Dev1 Galaxy Forum / acpi errors on each kernel update

Re: acpi errors on each kernel update

dummy@example.com (Head_on_a_Stick) — Sat, 27 Nov 2021 09:47:37 +0000

bai4Iej2need wrote:

seems I have to do more with newer kernel versions,
I like to have these fixed for this hardware and if possible within the operation of new kernel updates.

Have you considered informing the kernel developers? I don't think they can fix problems of which they are unaware. Be sure to check for matching extant issues on the kernel bug tracker first though.

Re: acpi errors on each kernel update

dummy@example.com (Altoid) — Fri, 26 Nov 2021 15:51:56 +0000

Hello:

bai4Iej2need wrote:

Thanks

You're welcome.

bai4Iej2need wrote:

I can/could 'iasl -c' a new DSDT ...

Yes.
Provided you were able enough and found your way around the obscure maze of DSDT stanzas.

Now, if the kernel you are using was compiled with ACPI_TABLE_UPGRADE=y you can still load an alternate DSDT table using the acpi_override method.

Here's the list of links I have kept, some may be dead by now.
There are one or two that have the recipe to load an alternate DSDT via acpi_override in one of these groups.

I used it and was able to load a new table successfully.

Override: <---- here
https://delta-xi.net/#056
https://wiki.archlinux.org/index.php/DSDT
https://cgc.com.kw/tag/dsdt-override/

Patches:
https://github.com/RehabMan/Laptop-DSDT … estart.txt
https://github.com/RehabMan/Intel-NUC-D … DT-PTS.dsl
https://www.tonymacx86.com/threads/shut … dt.229418/

GRUB:
https://github.com/tsgates/arch-wiki-ma … ki/DSDT.md

Then there's all these, some of which may have been worthwhile, but I can't recall which:

https://github.com/hirschmann/nbfc/wiki … k%27s-DSDT
https://cannibalcandy.wordpress.com/201 … acpi-woes/
https://sit1way.wordpress.com/2013/04/1 … uggy-dsdt/
http://blog.michael.kuron-germany.de/20 … compiling/
https://lists.gnu.org/archive/html/bug- … 00008.html
https://github.com/torvalds/linux/blob/ … pi/osi.txt
https://patchwork.kernel.org/patch/2831128/
https://www.kernel.org/doc/Documentatio … erride.txt
https://www.tldp.org/HOWTO/SCSI-2.4-HOWTO/kconfig.html
https://01.org/linux-acpi/documentation/overriding-dsdt
https://www.insanelymac.com/forum/topic … dt-editor/
http://wiki.osx86project.org/wiki/index.php/DSDT
http://acpi.sourceforge.net/dsdt/
https://gaugusch.at/kernel.shtml
https://init-6.bitbucket.io/content/201 … r-central/
https://gist.github.com/rschmukler/1101770

Unfortunately DSDT editing is the almost exclusive realm of Mac boxes/laptop owners who want to run Linux on them.
This means that most if not all the information to be found is for that hardware, albeit with quite a bit of overlapping with Wintel stuff.
A fortunate thing because help is not available (or offered) to non-Mac people.

Long story short: I was eventually able to fix most of the BIOS errors/warnings detected with the MS based tools I found on line and came up with a new DSDT.

Much to my chagrin and after a great many hours hard work, the new DSDT did not solve the problems the crap BIOS my Sun Ultra 24 has.
Save getting rid of most of the errors/warnings (not all), did nothing else for me, at least nothing noticeable.

Which meant that the Linux kernel was doing its workaround properly. 8^)

But I had to do it all over again when the kernel got upgraded and after I forgot to do it for a second time, I decided it was not worth the hassle.
It did not solve my Ultra 24 issues so that was it.

HTH.

Best,

Re: acpi errors on each kernel update

dummy@example.com (bai4Iej2need) — Fri, 26 Nov 2021 14:29:28 +0000

Thanks
I can/could 'iasl -c' a new DSDT , but once the kernel had a way to incorporate that in the boot , but it was lost after kernel 2.6
Best would be to have that integrated in the initramfs build.
Chances are nothing to have that in the BIOS EEPROM.

Re: acpi errors on each kernel update

dummy@example.com (Altoid) — Fri, 26 Nov 2021 12:16:12 +0000

Hello:

bai4Iej2need wrote:

... an old piece from Intel atoms time.
D945GCLF2D ...

All these warning are like/similar to the ones I get in dmesg when booting my Sun Ultra 24 WS:

groucho@devuan:~$ sudo dmesg | grep -i "acpi bios"
[    0.018306] ACPI BIOS Warning (bug): 32/64X length mismatch in FADT/Gpe0Block: 128/64 (20200925/tbfadt-569)
[    1.272873] ACPI BIOS Error (bug): Could not resolve symbol [\_SB._OSC.SUPP], AE_NOT_FOUND (20200925/psargs-330)

groucho@devuan:~$ sudo dmesg | grep -i "acpi error"
[    1.272882] ACPI Error: Aborting method \_SB._OSC due to previous error (AE_NOT_FOUND) (20200925/psparse-531)
groucho@devuan:~$

I run Devuan Beowulf on my Asus EEE 1000HE which has an Atom 280 processor and also get a number of ACPI warnings but it works properly.

Or so it seems. 8^/

groucho@eee-dev3:~$ sudo dmesg | grep -i "acpi warning"
[   14.086823] ACPI Warning: SystemIO range 0x0000000000000828-0x000000000000082F conflicts with OpRegion 0x0000000000000800-0x000000000000087F (\PMIO) (20180810/utaddress-213)
[   14.157275] ACPI Warning: SystemIO range 0x0000000000000828-0x000000000000082F conflicts with OpRegion 0x0000000000000800-0x000000000000087F (\_SB.PCI0.SBRG.PMS0) (20180810/utaddress-213)
[   14.226890] ACPI Warning: SystemIO range 0x00000000000004B0-0x00000000000004BF conflicts with OpRegion 0x0000000000000480-0x00000000000004BF (\_SB.PCI0.SBRG.GPBX) (20180810/utaddress-213)
[   14.290513] ACPI Warning: SystemIO range 0x0000000000000480-0x00000000000004AF conflicts with OpRegion 0x0000000000000480-0x00000000000004BF (\_SB.PCI0.SBRG.GPBX) (20180810/utaddress-213)
groucho@eee-dev3:~$

I understand that these ACPI errors/warnings are caused by badly written BIOSes.
ie: The result of a blatant lack of adherence to ACPI standards by BIOS writers, OEMs and their permanent bowing to the very lax ACPI compliance by Microsoft.

Many years ago this was labelled the "Wintel" way of doing things and the main reason most motherboards support MS OSs (from XP home to Vista) but not Linux.

From what I have read, the Linux kernel is supposed to be able to 'work around' these problems but this does not happen on all motherboards/BIOSes.
And as time passes by, a lot of this hardware, which would work properly with a correctly written BIOS, ends up falling by the wayside.

You may want to try booting a plain/basic Beowulf Devuan net install installation from a USB sick with ACPI disabled in the kernel command line ie; ACPI=off and see what (if any) errors/warnings you get from that via the install logs and dmesg.

Maybe your server can do well enough without ACPI?

Sorry but that's all I can come up with.

Best,

acpi errors on each kernel update

dummy@example.com (bai4Iej2need) — Fri, 26 Nov 2021 08:00:19 +0000

My home server is an old piece from Intel atoms time.
D945GCLF2D_ProductGuide02.pdf D945GCLF2D_QuickRef01.pdf

uname -a
Linux eaxah8eipiop 4.19.0-18-amd64 #1 SMP Debian 4.19.208-1 (2021-09-29) x86_64 GNU/Linux

cat /etc/devuan_version
beowulf

On each kernel update it reproduces acpi Errors which I have silenced the update before.

################### Logwatch 7.4.3 (04/27/16) #################### 
        Processing Initiated: Tue Sep  7 00:25:06 2021
.... 
WARNING:  Kernel Errors Present
    ACPI Error: AE_AML_LOOP_TIM ...:  635 Time(s)
    ACPI Error: Method parse/ex ...:  1905 Time(s)

This usually occupies one core of 4 for 100%

################### Logwatch 7.4.3 (04/27/16) #################### 
        Processing Initiated: Fri Nov 26 00:25:05 2021
  --------------------- Kernel Begin ------------------------

 WARNING:  Kernel Errors Present
    ACPI Error: AE_AML_LOOP_TIM ...:  20 Time(s)
    ACPI Error: Method parse/ex ...:  60 Time(s)

/ # ls -l vmlinuz*
lrwxrwxrwx 1 root root 28 Oct 11 00:49 vmlinuz -> boot/vmlinuz-4.19.0-18-amd64
lrwxrwxrwx 1 root root 28 Oct 11 00:49 vmlinuz.old -> boot/vmlinuz-4.19.0-17-amd64

So Update was on Oct 11.

grep ACPI dmesg_ACPI_Warning_Error.txt.1616445744 
[    4.119887] ACPI Warning: SystemIO range 0x0000000000000428-0x000000000000042F conflicts with OpRegion 0x0000000000000428-0x0000000000000429 (\GPE0) (20180810/utaddress-213)
[    4.119907] ACPI: This conflict may cause random problems and system instability
[    4.119910] ACPI: If an ACPI driver is available for this device, you should use it instead of the native driver
[    4.119920] ACPI Warning: SystemIO range 0x0000000000000500-0x000000000000052F conflicts with OpRegion 0x000000000000050C-0x000000000000050F (\IGPO) (20180810/utaddress-213)
[    4.119933] ACPI: This conflict may cause random problems and system instability
[    4.119936] ACPI: If an ACPI driver is available for this device, you should use it instead of the native driver
[ 4294.391155] ACPI Warning: SystemIO range 0x0000000000000684-0x0000000000000684 conflicts with OpRegion 0x0000000000000680-0x00000000000006FF (\_SB.PCI0.LPC.RUNT) (20180810/utaddress-213)
[ 4294.391171] ACPI Warning: SystemIO range 0x0000000000000684-0x0000000000000684 conflicts with OpRegion 0x0000000000000680-0x00000000000006FF (\_SB.PCI0.LPC.RTIO) (20180810/utaddress-213)
[ 4294.391181] ACPI: This conflict may cause random problems and system instability
[ 4294.391183] ACPI: If an ACPI driver is available for this device, you should use it instead of the native driver
[ 4294.391186] ACPI Warning: SystemIO range 0x00000000000006B3-0x00000000000006B6 conflicts with OpRegion 0x0000000000000680-0x00000000000006FF (\_SB.PCI0.LPC.RUNT) (20180810/utaddress-213)
[ 4294.391196] ACPI Warning: SystemIO range 0x00000000000006B3-0x00000000000006B6 conflicts with OpRegion 0x0000000000000680-0x00000000000006FF (\_SB.PCI0.LPC.RTIO) (20180810/utaddress-213)
[ 4294.391205] ACPI: This conflict may cause random problems and system instability
[ 4294.391207] ACPI: If an ACPI driver is available for this device, you should use it instead of the native driver
[ 4294.391209] ACPI Warning: SystemIO ragrep ACPI dmesg_ACPI_Warning_Error.txt.1616445744 
[    4.119887] ACPI Warning: SystemIO range 0x0000000000000428-0x000000000000042F conflicts with OpRegion 0x0000000000000428-0x0000000000000429 (\GPE0) (20180810/utaddress-213)
[    4.119907] ACPI: This conflict may cause random problems and system instability
[    4.119910] ACPI: If an ACPI driver is available for this device, you should use it instead of the native driver
[    4.119920] ACPI Warning: SystemIO range 0x0000000000000500-0x000000000000052F conflicts with OpRegion 0x000000000000050C-0x000000000000050F (\IGPO) (20180810/utaddress-213)
[    4.119933] ACPI: This conflict may cause random problems and system instability
[    4.119936] ACPI: If an ACPI driver is available for this device, you should use it instead of the native driver
[ 4294.391155] ACPI Warning: SystemIO range 0x0000000000000684-0x0000000000000684 conflicts with OpRegion 0x0000000000000680-0x00000000000006FF (\_SB.PCI0.LPC.RUNT) (20180810/utaddress-213)
[ 4294.391171] ACPI Warning: SystemIO range 0x0000000000000684-0x0000000000000684 conflicts with OpRegion 0x0000000000000680-0x00000000000006FF (\_SB.PCI0.LPC.RTIO) (20180810/utaddress-213)
[ 4294.391181] ACPI: This conflict may cause random problems and system instability
[ 4294.391183] ACPI: If an ACPI driver is available for this device, you should use it instead of the native driver
[ 4294.391186] ACPI Warning: SystemIO range 0x00000000000006B3-0x00000000000006B6 conflicts with OpRegion 0x0000000000000680-0x00000000000006FF (\_SB.PCI0.LPC.RUNT) (20180810/utaddress-213)
[ 4294.391196] ACPI Warning: SystemIO range 0x00000000000006B3-0x00000000000006B6 conflicts with OpRegion 0x0000000000000680-0x00000000000006FF (\_SB.PCI0.LPC.RTIO) (20180810/utaddress-213)
[ 4294.391205] ACPI: This conflict may cause random problems and system instability
[ 4294.391207] ACPI: If an ACPI driver is available for this device, you should use it instead of the native driver
[ 4294.391209] ACPI Warning: SystemIO range 0x00000000000006D6-0x00000000000006DC conflicts with OpRegion 0x0000000000000680-0x00000000000006FF (\_SB.PCI0.LPC.RUNT) (20180810/utaddress-213)
[ 4294.391219] ACPI Warning: SystemIO range 0x00000000000006D6-0x00000000000006DC conflicts with OpRegion 0x0000000000000680-0x00000000000006FF (\_SB.PCI0.LPC.RTIO) (20180810/utaddress-213)
[ 4294.391228] ACPI: This conflict may cause random problems and system instability
[ 4294.391229] ACPI: If an ACPI driver is available for this device, you should use it instead of the native driver
[31253.254264] ACPI Error: Method parse/execution failed \_SB.PCI0.LPC.SMBR, AE_AML_LOOP_TIMEOUT (20180810/psparse-516)
[31253.254294] ACPI Error: Method parse/execution failed \_SB.PCI0.LPC.INIT, AE_AML_LOOP_TIMEOUT (20180810/psparse-516)
[31253.254310] ACPI Error: Method parse/execution failed \_GPE._L00, AE_AML_LOOP_TIMEOUT (20180810/psparse-516)
[31253.254329] ACPI Error: AE_AML_LOOP_TIMEOUT, while evaluating GPE method [_L00] (20180810/evgpe-515)
nge 0x00000000000006D6-0x00000000000006DC conflicts with OpRegion 0x0000000000000680-0x00000000000006FF (\_SB.PCI0.LPC.RUNT) (20180810/utaddress-213)
[ 4294.391219] ACPI Warning: SystemIO range 0x00000000000006D6-0x00000000000006DC conflicts with OpRegion 0x0000000000000680-0x00000000000006FF (\_SB.PCI0.LPC.RTIO) (20180810/utaddress-213)
[ 4294.391228] ACPI: This conflict may cause random problems and system instability
[ 4294.391229] ACPI: If an ACPI driver is available for this device, you should use it instead of the native driver
[31253.254264] ACPI Error: Method parse/execution failed \_SB.PCI0.LPC.SMBR, AE_AML_LOOP_TIMEOUT (20180810/psparse-516)
[31253.254294] ACPI Error: Method parse/execution failed \_SB.PCI0.LPC.INIT, AE_AML_LOOP_TIMEOUT (20180810/psparse-516)
[31253.254310] ACPI Error: Method parse/execution failed \_GPE._L00, AE_AML_LOOP_TIMEOUT (20180810/psparse-516)
[31253.254329] ACPI Error: AE_AML_LOOP_TIMEOUT, while evaluating GPE method [_L00] (20180810/evgpe-515)

GRUB_CMDLINE_LINUX="processor.nocst=1 acpi_mask_gpe=0x13"
# acpi_mask_gpe=0x13 : no eventhandler exists for gpe13, gpe13 consumes one core @ 100%

/etc/default/grub

6:GRUB_DEFAULT=0
7:GRUB_TIMEOUT=5
8:GRUB_DISTRIBUTOR=`lsb_release -i -s 2> /dev/null || echo Debian`
9:GRUB_CMDLINE_LINUX_DEFAULT="sysrq_always_enabled log_buf_len=1M consoleblank=0 keep_bootcon panic=120 rootwait=60 acpi_enforce_resources=lax apparmor=1 security=apparmor quiet kernel.printk=2"
11:GRUB_CMDLINE_LINUX="processor.nocst=1 acpi_mask_gpe=0x13"

edit : these are a set of four messages which come regularly in dmesg

[466942.209754] ACPI Error: Method parse/execution failed \_SB.PCI0.LPC.SMBR, AE_AML_LOOP_TIMEOUT (20180810/psparse-516)
[466942.209784] ACPI Error: Method parse/execution failed \_SB.PCI0.LPC.INIT, AE_AML_LOOP_TIMEOUT (20180810/psparse-516)
[466942.209800] ACPI Error: Method parse/execution failed \_GPE._L00, AE_AML_LOOP_TIMEOUT (20180810/psparse-516)
[466942.209820] ACPI Error: AE_AML_LOOP_TIMEOUT, while evaluating GPE method [_L00] (20180810/evgpe-515)

seems I have to do more with newer kernel versions,
I like to have these fixed for this hardware and if possible within the operation of new kernel updates.