| View previous topic :: View next topic |
| Author |
Message |
Gentoopc
Apprentice
Joined: 25 Dec 2017
Posts: 296
|
|
| Back to top |
|
 |
eccerr0r
Watchman
Joined: 01 Jul 2004
Posts: 9691
Location: almost Mile High in the USA
|
 Posted: Thu Jul 18, 2019 11:38 am Post subject: |
 |
|
Need clarification: so you see the kernel starting init or not? How far did it get? Can you see where it stopped and rebooted (it probably will be hard to tell because it blinks out fast?)
I assume the dmesg you see in https://pastebin.com/aUQTxDM8 is from a miniboot CD and not your actual setup?
What system is this? Is there a reason you're using 3.16.70 in your grub config?
Since you have CONFIG_PANIC_TIMEOUT at 0, it's not a panic causing a reboot... hmm...
_________________
Intel Core i7 2700K/Radeon R7 250/24GB DDR3/256GB SSD
What am I supposed watching? |
|
| Back to top |
|
|
Gentoopc
Apprentice
Joined: 25 Dec 2017
Posts: 296
|
| Posted: Thu Jul 18, 2019 5:18 pm Post subject: |
|
|
you are right .... dmesg is taken from the mini installation disk. I only see the grub menu at startup and the loading of the kernel for about one second.
please tell me what information to provide for you to find the reason? What command should I enter into the terminal to provide you with the data? |
|
| Back to top |
|
|
Gentoopc
Apprentice
Joined: 25 Dec 2017
Posts: 296
|
| Posted: Thu Jul 18, 2019 5:20 pm Post subject: |
|
|
| the kernel I use this version because it seems easier to me in the assembly. I could be wrong. |
|
| Back to top |
|
|
NeddySeagoon
Administrator
Joined: 05 Jul 2003
Posts: 54300
Location: 56N 3W
|
| Posted: Thu Jul 18, 2019 5:59 pm Post subject: |
|
|
Gentoopc,
This looks like an illegal instruction exception early in the boot process.
From dmesg you have a | Code: | | [ 0.391076] smpboot: CPU0: Intel(R) Xeon(R) CPU E5430 @ 2.66GHz (family: 0x6, model: 0x17, stepping: 0x6) |
but your kernel is built with | Code: | | CONFIG_GENERIC_CPU=y |
so it should run on any amd64 CPU.
That might point the finger at the initrd being built for the wrong CPU but by the time the initrd is being executed, the illegal instruction exception handler is available, so the system would panic rather than reboot.
What does output?
_________________
Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
| Back to top |
|
|
eccerr0r
Watchman
Joined: 01 Jul 2004
Posts: 9691
Location: almost Mile High in the USA
|
| Posted: Fri Jul 19, 2019 2:28 pm Post subject: |
|
|
I'd suggest trying to build a new 4.19 stable kernel from scratch and see what it does.
I'm hoping there's no newer gcc sillyness that would stop it from building older kernels properly...
_________________
Intel Core i7 2700K/Radeon R7 250/24GB DDR3/256GB SSD
What am I supposed watching? |
|
| Back to top |
|
|
Gentoopc
Apprentice
Joined: 25 Dec 2017
Posts: 296
|
| Posted: Fri Jul 19, 2019 5:26 pm Post subject: |
|
|
| NeddySeagoon wrote: | Gentoopc,
This looks like an illegal instruction exception early in the boot process.
From dmesg you have a | Code: | | [ 0.391076] smpboot: CPU0: Intel(R) Xeon(R) CPU E5430 @ 2.66GHz (family: 0x6, model: 0x17, stepping: 0x6) |
but your kernel is built with | Code: | | CONFIG_GENERIC_CPU=y |
so it should run on any amd64 CPU.
That might point the finger at the initrd being built for the wrong CPU but by the time the initrd is being executed, the illegal instruction exception handler is available, so the system would panic rather than reboot.
What does output? |
https://pastebin.com/TAFPEB0c |
|
| Back to top |
|
|
Gentoopc
Apprentice
Joined: 25 Dec 2017
Posts: 296
|
| Posted: Fri Jul 19, 2019 5:27 pm Post subject: |
|
|
| eccerr0r wrote: | I'd suggest trying to build a new 4.19 stable kernel from scratch and see what it does.
I'm hoping there's no newer gcc sillyness that would stop it from building older kernels properly... |
I will try |
|
| Back to top |
|
|
NeddySeagoon
Administrator
Joined: 05 Jul 2003
Posts: 54300
Location: 56N 3W
|
| Posted: Fri Jul 19, 2019 7:47 pm Post subject: |
|
|
Gentoopc,
| Code: | | CFLAGS="-O2 -march=native -pipe -ffast-math" |
You know that -ffast-math will sometimes produce incorrect floating point results?
Its a design feature and is the price you pay for it being fast.
It should never be used globally. Rather it should only be used where the speed/accuracy tradeoff has been proven to be acceptable.
I don't think its a part of your problem just now. The kernel certainly does not use any floating point.
I don't know about the initrd.
_________________
Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
| Back to top |
|
|
eccerr0r
Watchman
Joined: 01 Jul 2004
Posts: 9691
Location: almost Mile High in the USA
|
| Posted: Fri Jul 19, 2019 11:07 pm Post subject: |
|
|
BTW do you see the usual kernel spew (starting with "uncompressing kernel...") just before reboot happens, or does it just goes "loading kernel... loading initramfs..." (messages from bootloader) and then reboots? Yeah it'd take a fast eye to tell but it's a clue...
_________________
Intel Core i7 2700K/Radeon R7 250/24GB DDR3/256GB SSD
What am I supposed watching? |
|
| Back to top |
|
|
Gentoopc
Apprentice
Joined: 25 Dec 2017
Posts: 296
|
| Posted: Fri Jul 19, 2019 11:40 pm Post subject: |
|
|
| Thanks for the help guys .... the problem is solved by installing the kernel version 5.0, it was on my goole disk, everything works))))) thanks |
|
| Back to top |
|
|
tippincows
n00b
Joined: 05 Jan 2013
Posts: 9
|
| Posted: Sun Dec 15, 2019 11:39 am Post subject: Upgraded to kernel Still same thing |
|
|
I am having the same issue from what i can see.
during install emerge pulled kernel 4.19 i would boot with no problems, and get to just after the frame buffer console kicked in. The tux logo would show up and i would get an instant reboot with no warning or error message i can decipher from the kernel text.
My info is below which made me think my situation is similar.
Dell Poweredge 2950
Exsi 5 installing Gentoo in virtual environment.
Linux livecd 4.19.72-gentoo #1 SMP Thu Oct 17 02:49:00 UTC 2019 x86_64 Intel(R) Xeon(R) CPU 5150 @ 2.66GHz GenuineIntel GNU/Linux
ii booted back into the live cd and added the accept keywords="~amd64" to make.conf in order to upgrade from kernel 4.19 to 5.4.3
after upgrading and booting into kernel 5.4 my issue is still persisting. i will post here any solutions i find.
Here is screenshot of kernel text, right before reboot.
https://i.imgur.com/1SRZIxe.png |
|
| Back to top |
|
|
NeddySeagoon
Administrator
Joined: 05 Jul 2003
Posts: 54300
Location: 56N 3W
|
| Posted: Sun Dec 15, 2019 12:15 pm Post subject: |
|
|
tippincows,
There are no clues in your screenshot.
Please post the output of and the first CPU in /proc/cpuinfo
Put your kernel .config onto a pastebin site.
A kernel upgrade does not normally fix this issue.
If its an Illegal Instruction error, there won't be any messages or logs.
That would require the kernel to continue after the Illegal Instruction rather than reboot.
_________________
Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
| Back to top |
|
|
tippincows
n00b
Joined: 05 Jan 2013
Posts: 9
|
| Posted: Sun Dec 15, 2019 6:05 pm Post subject: |
|
|
Emerge --info - https://pastebin.com/Ws1atnnT
proc/cpuinfo - https://pastebin.com/6mPPYi2a
kernel .config - https://pastebin.com/F3Fa4Hi6
make.conf - https://pastebin.com/nhhidrWJ
Thank you for the prompt reply. You are right there is no clue or anything in the kernel boot sequence that i can see that would point me in any sort of direction to fix.
All i know at this point is im getting constant reboots right at that point., which is making it tricky but not impossible to troubleshoot.
I haven't been able to successfully install Gentoo on this dell power edge running esxi i have yet.
Here is the troubleshooting steps i have done so far.
1.Changing my kernel type in make.conf from march=native to something more specific, in this case core2 as per matching the proc/cpuinfo with my cpu in https://wiki.gentoo.org/wiki/Safe_CFLAGS
2. re compiled 4.19 kernel changing the processor type from generic-x86_64 to both new xeon and old xeon options.
3. added ACCEPT_KEYWORD="~amd64" variable to make.conf to upgrade to kernel 5.4.
4. checked kernel config a few times to make sure LSI and raid drivers were enabled for perc5i (the raid controller i have)
Grub is able to boot, and able to boot the kernel even past the initramfs stage, but i get no indication from the original screenshot that the kernel was able to find the root block. thats my only guess.
Thank you |
|
| Back to top |
|
|
NeddySeagoon
Administrator
Joined: 05 Jul 2003
Posts: 54300
Location: 56N 3W
|
| Posted: Sun Dec 15, 2019 6:51 pm Post subject: |
|
|
tippincows,
Run the command | Code: | | gcc -### -E - -march=native 2>&1 | sed -r '/cc1/!d;s/(")|(^.* - )|( -mno-[^\ ]+)//g' |
in your chroot.
It will tell what gcc thinks -march=nate means. I get.
| Code: | gcc -### -E - -march=native 2>&1 | sed -r '/cc1/!d;s/(")|(^.* - )|( -mno-[^\ ]+)//g'
-march=amdfam10 -mmmx -m3dnow -msse -msse2 -msse3 -msse4a -mcx16 -msahf -mpopcnt -mabm -mlzcnt -mprfchw -mfxsr --param l1-cache-size=64 --param l1-cache-line-size=64 --param l2-cache-size=512 -mtune=amdfam10 |
On amd64, gcc is pretty good with -march=native on older CPUs.
| Quote: | | ACCEPT_KEYWORD="~amd64" variable to make.conf |
Its safer to do that on a per package basis. make.conf applies to your entire install. See this wiki page and revert your ACCEPT_KEYWORD change in make.conf
| Quote: | | 4. checked kernel config a few times to make sure LSI and raid drivers were enabled for perc5i (the raid controller i have) |
I was about to ask about that :)
From memory, you also have some ordinary SATA ports too.
Please post your output. Where is the root filesystem connected on your system?
As your system reboots rather than panics and halts, I don't think its a root filesystem issue.
_________________
Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
| Back to top |
|
|
tippincows
n00b
Joined: 05 Jan 2013
Posts: 9
|
| Posted: Sun Dec 15, 2019 7:40 pm Post subject: |
|
|
| Code: | livecd /usr/src/linux # gcc -### -E - -march=native 2>&1 | sed -r '/cc1/!d;s/(")|(^.* - )|( -mno-[^\ ]+)//g'
-march=core2 -mmmx -msse -msse2 -msse3 -mssse3 -mcx16 -msahf -mfxsr --param l1-cache-size=32 --param l1-cache-line-size=64 --param l2-cache-size=4096 -mtune=core2 |
Changed ACCEPT_KEYWORDS from make.conf to "amd64" (removed ~)
added ~amd64 to /etc/portage/package/accept_keywords/sys-keynel/gentoo-sources
and tested to make sure kernel 5.4 still emerges. (good to go) i didnt emerge anything else since i changed my keyword to ~ globally (thankfully)
| Code: | | lspci-nnk - https://pastebin.com/J1TL9Mvf |
| Quote: | | Where is the root filesystem connected on your system? |
root filesystem is located on /dev/sda4 (partition scheme relatively identical to handbook)
| Code: | livecd /etc/portage/package.accept_keywords/sys-kernel # df -h
Filesystem Size Used Avail Use% Mounted on
/dev/sda4 391G 6.2G 365G 2% /
/dev/sda2 233M 53M 164M 25% /boot
udev 10M 4.0K 10M 1% /dev
shm 2.0G 0 2.0G 0% /dev/shm
cgroup_root 10M 0 10M 0% /sys/fs/cgroup
livecd /etc/portage/package.accept_keywords/sys-kernel #
Disk: /dev/sda
Size: 400 GiB, 429496729600 bytes, 838860800 sectors
Label: dos, identifier: 0xd997d47b
Device Boot Start End Sectors Size Id Type
>> /dev/sda1 2048 264191 262144 128M 83 Linux
/dev/sda2 264192 763903 499712 244M 83 Linux
/dev/sda3 763904 4958207 4194304 2G 83 Linux
/dev/sda4 4958208 838860799 833902592 397.7G 83 Linux
|
Currently checking kernel config to make sure mptspi is enabled in kernel from my lspci output.
| Code: | 00:10.0 SCSI storage controller [0100]: Broadcom / LSI 53c1030 PCI-X Fusion-MPT Dual Ultra320 SCSI [1000:0030] (rev 01)
Subsystem: VMware LSI Logic Parallel SCSI Controller [15ad:1976]
Kernel driver in use: mptspi |
[Moderator edit: added [code] tags to preserve output layout. -Hu] |
|
| Back to top |
|
|
tippincows
n00b
Joined: 05 Jan 2013
Posts: 9
|
| Posted: Sun Dec 15, 2019 7:58 pm Post subject: |
|
|
Got it.
I googled mptspi as i couldn't find it in menuconfig using / function. Found out from the folowing link that i needed fusionMPT device support enabled in kernel for SAS.
After compiling and rebooting i am now at login prompt.
source - http://bruceblinn.com/linuxinfo/ConfigKernel-VMware.html
I would Mark solved if i didn't hijack someone elses thread 
Thank you. |
|
| Back to top |
|
|
NeddySeagoon
Administrator
Joined: 05 Jul 2003
Posts: 54300
Location: 56N 3W
|
| Posted: Sun Dec 15, 2019 8:00 pm Post subject: |
|
|
tippincows,
-march=core2 is what you have now, so that's no change.
| Code: | 00:10.0 SCSI storage controller [0100]: Broadcom / LSI 53c1030 PCI-X Fusion-MPT Dual Ultra320 SCSI [1000:0030] (rev 01)
Subsystem: VMware LSI Logic Parallel SCSI Controller [15ad:1976]
Kernel driver in use: mptspi |
Fusion-MPT. Thats on its own menu. The menu is off in your kernel.
What you are showing in lspci is the emulated hardware as seen by a VMware guest operating system.
Is that what you intended to show?
I thought you were installing on real hardware?
_________________
Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
| Back to top |
|
|
tippincows
n00b
Joined: 05 Jan 2013
Posts: 9
|
| Posted: Sun Dec 15, 2019 8:36 pm Post subject: |
|
|
You are right the FusionMPT menu on menuconfig was turned off. I enabled it and Selected the option for SAS, allowing my system to boot properly, and stop boot looping.
Yes, i meant to post that lspci output. I am installing in a virtual environment using Esxi on my Dell power edge, sorry i should of emphasized that point a bit more. I know how important that is to solving any sort of issue.
I bookmarked that link from my previous post because there is other modules listed that i want to make sure i have enabled specifically for running Gentoo in vmware. |
|
| Back to top |
|
|
|
Installing Gentoo
