EQ overflow FreeBSD 12.0 + nVidia 1660Ti with 430 driver + Ryzen 2400G

It might be worth asking over at NVidia as this issue has been posted a lot on recent cards and the newest version of the Linux driver. Seems like it could be a bug.
 
It might be worth asking over at NVidia as this issue has been posted a lot on recent cards and the newest version of the Linux driver. Seems like it could be a bug.
OK might do that. By the way, do you think it's worth trying 418 driver or whatever the earlier one with 1660ti support is?

Thanks for looking at it.
 
It's not a single issue. "GPU has fallen off the bus" is a generic catch-it-all error message.

I know, I'm just saying it seems to be a common issue across FreeBSD and Linux with the latest Nvidia cards 1070, 1080 and more recent. It may well be a bug in the driver as a lot of ArchLinux users have also reported this issue on Arch Forums with the newer drivers and cards as well which is why it may be best to report it to the Nvidia forums.
 
Seems like this is a driver bug that has been affecting a small percentage of users and that hasn't been fixed for the better part of a decade. I'm wondering now if I've got a valid case for an RMA... Ideally to get an AMD Navi, which I guess is more FreeBSD friendly, in part exchange.
Also wondered if an alternative would be to run both cards (1660Ti and Ryz 2400g) and switch-over the display while running FreeBSD (other ssd has Win10 for games).
 
Which PSU model do you have, by the way?

Seems like this is a driver bug that has been affecting a small percentage of users and that hasn't been fixed for the better part of a decade.

Nope. It's not a single software bug or a hardware issue. It's just a "something is not right" message with pretty much unlimited number of causes.

I'm wondering now if I've got a valid case for an RMA...

Can you get another 1660 Ti? Preferably a different model.
 
Maybe
Which PSU model do you have, by the way?



Nope. It's not a single software bug or a hardware issue. It's just a "something is not right" message with pretty much unlimited number of causes.



Can you get another 1660 Ti? Preferably a different model.
Well, it's worth a try. Maybe could get something without OC. It's just I thought I heard that AMD cards work better in Freebsd/Linux.

PSU is a Corsair 750M. According to the power reqs. calculator I used, it's about twice as much as I need.
 
Maybe

Well, it's worth a try. Maybe could get something without OC. It's just I thought I heard that AMD cards work better in Freebsd/Linux.

PSU is a Corsair 750M. According to the power reqs. calculator I used, it's about twice as much as I need.
1660Ti is, of course, very low power consumption, at 120W.
 
It's just I thought I heard that AMD cards work better in Freebsd/Linux.

No idea, I don't trust AMD fans. You have a Ryzen 2400G APU, thus you can assess the driver (amdgpu) performance and stability yourself.

PSU is a Corsair 750M. According to the power reqs. calculator I used, it's about twice as much as I need.

Ah, ok. There were some reports of issues with Seasonic Focus PSUs.
 
No idea, I don't trust AMD fans. You have a Ryzen 2400G APU, thus you can assess the driver (amdgpu) performance and stability yourself.
I assume you mean fans in the sense of "advocates", rather than "cooling solution". If the latter, then I believe that's manufacturer-dependent. I've always thought NVidia cards are better made but, apparently, the AMD drivers are better supported in FreeBSD/Linux.

I've heard that 2400G works well in FreeBSD but it would be easier to use the same card in both FreeBSD and Win10 - the reason I have got a med/high spec card is so I can play modern games in Win10. Switching back to 2400G on both OSs is not a good option.
 
Thanks all for your help. Going to try for an RMA. Will post on NVidia boards when I get the chance. 👍
 
I have Nvidia GTX 1660 Zotac card.Gforce driver 430.34 compiled by source Gnome and KDE freezes, BUT not the Mate Desktop !! It is working with no problems . I do not believe that is a problem with the nvidia driver. I think some Desktops not work with the new Driver properly.
 
I have Nvidia GTX 1660 Zotac card.Gforce driver 430.34 compiled by source Gnome and KDE freezes, BUT not the Mate Desktop !! It is working with no problems . I do not believe that is a problem with the nvidia driver. I think some Desktops not work with the new Driver properly.
That's interesting. I will try installing Mate Destkop when I get the chance.
 
hello,

not to hi jack the thread, but i'm thinking since this is pretty much the same thing i'd just jump in here...

i have a geforce gtx 1050 on a freebsd 12 box (12.0-RELEASE-p10 FreeBSD 12.0-RELEASE-p10 GENERIC amd64) and with some games (games/warzone2100 and games/endless-sky for now) the kernel panics after maybe a minute. using kgdb, here's what i get:


Code:
# kgdb /boot/kernel/kernel /var/crash/vmcore.9

GNU gdb (GDB) 8.3 [GDB v8.3 for FreeBSD]

Copyright (C) 2019 Free Software Foundation, Inc.

License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>

This is free software: you are free to change and redistribute it.

There is NO WARRANTY, to the extent permitted by law.

Type "show copying" and "show warranty" for details.

This GDB was configured as "x86_64-portbld-freebsd12.0".

Type "show configuration" for configuration details.

For bug reporting instructions, please see:

<http://www.gnu.org/software/gdb/bugs/>.

Find the GDB manual and other documentation resources online at:

    <http://www.gnu.org/software/gdb/documentation/>.



For help, type "help".       

Type "apropos word" to search for commands related to "word"...

Reading symbols from /boot/kernel/kernel...

Reading symbols from /usr/lib/debug//boot/kernel/kernel.debug...

                             

Unread portion of the kernel message buffer:

NVRM: GPU at PCI:0000:01:00: GPU-66172491-d2d9-dd3b-e25e-8cf4ef024576

NVRM: GPU Board Serial Number:

NVRM: Xid (PCI:0000:01:00): 8, Channel 00000009

                             

                             

Fatal trap 12: page fault while in kernel mode

cpuid = 2; apic id = 12      

fault virtual address   = 0x0

fault code              = supervisor read data, page not present

instruction pointer     = 0x20:0xffffffff82de73a4

stack pointer           = 0x28:0xfffffe00004d1430

frame pointer           = 0x28:0xfffffe004d1fca88

code segment            = base 0x0, limit 0xfffff, type 0x1b

                        = DPL 0, pres 1, long 1, def32 0, gran 1

processor eflags        = interrupt enabled, resume, IOPL = 0

current process         = 12 (swi4: clock (0))

trap number             = 12 

panic: page fault            

cpuid = 2                    

time = 1566588099            

KDB: stack backtrace:        

#0 0xffffffff80be78d7 at kdb_backtrace+0x67

#1 0xffffffff80b9b4b3 at vpanic+0x1a3

#2 0xffffffff80b9b303 at panic+0x43

#3 0xffffffff81074bff at trap_fatal+0x35f

#4 0xffffffff81074c59 at trap_pfault+0x49

#5 0xffffffff8107427e at trap+0x29e

#6 0xffffffff8104f625 at calltrap+0x8

Uptime: 13m4s

Dumping 657 out of 8031 MB:..3%..13%..22%..32%..42%..52%..61%..71%..81%..91%



__curthread () at ./machine/pcpu.h:234

234             __asm("movq %%gs:%1,%0" : "=r" (td)

(kgdb) backtrace

#0  __curthread () at ./machine/pcpu.h:234

#1  doadump (textdump=<optimized out>) at /usr/src/sys/kern/kern_shutdown.c:366

#2  0xffffffff80b9b09b in kern_reboot (howto=260) at /usr/src/sys/kern/kern_shutdown.c:446

#3  0xffffffff80b9b513 in vpanic (fmt=<optimized out>, ap=0xfffffe00004d1180) at /usr/src/sys/kern/kern_shutdown.c:872

#4  0xffffffff80b9b303 in panic (fmt=<unavailable>) at /usr/src/sys/kern/kern_shutdown.c:799

#5  0xffffffff81074bff in trap_fatal (frame=0xfffffe00004d1370, eva=0) at /usr/src/sys/amd64/amd64/trap.c:929

#6  0xffffffff81074c59 in trap_pfault (frame=0xfffffe00004d1370, usermode=0) at /usr/src/sys/amd64/amd64/trap.c:765

#7  0xffffffff8107427e in trap (frame=0xfffffe00004d1370) at /usr/src/sys/amd64/amd64/trap.c:441

#8  <signal handler called>

#9  0xffffffff82de73a4 in _nv008794rm () from /boot/modules/nvidia.ko

#10 0xfffffe004d2bd000 in ?? ()

#11 0xffffffff82de73b9 in _nv008790rm () from /boot/modules/nvidia.ko

#12 0xfffff801525cbd80 in ?? ()

#13 0xffffffff82de7237 in _nv008791rm () from /boot/modules/nvidia.ko

#14 0xfffffe004d1fca88 in ?? ()

#15 0xfffffe004d1fcb00 in ?? ()

#16 0x0000000000000000 in ?? ()

(kgdb) list *0xffffffff82de73a4

(kgdb)

the similarity to the OP's problem is that once i actually saw X throwing all these (EE)s, so this and also the kernel panic backtrace mentioning nvidia module makes me think i'm having the same issue. seeing the (EE)s was under the condition that i'd be able to quickly type ctrl+alt+del once the screen freezes, sometimes that would work and un-freeze the screen after a couple of seconds, after which everything would work alright for as long as i would be willing to test (say, at least another hour). oh, and the driver installed is the x11/nvidia-driver. also, i compiled the 430.40 tarball from nvidia.com, then it would be more time until the kernel panicked, but it still would happen eventually. downside of compiling the 430.40 was that i would get up to mountroot> and then have to mount /dev/ada0p3 manually or comment out all the nvidia modules in /boot/loader.conf to have it booting alright, so that's a bummer. just for the heck of it i also replaced the ram with a brand new one, still the same thing.

not sure if there's anything that could be done about it, i've been running this machine for more than two years now (initially with 375 driver from nvidia.com and for over a year now with x11/nvidia-driver and never had a problem, up until now, i don't recall exactly when it started but i'm almost sure after i upgraded userland like 2 weeks ago and x11/nvidia-driver was part of that.
 
an interesting detail i figured yesterday is that the system actually only freezes in fullscreen mode, i had glxspheres64 running for a while with no problem then i switched the window to fullscreen mode (alt+f11, i'm running icewm btw) and after 10 or so seconds screen freezes and i get a reboot after kernel panic.
 
Hello.

I am getting EQ overflow errors in my Xorg log using the nVidia 430 driver. These correspond to my system freezing-up. The mouse and keyboard work for about 10 seconds before this happens.
...
...

this i misinterpreted then as the same thing, sorry for that. in my case the mouse keeps working as well for some seconds after the screen freezes, haven't figured the keyboard.
for now i just avoid entering fullscreen mode...
 
after updating the x11/nvidia-driver to version 440 the problem is gone. only i had to use

kld_list="nvidia-modeset"

in /etc/rc.conf instead of

nvidia_load="YES"
nvidia_name="nvidia"
nvidia_modeset_load="YES"
nvidia_modeset_name="nvidia-modeset"

in /boot/loader.conf. if i wouldn't, then X would hang at startup.
 
after updating the x11/nvidia-driver to version 440 the problem is gone. only i had to use

kld_list="nvidia-modeset"

in /etc/rc.conf instead of

nvidia_load="YES"
nvidia_name="nvidia"
nvidia_modeset_load="YES"
nvidia_modeset_name="nvidia-modeset"

in /boot/loader.conf. if i wouldn't, then X would hang at startup.

Thanks for letting me know. I'm not able to test this for a couple of weeks but hopefully the port updates will solve my issue as well.
 
Back
Top