http://bugzilla.opensuse.org/show_bug.cgi?id=1173085 http://bugzilla.opensuse.org/show_bug.cgi?id=1173085#c1 --- Comment #1 from Olav Reinert --- Created attachment 838908 --> http://bugzilla.opensuse.org/attachment.cgi?id=838908&action=edit Kernel messages -- You are receiving this mail because: You are the assignee for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1173085 http://bugzilla.opensuse.org/show_bug.cgi?id=1173085#c2 Takashi Iwai changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |seroton10@gmail.com, | |tiwai@suse.com Flags| |needinfo?(seroton10@gmail.c | |om) --- Comment #2 from Takashi Iwai --- Do you still have Leap 15.1 system or kernel on your machine? If not, could you try to install Leap 15.1 kernel on top of your 15.2 system with zypper in --oldpackage option and check whether the device shows up properly? We need some direct comparison of kernels while keeping the rest, to exclude other possibilities. If Leap 15.1 kernel really works, please upload dmesg output and hwinfo from from there. -- You are receiving this mail because: You are the assignee for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1173085 http://bugzilla.opensuse.org/show_bug.cgi?id=1173085#c4 --- Comment #4 from Olav Reinert --- Created attachment 838938 --> http://bugzilla.opensuse.org/attachment.cgi?id=838938&action=edit Kernel messages on kernel-default-4.12.14-lp151.28.52.1.x86_64 -- You are receiving this mail because: You are the assignee for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1173085 http://bugzilla.opensuse.org/show_bug.cgi?id=1173085#c6 --- Comment #6 from Takashi Iwai --- Thanks. Could you try to re-probe the PCI device? e.g. echo 1 > /devices/pci0000:00/0000:00:01.2/0000:02:00.0/0000:03:04.0/0000:06:00.1/remove echo 1 > devices/pci0000:00/0000:00:01.2/0000:02:00.0/0000:03:04.0/rescan Often it's something wrongly powered at a fresh boot and the forcible rescan helps. -- You are receiving this mail because: You are the assignee for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1173085 http://bugzilla.opensuse.org/show_bug.cgi?id=1173085#c8 --- Comment #8 from Takashi Iwai --- OK thanks for checking. Meanwhile I noticed that your chip hasn't been supported properly on r8169, and its support was added very recently (5.5). Maybe it worked casually in the past although it wasn't supported. I'll check whether the support patch can be backported to Leap 15.2 kernel easily. -- You are receiving this mail because: You are the assignee for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1173085 http://bugzilla.opensuse.org/show_bug.cgi?id=1173085#c14 --- Comment #14 from Olav Reinert --- (In reply to Takashi Iwai from comment #11)

You might need to unload realtek kernel module (it's phy driver) beforehand.

The "realtek" module isn't loaded at any time, neither before nor after the remove+rescan (for which I don't show log entries here, they are the same as reported earlier): bb:~ # modinfo realtek filename: /lib/modules/5.3.18-lp152.1.gd4f520d-default/kernel/drivers/net/phy/realtek.ko license: GPL author: Johnson Leung description: Realtek PHY driver suserelease: openSUSE Leap 15.2 srcversion: 115992CBFCAF74B6BF13EE3 alias: mdio:0000000000011100110010?????????? depends: libphy retpoline: Y intree: Y name: realtek vermagic: 5.3.18-lp152.1.gd4f520d-default SMP mod_unload modversions sig_id: PKCS#7 signer: home:tiwai OBS Project sig_key: 30:4F:31:1F:30:1D:06:03:55:04:03:0C:16:68:6F:6D:65:3A:74:69: 77:61:69:20:4F:42:53:20:50:72:6F:6A:65:63:74:31:2C:30:2A:06: 09:2A:86:48:86:F7:0D:01:09:01:16:1D:68:6F:6D:65:3A:74:69:77: 61:69:40:62:75:69:6C:64:2E:6F:70:65:6E:73:75:73:65:2E:6F:72: 67:02:09:00:87:F7:87:C1:D0:A0:DB:59 sig_hashalgo: sha256 signature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bb:~ # rmmod realtek rmmod: ERROR: Module realtek is not currently loaded bb:~ # echo 1

/sys/devices/pci0000:00/0000:00:01.2/0000:02:00.0/0000:03:04.0/0000:06:00.1/remove bb:~ # rmmod realtek rmmod: ERROR: Module realtek is not currently loaded bb:~ # echo 1 /sys/devices/pci0000:00/0000:00:01.2/0000:02:00.0/0000:03:04.0/rescan bb:~ # rmmod realtek rmmod: ERROR: Module realtek is not currently loaded

-- You are receiving this mail because: You are the assignee for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1173085 http://bugzilla.opensuse.org/show_bug.cgi?id=1173085#c16 --- Comment #16 from Olav Reinert --- (In reply to Takashi Iwai from comment #12)

Also, could you confirm that 5.7.x kernel in OBS Kernel:stable works? If that's confirmed, I can try to backport up to 5.7.

It works With kernel-default-5.7.3-1.1.g44c4af0.x86_64 from OBS Kernel:stable. -- You are receiving this mail because: You are the assignee for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1173085 http://bugzilla.opensuse.org/show_bug.cgi?id=1173085#c18 --- Comment #18 from Olav Reinert --- (In reply to Takashi Iwai from comment #13)

FYI, I created another test kernel in OBS home:tiwai:bsc1173085-2 repo containing more backports. If 5.7 kernel worked, please give this a try.

This kernel boots just fine, but doesn't solve the problem with the Realtek NIC. -- You are receiving this mail because: You are the assignee for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1173085 http://bugzilla.opensuse.org/show_bug.cgi?id=1173085#c20 Takashi Iwai changed: What |Removed |Added ---------------------------------------------------------------------------- Flags| |needinfo?(seroton10@gmail.c | |om) --- Comment #20 from Takashi Iwai --- Thanks. Then it might be something wrong in the phy side. I found a bug with the doubly entries, and that might be the cause of "RTL8208" error messages. I'm build yet again Leap 15.2 kernel in OBS home:tiwai:bsc1173085-3 repo with the correction. Even if this still shows the same error, it'll indicate the phy number in the error message, so it might help. Let's see. -- You are receiving this mail because: You are the assignee for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1173085 http://bugzilla.opensuse.org/show_bug.cgi?id=1173085#c22 Olav Reinert changed: What |Removed |Added ---------------------------------------------------------------------------- Flags|needinfo?(seroton10@gmail.c | |om) | --- Comment #22 from Olav Reinert --- Created attachment 838976 --> http://bugzilla.opensuse.org/attachment.cgi?id=838976&action=edit dmesg using patched kernel-default-5.3.18-lp152.1.1.g4cce4f9 -- You are receiving this mail because: You are the assignee for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1173085 Takashi Iwai changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |tbogendoerfer@suse.com -- You are receiving this mail because: You are the assignee for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1173085 http://bugzilla.opensuse.org/show_bug.cgi?id=1173085#c25 Olav Reinert changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |kernel-bugs@opensuse.org Flags| |needinfo?(kernel-bugs@opens | |use.org) --- Comment #25 from Olav Reinert --- (In reply to Takashi Iwai from comment #23)

Thanks, finally good to hear :)

The only slight concern is that the new kernel shows an error at probe: [ 6.526867] r8169 0000:06:00.1 (unnamed net_device) (uninitialized): rtl_txcfg_empty_cond == 0 (loop: 42, delay: 100).

while it isn't shown in 5.7.1 kernel. But it seems like a one-off thing, so maybe not too serious.

It seems I was too quick to declare success, unfortunately. With your kernel-default-5.3.18-lp152.1.1.g4cce4f9, the Realtek interface is indeed detected, but because wicked fails to assign an IP-address to it, it remains in state DOWN: # ifstatus eth1 eth1 setup-in-progress link: #2, state device-up, mtu 1500 type: ethernet, hwaddr 34:97:f6:32:81:3f config: compat:suse:/etc/sysconfig/network/ifcfg-eth1 Please let me know what other debug information you need to diagnose this. (In contrast, with kernel-default-5.7.3-1.1.g44c4af0 from OBS Kernel:stable, eth1 is initialized correctly, and the system is able to send and receive network traffic with it.) -- You are receiving this mail because: You are on the CC list for the bug. You are the assignee for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1173085 http://bugzilla.opensuse.org/show_bug.cgi?id=1173085#c27 Olav Reinert changed: What |Removed |Added ---------------------------------------------------------------------------- Flags|needinfo?(kernel-bugs@opens | |use.org) | --- Comment #27 from Olav Reinert --- (In reply to Takashi Iwai from comment #26)

Then we may try another approach. Instead of backport tons of patches, revert / work around the change that caused the probe failure.

I'm building another test kernel in OBS home:tiwai:bsc1173085-3 repo with a downstream patch that allows the default MAC version assignment for your chip. It's a very small change and less intrusive. Let's see whether it works.

This fix (in your kernel-default-5.3.18-8.1.g603a44a.x86_64 package) works! The Realtek NIC is detected, and wicked is able to bring it UP as expected. Even though the "unknown chip XID" error is still present in the kernel messages, I am satisfied with this solution. -- You are receiving this mail because: You are on the CC list for the bug. You are the assignee for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1173085 http://bugzilla.opensuse.org/show_bug.cgi?id=1173085#c29 Takashi Iwai changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution|--- |FIXED --- Comment #29 from Takashi Iwai --- Finally a good news :) FWIW, below is the downstream fix patch for SLE15-SP2 / Leap 15.2. I pushed this for the merge to SLE15-SP2-UPDATE branch, which will be used for the first Leap 15.2 update kernel. Thanks for the patient testing! -- You are receiving this mail because: You are on the CC list for the bug. You are the assignee for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1173085 http://bugzilla.opensuse.org/show_bug.cgi?id=1173085#c31 --- Comment #31 from Olav Reinert --- (In reply to Takashi Iwai from comment #29)

Thanks for the patient testing!

No problem - thanks for your hard work creating patches. It's good to know that my hardware will be supported by Leap 15.2, eventually. -- You are receiving this mail because: You are on the CC list for the bug. You are the assignee for the bug.