| View previous topic :: View next topic |
| Author |
Message |
DeIM
Guru
Joined: 11 Apr 2006
Posts: 429
|
 Posted: Sun Jan 29, 2023 6:59 pm Post subject: [solved] kernel upgrade 6.1.0 -> 6.1.8 not booting |
 |
|
new GCC and SCHED_BMQ as alt scheduler.
What could cause unbootable kernel?
Thanks for opinion
| Code: | # diff linux-6.1.0-gentoo/.config linux-6.1.8-gentoo/.config
3c3
< # Linux/x86 6.1.0-gentoo Kernel Configuration
---
> # Linux/x86 6.1.8-gentoo Kernel Configuration
5c5
< CONFIG_CC_VERSION_TEXT="gcc (Gentoo 11.3.0 p7) 11.3.0"
---
> CONFIG_CC_VERSION_TEXT="gcc (Gentoo 12.2.1_p20230121-r1 p10) 12.2.1 20230121"
7c7
< CONFIG_GCC_VERSION=110300
---
> CONFIG_GCC_VERSION=120201
10c10
< CONFIG_AS_VERSION=23800
---
> CONFIG_AS_VERSION=23900
12c12
< CONFIG_LD_VERSION=23800
---
> CONFIG_LD_VERSION=23900
137d136
< CONFIG_SCHED_CORE=y
188c187,189
< # CONFIG_UCLAMP_TASK is not set
---
> CONFIG_SCHED_ALT=y
> CONFIG_SCHED_BMQ=y
> # CONFIG_SCHED_PDS is not set
194a196
> CONFIG_GCC11_NO_ARRAY_BOUNDS=y
195a198
> CONFIG_CC_NO_ARRAY_BOUNDS=y
197,198d199
< CONFIG_NUMA_BALANCING=y
< CONFIG_NUMA_BALANCING_DEFAULT_ENABLED=y
208,209d208
< CONFIG_CFS_BANDWIDTH=y
< # CONFIG_RT_GROUP_SCHED is not set
231d229
< CONFIG_SCHED_AUTOGROUP=y
7554a7553,7555
> CONFIG_CC_HAS_AUTO_VAR_INIT_PATTERN=y
> CONFIG_CC_HAS_AUTO_VAR_INIT_ZERO_BARE=y
> CONFIG_CC_HAS_AUTO_VAR_INIT_ZERO=y
7556,7558c7557,7558
< # CONFIG_GCC_PLUGIN_STRUCTLEAK_USER is not set
< # CONFIG_GCC_PLUGIN_STRUCTLEAK_BYREF is not set
< # CONFIG_GCC_PLUGIN_STRUCTLEAK_BYREF_ALL is not set
---
> # CONFIG_INIT_STACK_ALL_PATTERN is not set
> # CONFIG_INIT_STACK_ALL_ZERO is not set |
Last edited by DeIM on Sun Mar 12, 2023 2:04 pm; edited 1 time in total |
|
| Back to top |
|
 |
DeIM
Guru
Joined: 11 Apr 2006
Posts: 429
|
| Posted: Sun Jan 29, 2023 8:42 pm Post subject: |
|
|
| Using direct efi loading by UEFI BIOS - same as 6.1.0 but 6.1.8 doesn't load at all - it stuck on laptop logo no boot oputput. |
|
| Back to top |
|
|
logrusx
Veteran
Joined: 22 Feb 2018
Posts: 1532
|
| Posted: Mon Jan 30, 2023 9:18 am Post subject: |
|
|
What CPU are you running?
Best Regards,
Georgi |
|
| Back to top |
|
|
NeddySeagoon
Administrator
Joined: 05 Jul 2003
Posts: 54236
Location: 56N 3W
|
| Posted: Mon Jan 30, 2023 11:13 am Post subject: |
|
|
DeIM,
Past your and pastebin the full kernel config file.
Confirm that you used to get a legal config for the new kernel.
Making a kernel with an illegal config file can fail in all sots of weird and wonderful ways that nobody has ever seen.
"no boot oputput" does not mean not booting. It may be that its not drawing the console, so it boots properly but you can't see it.
If sshd is set up, you may be able to ssh into the system and get dmesg. That will be full of useful diagnostic messages.
_________________
Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
| Back to top |
|
|
DeIM
Guru
Joined: 11 Apr 2006
Posts: 429
|
| Posted: Tue Jan 31, 2023 7:33 pm Post subject: |
|
|
I did now make clean then unset alt_sched drivers as was in 6.1.0 and 6.1.8 is bootable now - but I can't change backlit of display - strange I supposed it's not so big change in minor version. When I boot 6.1.0 backlit can be changed well.
| Code: | $ xbacklight -1
/usr/lib/python-exec/python3.10/xbacklight: [Errno 13] Permission denied: '/sys/class/leds/phy0-led/brightness'
$ ls -l /sys/class/leds/phy0-led/brightness
-rw-r--r-- 1 root root 4096 31. led 19.54 /sys/class/leds/phy0-led/brightness
$ groups
disk wheel audio cdrom video cdrw usb users polkitd flutter plugdev deim |
I confirm before first report I did make oldconfig - and set there alt_sched - SCHED_BMQ as default.
It's Intel Pentium N3700 Acer Travelmate B-116M
| Code: | # lspci -nnk
00:00.0 Host bridge [0600]: Intel Corporation Atom/Celeron/Pentium Processor x5-E8000/J3xxx/N3xxx Series SoC Transaction Register [8086:2280] (rev 21)
Subsystem: Acer Incorporated [ALI] Atom/Celeron/Pentium Processor x5-E8000/J3xxx/N3xxx Series SoC Transaction Register [1025:1043]
Kernel driver in use: iosf_mbi_pci
00:02.0 VGA compatible controller [0300]: Intel Corporation Atom/Celeron/Pentium Processor x5-E8000/J3xxx/N3xxx Integrated Graphics Controller [8086:22b1] (rev 21)
Subsystem: Acer Incorporated [ALI] Atom/Celeron/Pentium Processor x5-E8000/J3xxx/N3xxx Integrated Graphics Controller [1025:1043]
Kernel driver in use: i915
00:0b.0 Signal processing controller [1180]: Intel Corporation Atom/Celeron/Pentium Processor x5-E8000/J3xxx/N3xxx Series Power Management Controller [8086:22dc] (rev 21)
Subsystem: Acer Incorporated [ALI] Atom/Celeron/Pentium Processor x5-E8000/J3xxx/N3xxx Series Power Management Controller [1025:1043]
Kernel driver in use: proc_thermal
Kernel modules: processor_thermal_device_pci_legacy
00:13.0 SATA controller [0106]: Intel Corporation Atom/Celeron/Pentium Processor x5-E8000/J3xxx/N3xxx Series SATA Controller [8086:22a3] (rev 21)
Subsystem: Acer Incorporated [ALI] Atom/Celeron/Pentium Processor x5-E8000/J3xxx/N3xxx Series SATA Controller [1025:1043]
Kernel driver in use: ahci
00:14.0 USB controller [0c03]: Intel Corporation Atom/Celeron/Pentium Processor x5-E8000/J3xxx/N3xxx Series USB xHCI Controller [8086:22b5] (rev 21)
Subsystem: Acer Incorporated [ALI] Atom/Celeron/Pentium Processor x5-E8000/J3xxx/N3xxx Series USB xHCI Controller [1025:1043]
Kernel driver in use: xhci_hcd
00:1a.0 Encryption controller [1080]: Intel Corporation Atom/Celeron/Pentium Processor x5-E8000/J3xxx/N3xxx Series Trusted Execution Engine [8086:2298] (rev 21)
Subsystem: Acer Incorporated [ALI] Atom/Celeron/Pentium Processor x5-E8000/J3xxx/N3xxx Series Trusted Execution Engine [1025:1043]
Kernel driver in use: mei_txe
00:1b.0 Audio device [0403]: Intel Corporation Atom/Celeron/Pentium Processor x5-E8000/J3xxx/N3xxx Series High Definition Audio Controller [8086:2284] (rev 21)
Subsystem: Acer Incorporated [ALI] Atom/Celeron/Pentium Processor x5-E8000/J3xxx/N3xxx Series High Definition Audio Controller [1025:1043]
Kernel driver in use: snd_hda_intel
Kernel modules: snd_hda_intel
00:1c.0 PCI bridge [0604]: Intel Corporation Atom/Celeron/Pentium Processor x5-E8000/J3xxx/N3xxx Series PCI Express Port #1 [8086:22c8] (rev 21)
Subsystem: Acer Incorporated [ALI] Atom/Celeron/Pentium Processor x5-E8000/J3xxx/N3xxx Series PCI Express Port [1025:1043]
Kernel driver in use: pcieport
00:1c.2 PCI bridge [0604]: Intel Corporation Atom/Celeron/Pentium Processor x5-E8000/J3xxx/N3xxx Series PCI Express Port #3 [8086:22cc] (rev 21)
Subsystem: Acer Incorporated [ALI] Atom/Celeron/Pentium Processor x5-E8000/J3xxx/N3xxx Series PCI Express Port [1025:1043]
Kernel driver in use: pcieport
00:1c.3 PCI bridge [0604]: Intel Corporation Atom/Celeron/Pentium Processor x5-E8000/J3xxx/N3xxx Series PCI Express Port #4 [8086:22ce] (rev 21)
Subsystem: Acer Incorporated [ALI] Atom/Celeron/Pentium Processor x5-E8000/J3xxx/N3xxx Series PCI Express Port [1025:1043]
Kernel driver in use: pcieport
00:1f.0 ISA bridge [0601]: Intel Corporation Atom/Celeron/Pentium Processor x5-E8000/J3xxx/N3xxx Series PCU [8086:229c] (rev 21)
Subsystem: Acer Incorporated [ALI] Atom/Celeron/Pentium Processor x5-E8000/J3xxx/N3xxx Series PCU [1025:1043]
Kernel driver in use: lpc_ich
00:1f.3 SMBus [0c05]: Intel Corporation Atom/Celeron/Pentium Processor x5-E8000/J3xxx/N3xxx SMBus Controller [8086:2292] (rev 21)
Subsystem: Acer Incorporated [ALI] Atom/Celeron/Pentium Processor x5-E8000/J3xxx/N3xxx SMBus Controller [1025:1043]
Kernel driver in use: i801_smbus
02:00.0 Network controller [0280]: Intel Corporation Wireless 7265 [8086:095a] (rev 59)
Subsystem: Intel Corporation Dual Band Wireless-AC 7265 [8086:5010]
Kernel driver in use: iwlwifi
Kernel modules: iwlwifi
03:00.0 Ethernet controller [0200]: Realtek Semiconductor Co., Ltd. RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller [10ec:8168] (rev 15)
Subsystem: Acer Incorporated [ALI] RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller [1025:1043]
Kernel driver in use: r8169
Kernel modules: r8169 |
fully working:
6.1.0 .config:
http://dpaste.com/EM4SX5VRE
partially working actual state:
6.1.8 .config:
http://dpaste.com/EPMGUGZ9Z
This state never happened before. There was 3 changes - gcc, alt_sched and new kernel version.
Thank you all for replies |
|
| Back to top |
|
|
DeIM
Guru
Joined: 11 Apr 2006
Posts: 429
|
| Posted: Tue Jan 31, 2023 7:40 pm Post subject: |
|
|
| Configs are quite permissive since long time debugging SD card reader, then gave up and started from default config and still didn't find motivation to remove unused drivers (SD card reader is working now but it's not quality one and doesn't read every SD card I put in but driver is OK). |
|
| Back to top |
|
|
krumpf
Apprentice
Joined: 15 Jul 2018
Posts: 175
|
| Posted: Thu Feb 02, 2023 8:02 am Post subject: Re: kernel upgrade 6.1.0 -> 6.1.8 not booting |
|
|
| DeIM wrote: | new GCC and SCHED_BMQ as alt scheduler.
What could cause unbootable kernel?
Thanks for opinion |
Dunno if it's related, but each time I tried compiling the kernel with SCHED_BMQ, I got an error (note: I got WERROR=y), with GCC 11 or GCC 12
| Code: | CC fs/notify/mark.o
kernel/sched/alt_core.c: In function ‘select_task_rq’:
kernel/sched/alt_core.c:1907:1: error: the frame size of 2056 bytes is larger than 2048 bytes [-Werror=frame-larger-than=]
1907 | }
| ^
CC crypto/asymmetric_keys/x509_cert_parser.o |
If SCHED_ALT is not set, the kernel compile without trouble. |
|
| Back to top |
|
|
logrusx
Veteran
Joined: 22 Feb 2018
Posts: 1532
|
| Posted: Thu Feb 02, 2023 7:33 pm Post subject: |
|
|
I think you need to be running acpi cpufreq and you've selected intel p-state as compiled-in which locks it as the default.
Best Regards,
Georgi
Last edited by logrusx on Fri Feb 03, 2023 12:27 pm; edited 1 time in total |
|
| Back to top |
|
|
Goverp
Advocate
Joined: 07 Mar 2007
Posts: 2006
|
| Posted: Fri Feb 03, 2023 11:06 am Post subject: Re: kernel upgrade 6.1.0 -> 6.1.8 not booting |
|
|
| krumpf wrote: | ...
Dunno if it's related, but each time I tried compiling the kernel with SCHED_BMQ, I got an error (note: I got WERROR=y), with GCC 11 or GCC 12
| Code: | CC fs/notify/mark.o
kernel/sched/alt_core.c: In function ‘select_task_rq’:
kernel/sched/alt_core.c:1907:1: error: the frame size of 2056 bytes is larger than 2048 bytes [-Werror=frame-larger-than=]
1907 | }
| ^
CC crypto/asymmetric_keys/x509_cert_parser.o |
... |
I've been compiling with SCHED_BMQ for some time. That message appears for me too; it only started appearing with kernel 6.1 as far as I remember. The resulting kernels appear to work fine.
_________________
Greybeard |
|
| Back to top |
|
|
logrusx
Veteran
Joined: 22 Feb 2018
Posts: 1532
|
| Posted: Fri Feb 03, 2023 1:24 pm Post subject: Re: kernel upgrade 6.1.0 -> 6.1.8 not booting |
|
|
| krumpf wrote: | | DeIM wrote: | new GCC and SCHED_BMQ as alt scheduler.
What could cause unbootable kernel?
Thanks for opinion |
Dunno if it's related, but each time I tried compiling the kernel with SCHED_BMQ, I got an error (note: I got WERROR=y), with GCC 11 or GCC 12
| Code: | CC fs/notify/mark.o
kernel/sched/alt_core.c: In function ‘select_task_rq’:
kernel/sched/alt_core.c:1907:1: error: the frame size of 2056 bytes is larger than 2048 bytes [-Werror=frame-larger-than=]
1907 | }
| ^
CC crypto/asymmetric_keys/x509_cert_parser.o |
If SCHED_ALT is not set, the kernel compile without trouble. |
| Goverp wrote: | | krumpf wrote: | ...
Dunno if it's related, but each time I tried compiling the kernel with SCHED_BMQ, I got an error (note: I got WERROR=y), with GCC 11 or GCC 12
| Code: | CC fs/notify/mark.o
kernel/sched/alt_core.c: In function ‘select_task_rq’:
kernel/sched/alt_core.c:1907:1: error: the frame size of 2056 bytes is larger than 2048 bytes [-Werror=frame-larger-than=]
1907 | }
| ^
CC crypto/asymmetric_keys/x509_cert_parser.o |
... |
I've been compiling with SCHED_BMQ for some time. That message appears for me too; it only started appearing with kernel 6.1 as far as I remember. The resulting kernels appear to work fine. |
I just compiled 6.1.8 and don't see such an error, which btw is a result of -Werror=frame-larger-than=2048, which in turn might be something form your make.conf or hardening practice. However it doesn't seem to be related to BMQ. In few minutes I'll report if the system booted. Note that I've selected acpi-cpufreq as I believe amd and intel p-state do not support cgroups which in turn makes the kernel with BMQ scheduler hang.
UPDATE: nope, couldn't make it boot either. I also remember there were other people who couldn't.
Best Regards,
Georgi |
|
| Back to top |
|
|
DeIM
Guru
Joined: 11 Apr 2006
Posts: 429
|
| Posted: Sun Feb 05, 2023 12:24 pm Post subject: |
|
|
Tried 6.1.9 with clang 15.0.6 - it took significantly more time than gcc.
Tried with BMQ and thin LTO, removing some of forced kernel args. Still not booting
.config:
http://dpaste.com/CY2TP7LHQ
| Code: | # diff linux-6.1.0-gentoo/.config linux-6.1.9-gentoo/.config
3c3
< # Linux/x86 6.1.0-gentoo Kernel Configuration
---
> # Linux/x86 6.1.9-gentoo Kernel Configuration
5,13c5,13
< CONFIG_CC_VERSION_TEXT="gcc (Gentoo 11.3.0 p7) 11.3.0"
< CONFIG_CC_IS_GCC=y
< CONFIG_GCC_VERSION=110300
< CONFIG_CLANG_VERSION=0
< CONFIG_AS_IS_GNU=y
< CONFIG_AS_VERSION=23800
< CONFIG_LD_IS_BFD=y
< CONFIG_LD_VERSION=23800
< CONFIG_LLD_VERSION=0
---
> CONFIG_CC_VERSION_TEXT="clang version 15.0.6"
> CONFIG_GCC_VERSION=0
> CONFIG_CC_IS_CLANG=y
> CONFIG_CLANG_VERSION=150006
> CONFIG_AS_IS_LLVM=y
> CONFIG_AS_VERSION=150006
> CONFIG_LD_VERSION=0
> CONFIG_LD_IS_LLD=y
> CONFIG_LLD_VERSION=150006
17a18
> CONFIG_TOOLS_SUPPORT_RELR=y
137d137
< CONFIG_SCHED_CORE=y
188c188,190
< # CONFIG_UCLAMP_TASK is not set
---
> CONFIG_SCHED_ALT=y
> CONFIG_SCHED_BMQ=y
> # CONFIG_SCHED_PDS is not set
194c196,197
< CONFIG_CC_IMPLICIT_FALLTHROUGH="-Wimplicit-fallthrough=5"
---
> CONFIG_CC_IMPLICIT_FALLTHROUGH="-Wimplicit-fallthrough"
> CONFIG_GCC11_NO_ARRAY_BOUNDS=y
197,198d199
< CONFIG_NUMA_BALANCING=y
< CONFIG_NUMA_BALANCING_DEFAULT_ENABLED=y
208,209d208
< CONFIG_CFS_BANDWIDTH=y
< # CONFIG_RT_GROUP_SCHED is not set
231d229
< CONFIG_SCHED_AUTOGROUP=y
512c510
< CONFIG_CMDLINE="root=PARTUUID=2f672aab-ca50-4be8-a346-2908e3b36c9e acpi_backlight=vendor acpi_osi=linux i8042.nopnp i8042.nomux i8042.reset i915.fastboot=1 modprobe.blacklist=pinctrl_cherryview i915.enable_psr=0 intel_idle.max_cstate=2 quiet"
---
> CONFIG_CMDLINE="root=PARTUUID=2f672aab-ca50-4be8-a346-2908e3b36c9e i8042.nopnp i8042.nomux i8042.reset quiet"
684d681
< # CONFIG_X86_X32_ABI is not set
789a787,788
> CONFIG_LTO=y
> CONFIG_LTO_CLANG=y
792c791,794
< CONFIG_LTO_NONE=y
---
> CONFIG_HAS_LTO_CLANG=y
> # CONFIG_LTO_NONE is not set
> # CONFIG_LTO_CLANG_FULL is not set
> CONFIG_LTO_CLANG_THIN=y
864,865d865
< CONFIG_GCC_PLUGINS=y
< # CONFIG_GCC_PLUGIN_LATENT_ENTROPY is not set
7554a7555,7557
> CONFIG_CC_HAS_AUTO_VAR_INIT_PATTERN=y
> CONFIG_CC_HAS_AUTO_VAR_INIT_ZERO_ENABLER=y
> CONFIG_CC_HAS_AUTO_VAR_INIT_ZERO=y
7556,7559c7559,7560
< # CONFIG_GCC_PLUGIN_STRUCTLEAK_USER is not set
< # CONFIG_GCC_PLUGIN_STRUCTLEAK_BYREF is not set
< # CONFIG_GCC_PLUGIN_STRUCTLEAK_BYREF_ALL is not set
< # CONFIG_GCC_PLUGIN_STACKLEAK is not set
---
> # CONFIG_INIT_STACK_ALL_PATTERN is not set
> # CONFIG_INIT_STACK_ALL_ZERO is not set
7562,7563d7562
< CONFIG_CC_HAS_ZERO_CALL_USED_REGS=y
< # CONFIG_ZERO_CALL_USED_REGS is not set
7565a7565
> CONFIG_CC_HAS_RANDSTRUCT=y
7568d7567
< # CONFIG_RANDSTRUCT_PERFORMANCE is not set
8043d8041
< # CONFIG_READABLE_ASM is not set
8045d8042
< # CONFIG_DEBUG_SECTION_MISMATCH is not set
8123a8121
> CONFIG_CC_HAS_KASAN_SW_TAGS=y
8128a8127,8128
> CONFIG_HAVE_KMSAN_COMPILER=y
> # CONFIG_KMSAN is not set
8276c8276
< CONFIG_FTRACE_MCOUNT_USE_CC=y
---
> CONFIG_FTRACE_MCOUNT_USE_OBJTOOL=y |
I use BMQ on i7-4770K without problems |
|
| Back to top |
|
|
DeIM
Guru
Joined: 11 Apr 2006
Posts: 429
|
| Posted: Sat Feb 25, 2023 7:29 am Post subject: |
|
|
ok tried 6.2.0 - seem no BMQ available in experimantal.
But it's usable for me now - removed acpi_backlight=vendor from kernel commandline and acpilight works well now.
https://bbs.archlinux.org/viewtopic.php?id=282925 |
|
| Back to top |
|
|
DeIM
Guru
Joined: 11 Apr 2006
Posts: 429
|
| Posted: Sun Mar 12, 2023 2:03 pm Post subject: |
|
|
After use v 6.2.0 without BMQ patch I tried 6.2.5 with BMQ patch included and it works now!
Line from dmesg:
| Code: | | [ 0.147609] sched/bmq: BMQ CPU Scheduler v6.2-r0 by Alfred Chen. |
|
|
| Back to top |
|
|
|
Kernel & Hardware
