mkinitcpio autodetect fails to run leading into error on booting

cminus · 2018-03-16 16:22:51

After last kernel update, I can't boot getting the following error message

starting version 238
mount: /new_root: can't find UUID=d67fb139-67ab-4ec1-808d-63367fad91f0.
You are now being dropped into an emergency shell.
sh: can't access tty: job control turned off
[rootfs ]#

And nothing more. even typing anything doesn't appear or have any effect on this shell.

I was able to boot using old initramfs option.

Here is the output of blkid

sudo blkid
/dev/sda1: UUID="A4AC-FD2C" TYPE="vfat" PARTLABEL="EFI system partition" PARTUUID="21a768ab-0728-4541-b32f-47cd560122fc"
/dev/sda2: PARTLABEL="Microsoft reserved partition" PARTUUID="6608935f-16a3-4218-905d-075b606953e4"
/dev/sda3: UUID="223C464B3C461A65" TYPE="ntfs" PARTLABEL="Basic data partition" PARTUUID="ffbd2d93-619b-443d-aaf1-d458b295dd8f"
/dev/sda4: UUID="d67fb139-67ab-4ec1-808d-63367fad91f0" TYPE="ext4" PARTLABEL="ArchBoot" PARTUUID="2fe92116-4505-410e-872b-53da9d73de19"
/dev/sda5: UUID="d34a569f-a3f3-4d7c-9884-32b3a6d7794d" TYPE="ext4" PARTLABEL="Home" PARTUUID="7ddaf3ee-6632-4b97-8f99-d5243833f320"
/dev/sda6: LABEL="Games" UUID="269FE1237DDA4770" TYPE="ntfs" PTTYPE="dos" PARTUUID="e72ced68-060c-4fa7-a4d0-a63d18d0d166"
/dev/sda7: LABEL="Programs" UUID="53541BC9174B0171" TYPE="ntfs" PTTYPE="dos" PARTUUID="b7ccbf4a-f706-4e4c-91ac-c6448d919404"
/dev/sda8: LABEL="Learning" UUID="643017F03017C846" TYPE="ntfs" PARTLABEL="Basic data partition" PARTUUID="522a8bc6-d5d3-49d4-afc9-bf6f9ef72de9"
/dev/sda9: UUID="710e02c5-9e41-4496-b3a3-36b37dd06af4" TYPE="swap" PARTUUID="9cac226e-b63d-4281-924e-52adb09625ef"
/dev/sda10: LABEL="Media & Services" UUID="1D6AAD275BE887C9" TYPE="ntfs" PARTLABEL="Apple_HFS_Untitled_2" PARTUUID="997a256d-f76c-49a2-8cf4-0b0fb32e25cc"
/dev/sda11: LABEL="Work" UUID="3BF86273567F397C" TYPE="ntfs" PARTUUID="ecb9834e-3ec9-4bb5-950f-a2b094760f76"
/dev/sda12: LABEL="srv" UUID="81a8da8e-bc77-4aa7-b5b0-1b54175b692c" TYPE="ext4" PARTUUID="f0f9f31b-4fb6-4c24-9b9d-64b366467f66"
/dev/sda13: LABEL="SAMSUNG_REC2" UUID="742ABED12ABE901A" TYPE="ntfs" PARTLABEL="Basic data partition" PARTUUID="95166a79-939a-4cf6-86e1-027e00783f22"
/dev/sda14: LABEL="SAMSUNG_REC" UUID="80EF-FAA3" TYPE="vfat" PARTLABEL="Basic data partition" PARTUUID="6e29484f-8a57-42aa-4173-636c65706975"

Note that /dev/sda4 has the same UUID that it can't find.

And here is the output of /etc/fstab

# /etc/fstab: static file system information
#
# <file system> <dir> <type> <options> <dump> <pass>
# /dev/sda4
UUID=d67fb139-67ab-4ec1-808d-63367fad91f0 / ext4 rw,relatime,data=ordered 0 1
# /dev/sda5
UUID=d34a569f-a3f3-4d7c-9884-32b3a6d7794d /home ext4 rw,relatime,data=ordered 0 2

#/dev/sda14 Server
UUID=81a8da8e-bc77-4aa7-b5b0-1b54175b692c /srv ext4 auto,rw 0 0
# EFI Partition
UUID=A4AC-FD2C /boot vfat defaults 0 1

According to some post, I have to run mkinitcpio -p linux, but the auto detect part fails with the following output

==> Building image from preset: /etc/mkinitcpio.d/linux.preset: 'default'
-> -k /boot/vmlinuz-linux -c /etc/mkinitcpio.conf -g /boot/initramfs-linux.img
==> Starting build: 4.15.9-1-ARCH
-> Running build hook: [base]
-> Running build hook: [udev]
-> Running build hook: [autodetect]
find: ‘sort’ terminated by signal 11
modprobe: ERROR: missing parameters. See -h.
-> Running build hook: [modconf]
-> Running build hook: [block]
-> Running build hook: [filesystems]
-> Running build hook: [keyboard]
-> Running build hook: [fsck]
==> Generating module dependencies
==> Creating gzip-compressed initcpio image: /boot/initramfs-linux.img
==> Image generation successful
==> Building image from preset: /etc/mkinitcpio.d/linux.preset: 'fallback'
-> -k /boot/vmlinuz-linux -c /etc/mkinitcpio.conf -g /boot/initramfs-linux-fallback.img -S autodetect
==> Starting build: 4.15.9-1-ARCH
-> Running build hook: [base]
-> Running build hook: [udev]
-> Running build hook: [modconf]
-> Running build hook: [block]
==> WARNING: Possibly missing firmware for module: wd719x
==> WARNING: Possibly missing firmware for module: aic94xx
-> Running build hook: [filesystems]
-> Running build hook: [keyboard]
-> Running build hook: [fsck]
==> Generating module dependencies
==> Creating gzip-compressed initcpio image: /boot/initramfs-linux-fallback.img
==> Image generation successful

I believe that's reason for boot failure, as it doesn't add the required modules to initramfs image when created, am i right ?

And any help to go further would be appreciated.

Thanks

Last edited by cminus (2018-03-16 16:33:22)

loqs · 2018-03-16 16:29:36

UUID=d67fb139067ab-ec1-808d-63367fad91f0
UUID="d67fb139-67ab-4ec1-808d-63367fad91f0"

The UUID's are not the same, transcription error? Please use code tags not quote tags. What is the contents of /etc/mkinitcpio.conf?

Slithery · 2018-03-16 16:30:08

If you mean the WARNING lines then no, they have nothing to do with your problem. Every Arch installation gets these unless you have installed the drivers for the specific mentioned hardware (you'd know if you needed them).

loqs · 2018-03-16 16:31:29

@Slithery

  -> Running build hook: [autodetect]
find: ‘sort’ terminated by signal 11
modprobe: ERROR: missing parameters. See -h.

cminus · 2018-03-16 16:34:06

loqs wrote:

UUID=d67fb139067ab-ec1-808d-63367fad91f0
UUID="d67fb139-67ab-4ec1-808d-63367fad91f0"
The UUID's are not the same, transcription error? Please use code tags not quote tags. What is the contents of /etc/mkinitcpio.conf?

It was a typo as I got the first one from mobile camera shot.. I double checked them

Slithery · 2018-03-16 16:34:46

Sorry, I missed that bit.

Ignore me cminus

cminus · 2018-03-16 16:35:32

Slithery wrote:

Sorry, I missed that bit.
Ignore me cminus

It's okay.
About the warnings, I've been living with that for 5 years and still good

frostschutz · 2018-03-16 16:47:25

Is your /sys not mounted properly?

find /sys/devices -name uevent
find /sys/devices -name uevent -exec sort -u '{}' +

If those look fine, then patch /lib/initcpio/build/autodetect (add set -x at start, and set +x at end of build() function)

--- autodetect	2017-10-01 20:21:09.000000000 +0200
+++ autodetect.modified	2018-03-16 17:45:30.676958063 +0100
@@ -1,6 +1,7 @@
 #!/bin/bash
 
 build() {
+    set -x
     local m=
     local -a md_devs mods
 
@@ -55,6 +56,7 @@
     if (( ${#_autodetect_cache[*]} )); then
         quiet "caching %d modules" "${#_autodetect_cache[*]}"
     fi
+    set +x
 }
 
 help() {

when you run mkinitcpio, it will print some more debug output to help figure out what is going on

cminus · 2018-03-16 16:54:13

@frostschutz, the first commands seems to run good

Here is a sample output

/sys/devices/virtual/workqueue/uevent
/sys/devices/virtual/dmi/id/uevent
/sys/devices/virtual/input/mice/uevent
/sys/devices/virtual/vc/vcsa5/uevent
/sys/devices/virtual/vc/vcs6/uevent
/sys/devices/virtual/vc/vcsa3/uevent
/sys/devices/virtual/vc/vcs4/uevent
/sys/devices/virtual/vc/vcsa1/uevent
/sys/devices/virtual/vc/vcs2/uevent
/sys/devices/virtual/vc/vcs/uevent
/sys/devices/virtual/vc/vcsa6/uevent
/sys/devices/virtual/vc/vcsa4/uevent
/sys/devices/virtual/vc/vcs5/uevent
/sys/devices/virtual/vc/vcsa2/uevent
/sys/devices/virtual/vc/vcs3/uevent
/sys/devices/virtual/vc/vcsa/uevent
/sys/devices/virtual/vc/vcs1/uevent
/sys/devices/uncore_cbox_0/uevent

But the second one fails with signal 11

Arch~ % find /sys/devices -name uevent -exec sort -u '{}' +
find: ‘sort’ terminated by signal 11

frostschutz · 2018-03-16 17:00:07

What is your locale?

find /sys/devices -name uevent -print -exec sort -u '{}' \;

Does `echo anything | sort -u` work?

Last edited by frostschutz (2018-03-16 17:00:47)

cminus · 2018-03-16 17:08:59

frostschutz wrote:

What is your locale?
find /sys/devices -name uevent -print -exec sort -u '{}' \;
Does `echo anything | sort -u` work?

My locale is

en_US.utf8

Yes

Trying

  echo "rep\nrep" | sort -u

outputs

rep

Also Command

 find /sys/devices -name uevent -print -exec sort -u '{}' \;

Works well.

Last edited by cminus (2018-03-16 17:09:39)

frostschutz · 2018-03-16 17:21:08

You sure it works well? The problem might be somewhere in the middle.

Does cat {} + instead of sort -u {} + work?

I guess you could also use 'strace sort -u {} +' to get, well, quite a lot more output...

WorMzy · 2018-03-16 17:23:03

Please use code tags for terminal output/config files/etc.

https://wiki.archlinux.org/index.php/Co … s_and_code

cminus · 2018-03-16 17:59:09

frostschutz wrote:

You sure it works well? The problem might be somewhere in the middle.
Does cat {} + instead of sort -u {} + work?
I guess you could also use 'strace sort -u {} +' to get, well, quite a lot more output...

Not really, it process a few files then gets the same signal

INTERFACE=enp3s0
IFINDEX=2
DRIVER=r8169
PCI_CLASS=20000
PCI_ID=10EC:8168
PCI_SUBSYS_ID=144D:C0D1
PCI_SLOT_NAME=0000:03:00.0
MODALIAS=pci:v000010ECd00008168sv0000144Dsd0000C0D1bc02sc00i00
DRIVER=pcieport
PCI_CLASS=60400
PCI_ID=8086:1E16
PCI_SUBSYS_ID=144D:C0D1
PCI_SLOT_NAME=0000:00:1c.3
MODALIAS=pci:v00008086d00001E16sv0000144Dsd0000C0D1bc06sc04i00
DRIVER=i801_smbus
PCI_CLASS=C0500
PCI_ID=8086:1E22
PCI_SUBSYS_ID=144D:C0D1
PCI_SLOT_NAME=0000:00:1f.3
MODALIAS=pci:v00008086d00001E22sv0000144Dsd0000C0D1bc0Csc05i00
DRIVER=ivb_uncore
PCI_CLASS=60000
PCI_ID=8086:0154
PCI_SUBSYS_ID=144D:C0D1
PCI_SLOT_NAME=0000:00:00.0
MODALIAS=pci:v00008086d00000154sv0000144Dsd0000C0D1bc06sc00i00
DEVTYPE=usb_interface
PRODUCT=2232/1029/25
TYPE=239/2/1
INTERFACE=14/2/0
MODALIAS=usb:v2232p1029d0025dcEFdsc02dp01ic0Eisc02ip00in01
find: ‘cat’ terminated by signal 11

strace produces a list of syscalls, where I usually get lost.

As I currently don't have the time for more investigation, I think I will quit using this Arch for a while.

frostschutz · 2018-03-16 18:02:49

Well, I expected the "terminated by signal 11" problem to appear somewhere in the middle when you -print -exec \;

Then you'd know which file is causing it (as that prints the filename first).

Anything in dmesg by the way?

cminus · 2018-03-16 18:04:42

Same message repeated over and over

[ 6495.367857] RIP: __x86_indirect_thunk_rax+0x10/0x20 RSP: ffffaa8503cdfd20
[ 6495.367859] ---[ end trace b80494dee153917b ]---
[ 6498.565575] general protection fault: 0000 [#39] PREEMPT SMP PTI
[ 6498.565580] Modules linked in: fuse rfcomm bbswitch(O) ccm bnep nls_iso8859_1 nls_cp437 vfat fat snd_hda_codec_hdmi arc4 snd_hda_codec_realtek intel_rapl x86_pkg_temp_thermal snd_hda_codec_generic intel_powerclamp iwldvm mac80211 iTCO_wdt iTCO_vendor_support joydev mousedev mxm_wmi coretemp snd_hda_intel kvm_intel i915 btusb snd_hda_codec snd_hda_core btrtl btbcm snd_hwdep btintel snd_pcm bluetooth snd_timer kvm i2c_algo_bit drm_kms_helper snd iwlwifi drm irqbypass soundcore ecdh_generic crct10dif_pclmul intel_gtt mei_me cfg80211 agpgart syscopyarea sysfillrect sysimgblt ghash_clmulni_intel input_leds intel_cstate intel_uncore rfkill r8169 fb_sys_fops led_class mii mei evdev i2c_i801 intel_rapl_perf pcspkr psmouse lpc_ich shpchp rtc_cmos mac_hid battery ac wmi sg crypto_user ip_tables x_tables ext4
[ 6498.565632]  crc16 mbcache jbd2 fscrypto hid_generic usbhid hid sr_mod cdrom sd_mod serio_raw atkbd libps2 crc32_pclmul crc32c_intel ahci xhci_pci libahci aesni_intel ehci_pci aes_x86_64 xhci_hcd ehci_hcd crypto_simd libata cryptd glue_helper usbcore scsi_mod usb_common i8042 serio [last unloaded: videobuf2_memops]
[ 6498.565655] CPU: 1 PID: 1458 Comm: cat Tainted: G      D    O     4.15.9-1-ARCH #1
[ 6498.565657] Hardware name: SAMSUNG ELECTRONICS CO., LTD. 550P5C/550P7C/NP550P5C-S03AE, BIOS P09ABI.026.141024.dg 10/24/2014
[ 6498.565664] RIP: 0010:__x86_indirect_thunk_rax+0x10/0x20
[ 6498.565666] RSP: 0018:ffffaa8503c6bd20 EFLAGS: 00010202
[ 6498.565668] RAX: 7d89c80948000002 RBX: ffff8e761e722020 RCX: 0000000000000012
[ 6498.565670] RDX: 0000000000000000 RSI: ffff8e758ac6b000 RDI: ffff8e761e722010
[ 6498.565672] RBP: ffff8e758ac6b000 R08: ffff8e762552430f R09: ffff8e758ac6b136
[ 6498.565674] R10: 0000000000003000 R11: ffffffffb9e926e7 R12: ffff8e761e722010
[ 6498.565675] R13: ffff8e758ac6b000 R14: 0000000000000001 R15: ffff8e75be34a800
[ 6498.565678] FS:  00007f66e29c1500(0000) GS:ffff8e762f240000(0000) knlGS:0000000000000000
[ 6498.565680] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 6498.565681] CR2: 0000000001c9f000 CR3: 00000001b05a4003 CR4: 00000000001606e0
[ 6498.565683] Call Trace:
[ 6498.565690]  ? dev_uevent+0x9b/0x2f0
[ 6498.565694]  ? kmem_cache_alloc_trace+0x16c/0x1c0
[ 6498.565696]  ? uevent_show+0xa1/0x100
[ 6498.565699]  ? dev_attr_show+0x1f/0x50
[ 6498.565702]  ? kernfs_seq_start+0x2d/0x90
[ 6498.565705]  ? sysfs_kf_seq_show+0x9b/0x100
[ 6498.565708]  ? seq_read+0xee/0x480
[ 6498.565712]  ? __vfs_read+0x36/0x170
[ 6498.565716]  ? vfs_read+0x91/0x130
[ 6498.565720]  ? SyS_read+0x52/0xc0
[ 6498.565724]  ? do_syscall_64+0x74/0x190
[ 6498.565727]  ? entry_SYSCALL_64_after_hwframe+0x3d/0xa2
[ 6498.565730] Code: 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 e8 07 00 00 00 f3 90 0f ae e8 eb f9 48 89 04 24 <c3> 0f 1f 44 00 00 66 2e 0f 1f 84 00 00 00 00 00 e8 07 00 00 00 
[ 6498.565766] RIP: __x86_indirect_thunk_rax+0x10/0x20 RSP: ffffaa8503c6bd20
[ 6498.565769] ---[ end trace b80494dee153917c ]---

frostschutz · 2018-03-16 18:09:49

Well, it may be random, but the indirect_thunk stuff sounds like retpoline spectre patches to me. Maybe someone messed those up again.

Just for kicks, see if it still happens without microcode update, and kernel parameters pti=off spectre_v2=off

loqs · 2018-03-16 18:26:14

Indirect thunk is retpoline however looking at https://cdn.kernel.org/pub/linux/kernel … Log-4.15.9 the only retpoline change was for BPF.
The latest microcode update did include fixed updates for IBPB and IBRS and additional CPU's gained IBPB and IBRS support.

arcdmp · 2018-03-30 23:45:37

I've got a near-identical problem, but I've caught it before rebooting.
mkinicpio fails on the "autodetect" hook, but with a "signal 9" error:

  -> Running build hook: [autodetect]
find: ‘sort’ terminated by signal 9
modprobe: ERROR: missing parameters. See -h.

Running either of these commands:

find /sys/devices -name uevent -exec sort -u '{}' +
sudo find /sys/devices -name uevent -exec sort -u '{}' +

also results in the same error:

find: ‘sort’ terminated by signal 9

Also notable, I just started seeing kernel paging segfault dumps in dmesg.
These keep repeating, from 1 hour to 10 minutes between each crash. Oddly enough though, my machine still looks and works fine (until I eventually reboot, that is).
The error seems to be caused by something called "dev_uevent", not "indirect_thunk":

[  +6.677429] BUG: unable to handle kernel paging request at ffffffffc04a81f8
[  +0.000010] IP: dev_uevent+0x7c/0x2e0
[  +0.000001] PGD 1cb00e067 P4D 1cb00e067 PUD 1cb010067 PMD 221478067 PTE 0
[  +0.000005] Oops: 0000 [#14] SMP PTI
[  +0.000002] Modules linked in: xt_REDIRECT nf_nat_redirect xt_tcpudp nf_nat_pptp nf_nat_proto_gre nf_c
onntrack_pptp nf_conntrack_proto_gre iptable_filter ipt_MASQUERADE nf_nat_masquerade_ipv4 iptable_nat nf
_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 nf_nat nf_conntrack libcrc32c crc32c_generic ccm intel_rapl a
rc4 snd_hda_codec_realtek x86_pkg_temp_thermal intel_powerclamp snd_hda_codec_generic coretemp kvm_intel
 snd_hda_intel snd_hda_codec joydev kvm mousedev snd_hda_core iTCO_wdt ath9k_htc nouveau iTCO_vendor_sup
port i915 alx snd_hwdep snd_pcm asus_nb_wmi asus_wmi ttm sparse_keymap ath9k_common wmi_bmof mxm_wmi ath
9k_hw btusb btrtl snd_timer btbcm i2c_algo_bit btintel snd bluetooth irqbypass soundcore drm_kms_helper
drm ath mac80211 ecdh_generic intel_gtt agpgart intel_cstate intel_rapl_perf syscopyarea
[  +0.000041]  sysfillrect cfg80211 sysimgblt fb_sys_fops rfkill mdio evdev input_leds lpc_ich psmouse i
2c_i801 led_class pcspkr shpchp mei_me mei mac_hid thermal video battery button ac wmi sch_fq_codel cryp
to_user acpi_call(O) ip_tables x_tables ext4 crc16 mbcache jbd2 fscrypto algif_skcipher af_alg dm_crypt
dm_mod sr_mod sd_mod cdrom uas usb_storage crct10dif_pclmul crc32_pclmul crc32c_intel ghash_clmulni_inte
l pcbc serio_raw atkbd libps2 ahci libahci aesni_intel xhci_pci aes_x86_64 crypto_simd glue_helper ehci_
pci libata xhci_hcd ehci_hcd cryptd scsi_mod usbcore usb_common i8042 serio [last unloaded: iwlwifi]
[  +0.000042] CPU: 1 PID: 20652 Comm: sort Tainted: G      D W  O    4.14.30-1-lts #1           [0/1918]
[  +0.000001] Hardware name: ASUSTeK COMPUTER INC. N56VM/N56VM, BIOS N56VM.206 04/13/2012
[  +0.000002] task: ffff9d64ab059d00 task.stack: ffffb4ca03d50000
[  +0.000003] RIP: 0010:dev_uevent+0x7c/0x2e0
[  +0.000001] RSP: 0018:ffffb4ca03d53d38 EFLAGS: 00010286
[  +0.000003] RAX: ffffffffc04a81c0 RBX: ffff9d65a1f28820 RCX: 0000000000000012
[  +0.000001] RDX: 0000000000000000 RSI: ffff9d63d6e8d12e RDI: 0000000000000000
[  +0.000002] RBP: ffff9d63d6e8d000 R08: ffff9d65a50d32f7 R09: ffff9d63d6e8d136
[  +0.000001] R10: 0000000000003000 R11: ffffffff88e3a789 R12: ffff9d65a1f28810
[  +0.000002] R13: ffff9d63d6e8d000 R14: 0000000000000001 R15: ffff9d64d1f80280
[  +0.000002] FS:  00007ff99a16cb80(0000) GS:ffff9d65aee40000(0000) knlGS:0000000000000000
[  +0.000002] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  +0.000002] CR2: ffffffffc04a81f8 CR3: 00000001c9b28003 CR4: 00000000001606e0
[  +0.000001] Call Trace:
[  +0.000006]  ? kmem_cache_alloc_trace+0x140/0x190
[  +0.000004]  uevent_show+0xa1/0x100
[  +0.000003]  dev_attr_show+0x1f/0x50
[  +0.000003]  ? kernfs_seq_start+0x2d/0x90
[  +0.000003]  sysfs_kf_seq_show+0x9d/0x120
[  +0.000004]  seq_read+0xee/0x480
[  +0.000004]  __vfs_read+0x36/0x170
[  +0.000002]  vfs_read+0x89/0x130
[  +0.000003]  SyS_read+0x52/0xc0
[  +0.000004]  do_syscall_64+0x67/0x120
[  +0.000004]  entry_SYSCALL_64_after_hwframe+0x3d/0xa2
[  +0.000002] RIP: 0033:0x7ff9996073a1
[  +0.000002] RSP: 002b:00007fff680e1d08 EFLAGS: 00000246 ORIG_RAX: 0000000000000000
[  +0.000002] RAX: ffffffffffffffda RBX: 00005625c720ac80 RCX: 00007ff9996073a1
[  +0.000002] RDX: 00000000002fa000 RSI: 00007ff9810ba40c RDI: 0000000000000003
[  +0.000001] RBP: 00007ff9998ce240 R08: 0000000000000003 R09: 0000000000000077
[  +0.000002] R10: 00005625c7208010 R11: 0000000000000246 R12: 00007ff9810ba40c
[  +0.000001] R13: 00000000002fafd8 R14: 0000000000000d68 R15: 00000000002fafd8
[  +0.000002] Code: 89 ef e8 18 d5 1c 00 48 8b 43 78 48 85 c0 74 12 48 8b 10 48 c7 c6 bf a7 e3 88 48 89 ef e8 fd d4 1c 00 48 8b 43 70 48 85 c0 74 18 <48> 8b 40 38 48 85 c0 74 0f 48 89 ee 4c 89 e7 e8 50 aa 57 00 85
[  +0.000031] RIP: dev_uevent+0x7c/0x2e0 RSP: ffffb4ca03d53d38
[  +0.000001] CR2: ffffffffc04a81f8
[  +0.000002] ---[ end trace b5241eaa3bfbfd4d ]---

I did unload iwlwifi and its dependency, iwldvm, since I use ath9k_htc. I did this only to reduce the hundreds of lines in dmesg of:

[191900.308036] iwlwifi 0000:03:00.0: Radio type=0x2-0x0-0x0
[191900.595823] iwlwifi 0000:03:00.0: Radio type=0x2-0x0-0x0
[191900.696612] iwlwifi 0000:03:00.0: Radio type=0x2-0x0-0x0
[191900.987620] iwlwifi 0000:03:00.0: Radio type=0x2-0x0-0x0

But maybe that was a mistake.

Also probably unrelated, but hundreds of these error messages keep filling up dmesg/journalctl:

[drm:intel_pipe_update_end [i915]] *ERROR* Atomic update failure on pipe A (start=5076691 end=5076692) time 390 us, min 1073, max 1079, scanline start 1071, end 1092

I can't find anything about how to fix it. As far as I can tell, it appears to be something wrong with Intel's i915 video driver.

What steps can I take to safely generate the autodetect hook?
I tried downgrading the kernel (linux-lts) to the previous version (linux-lts-4.14.30-1), but the exact same error still happens.
Is it possible to manually create the autodetect hook file? I could do it, but there's nothing to work off of in /lib/initcpio/build/autodetect.

At this point I'm sort of held hostage by my machine unless I can fix this or it crashes.

loqs · 2018-03-31 00:23:21

Do you have boot media that is known to work from which you could try regenerating the initrd if the system has to be restarted or crashes and will not boot because the generated initrd is bad?
Edit:
It does look as if it is going to be impossible to use the autodetect hook as the hook enumerates through /sys/devices that triggers the OOPS.
If you remove the autodetect hook does mkinitrd then execute successfully?

Last edited by loqs (2018-03-31 00:33:56)

arcdmp · 2018-03-31 00:38:11

Yes, I've got the USB drive media originally used to install the system, and I could mount the drives and chroot into the system.

Edit: Yes, removing "autodetect" results in a successful image generation. However, I'm using a full system encryption, would that cause any new issues?
Edit 2: I should be more clear, it never really "failed", it simply printed out that autodetect generation failed, and then continued to successfully generate the rest of the modules.

Last edited by arcdmp (2018-03-31 01:19:52)

arcdmp · 2018-03-31 02:04:23

I decided to debug mkinitcpio with strace, and came across something interesting:

9205  open("/usr/share/locale/en_US.UTF-8/LC_MESSAGES/findutils.mo", O_RDONLY) = -1 ENOENT (No such file or directory)
9205  open("/usr/share/locale/en_US.utf8/LC_MESSAGES/findutils.mo", O_RDONLY) = -1 ENOENT (No such file or directory)
9205  open("/usr/share/locale/en_US/LC_MESSAGES/findutils.mo", O_RDONLY) = -1 ENOENT (No such file or directory)
9205  open("/usr/share/locale/en.UTF-8/LC_MESSAGES/findutils.mo", O_RDONLY) = -1 ENOENT (No such file or directory)
9205  open("/usr/share/locale/en.utf8/LC_MESSAGES/findutils.mo", O_RDONLY) = -1 ENOENT (No such file or directory)
9205  open("/usr/share/locale/en/LC_MESSAGES/findutils.mo", O_RDONLY) = -1 ENOENT (No such file or directory)
9205  openat(AT_FDCWD, "/usr/lib/charset.alias", O_RDONLY|O_NOFOLLOW) = -1 ENOENT (No such file or directory)
9205  write(2, "find: ", 6)             = 6
9205  write(2, "\342\200\230sort\342\200\231 terminated by signal "..., 33) = 33
9205  write(2, "\n", 1)                 = 1

The core problem seems that "find" can't find specific locale files inside the directory "/usr/share/locale/".
The only subdirectories from "locale/" that exist are:

en
en_US

These subdirectories don't even exist:

en_US.UTF-8
en_US.utf8
en.UTF-8
en.utf8

And (except for "/usr/lib/charset.alias"), the file that they're all looking for is "findutils.mo".
This is odd, considering I've never messed with locale settings since first installing Arch.

Running "locale -a" (to print the list of enabled locales) results in:

C
en_US
en_US.iso88591
en_US.utf8
POSIX

These are both uncommented in "/etc/locale.gen":

en_US.UTF-8 UTF-8  
en_US ISO-8859-1

Running just "locale" shows:

LANG=en_US.UTF-8
LC_CTYPE="en_US.UTF-8"
LC_NUMERIC="en_US.UTF-8"
LC_TIME="en_US.UTF-8"
LC_COLLATE="en_US.UTF-8"
LC_MONETARY="en_US.UTF-8"
LC_MESSAGES="en_US.UTF-8"
LC_PAPER="en_US.UTF-8"
LC_NAME="en_US.UTF-8"
LC_ADDRESS="en_US.UTF-8"
LC_TELEPHONE="en_US.UTF-8"
LC_MEASUREMENT="en_US.UTF-8"
LC_IDENTIFICATION="en_US.UTF-8"
LC_ALL=

And re-running "locale-gen" doesn't change any of these results. This seems like something went wrong with generating the UTF8 locales, but I have no idea where to go from here.

Edit: I also discovered that the "findutils.mo" file exists in most of the other language subdirectories for some reason.

Last edited by arcdmp (2018-03-31 02:09:10)

seth · 2018-03-31 07:15:38

The "missing" locales are hardly crucial - the process gets SIGKILL'd by the kernel, ie. the latter is really unhappy w/ you accessing sth. in sysfs - the problem occured long before in the strace (can you paste it somewhere? No guarantee that it's helpful - can you just "ls /sys"?)

loqs · 2018-03-31 08:56:38

https://elixir.bootlin.com/linux/v4.15. … ore.c#L967 seems to trigger an OOPS in a kmalloc request.
Edit:
Pass this upstream to the linux-mm or linux-kernel mailing lists?
Edit2:
fixed link

Last edited by loqs (2018-03-31 15:05:50)

arcdmp · 2018-03-31 14:04:31

seth wrote:

The "missing" locales are hardly crucial - the process gets SIGKILL'd by the kernel, ie. the latter is really unhappy w/ you accessing sth. in sysfs - the problem occured long before in the strace (can you paste it somewhere? No guarantee that it's helpful - can you just "ls /sys"?)

The whole strace file is pretty large (125 Mb), so I uploaded it to https://transfer.sh/. To get it, you can curl/wget it directly from https://transfer.sh/SaK01/strace_mkinitcpio.txt

Both

ls /sys
ls /sys/devices

work fine and just list the directories contained, did you mean inside the strace file instead?

loqs wrote:

https://elixir.bootlin.com/linux/v4.15. … ore.c#L967 seems to trigger an OOPS in a kmalloc request.
Edit:
Pass this upstream to the linux-mm or linux-kernel mailing lists?

I can't seem to see this link ("This file does not exist." error). Is this a file that might have moved?

Arch Linux

#1 2018-03-16 16:22:51

mkinitcpio autodetect fails to run leading into error on booting

#2 2018-03-16 16:29:36

Re: mkinitcpio autodetect fails to run leading into error on booting

#3 2018-03-16 16:30:08

Re: mkinitcpio autodetect fails to run leading into error on booting

#4 2018-03-16 16:31:29

Re: mkinitcpio autodetect fails to run leading into error on booting

#5 2018-03-16 16:34:06

Re: mkinitcpio autodetect fails to run leading into error on booting

#6 2018-03-16 16:34:46

Re: mkinitcpio autodetect fails to run leading into error on booting

#7 2018-03-16 16:35:32

Re: mkinitcpio autodetect fails to run leading into error on booting

#8 2018-03-16 16:47:25

Re: mkinitcpio autodetect fails to run leading into error on booting

#9 2018-03-16 16:54:13

Re: mkinitcpio autodetect fails to run leading into error on booting

#10 2018-03-16 17:00:07

Re: mkinitcpio autodetect fails to run leading into error on booting

#11 2018-03-16 17:08:59

Re: mkinitcpio autodetect fails to run leading into error on booting

#12 2018-03-16 17:21:08

Re: mkinitcpio autodetect fails to run leading into error on booting

#13 2018-03-16 17:23:03

Re: mkinitcpio autodetect fails to run leading into error on booting

#14 2018-03-16 17:59:09

Re: mkinitcpio autodetect fails to run leading into error on booting

#15 2018-03-16 18:02:49

Re: mkinitcpio autodetect fails to run leading into error on booting

#16 2018-03-16 18:04:42

Re: mkinitcpio autodetect fails to run leading into error on booting

#17 2018-03-16 18:09:49

Re: mkinitcpio autodetect fails to run leading into error on booting

#18 2018-03-16 18:26:14

Re: mkinitcpio autodetect fails to run leading into error on booting

#19 2018-03-30 23:45:37

Re: mkinitcpio autodetect fails to run leading into error on booting

#20 2018-03-31 00:23:21

Re: mkinitcpio autodetect fails to run leading into error on booting

#21 2018-03-31 00:38:11

Re: mkinitcpio autodetect fails to run leading into error on booting

#22 2018-03-31 02:04:23

Re: mkinitcpio autodetect fails to run leading into error on booting

#23 2018-03-31 07:15:38

Re: mkinitcpio autodetect fails to run leading into error on booting

#24 2018-03-31 08:56:38

Re: mkinitcpio autodetect fails to run leading into error on booting

#25 2018-03-31 14:04:31

Re: mkinitcpio autodetect fails to run leading into error on booting

Board footer