Switching from X to console results in black console in 169.07

[Maxim Levitsky]

169.09 doesn't fix this.

[hussam]

Sorry for spamming but at least for the time being, can someone from Nvidia simply acknowledge that they are aware of the bug?

[nael]

Originally Posted by hussam

Sorry for spamming but at least for the time being, can someone from Nvidia simply acknowledge that they are aware of the bug?

It's really annoying the absence of an answer with this bug

[lloeki]

I'd l'ike to report I'm hitting this bug too (or a variation of if).

- usually, VT switching works correctly.
- if I suspend and resume, I am 99% sure to have blank consoles, where I can blindly type stuff.
- if I init 3 and hibernate/thaw then consoles are back, even after init 5.
- I'm unsure what would happen if I let X run while hibernating, or if I hibernate again.

that's on a asus w7j with a go7400

[lguerb]

I do not know if nvidia developers are aware of this, even if it has discussed a hundred of times already. I have similar problems (GeForce Go 6100)

1) 100.14.19 works fine without framebuffer. If I have a framebuffer, then switching to vty's will result in some corrupted lines and such. Still readable but definitely not good.

2) 169.X work fine in X also, but switching to vty's results in totally corrupted fonts, moving like they where floating in water.. and this happens either with framebuffer or without it.

Also, this is not a Linux issue, since I have the same results in FreeBSD. It's definitely something in the driver that they think they're doing ok, but it's just not ok.

An answer from a developer, at least telling me what can I do for help, or what else information I can give for the bug, would be desirable at this point, since this is becoming annoying.

[lloeki]

I'll have to experiment more, but in the process to accelerate hibernation with vanilla kernel, I tried uswsusp.

along with s2disk, with it you also have s2ram (for obviously suspending to ram). out of curiosity, I tried it. it seems that its switch to console prior to suspending and back again makes console VTs not blank upon resume.
I'm using this right now (both s2disk and s2ram) to gether with opensuse-patched pm-utils (to support uswsusp instead of vanilla), and will report if I hit console blanking again.

still, if it happens to work reliably, this is only a workaround.

[lguerb]

I have some small news. Using the new uvesafb from the kernel 2.6.24, 100.14.19 works perfect with it, I can switch to console and see no corruption at all. This is good. (If it helps the developers, 100.14.19 worked good on the FreeBSD framebuffer too).

Nevertheless, 169.X drivers still corrupt everything. This is bad.

[ohcibi]

i'm experiencing the same problem without doing anything special.

i just boot my computer and once X is started, the ttys are not available anymore, even if i shut down X. ive talked with some people i know and they do not have this problem, but they have different nvidia-chipsets. so maybe it would help gathering the chipset on which the problem occurs...

i'm using geforce 6200 here.

[lloeki]

bang. it's back. blackness after a suspend. (note that I'm currently using uvesafb, but it occured nonetheless with vesafb and with plain vga)

[Maxim Levitsky]

This bug is finally fixed in 169.12


Note that some reporters here were describing other bugs,
I was telling about a blank VT after I do a suspend to ram, and then restart X. otherwise the VTs did work here.

[codermotor]

169.12 does not fix the problem for me. I reported to Nvidia with no response.

My symptoms are:
o After starting X (via startx), either switching to a console, or upon
exiting X, the console is black.
o The system is still responsive (i.e. the CLI works).
o Switching back to X (Alt+F7) displays GUI (desktop)
o Resolution: reboot system.
o Prevention: use NV driver

This is with an up-to-date Fedora 7, on an Asus Z71V, with GeForce Go 6600. This problem has existed for quite a few versions of the Nvidia driver.

[Maxim Levitsky]

Originally Posted by Maxim Levitsky

This bug is finally fixed in 169.12


Note that some reporters here were describing other bugs,
I was telling about a blank VT after I do a suspend to ram, and then restart X. otherwise the VTs did work here.

Yes, on your system console doesn't work, but this isn't this bug.
On my system consoles did work always. I can switch to them after I start x, and if I stop X
I can switch to them even if I suspend to ram/disk

The only situation when they didn't work, and do work now, was when

I suspend system to ram, then restart X, and then consoles didn't work