You are not logged in.
a little question, someone with haswell desktop are having problems like freezes? kernel freezes totally and I acn't do nothing more than reset.. with the generic linux ck and the linux ck haswell since the 4.7.2-2 , the bfs 480 patch
I'm on a Haswell laptop, rather than a desktop, but am experiencing no freezes here. In case it could help, my /etc/mkinitcpio.conf contains:
[…]
MODULES="i915 crc32_generic crc32-pclmul" # 4.6 and later
HOOKS="base udev autodetect modconf block filesystems keyboard fsck"
[…]
…and /boot/loader/entries/arch-ck.conf contains the options:
options root=/dev/sda2 rw elevator=bfq nmi_watchdog=0
Also, no signature difficulties here, nomorewindows, so maybe it's a local configuration issue.
@archun: Intel® Core™ i5-4210M • [GPU] Intel® HD Graphics 4600 • [Kernel] linux-ck-haswell
Handmade.Network • GitLab
The Life and Times of Miblo del Carpio
Offline
lscpu
Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Byte Order: Little Endian
CPU(s): 8
On-line CPU(s) list: 0-7
Thread(s) per core: 2
Core(s) per socket: 4
Socket(s): 1
NUMA node(s): 1
Vendor ID: GenuineIntel
CPU family: 6
Model: 60
Model name: Intel(R) Core(TM) i7-4770 CPU @ 3.40GHz
Stepping: 3
CPU MHz: 3400.830
CPU max MHz: 3900.0000
CPU min MHz: 800.0000
BogoMIPS: 6800.36
Virtualization: VT-x
L1d cache: 32K
L1i cache: 32K
L2 cache: 256K
L3 cache: 8192K
NUMA node0 CPU(s): 0-7
Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc aperfmperf eagerfpu pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm epb tpr_shadow vnmi flexpriority ept vpid fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid xsaveopt dtherm ida arat pln pts
Getting complete freezes last days playing 3d games or videos in browser - chrome (nvidia gts 450, ck as well repo). SysRq+"REISUB" do not help at all (have that enabled, bcs nvidia is buggy). Only reset. Wanted to make new topic, but luckelly got here.
Upd: if that matters, have SSD + HDDs for the years, and fuse with zipped fs for some folders in my acc. But hangs was on video-related things only (usually, except games and browser I do tons of GCC), so maybe something with kernel + nvidia.
Upd2: tried sudo journalctl --since yesterday
it's pretty clean - regular messages - no errors to take in account..but some like
systemd-udevd[233]: specified group 'plugdev' unknown
ksysguardd[675]: cannot open config file '"/etc/ksysguarddrc"'
Sep 11 01:09:27 moy-super akonadi_control[682]: Cannot remove old log file - running on a readonly filesystem ma
Sep 11 01:09:27 moy-super akonadi_control[682]: [
0: /usr/bin/akonadi_control() [0x42f366]
...nothing special
Any idea how to get log of wtf is going on ?
Last edited by alexzk (2016-09-12 07:26:49)
Offline
this crc32_generic crc32-pclmul and this nmi_watchdog=0.. is for..? I mean what is the function of this modules and boot kernel parameter? I have 2 laptops and 1 desktop, the laptops have amd cpu and intel cpu but the intel has 7 or 8 years, in the only pc I have freeze many times gaming is the haswelll desktop since 4.7.2-2(bfs480 update) and in the journalctl there isn't any weird error or nothing like this..
Offline
other thing is that I have hd intel graphics..
Offline
this crc32_generic crc32-pclmul and this nmi_watchdog=0.. is for..? I mean what is the function of this modules and boot kernel parameter? I have 2 laptops and 1 desktop, the laptops have amd cpu and intel cpu but the intel has 7 or 8 years, in the only pc I have freeze many times gaming is the haswelll desktop since 4.7.2-2(bfs480 update) and in the journalctl there isn't any weird error or nothing like this..
The crc32* modules are needed in linux 4.6 and onwards if your /boot is on an f2fs partition, as discussed in this thread, and disabling the watchdog supposedly reduces power consumption. Since you didn't need the crc* modules before, they probably won't help you now.
Have you enabled Intel microcode updates? If not, then enabling them may fix the issue. If it's still not fixed, it may be best to post a new topic in the Kernel & Hardware sub-forum if there isn't an appropriate topic already in discussion.
Last edited by Miblo (2016-09-12 15:55:49)
@archun: Intel® Core™ i5-4210M • [GPU] Intel® HD Graphics 4600 • [Kernel] linux-ck-haswell
Handmade.Network • GitLab
The Life and Times of Miblo del Carpio
Offline
agm28011997 wrote:this crc32_generic crc32-pclmul and this nmi_watchdog=0.. is for..? I mean what is the function of this modules and boot kernel parameter? I have 2 laptops and 1 desktop, the laptops have amd cpu and intel cpu but the intel has 7 or 8 years, in the only pc I have freeze many times gaming is the haswelll desktop since 4.7.2-2(bfs480 update) and in the journalctl there isn't any weird error or nothing like this..
The crc32* modules are needed in linux 4.6 and onwards if your /boot is on an f2fs partition, as discussed in this thread, and disabling the watchdog supposedly reduces power consumption. Since you didn't need the crc* modules before, they probably won't help you now.
Have you enabled Intel microcode updates? If not, then enabling them may fix the issue. If it's still not fixed, it may be best to post a new topic in the Kernel & Hardware sub-forum if there isn't an appropriate topic already in discussion.
yes I had the micrcode update early feature enabled, and I use the boot partition, but in efi... the problem is in the bfs patch I think because with the change of bfs 472 (with the kernel 4.7.2-1 I hadn't any type of problem)to 480 and to 490 the freezes started , what I see is that the problems are in haswell desktops only.. I don't know why and it is very annoying
Offline
@agm28011997
Pls see my post (#10) here:
https://bbs.archlinux.org/viewtopic.php?id=216966
for similar sudden and hard freezes with linux-ck-silvermont 4.7.3-[2-4]. So it appears NOT to be specific to haswell desktops.
HOWEVER, there are similar reports of sudden and hard freezes on the stock (i.e., 4.7.2-1-ARCH) kernel too:
https://bbs.archlinux.org/viewtopic.php?id=217015
so this patchset might not be solely to blame. Hopefully the dust settles quickly on this bug since nothing is more "fun" then waiting for the next freeze on either kernel.....
Halocaridina
4X Intel(R) Xeon(R) CPU E7-8870 @ 2.40GHz; 256GiB; 11TB EXT4; SL6.8
2X Intel(R) Xeon(R) CPU E5-2680 @ 2.70GHz; 256GiB; 3.4TB EXT4; SL6.8
Thinkpad X250; Broadwell-ULT Core i5; 8GiB, 525GB Crucial_CT525MX3; Arch
Thinkpad T440S; Haswell-ULT Core i7; 12GiB, 512GB SanDisk X210; Arch
Offline
I see the posts and I posted my messages, PD con kolivas and gravyky thanks for your work, I desire to be more helpful but I only can test the packages..
Offline
nothing the freezes come browsing only too, I haven't put any game in the background, the freeze come when I was commenting in the ck blogspot
Offline
Looking at my package files, it wasn't big enough to be the complete file, so maybe the broken downloading is causing part of my problem...14MB instead of ~50MB, but the file isn't even a part file, so it thinks it's complete...Deleting the incomplete file and replacing it with a fresh download seems to have corrected some of my problem....may have to do with my proxy server....
I may have to CONSOLE you about your usage of ridiculously easy graphical interfaces...
Look ma, no mouse.
Offline
Oops
Upgraded linux-ck-k10 (4.7.3-4 -> 4.7.3-5) & linux-ck-k10-headers (4.7.3-4 -> 4.7.3-5) earlier today.
Boot failed, (EE) NVIDIA: Failed to initialize the NVIDIA kernel module.
4.7.2-1-ARCH/nvidia 370.28-1 works ok... .. .
repo-ck/nvidia-ck-k10 370.28-2 (ck-k10) installed.
Al Einstein: "Man soll die Dinge so einfach machen wie möglich ~ aber nicht einfacher." (Things should be as simple as possible ~ but not too simple.) ~ Al (Einstein) war ein Cousin von Albert, "Al" ist die Abkürzung für Aloysius
Offline
Me too.
Otherwise linux-ck 4.7.3-5 is finally stable but graphical login fails with that nvidia error. I haven't had time yet to check how to fix these module signature errors but I will later.
Thanks graysky for making new versions during these problems.
Duettaeánn aef cirrán Cáerme Gláeddyv. Yn á esseáth.
Offline
Me too.
Otherwise linux-ck 4.7.3-5 is finally stable but graphical login fails with that nvidia error. I haven't had time yet to check how to fix these module signature errors but I will later.
Thanks graysky for making new versions during these problems.
Just delete package nvidia-ck* from pacman cache, then download it again and reinstall using text console before GUI.
It can be done by yaourt in 2 runs, 1st time it will ask to delete cached because of wrong sign.
I believe this happens when wrong configured/used DKMS, will be happening each time kernel upgraded.
Last edited by alexzk (2016-09-15 13:37:19)
Offline
mclang wrote:Me too.
Otherwise linux-ck 4.7.3-5 is finally stable but graphical login fails with that nvidia error. I haven't had time yet to check how to fix these module signature errors but I will later.
Thanks graysky for making new versions during these problems.
Just delete package nvidia-ck* from pacman cache, then download it again and reinstall using text console before GUI.
It can be done by yaourt in 2 runs, 1st time it will ask to delete cached because of wrong sign.
I believe this happens when wrong configured/used DKMS, will be happening each time kernel upgraded.
I figured this out by my own and now everything works.
But thanks anyways! It is nice that not everybody tell to read back last ten pages or so to find solution. Not that I have been told to do so... But you know
Duettaeánn aef cirrán Cáerme Gláeddyv. Yn á esseáth.
Offline
4.7.4-1-ck
hangs again (seemed fixed with 4.7.3-last)
Offline
Apparently it's a problem with polipo. It stores an incomplete version of the package file when it doesn't get downloaded all at once. So removing it from pacman cache doesn't help. Polipo just gives me the incomplete version it stores and can't download the remaining file. When I rename the pkg file to have the part extension, (polipo) reports that it is a http server that doesn't support byte ranges. So if I change http_proxy or just clear it, then it can download the remaining part of the file.
I may have to CONSOLE you about your usage of ridiculously easy graphical interfaces...
Look ma, no mouse.
Offline
It seems we have at least one mirror of repo-ck now. In the past I have asked that users not mirror the packages so I could maintain a somewhat accurate count of package popularity. Thank you to everyone who has written me privately with generous offers to spin up a mirror by the way Anyway, since we now have one out there and since godaddy is hit-or-miss actually writing out something as basic as a fricking server log, users have experienced some problems with partial downloads, etc. please feel free to mirror repo-ck if you wish. I will continue to support it and push packages there. Since I sign the packages, it doesn't matter where you get them from mirror-wise, so long as you've enabled pacman checking, you should be assured that you're getting what I built.
Last edited by graysky (2016-10-05 20:38:54)
CPU-optimized Linux-ck packages @ Repo-ck • AUR packages • Zsh and other configs
Offline
I'm having problems upgrading to linux-ck-sandybridge 4.8.2-1:
virtualbox-ck-host-modules-sandybridge: installing linux-ck-sandybridge (4.8.2-1) breaks dependency 'linux-ck-sandybridge<4.8'
This doesn't seem like a problem on my end, but if it is what am I doing wrong, and what could I do to fix it?
Last edited by DemonicSavage (2016-10-20 14:18:22)
Offline
@Demonic - My mirrorlist was out-of-date it seems as the build chroot did not pull down 5.1.8 to build against. Fixed now, thanks for reporting.
Last edited by graysky (2016-10-20 18:52:02)
CPU-optimized Linux-ck packages @ Repo-ck • AUR packages • Zsh and other configs
Offline
Just re-installed Arch and added the repo-ck repository. As recommended un-commented the XferCommend for wget and it didn't seem to work so well. Several error messages. Switched to the one from the wget wiki and it works much better.
XferCommand = /usr/bin/wget --passive-ftp -q --show-progress -c -O %o %u
@glavin: Thank you, will do.
@glavin: Thanks, made the change and it works great.
Last edited by cowlick (2016-10-21 23:59:10)
DELL Inspiron 14-3452, 32GB emmc, 4 GB RAM
Offline
You can try the new mirror for repo-ck, works great. Just add the following server to pacman.conf
[repo-ck]
Server = https://mirror.archlinux.no/repo-ck/os/$arch
https://wiki.archlinux.org/index.php/Repo-ck#Mirrors
Last edited by glavin (2016-10-21 11:59:48)
Speak your mind even if your voice shakes.
Offline
Since last couple of days on 4.8 kernels, using silvermont, my machine, dell 3147 - pentium 3530, machine will not boot, get error something like cpu0 stuck, nmi error. downgraded to 4.7.8 which runs well. the stock kernel in the 4.8 series boots normally.
Offline
@glenb77 - Please post to ck's blog. Note that 4.8.4-1-ck does not use ck3 (MuQSS 0.115) it uses ck1 (bfs 0.512).
CPU-optimized Linux-ck packages @ Repo-ck • AUR packages • Zsh and other configs
Offline
Since last couple of days on 4.8 kernels, using silvermont, my machine, dell 3147 - pentium 3530, machine will not boot, get error something like cpu0 stuck, nmi error. downgraded to 4.7.8 which runs well. the stock kernel in the 4.8 series boots normally.
I'm having a similar issue with nehalem. I am able to boot but the system gives me similar errors and locks up after starting mpd. Sticking with 4.7.8-1 for now. (Intel i5-520M)
Offline
@Justin - Please post to ck's blog if you haven't already done so.
CPU-optimized Linux-ck packages @ Repo-ck • AUR packages • Zsh and other configs
Offline