Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
more than 832MB RAM => blank framebuffer screen
View unanswered posts
View posts from last 24 hours

Goto page 1, 2  Next  
Reply to topic    Gentoo Forums Forum Index Kernel & Hardware
View previous topic :: View next topic  
Author Message
The One
n00b
n00b


Joined: 24 Sep 2002
Posts: 20

PostPosted: Mon Feb 17, 2003 5:34 pm    Post subject: more than 832MB RAM => blank framebuffer screen Reply with quote

hi

today i've upgraded to 1GB RAM (from 512MB) and now my vesa framebuffer doesn't work anymore.
my system boots but the screen is blank, i know that the system is running, because i can type blindly and reboot the system.
after hours of googling i found no solution to this problem with the framebuffer, only one guy with the same problem, but he just disabled the framebuffer and used 640x480 textmode which i really don't wanna use.

the framebuffer works again when i choose the kernel parameter mem=832M

i've recompiled the kernel many times now, chose highmem support and so on, but nothing works.

does the vesa framebuffer not work with more than 832MB RAM??

thx,
the one

hardware:
msi sis745
gf 4 ti4200
1GB ram
and much other stuff
---
gentoo 2.4.19-r10-kernel
Back to top
View user's profile Send private message
Carlos
Guru
Guru


Joined: 07 Aug 2002
Posts: 458
Location: Providence, RI

PostPosted: Tue Feb 18, 2003 4:36 am    Post subject: Reply with quote

Maybe some of your RAM is bad; you might want to try out memtest86.
_________________
Man must shape his tools lest they shape him.
Back to top
View user's profile Send private message
The One
n00b
n00b


Joined: 24 Sep 2002
Posts: 20

PostPosted: Tue Feb 18, 2003 1:34 pm    Post subject: Reply with quote

well, i don't think that my RAM is bad, but i'll give it a try later...

thx,
the one
Back to top
View user's profile Send private message
acidreign
Tux's lil' helper
Tux's lil' helper


Joined: 21 Apr 2002
Posts: 122
Location: Brisbane, Australia

PostPosted: Tue Feb 18, 2003 5:19 pm    Post subject: Working here. Reply with quote

I am using a dual 2.0ghz AMD system (Tyan760 chipset).
1gb ddr ram.

no problems here with framebuffer. Does your video card take memory from your system ram ?
Back to top
View user's profile Send private message
The One
n00b
n00b


Joined: 24 Sep 2002
Posts: 20

PostPosted: Wed Feb 19, 2003 1:39 am    Post subject: Reply with quote

not that i know of, but i'm using 256MB agpgart size in bios, maybe that's too high...hmm, gonna test tomorrow

the one
Back to top
View user's profile Send private message
Turambar
n00b
n00b


Joined: 04 Oct 2002
Posts: 7
Location: Bay Area, CA

PostPosted: Thu Feb 20, 2003 8:32 am    Post subject: Need "High Memory Support" in kernel? Reply with quote

How was your kernel compiled?

I just dropped in another 512MB of RAM (for a new total of 1GB) and got the same problem. After running tests on the RAM, using each of the modules separately, etc., I determined two things:

1) The RAM is fine.
2) While fb works with either one, it yields the following dmesg error with both:

Code:
vesafb: abort, cannot ioremap video memory 0x8000000 @ 0xd8000000


I'm recompiling the kernel with the following (additional) options:

Code:
# CONFIG_NOHIGHMEM is not set
CONFIG_HIGHMEM4G=y
# CONFIG_HIGHMEM64G is not set
CONFIG_HIGHMEM=y
CONFIG_HIGHPTE=y
CONFIG_1GB=y
# CONFIG_2GB is not set
# CONFIG_3GB is not set
# CONFIG_05GB is not set
CONFIG_HIGHIO=y


I hope this solves the problem.

UPDATE: This did not solve the problem. In fact it compounded it by giving me a kernel panic in the blank framebuffer...
Back to top
View user's profile Send private message
deviljelly
Apprentice
Apprentice


Joined: 19 Feb 2003
Posts: 193
Location: Netherlands

PostPosted: Thu Feb 20, 2003 9:48 am    Post subject: same for me Reply with quote

I'm having the same problem, I dropped and extra 512 stick into my box an now have 1GB, since then none of my kernels seem to be able to support the frame buffer....

I have no solution yet, but I'll keep trying

Paul
Back to top
View user's profile Send private message
The One
n00b
n00b


Joined: 24 Sep 2002
Posts: 20

PostPosted: Thu Feb 20, 2003 7:16 pm    Post subject: Reply with quote

Carlos was right, one of my sticks was really bad though it was the newest one (only had it a day) of my 2x256 and 1x512. now i have new brand RAM (tested it and it is fine), but the problem persists :(
now i have 128MB agpgart size, so that can't be the prob either.
my framebuffer only works with the mem=832M (or lower) kernel option.
i also tried to boot my older kernel, a vanilla 2.4.19 with ac4 patch, but the same behaviour.
@Turambar
i compiled my kernel many times, with and without high memory support but it seems that it doesn't have any effect.

which kernels are you using?

the one
Back to top
View user's profile Send private message
deviljelly
Apprentice
Apprentice


Joined: 19 Feb 2003
Posts: 193
Location: Netherlands

PostPosted: Thu Feb 20, 2003 7:34 pm    Post subject: I think I have it working Reply with quote

I rebuilt the kernel and removed the radeon fb driver option leaving just the VESA one, I also changed the kernel back to 3GB which fixed mu missing 100KB.

CONFIG_NOHIGHMEM=y
# CONFIG_HIGHMEM4G is not set
# CONFIG_HIGHMEM64G is not set
# CONFIG_1GB is not set
# CONFIG_2GB is not set
CONFIG_3GB=y
# CONFIG_05GB is not set


#
# Frame-buffer support
#
CONFIG_FB=y
CONFIG_DUMMY_CONSOLE=y
# CONFIG_FB_LOGO_TUX is not set
# CONFIG_FB_LOGO_LARRY is not set
CONFIG_FB_LOGO_GENTOO=y
# CONFIG_FB_LOGO_POWEREDBY is not set
# CONFIG_FB_RIVA is not set
# CONFIG_FB_CLGEN is not set
# CONFIG_FB_PM2 is not set
# CONFIG_FB_PM3 is not set
# CONFIG_FB_CYBER2000 is not set
CONFIG_FB_VESA=y
# CONFIG_FB_VGA16 is not set
# CONFIG_FB_HGA is not set
CONFIG_VIDEO_SELECT=y
# CONFIG_FB_MATROX is not set
# CONFIG_FB_ATY is not set
CONFIG_FB_RADEON=m
# CONFIG_FB_ATY128 is not set
# CONFIG_FB_SIS is not set
# CONFIG_FB_NEOMAGIC is not set
# CONFIG_FB_3DFX is not set
# CONFIG_FB_VOODOO1 is not set
# CONFIG_FB_TRIDENT is not set
# CONFIG_FB_VIRTUAL is not set
# CONFIG_FBCON_ADVANCED is not set
CONFIG_FBCON_CFB8=y
CONFIG_FBCON_CFB16=y
CONFIG_FBCON_CFB24=y
CONFIG_FBCON_CFB32=y
# CONFIG_FBCON_FONTWIDTH8_ONLY is not set
# CONFIG_FBCON_FONTS is not set
CONFIG_FONT_8x8=y
CONFIG_FONT_8x16=y
Back to top
View user's profile Send private message
The One
n00b
n00b


Joined: 24 Sep 2002
Posts: 20

PostPosted: Thu Feb 20, 2003 10:34 pm    Post subject: Reply with quote

my config is exactly the same (i never had the rivafb for my GF4 compiled in, only vesa), what is a bit strange is that if i choose 1GB user address space size, then it is 3GB in the .config file, if i choose 3GB then CONFIG_1GB is enabled, well

but it doesn't work for me, still blank screen...

the one
Back to top
View user's profile Send private message
Turambar
n00b
n00b


Joined: 04 Oct 2002
Posts: 7
Location: Bay Area, CA

PostPosted: Fri Feb 21, 2003 1:35 am    Post subject: fb works, kernel panics Reply with quote

I've made some progress. I changed the User address space size from it's default of 1GB to 2GB, and got further along. If the kernel options (linux-2.4.20-gentoo-r1 w/bootsplash w/lm_sensors) are changed to this:

Code:
# CONFIG_NOHIGHMEM is not set
CONFIG_HIGHMEM4G=y
# CONFIG_HIGHMEM64G is not set
CONFIG_HIGHMEM=y
CONFIG_HIGHPTE=y
# CONFIG_1GB is not set
CONFIG_2GB=y
# CONFIG_3GB is not set
# CONFIG_05GB is not set
CONFIG_HIGHIO=y


Then the fb now works exactly as it did - the bootsplash image works, everything looks great! And then the kernel panics on the partition check section. :(

Code:
Partition check:
 /dev/ide/host0/bus0/target0/lun0: kernel BUG in sched.c:1131!
invalid operand:0000
...
Process swapper (pid: 0, stackpage=8047f000)
...
 <0>Kernel panic: Aiee, killing interrupt handler!
In interrupt handler - not syncing


I guess this is progress. =/ Time to Google for bugs in sched.c.
Back to top
View user's profile Send private message
The One
n00b
n00b


Joined: 24 Sep 2002
Posts: 20

PostPosted: Fri Feb 21, 2003 1:45 pm    Post subject: Reply with quote

hm, i'll try your options, you could deactivate the preemptible patch (in menu processor type and features).
if that doesn't work i'll try one of the latest ac-kernels (i guess it's 2.4.21-pre4-ac4)

the one
Back to top
View user's profile Send private message
Turambar
n00b
n00b


Joined: 04 Oct 2002
Posts: 7
Location: Bay Area, CA

PostPosted: Fri Feb 21, 2003 9:25 pm    Post subject: Preemptible kernel disable problems Reply with quote

I tried disabling the preemptible kernel option. It partially builds, but fails when linking in fs.o, due to a missing function. make (dist)clean doesn't help. I guess my next step along these lines is to make really clean by removing the directory, reemerging the kernel sources and patches, then testing how preemption affects the fb/RAM problem.

I've seen some other posts on the lkml about sched.c bugs wrt USB, though no one else seems to hit the same problem I've seen. Does everyone else here have USB enabled?

I rebuilt the kernel with debugging options enabled, so I'll try to find why I'm getting an oops before I completely start over.
Back to top
View user's profile Send private message
The One
n00b
n00b


Joined: 24 Sep 2002
Posts: 20

PostPosted: Sat Feb 22, 2003 1:34 pm    Post subject: Reply with quote

everything (your config!) is working for me now!! :D :D :D

here's my current kernel config:
Code:

# CONFIG_NOHIGHMEM is not set
CONFIG_HIGHMEM4G=y
# CONFIG_HIGHMEM64G is not set
CONFIG_HIGHMEM=y
# CONFIG_1GB is not set
CONFIG_2GB=y
# CONFIG_3GB is not set
# CONFIG_05GB is not set
# CONFIG_MATH_EMULATION is not set
CONFIG_MTRR=y
# CONFIG_SMP is not set
# CONFIG_PREEMPT is not set
# CONFIG_PREEMPT_TIMES is not set
CONFIG_X86_UP_APIC=y
CONFIG_X86_UP_IOAPIC=y
CONFIG_X86_LOCAL_APIC=y
CONFIG_X86_IO_APIC=y


this is with gentoo kernel 2.4.19-r10 as i mentioned in my first post. i didn't test whether it works with preempt=y but at the moment i don't care ;))
bootsplash also works!
output of free:
# free
total used free shared buffers cached
Mem: 1030076 250644 779432 0 10348 119660
-/+ buffers/cache: 120636 909440
Swap: 1124508 0 1124508

the best thing is that my WLAN card is working again now. since my memory upgrade cardmgr just didn't do anything anymore, now without changing the config it is also working again ;)

i hope you get it to work with your kernel!
(i have USB enabled, i'm using 4-6 USB devices)

thx!
the one
Back to top
View user's profile Send private message
deviljelly
Apprentice
Apprentice


Joined: 19 Feb 2003
Posts: 193
Location: Netherlands

PostPosted: Sat Feb 22, 2003 5:18 pm    Post subject: Reply with quote

cool, have fun, mine is working too but as I said the ATI dirvers screw it up.....I'm sure they'll fix them soon, they are in effect the FireGL drivers and they have just got a beating on tomshardware so I think ATI will put more focus on them.

Paul
Back to top
View user's profile Send private message
Turambar
n00b
n00b


Joined: 04 Oct 2002
Posts: 7
Location: Bay Area, CA

PostPosted: Thu Mar 13, 2003 5:26 am    Post subject: Finally working... Reply with quote

I hate it when the Real World imposes on my pet projects...

There was a problem with both the preemptible kernel and lm_sensors which caused kernel oops at different points. Although both of these (along with usb, IEEE 1394, bluetooth, fb, etc.) had no problems when high memory support was disabled, they do not behave well with it on. I'm using the gentoo-sources, so that increases the possible points of failure. If some kernel guru (with lots of free time) feels like troubleshooting this one, I'll be happy to send my .config file along. I think I'll wait for the next kernel revision, though.

The good side is that the framebuffer, bootsplash, and memory are all working perfectly. To celebrate, I include some gratuitous system stats:

Code:
# free -m
             total       used       free     shared    buffers     cached
Mem:          1005        309        696          0         45        158
-/+ buffers/cache:        105        900
Swap:         2000          0       2000

and
Code:
# df -B G 
Filesystem           1G-blocks      Used Available Use% Mounted on
rootfs                      36        31         3  92% /
/dev/root                   36        31         3  92% /
tmpfs                        1         1         1   2% /var/state/init.d
/dev/hdb6                  186         1       186   1% /mnt/burn


Note: Some OS force their users to go through gyrations (http://wdc.custhelp.com/cgi-bin/wdc.cfg/php/enduser/std_adp.php?p_faqid=533) to get large drives to be fully recognized, but Linux behaves well.
Back to top
View user's profile Send private message
phlashback
Tux's lil' helper
Tux's lil' helper


Joined: 27 Jul 2002
Posts: 142
Location: Normal

PostPosted: Wed Apr 09, 2003 6:51 pm    Post subject: Reply with quote

For anybody else who is having this problem....

I belive that this link will be of some help, it seems a bit more scientific
http://www.nvnews.net/vbulletin/showthread.php?s=&postid=100543#post100543


Good luck
Back to top
View user's profile Send private message
phlashback
Tux's lil' helper
Tux's lil' helper


Joined: 27 Jul 2002
Posts: 142
Location: Normal

PostPosted: Wed Apr 09, 2003 10:27 pm    Post subject: Reply with quote

Follow up....

for those that don't feel like clicking on a link here is what needs to be done

Change: /usr/src/{KERNEL}/drivers/video/vesafb.c

for 2.4.20 (and presumably earlier, dont have them to test and dont want to) :wink:
From:
Code:
video_size = screen_info.lfb_size * 65536;


To:
Code:
video_size = screen_info.lfb_width * screen_info.lfb_height * video_bpp;


for 2.5.67 / 2.5.67-mm1 (same not as to earlier)
Frrom:
Code:
vesafb_fix.smem_len = screen_info.lfb_size * 65536;


To:
Code:
esafb_fix.smem_len = screen_info.lfb_width * screen_info.lfb_height * vesafb_defined.bits_per_pixel;


I hope that this help somebody out thier.

:twisted: I Take no responsibility for the damage that may be caused to your system. I am using it right now, and I have had no problems yet. I am not a kernel hacker! :twisted:

Once again good luck
Back to top
View user's profile Send private message
bludger
Guru
Guru


Joined: 09 Apr 2003
Posts: 389

PostPosted: Fri Apr 11, 2003 5:52 am    Post subject: Reply with quote

I applied the patch that you quote from the kernel mailing list, but it made no difference at all. I have an Acer 513T laptop (Intel Celeron - family 6 model 6 stepping 10) with 186MB RAM. I currently have acpi enabled, but tried also with acpi=off.

Any ideas?
Back to top
View user's profile Send private message
jbygden
n00b
n00b


Joined: 09 Apr 2003
Posts: 4

PostPosted: Fri Apr 11, 2003 9:29 pm    Post subject: Reply with quote

phlashback wrote:
Follow up....

for those that don't feel like clicking on a link here is what needs to be done

Change: /usr/src/{KERNEL}/drivers/video/vesafb.c

for 2.4.20 (and presumably earlier, dont have them to test and dont want to) :wink:
From:
Code:
video_size = screen_info.lfb_size * 65536;


To:
Code:
video_size = screen_info.lfb_width * screen_info.lfb_height * video_bpp;



...snip...
Quote:

I hope that this help somebody out thier.


It sure did make a world of difference, not only your troubleshooting, but also that you made this little codechange avaliable here - saving me (and probably others, equally lazy bastards ;-) the pain of finding it out ourselves.

THANKS a lot!!

/jb
Back to top
View user's profile Send private message
phlashback
Tux's lil' helper
Tux's lil' helper


Joined: 27 Jul 2002
Posts: 142
Location: Normal

PostPosted: Sun Apr 20, 2003 12:15 am    Post subject: Reply with quote

bludger:
What problem are you having.... do you have any dmesg output.
I don't belive that this is related to your problem, as it seems to affect machines with large mounts of memory... 186MB is not large.

I am happy to "try" and help, but I need a little more info :)


jbygden:
You are welcome.
I was actually a lot of fun finding the solution, but much more rewarding to see it has actually helped someone :lol:

btw...
I have posted patches that you can download from:
https://bugs.gentoo.org/show_bug.cgi?id=19061

now no need to hand edit (I hope) :wink:

I store the patches in /usr/src/

assuming that your kernel sources are in /usr/src/linux
this is how i do it.

Code:
# cd /usr/src/linux
# patch --dry-run -p1 -i ../2.4-large_mem_vesa.patch
patch
patching file drivers/video/vesafb.c
#patch -p1 -i ../2.4-large_mem_vesa.patch
patch
patching file drivers/video/vesafb.c
#


the first patch command is optional, I just like checking first befor I change the kernel sources :)

if at all possible I am trying to get a list of videocards (manufacturers, and amount of memory included) which suffer from this.

Happy gentooing :P
Back to top
View user's profile Send private message
taskara
Advocate
Advocate


Joined: 10 Apr 2002
Posts: 3763
Location: Australia

PostPosted: Tue Apr 22, 2003 3:21 am    Post subject: Reply with quote

thank you, thank you, thank you, thank you, thank you!!

I have been struggling with this stupid problem for over a week.. posted on every framebuffer thread I could find! ;) finally Phlashback sent me here.. this sounds like EXACTLY the problem I'm having.. will try it as soooon as I am home!!

here's hoping!!

(and thanks - after lots of frustration, you've helped restore some faith in the people of the gentoo forums! :D)
_________________
Kororaa install method - have Gentoo up and running quickly and easily, fully automated with an installer!
Back to top
View user's profile Send private message
taskara
Advocate
Advocate


Joined: 10 Apr 2002
Posts: 3763
Location: Australia

PostPosted: Tue Apr 22, 2003 12:00 pm    Post subject: Reply with quote

well this did indeed get my fb working!

HOWEVER, now when I try and start X, I get the same blank screen fb was giving me before!

any thoughts?
_________________
Kororaa install method - have Gentoo up and running quickly and easily, fully automated with an installer!
Back to top
View user's profile Send private message
bludger
Guru
Guru


Joined: 09 Apr 2003
Posts: 389

PostPosted: Tue Apr 22, 2003 12:10 pm    Post subject: Reply with quote

phlashback wrote:
bludger:
What problem are you having.... do you have any dmesg output.
I don't belive that this is related to your problem, as it seems to affect machines with large mounts of memory... 186MB is not large.


I have the same symptoms as described - simply a system freeze on bootup. Unfortunately I can't get any dmesg output at this stage of the boot. If you can give me any hints as to how I might get more info, let me know. In the meantime, I can live without a boot splash screen on this laptop.
Back to top
View user's profile Send private message
phlashback
Tux's lil' helper
Tux's lil' helper


Joined: 27 Jul 2002
Posts: 142
Location: Normal

PostPosted: Wed Apr 23, 2003 12:48 am    Post subject: Reply with quote

Quote:
I have the same symptoms as described - simply a system freeze on bootup. Unfortunately I can't get any dmesg output at this stage of the boot. If you can give me any hints as to how I might get more info, let me know. In the meantime, I can live without a boot splash screen on this laptop.

:?

maby I missed something...... but the symptoms you are having dont seem to be the same.

"system freeze" (at least what i think you mean by that term) is a completly non functional system

it is possible i ws not clear, but in my case, and others (as far as I know) do not have a frozen system, rather just console display being dead (ie system boots totaly, and you can login all be it blindly :wink:, and if you have a configured xserver, x loads properly, and displays just as x always had.)

btw... are you using bootsplash... if so disable the bootsplash, and see...

I apologize if i sound a bit rash, just got home from my new job and spent all day in training for a "scripted" (ie it should work, but this is a pilot program... in other words.. my team and I are the debuggers) XP rollout *BLAH*... atleast after 2 years of looking a decent paycheck again :wink:

i recomend tryig to loginto the darkend box (if you had disk activity during boot) after normal startup time, and try to startx... if your system loads x then it wasnt frozen, and you could just open up an xterm shell and type demsg.... otherwise i have no clue... at this moment..... unfortunatly my trouble shooting time has been stomped on by new resposiblities :D

I hope you get your issue resloved.. keep us posted
Back to top
View user's profile Send private message
Display posts from previous:   
Reply to topic    Gentoo Forums Forum Index Kernel & Hardware All times are GMT
Goto page 1, 2  Next
Page 1 of 2

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum