Page 59 of 135

Porteus Kernel Builder

Posted: 26 May 2020, 13:44
by Blaze
Hi johnny.mau,
do you want to disable all options in these sections?

Code: Select all

<*> Serial ATA and Parallel ATA drivers (libata)
[*] Multiple devices driver support (RAID and LVM)
Image

Thanks.

Porteus Kernel Builder

Posted: 11 Jun 2020, 18:02
by Blaze
[Updated kernel Examples]
==== AUFS Kernel only for Porteus ====
Simple package (vmlinuz, 000-kernel.xzm, 06-crippled_sources-NNN-XXbit.xzm)

[5.7.2] <-- NEW : "All patches" patching was done.
32bit-ALL-kernel5.7.2.tar (89 M)
http://www.mediafire.com/file/hwpkyttzg ... l5.7.2.tar
md5sum: a50ea3a4ad5ee906afeff4dd04bad348 32bit-ALL-kernel5.7.2.tar

[5.7.2] <-- NEW : "All patches" patching was done.
64bit-ALL-kernel5.7.2.tar (94 M)
http://www.mediafire.com/file/g5ub39j5p ... l5.7.2.tar
md5sum: 86f103f5025eca92b0255e29b64627d5 64bit-ALL-kernel5.7.2.tar

Note : Compiler that was used.
for kernel 5.7.2 : gcc-9.3.0-x86_64

Porteus Kernel Builder

Posted: 15 Jun 2020, 12:48
by dorohov.alex77
Blaze Thanks so much for your support.)

Porteus Kernel Builder

Posted: 24 Jun 2020, 14:10
by maia
Hello Blaze.

I have used the NEKO script for a long time to download the version and apply the patches.

Usually use the same .config that I created with the localmodconfig function. It worked fine until version 5.7.2

When I compiled version 5.7.4 and 5.7.5 I did not notice any errors during the compilation but my WIFI card no longer activated. I am not identifying the flaw in the .config file because the ath9k_htc module compiles and goes up and the firmware is read.

I tried the 5.7.2 kernel that you posted and WIFI works satisfactorily, but I notice in dmesg a constant error message that I couldn't identify the cause and I don't know if it's serious: buggy seq_file .next function 0xffffffff8115402e did not updated position index.

Could you send me your .config for me to compare with mine and try to identify the failure?

Thankful Maia

Porteus Kernel Builder

Posted: 24 Jun 2020, 16:28
by Blaze
Hi maia.

I use config from neko by default.
You can download it from here.

I have the same warning as you

Code: Select all

buggy seq_file .next function 0xffffffff81154272 did not updated position index
[Updated kernel Examples]
==== AUFS Kernel only for Porteus ====
Simple package (vmlinuz, 000-kernel.xzm, 06-crippled_sources-NNN-XXbit.xzm)

[5.7.5] <-- NEW : "All patches" patching was done.
32bit-ALL-kernel5.7.5.tar (89 M)
http://www.mediafire.com/file/gcgkanx06 ... l5.7.5.tar
4858f839f7330c1886feb932455eb42 32bit-ALL-kernel5.7.5.tar

[5.7.5] <-- NEW : "All patches" patching was done.
64bit-ALL-kernel5.7.5.tar (94 M)
http://www.mediafire.com/file/v2yx2ps7l ... l5.7.5.tar
89d837c4d20c3fd0ab404dc9c9f827da 64bit-ALL-kernel5.7.5.tar

Note : Compiler that was used.
for kernel 5.7.5 : gcc-9.3.0-x86_64

Porteus Kernel Builder

Posted: 24 Jun 2020, 17:14
by maia
Unfortunately it didn't work.
I will continue doing some tests but I believe the problem is with the version above 5.7.4.

Note the difference between the two kernels:

5.7.5 Kernel

[ 13.175827] usb 3-2: ath9k_htc: Transferred FW: ath9k_htc/htc_9271-1.4.0.fw, size: 51008
[ 13.442181] ath9k_htc 3-2:1.0: ath9k_htc: HTC initialized with 33 credits
[ 42.941591] ath9k_htc: Failed to initialize the device
[ 42.941765] usb 3-2: ath9k_htc: USB layer deinitialized
[ 153.035557] usb 3-2: ath9k_htc: Firmware ath9k_htc/htc_9271-1.4.0.fw requested
[ 153.320631] usb 3-2: ath9k_htc: Transferred FW: ath9k_htc/htc_9271-1.4.0.fw, size: 51008
[ 153.572385] ath9k_htc 3-2:1.0: ath9k_htc: HTC initialized with 33 credits
[ 183.037287] ath9k_htc: Failed to initialize the device
[ 183.037464] usb 3-2: ath9k_htc: USB layer deinitialized



5.7.2 Kernel

[ 35.120292] usb 3-2: Product: USB2.0 WLAN
[ 35.120294] usb 3-2: Manufacturer: ATHEROS
[ 35.120297] usb 3-2: SerialNumber: 12345
[ 35.691184] usb 3-2: ath9k_htc: Firmware ath9k_htc/htc_9271-1.4.0.fw requested
[ 35.692156] usbcore: registered new interface driver ath9k_htc
[ 35.983500] usb 3-2: ath9k_htc: Transferred FW: ath9k_htc/htc_9271-1.4.0.fw, size: 51008
[ 36.235260] ath9k_htc 3-2:1.0: ath9k_htc: HTC initialized with 33 credits
[ 36.501961] ath9k_htc 3-2:1.0: ath9k_htc: FW Version: 1.4
[ 36.501965] ath9k_htc 3-2:1.0: FW RMW support: On
[ 36.501967] ath: EEPROM regdomain: 0x809c
[ 36.501968] ath: EEPROM indicates we should expect a country code
[ 36.501970] ath: doing EEPROM country->regdmn map search
[ 36.501971] ath: country maps to regdmn code: 0x52
[ 36.501973] ath: Country alpha2 being used: CN
[ 36.501974] ath: Regpair used: 0x52
[ 36.506589] ieee80211 phy1: Atheros AR9271 Rev:1


Anyway, Blaze thanks for the help.

Porteus Kernel Builder

Posted: 24 Jun 2020, 18:01
by Blaze
maia, I sent a message to neko.
If something comes from him, I will let you know.

Porteus Kernel Builder

Posted: 24 Jun 2020, 18:21
by maia
TKS :)

Porteus Kernel Builder

Posted: 24 Jun 2020, 21:25
by donald
Seems to be an upstream issue
Confirmed for 5.7.3 (where ath9k changes were made)
Works fine after downgrading to 5.7.2
Quotes from
https://bugzilla.kernel.org/show_bug.cgi?id=208251

Porteus Kernel Builder

Posted: 24 Jun 2020, 21:41
by maia
Thank you Donald.
I will temporarily go back a few steps until version 5.6.
In 5.6 it worked smoothly, without strange error warnings, in which I cannot understand if it is serious or not. :)

Porteus Kernel Builder

Posted: 25 Jun 2020, 17:47
by Blaze
maia, I got a response from neko and a new mkKernel 20.06.22-1 with linux-firmware-20200519.8ba6fa6-noarch-1plus.xzm

Can you try a new kernel 5.7.6 ?

[Updated kernel Examples]
==== AUFS Kernel only for Porteus ====
Simple package (vmlinuz, 000-kernel.xzm, 06-crippled_sources-NNN-XXbit.xzm)

[5.7.6] <-- NEW : "All patches" patching was done.
32bit-ALL-kernel5.7.6.tar (89 M)
http://www.mediafire.com/file/ejn0cqfe8 ... l5.7.6.tar
e14431fb4410c5b782724ff8dc10d07d 32bit-ALL-kernel5.7.6.tar

[5.7.6] <-- NEW : "All patches" patching was done.
64bit-ALL-kernel5.7.6.tar (94 M)
http://www.mediafire.com/file/af3d8lqej ... l5.7.6.tar
01c6495171c4403fd5bbf53067033a52 64bit-ALL-kernel5.7.6.tar

Note : Compiler that was used.
for kernel 5.7.6 : gcc-10.1.0-x86_64

Отправлено спустя 14 часов 44 минуты 14 секунд:

Code: Select all

buggy seq_file .next function 0xffffffff81152e71 did not updated position index
is still in dmesg and it seems like it's considered normal.
More info is here https://github.com/torvalds/linux/commi ... ee92e49ac4 and here https://www.spinics.net/lists/mm-commits/msg143561.html
Image

Porteus Kernel Builder

Posted: 26 Jun 2020, 16:18
by Ed_P
Will you, or neko :) , be adding the seq_read-info-message-about-buggy-next-functions-fix.patch fix referenced in your link to the next Porteus kernel?

Porteus Kernel Builder

Posted: 26 Jun 2020, 16:40
by Blaze
Ed_P, I checked source code of fs/seq_file.c kernel 5.7.6 and this fix was applied by default.

Porteus Kernel Builder

Posted: 26 Jun 2020, 21:44
by Ed_P
This is what I see with 5.7.6.

Code: Select all

Jun 25 12:16:13 porteus syslogd 1.5.1: restart.
Jun 25 12:16:14 porteus kernel: klogd 1.5.1, log source = /proc/kmsg started.
Jun 25 12:16:14 porteus kernel: [    0.000000] Linux version 5.7.6-porteus (root@porteus) (gcc version 10.1.0 (GCC), GNU ld (GNU Binutils) 2.34.0) #1 SMP Thu Jun 25 19:48:39 UTC 2020
Jun 25 12:16:14 porteus kernel: [    0.000000] Kernel is locked down from Kernel configuration; see man kernel_lockdown.7
 :
 :
 :
 :

Jun 25 16:20:01 porteus kernel: [  232.974808] buggy seq_file .next function 0xffffffff81152e71 did not updated position index
Jun 25 16:20:01 porteus kernel: [  232.975475] buggy seq_file .next function 0xffffffff81152e71 did not updated position index
Jun 25 16:20:01 porteus kernel: [  232.976207] buggy seq_file .next function 0xffffffff81152e71 did not updated position index
Jun 25 16:20:02 porteus kernel: [  234.443635] buggy seq_file .next function 0xffffffff81152e71 did not updated position index
Jun 25 16:20:02 porteus kernel: [  234.444262] buggy seq_file .next function 0xffffffff81152e71 did not updated position index
Jun 25 16:20:02 porteus kernel: [  234.445449] buggy seq_file .next function 0xffffffff81152e71 did not updated position index
Jun 25 16:20:02 porteus kernel: [  234.445868] buggy seq_file .next function 0xffffffff81152e71 did not updated position index
Jun 25 16:20:02 porteus kernel: [  234.533245] buggy seq_file .next function 0xffffffff81152e71 did not updated position index
 :
 :
 :
 
Jun 25 16:35:42 porteus kernel: [ 1174.477815] buggy seq_file .next function 0xffffffff81152e71 did not updated position index
Jun 25 16:35:42 porteus kernel: [ 1174.479228] buggy seq_file .next function 0xffffffff81152e71 did not updated position index
Jun 25 16:35:42 porteus kernel: [ 1174.480690] buggy seq_file .next function 0xffffffff81152e71 did not updated position index
Jun 25 16:35:42 porteus kernel: [ 1174.482230] buggy seq_file .next function 0xffffffff81152e71 did not updated position index
Jun 25 16:35:42 porteus kernel: [ 1174.483925] buggy seq_file .next function 0xffffffff81152e71 did not updated position index
Jun 25 16:35:52 porteus kernel: [ 1184.511535] buggy seq_file .next function 0xffffffff81152e71 did not updated position index

Porteus Kernel Builder

Posted: 27 Jun 2020, 09:59
by maia
Hello Blaze.
I just tested the version now and it still has the same bug.

I'm glad I have two other wifi cards. :)

Thanks for all..



[ 28.074093] ath9k_htc 3-2:1.0: ath9k_htc: HTC initialized with 33 credits
[ 57.524032] ath9k_htc: Failed to initialize the device
[ 57.524200] usb 3-2: ath9k_htc: USB layer deinitialized