:1b:40
Dump Ethernet PHY
Offset Value
------ -----
####################page 0##################
0x00: 1040 7989 001c c800 0de1 0000 0064 2001
0x08: 0000 0200 0000 0000 0000 0000 0000 2000
Dump Extended Registers
Offset Value
------ -----
0x00: 00000000 00000000 00000000 00000000
0x10: 00000000 00000000 00000000 00000000
0x20: 00000000 00000000 00000000 00000000
0x30: 00000000 00000000 00000000 00000000
0x40: 00000000 00000000 00000000 00000000
0x50: 00000000 00000000 00000000 00000000
0x60: 00000000 00000000 00000000 00000000
0x70: 00000000 00000000 00000000 00000000
0x80: 00000000 00000000 00000000 00000000
0x90: 00000000 00000000 00000000 00000000
0xa0: 00000000 00000000 00000000 00000000
0xb0: 00000000 00000000 00000000 00000000
0xc0: 00000000 00000000 00080002 0000002f
0xd0: 0000005f 00001f92 00000000 001e0021
0xe0: 8603b254 0000401b 00100006 00000200
0xf0: b2540000 401b8603 00000000 00000000
Dump PCIE PHY
Offset Value
------ -----
0x00: a800 7a58 3cee 2e46 10f0 0004 0000 2600
0x08: 6443 3318 0cc3 60b0 11e1 0400 3713 9c20
0x10: 000c 4c00 fc01 0c81 de01 0000 0000 0000
0x18: 0000 fd01 0312 0ea1 0000 0000 24eb
Dump PCI Registers
Offset Value
------ -----
0x000: 816810ec 00100407 02000022 00800010
0x010: 0000d001 00000000 a1108004 00000000
0x020: a1100004 00000000 00000000 03441854
0x030: 00000000 00000040 00000000 000001ff
0x040: ffc35001 00000008 00000000 00000000
0x050: 00817005 fee01004 00000000 00004024
0x060: 00000000 00000000 00000000 00000000
0x070: 0202b010 05908ce0 001b5110 00477c12
0x080: 10120142 00000000 00000000 00000000
0x090: 00000000 000c181f 00000000 00000006
0x0a0: 00010000 00000000 00000000 00000000
0x0b0: 00030011 00000004 00000804 00000000
0x0c0: 00000000 00000000 00000000 00000000
0x0d0: 00000003 00000000 00000000 00000000
0x0e0: 00000000 00000000 00000000 00000000
0x0f0: 00000000 00000000 00000000 00000000
0x110: 00002081
0x70c: 27ffff01
Dump MAC Registers
Offset Value
------ -----
0x00: 54 b2 03 86 1b 40 00 00 00 00 00 00 80 00 00 00
0x10: 00 00 6f 79 01 00 00 00 a9 0c 06 00 00 00 00 00
0x20: 00 80 99 76 01 00 00 00 00 00 00 00 00 00 00 00
0x30: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
0x40: 80 0f a0 57 0e cf 02 00 00 00 00 00 00 00 00 00
0x50: 10 00 cf 9c 60 11 03 04 00 00 00 00 00 00 00 00
0x60: 00 00 00 00 00 00 00 01 70 f1 01 80 91 00 80 f0
0x70: 00 00 00 00 fc f0 00 bf 07 00 00 00 00 00 67 1e
0x80: eb 24 1e 80 00 00 00 00 00 00 00 00 00 00 00 00
0x90: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
0xa0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
0xb0: 00 00 00 00 00 00 00 00 00 20 0f d2 00 00 00 00
0xc0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
0xd0: e1 00 00 32 ac 00 01 00 00 00 f3 05 40 7f b1 00
0xe0: e1 20 51 5f 00 c0 99 76 01 00 00 00 3f 00 00 00
0xf0: 00 80 40 00 00 00 00 00 ff ff ff ff 00 00 00 00
Dump Tally Counter
Statistics Value
---------- -----
tx_packets 0
rx_packets 0
tx_errors 0
rx_missed 0
align_errors 0
tx_one_collision 0
tx_multi_collision 0
rx_unicast 0
rx_broadcast 0
rx_multicast 0
tx_aborted 0
tx_underun 0
>
> Hau
>>
>> Kai-Heng
>>
>>>
>>> Kai-Heng
>>>
>>>>
>>>>> Kai-Heng
>>>>>
>>>>>>
>>>>>> Andrew
>>>>>
>>>> Heiner
>>>
>>
>>
>> ------Please consider the environment before printing this e-mail.
>
>
>
> > On Feb 14, 2020, at 5:07 PM, Hau <[email protected]> wrote:
> >
> >> Chun-Hao,
> >>
> >>> On Jan 3, 2020, at 12:53, Kai-Heng Feng
> >>> <[email protected]>
> >> wrote:
> >>>
> >>>
> >>>
> >>>> On Jan 3, 2020, at 05:24, Heiner Kallweit <[email protected]>
> wrote:
> >>>>
> >>>> On 02.01.2020 17:46, Kai-Heng Feng wrote:
> >>>>> Hi Andrew,
> >>>>>
> >>>>>> On Jan 2, 2020, at 23:21, Andrew Lunn <[email protected]> wrote:
> >>>>>>
> >>>>>> On Thu, Jan 02, 2020 at 02:59:42PM +0800, Kai Heng Feng wrote:
> >>>>>>> Hi Heiner,
> >>>>>>>
> >>>>>>> There's an 8168fp/8117 chip has SFP+ port instead of RJ45, the
> >>>>>>> phy
> >> device ID matches "Generic FE-GE Realtek PHY" nevertheless.
> >>>>>>> The problems is that, since it uses SFP+, both BMCR and BMSR
> >>>>>>> read
> >> are always zero, so Realtek phylib never knows if the link is up.
> >>>>>>>
> >>>>>>> However, the old method to read through MMIO correctly shows
> the
> >> link is up:
> >>>>>>> static unsigned int rtl8169_xmii_link_ok(struct rtl8169_private
> >>>>>>> *tp) {
> >>>>>>> return RTL_R8(tp, PHYstatus) & LinkStatus; }
> >>>>>>>
> >>>>>>> Few ideas here:
> >>>>>>> - Add a link state callback for phylib like phylink's
> >> phylink_fixed_state_cb(). However there's no guarantee that other
> >> parts of this chip works.
> >>>>>>> - Add SFP+ support for this chip. However the phy device matches
> >>>>>>> to
> >> "Generic FE-GE Realtek PHY" which may complicate things.
> >>>>>>>
> >>>>>>> Any advice will be welcome.
> >>>>>>
> >>>>>> Hi Kai
> >>>>>>
> >>>>>> Is the i2c bus accessible?
> >>>>>
> >>>>> I don't think so. It seems to be a regular Realtek 8168 device
> >>>>> with generic
> >> PCI ID [10ec:8168].
> >>>>>
> >>>>>> Is there any documentation or example code?
> >>>>>
> >>>>> Unfortunately no.
> >>>>>
> >>>>>>
> >>>>>> In order to correctly support SFP+ cages, we need access to the
> >>>>>> i2c bus to determine what sort of module has been inserted. It
> >>>>>> would also be good to have access to LOS, transmitter disable,
> >>>>>> etc, from the SFP cage.
> >>>>>
> >>>>> Seems like we need Realtek to provide more information to support
> >>>>> this
> >> chip with SFP+.
> >>>>>
> >>>> Indeed it would be good to have some more details how this chip
> >>>> handles SFP+, therefore I add Hau to the discussion.
> >>>>
> >>>> As I see it the PHY registers are simply dummies on this chip. Or
> >>>> does this chip support both, PHY and SFP+? Hopefully SFP presence
> >>>> can be autodetected, we could skip the complete PHY handling in
> >>>> this case. Interesting would be which parts of the SFP interface
> >>>> are exposed
> >> how via (proprietary) registers.
> >>>> Recently the STMMAC driver was converted from phylib to phylink,
> >>>> maybe we have to do the same with r8169 one fine day. But w/o more
> >>>> details this is just speculation, much appreciated would be
> >>>> documentation from Realtek about the
> >>>> SFP+ interface.
> >>>>
> >>>> Kai, which hardware/board are we talking about?
> >>>
> >>> It's a regular Intel PC.
> >>>
> >>> The ethernet is function 1 of the PCI device, function 0 isn't bound
> >>> to any
> >> driver:
> >>> 02:00.0 Unassigned class [ff00]: Realtek Semiconductor Co., Ltd.
> >>> Device [10ec:816e] (rev 1a)
> >>> 02:00.1 Ethernet controller [0200]: Realtek Semiconductor Co., Ltd.
> >>> RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller
> >>> [10ec:8168] (rev 22)
> >>
> >> Would it be possible to share some info on SFP support?
> > Hi Kai-Heng,
> >
> > Could you use r8168 to dump hardware info with following command.
> > cat /proc/net/r8168/ethx/*
> >
> > I want to make sure which chip you use and try to add support it in
> r8168/r8169.
>
>
> Dump Driver Variable
> Variable Value
> ---------- -----
> MODULENAME r8168
> driver version 8.048.00-NAPI
> chipset 31
> chipset_name RTL8168FP/8111FP
> mtu 1500
> NUM_RX_DESC 0x400
> cur_rx 0x0
> dirty_rx 0x0
> NUM_TX_DESC 0x400
> cur_tx 0x0
> dirty_tx 0x0
> rx_buf_sz 0x5f3
> esd_flag 0x0
> pci_cfg_is_read 0x1
> rtl8168_rx_config 0xcf00
> cp_cmd 0x20e1
> intr_mask 0x115
> timer_intr_mask 0x4000
> wol_enabled 0x1
> wol_opts 0x20
> efuse_ver 0x3
> eeprom_type 0x0
> autoneg 0x1
> duplex 0x1
> speed 1000
> advertising 0x3f
> eeprom_len 0x0
> cur_page 0x0
> bios_setting 0x0
> features 0x2
> org_pci_offset_99 0x0
> org_pci_offset_180 0xf
> issue_offset_99_event 0x0
> org_pci_offset_80 0x42
> org_pci_offset_81 0x1
> use_timer_interrrupt 0x1
> HwIcVerUnknown 0x0
> NotWrRamCodeToMicroP 0x0
> NotWrMcuPatchCode 0x0
> HwHasWrRamCodeToMicroP 0x0
> sw_ram_code_ver 0x3
> hw_ram_code_ver 0x0
> rtk_enable_diag 0x0
> ShortPacketSwChecksum 0x0
> UseSwPaddingShortPkt 0x0
> RequireAdcBiasPatch 0x0
> AdcBiasPatchIoffset 0x0
> RequireAdjustUpsTxLinkPulseTiming 0x0
> SwrCnt1msIni 0x0
> HwSuppNowIsOobVer 0x1
> HwFiberModeVer 0x0
> HwFiberStat 0x0
> HwSwitchMdiToFiber 0x0
> NicCustLedValue 0xca9
> RequiredSecLanDonglePatch 0x0
> HwSuppDashVer 0x3
> DASH 0x1
> dash_printer_enabled 0x0
> HwSuppKCPOffloadVer 0x0
> speed_mode 0x3e8
> duplex_mode 0x1
> autoneg_mode 0x1
> advertising_mode 0x3f
> aspm 0x1
> s5wol 0x1
> s5_keep_curr_mac 0x0
> eee_enable 0x0
> hwoptimize 0x0
> proc_init_num 0x2
> s0_magic_packet 0x0
> HwSuppMagicPktVer 0x2
> HwSuppCheckPhyDisableModeVer 0x3
> HwPkgDet 0x6
> HwSuppGigaForceMode 0x1
> random_mac 0x0
> org_mac_addr 54:b2:03:86:1b:40
> perm_addr 54:b2:03:86:1b:40
> dev_addr 54:b2:03:86:1b:40
>
>
> Dump Ethernet PHY
>
> Offset Value
> ------ -----
>
> ####################page 0##################
>
> 0x00: 1040 7989 001c c800 0de1 0000 0064 2001
> 0x08: 0000 0200 0000 0000 0000 0000 0000 2000
>
> Dump Extended Registers
>
> Offset Value
> ------ -----
>
> 0x00: 00000000 00000000 00000000 00000000
> 0x10: 00000000 00000000 00000000 00000000
> 0x20: 00000000 00000000 00000000 00000000
> 0x30: 00000000 00000000 00000000 00000000
> 0x40: 00000000 00000000 00000000 00000000
> 0x50: 00000000 00000000 00000000 00000000
> 0x60: 00000000 00000000 00000000 00000000
> 0x70: 00000000 00000000 00000000 00000000
> 0x80: 00000000 00000000 00000000 00000000
> 0x90: 00000000 00000000 00000000 00000000
> 0xa0: 00000000 00000000 00000000 00000000
> 0xb0: 00000000 00000000 00000000 00000000
> 0xc0: 00000000 00000000 00080002 0000002f
> 0xd0: 0000005f 00001f92 00000000 001e0021
> 0xe0: 8603b254 0000401b 00100006 00000200
> 0xf0: b2540000 401b8603 00000000 00000000
>
> Dump PCIE PHY
>
> Offset Value
> ------ -----
>
> 0x00: a800 7a58 3cee 2e46 10f0 0004 0000 2600
> 0x08: 6443 3318 0cc3 60b0 11e1 0400 3713 9c20
> 0x10: 000c 4c00 fc01 0c81 de01 0000 0000 0000
> 0x18: 0000 fd01 0312 0ea1 0000 0000 24eb
>
> Dump PCI Registers
>
> Offset Value
> ------ -----
>
> 0x000: 816810ec 00100407 02000022 00800010
> 0x010: 0000d001 00000000 a1108004 00000000
> 0x020: a1100004 00000000 00000000 03441854
> 0x030: 00000000 00000040 00000000 000001ff
> 0x040: ffc35001 00000008 00000000 00000000
> 0x050: 00817005 fee01004 00000000 00004024
> 0x060: 00000000 00000000 00000000 00000000
> 0x070: 0202b010 05908ce0 001b5110 00477c12
> 0x080: 10120142 00000000 00000000 00000000
> 0x090: 00000000 000c181f 00000000 00000006
> 0x0a0: 00010000 00000000 00000000 00000000
> 0x0b0: 00030011 00000004 00000804 00000000
> 0x0c0: 00000000 00000000 00000000 00000000
> 0x0d0: 00000003 00000000 00000000 00000000
> 0x0e0: 00000000 00000000 00000000 00000000
> 0x0f0: 00000000 00000000 00000000 00000000
> 0x110: 00002081
> 0x70c: 27ffff01
>
> Dump MAC Registers
> Offset Value
> ------ -----
>
> 0x00: 54 b2 03 86 1b 40 00 00 00 00 00 00 80 00 00 00
> 0x10: 00 00 6f 79 01 00 00 00 a9 0c 06 00 00 00 00 00
> 0x20: 00 80 99 76 01 00 00 00 00 00 00 00 00 00 00 00
> 0x30: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> 0x40: 80 0f a0 57 0e cf 02 00 00 00 00 00 00 00 00 00
> 0x50: 10 00 cf 9c 60 11 03 04 00 00 00 00 00 00 00 00
> 0x60: 00 00 00 00 00 00 00 01 70 f1 01 80 91 00 80 f0
> 0x70: 00 00 00 00 fc f0 00 bf 07 00 00 00 00 00 67 1e
> 0x80: eb 24 1e 80 00 00 00 00 00 00 00 00 00 00 00 00
> 0x90: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> 0xa0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> 0xb0: 00 00 00 00 00 00 00 00 00 20 0f d2 00 00 00 00
> 0xc0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> 0xd0: e1 00 00 32 ac 00 01 00 00 00 f3 05 40 7f b1 00
> 0xe0: e1 20 51 5f 00 c0 99 76 01 00 00 00 3f 00 00 00
> 0xf0: 00 80 40 00 00 00 00 00 ff ff ff ff 00 00 00 00
>
> Dump Tally Counter
> Statistics Value
> ---------- -----
> tx_packets 0
> rx_packets 0
> tx_errors 0
> rx_missed 0
> align_errors 0
> tx_one_collision 0
> tx_multi_collision 0
> rx_unicast 0
> rx_broadcast 0
> rx_multicast 0
> tx_aborted 0
> tx_underun 0
>
>
Hi Kai-Heng,
Attached file is r8168 that I have add SFP+ support for rtl8168fp. If possible, please give it a try.
Hau
> >
> > Hau
> >>
> >> Kai-Heng
> >>
> >>>
> >>> Kai-Heng
> >>>
> >>>>
> >>>>> Kai-Heng
> >>>>>
> >>>>>>
> >>>>>> Andrew
> >>>>>
> >>>> Heiner
> >>>
> >>
> >>
> >>
Hi Hau,
> On Feb 19, 2020, at 22:22, Hau <[email protected]> wrote:
[snipped]
>
> Hi Kai-Heng,
>
> Attached file is r8168 that I have add SFP+ support for rtl8168fp. If possible, please give it a try.
I've already tested r8168 and it does support SFP+.
What we are discussing here is to support this chip properly in mainline kernel.
This is what we've discussed so far:
https://lore.kernel.org/lkml/[email protected]/
Kai-Heng
Hi Hau,
> On Feb 19, 2020, at 22:48, Kai-Heng Feng <[email protected]> wrote:
>
> Hi Hau,
>
>> On Feb 19, 2020, at 22:22, Hau <[email protected]> wrote:
>
> [snipped]
>
>>
>> Hi Kai-Heng,
>>
>> Attached file is r8168 that I have add SFP+ support for rtl8168fp. If possible, please give it a try.
>
> I've already tested r8168 and it does support SFP+.
>
> What we are discussing here is to support this chip properly in mainline kernel.
>
> This is what we've discussed so far:
> https://lore.kernel.org/lkml/[email protected]/
Is there anything you can share so we can support this chip properly in upstream?
Kai-Heng
>
> Kai-Heng
> Hi Hau,
>
> > On Feb 19, 2020, at 22:48, Kai-Heng Feng <[email protected]>
> wrote:
> >
> > Hi Hau,
> >
> >> On Feb 19, 2020, at 22:22, Hau <[email protected]> wrote:
> >
> > [snipped]
> >
> >>
> >> Hi Kai-Heng,
> >>
> >> Attached file is r8168 that I have add SFP+ support for rtl8168fp. If
> possible, please give it a try.
> >
> > I've already tested r8168 and it does support SFP+.
> >
> > What we are discussing here is to support this chip properly in mainline
> kernel.
> >
> > This is what we've discussed so far:
> > https://lore.kernel.org/lkml/2D8F5FFE-3EC3-480B-9D15-
> [email protected]/
>
> Is there anything you can share so we can support this chip properly in
> upstream?
>
Hi Kai-Heng,
For adding SFP+ support for rtl8168fp,
1.Some power saving features must be disabled, like APDLS/EEE/EEEPLUS...
2.PHY capability must be set to auto-negation.
I am kind of busy this week. I will try to add support for this chip into upstream next week.
Thanks,
Hau
> ------Please consider the environment before printing this e-mail.
> Hi Kai-Heng,
>
> For adding SFP+ support for rtl8168fp,
> 1.Some power saving features must be disabled, like APDLS/EEE/EEEPLUS...
> 2.PHY capability must be set to auto-negation.
>
> I am kind of busy this week. I will try to add support for this chip into upstream next week.
Is it possible to access the I2C bus? The GPIO lines from the SFP
socket like LOS, TX Disable?
Thanks
Andrew
> > Hi Kai-Heng,
> >
> > For adding SFP+ support for rtl8168fp, 1.Some power saving features
> > must be disabled, like APDLS/EEE/EEEPLUS...
> > 2.PHY capability must be set to auto-negation.
> >
> > I am kind of busy this week. I will try to add support for this chip into
> upstream next week.
>
> Is it possible to access the I2C bus? The GPIO lines from the SFP socket like
> LOS, TX Disable?
>
I am asking our hardware team. I will get back to you once I get the answer.
Thanks,
Hau
------Please consider the environment before printing this e-mail.
>
> > > Hi Kai-Heng,
> > >
> > > For adding SFP+ support for rtl8168fp, 1.Some power saving features
> > > must be disabled, like APDLS/EEE/EEEPLUS...
> > > 2.PHY capability must be set to auto-negation.
> > >
> > > I am kind of busy this week. I will try to add support for this chip
> > > into
> > upstream next week.
> >
> > Is it possible to access the I2C bus? The GPIO lines from the SFP
> > socket like LOS, TX Disable?
> >
This chip can get fiber info through I2C. It also can detect LOS or control TX Disable via GPIO.
Thanks,
Hau
------Please consider the environment before printing this e-mail.
> This chip can get fiber info through I2C. It also can detect LOS or control TX Disable via GPIO.
O.K. Good
You should be looking at using phylink to manage the SFP.
Andrew
On 06.03.2020 16:58, Andrew Lunn wrote:
>> This chip can get fiber info through I2C. It also can detect LOS or control TX Disable via GPIO.
>
I think the question is whether the chip exposes these signals for kernel use.
- Is there any register to read e.g. LOS status from?
- Is the I2C bus accessed only by chip firmware, or can the OS access the I2C bus somehow?
> O.K. Good
>
> You should be looking at using phylink to manage the SFP.
>
> Andrew
>