Is 10.3 i386 jinxed ?

[Manish Jain]

Hi,
I have a 5 year old Intel x86 Pentium-based laptop (Gateway NE56R). I
just finished installing FreeBSD 10.3 (i386) with Gnome on it. I was
expecting everything to go smoothly but the Gnome desktop is proving to
be a disaster.

This also was my first impression when installing Gnome 2 ("classic
Gnome desktop") on a Dell D630 laptop running FreeBSD i386 10.2 (started
with 10.0). There was some work to do, things to update and to install,
scripts to abuse, crazy things to do... but after that, everything
worked as intended. Sadly I didn't take notes on what I did, but I can
assure you that most of it was not obvious, logical, or acceptable. :-/
How did you install Gnome? Ports or packages? Which version?

Most of the desktop is trash, half the application icons
are missing and there are no panels.

You probably are missing some important dependency packages.

Even keyboard input typed into
gnome-terminal is not displayed correctly (some of the characters in
the
echo go missing).

That _really_ sounds strange.

All necessary entries (hald_enable, polkitd_enable,
dbus_enable, gnome_enable) are set to YES in rc.conf. I wonder what is
wrong ?

Using Gnome is wrong. ;-)

Is there any way I can get FreeBSD and Gnome working on the laptop ?

Reconsider using Gnome. Check if Lxde or Xfce will work better for you.

I reinstalled the whole system using xfce as the desktop. While things
have improved, the system is still not workable.

1) The installer 'forgot' to place the boot code on the MBR. I had to
use live CD + boot0cfg to fix that.
2) The first time X started, xfce4-panel crashed
3) The second time, xfce-terminal crashed
4) The third time, trying to launch firefox rebooted the system.

Overall, 10.3 i386 seems to have gremlins in the belly. What do I do ?
Try some linux distro ? This is exactly what I have been trying to avoid.

BTW, does anyone have any idea where can I get Windows XP drivers for
the NE56R laptop ? If nothing else works, XP remains my fall-back option.

Regards,
Manish Jain

[Polytropon]

Post by Manish Jain
I reinstalled the whole system using xfce as the desktop. While things
have improved, the system is still not workable.
1) The installer 'forgot' to place the boot code on the MBR. I had to
use live CD + boot0cfg to fix that.

Use MBR only if needed. Today's preference is to use GPT for
partitioning. Compare:

http://www.wonkity.com/~wblock/docs/html/disksetup.html

Use MBR only if you have a valid reason as it probably has
reached the state of "not supported anymore".

Post by Manish Jain
2) The first time X started, xfce4-panel crashed
3) The second time, xfce-terminal crashed
4) The third time, trying to launch firefox rebooted the system.

This (and especially the last problem) leads to the impression
that there is a _massive_ problem, it's probably not related to
the FreeBSD OS, and surely not to installed applications. Are
you sure there isn't a hardware problem lurking in the background?
Like... starting a web browser reboots the system... this simply
sounds totally wrong.

Post by Manish Jain
Overall, 10.3 i386 seems to have gremlins in the belly. What do I do ?

I have a 10.2 installation on an older Dell laptop, running Gnome 2,
no problems so far (but I haven't tried updating it yet, as I'm a
big fan of "never touch a running system", a symptom of lazyness
and "fat fingers"). :-)

Post by Manish Jain
Try some linux distro ? This is exactly what I have been trying to avoid.

Check out PC-BSD or try the Lxde desktop. Maybe you can also
settle for a less convoluted desktop, i.e. a window manager
that you like.

Post by Manish Jain
BTW, does anyone have any idea where can I get Windows XP drivers for
the NE56R laptop ?

From the manufacturer of course. :-)

Post by Manish Jain
If nothing else works, XP remains my fall-back option.

For offline use that's probably more or less okay. :-)

[Valeri Galtsev]

Post by Manish Jain

Hi,
I have a 5 year old Intel x86 Pentium-based laptop (Gateway NE56R). I
just finished installing FreeBSD 10.3 (i386) with Gnome on it. I was
expecting everything to go smoothly but the Gnome desktop is proving
to
be a disaster.

This also was my first impression when installing Gnome 2 ("classic
Gnome desktop") on a Dell D630 laptop running FreeBSD i386 10.2
(started
with 10.0). There was some work to do, things to update and to install,
scripts to abuse, crazy things to do... but after that, everything
worked as intended. Sadly I didn't take notes on what I did, but I can
assure you that most of it was not obvious, logical, or acceptable. :-/
How did you install Gnome? Ports or packages? Which version?

Most of the desktop is trash, half the application icons
are missing and there are no panels.

You probably are missing some important dependency packages.

Even keyboard input typed into
gnome-terminal is not displayed correctly (some of the characters in
the
echo go missing).

That _really_ sounds strange.

All necessary entries (hald_enable, polkitd_enable,
dbus_enable, gnome_enable) are set to YES in rc.conf. I wonder what is
wrong ?

Using Gnome is wrong. ;-)

Is there any way I can get FreeBSD and Gnome working on the laptop ?

Reconsider using Gnome. Check if Lxde or Xfce will work better for you.

I reinstalled the whole system using xfce as the desktop. While things
have improved, the system is still not workable.
1) The installer 'forgot' to place the boot code on the MBR. I had to
use live CD + boot0cfg to fix that.
2) The first time X started, xfce4-panel crashed
3) The second time, xfce-terminal crashed
4) The third time, trying to launch firefox rebooted the system.
Overall, 10.3 i386 seems to have gremlins in the belly. What do I do ?
Try some linux distro ? This is exactly what I have been trying to avoid.
BTW, does anyone have any idea where can I get Windows XP drivers for
the NE56R laptop ? If nothing else works, XP remains my fall-back option.

Hearing all that makes me think that this laptop likely has hardware
problems. I would definitely get some live system CD and check drive smart
status. I also would definitely run memtest86 (leaving it running for 24
hours at lest). memtest86 is not definitive for clearing suspicion about
RAM, RAM errors sometimes manifest themselves only under heavy load
whereas load is almost zero when memtest86 is run. I would check if the
laptop needs cleaning inside. If laptop body is easily flexed, and laptop
was carried around a lot, I also would suspect micro cracks in system
board ("motherboard" is common jargon for it for last couple of decades
;-) One thing I forgot to mention: poor condition of optical drive can
lead to poorly installed system, I guess.

Just my $0.02

Valeri

Post by Manish Jain
Regards,
Manish Jain
_______________________________________________
https://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to

++++++++++++++++++++++++++++++++++++++++
Valeri Galtsev
Sr System Administrator
Department of Astronomy and Astrophysics
Kavli Institute for Cosmological Physics
University of Chicago
Phone: [phone]
++++++++++++++++++++++++++++++++++++++++

[Warren Block]

I reinstalled the whole system using xfce as the desktop. While things have
improved, the system is still not workable.
1) The installer 'forgot' to place the boot code on the MBR. I had to use
live CD + boot0cfg to fix that.

boot0cfg is (mostly) for writing the dual-boot boot0 code. Use gpart
for ordinary booting.

2) The first time X started, xfce4-panel crashed
3) The second time, xfce-terminal crashed
4) The third time, trying to launch firefox rebooted the system.

This sounds unlike my experience with xfce.

Some of the Gateway NE56R models included Nvidia GPUs. Some special
thing might be needed to properly support that GPU.

I have an i386 image with xfce that is bootable on most systems, but I
have not tested it with Nvidia due to a mysterious lack of them among my
computers. It works fine on Intel and AMD video, though.

The RAM or power supply could be a problem with the Gateway.

[Polytropon]

Post by Warren Block

Post by Manish Jain
2) The first time X started, xfce4-panel crashed
3) The second time, xfce-terminal crashed
4) The third time, trying to launch firefox rebooted the system.

This sounds unlike my experience with xfce.
Some of the Gateway NE56R models included Nvidia GPUs. Some special
thing might be needed to properly support that GPU.

This might be at least a part of the occuring problems. I have tried
the "nv" and "nouveau" drivers with my nVidia GPU, but the best results
are from the (proprietary) "nvidia" driver which consists of a kernel
module and the X component. Maybe it helps checking the actual hardware
(GPU model number) and verify which driver will support it. However,
using the VESA driver should be sufficient for typical desktop use
(no "big 3D stuff").

Checking the other possible hardware-related problems should still be
on the to-do list. :-)

[Manish Jain]

Post by Polytropon

Post by Manish Jain
I reinstalled the whole system using xfce as the desktop. While things
have improved, the system is still not workable.
1) The installer 'forgot' to place the boot code on the MBR. I had to
use live CD + boot0cfg to fix that.

Use MBR only if needed. Today's preference is to use GPT for
http://www.wonkity.com/~wblock/docs/html/disksetup.html
Use MBR only if you have a valid reason as it probably has
reached the state of "not supported anymore".

Post by Manish Jain
2) The first time X started, xfce4-panel crashed
3) The second time, xfce-terminal crashed
4) The third time, trying to launch firefox rebooted the system.

This (and especially the last problem) leads to the impression
that there is a _massive_ problem, it's probably not related to
the FreeBSD OS, and surely not to installed applications. Are
you sure there isn't a hardware problem lurking in the background?
Like... starting a web browser reboots the system... this simply
sounds totally wrong.

Post by Manish Jain
Overall, 10.3 i386 seems to have gremlins in the belly. What do I do ?

I have a 10.2 installation on an older Dell laptop, running Gnome 2,
no problems so far (but I haven't tried updating it yet, as I'm a
big fan of "never touch a running system", a symptom of lazyness
and "fat fingers"). :-)

Post by Manish Jain
Try some linux distro ? This is exactly what I have been trying to avoid.

Check out PC-BSD or try the Lxde desktop. Maybe you can also
settle for a less convoluted desktop, i.e. a window manager
that you like.

Post by Manish Jain
BTW, does anyone have any idea where can I get Windows XP drivers for
the NE56R laptop ?
From the manufacturer of course. :-)
If nothing else works, XP remains my fall-back option.

For offline use that's probably more or less okay. :-)

I strongly suspect some hardware problem, quite possible RAM : there is
a page fault at boot time and the report is in /var/crash. I am running
memtest86+ for 24 hours, and will report results tomorrow. If memtest86+
detects no errors, I will paste the contents of the crash report for
some expert to analyse.

Regards
Manish Jain

[Manish Jain]

Post by Manish Jain

Post by Polytropon

Post by Manish Jain
I reinstalled the whole system using xfce as the desktop. While things
have improved, the system is still not workable.
1) The installer 'forgot' to place the boot code on the MBR. I had to
use live CD + boot0cfg to fix that.

Use MBR only if needed. Today's preference is to use GPT for
http://www.wonkity.com/~wblock/docs/html/disksetup.html
Use MBR only if you have a valid reason as it probably has
reached the state of "not supported anymore".

Post by Manish Jain
2) The first time X started, xfce4-panel crashed
3) The second time, xfce-terminal crashed
4) The third time, trying to launch firefox rebooted the system.

This (and especially the last problem) leads to the impression
that there is a _massive_ problem, it's probably not related to
the FreeBSD OS, and surely not to installed applications. Are
you sure there isn't a hardware problem lurking in the background?
Like... starting a web browser reboots the system... this simply
sounds totally wrong.

Post by Manish Jain
Overall, 10.3 i386 seems to have gremlins in the belly. What do I do ?

I have a 10.2 installation on an older Dell laptop, running Gnome 2,
no problems so far (but I haven't tried updating it yet, as I'm a
big fan of "never touch a running system", a symptom of lazyness
and "fat fingers"). :-)

Post by Manish Jain
Try some linux distro ? This is exactly what I have been trying to
avoid.

Check out PC-BSD or try the Lxde desktop. Maybe you can also
settle for a less convoluted desktop, i.e. a window manager
that you like.

Post by Manish Jain
BTW, does anyone have any idea where can I get Windows XP drivers for
the NE56R laptop ?
From the manufacturer of course. :-)
If nothing else works, XP remains my fall-back option.

For offline use that's probably more or less okay. :-)

I strongly suspect some hardware problem, quite possible RAM : there
is a page fault at boot time and the report is in /var/crash. I am
running memtest86+ for 24 hours, and will report results tomorrow. If
memtest86+ detects no errors, I will paste the contents of the crash
report for some expert to analyse.
Regards
Manish Jain

I ran memtest86+ for about 30 hours, doing 65 passes. No errors
detected. So I rebooted, but I continue to get page fault and launching
xfce-terminal rebooted the system. I am pasting the top few lines of the
crash report here. If anybody would like the complete report (something
like 2500 lines), please email me separately. If anybody would further
like to have the core file itself (something like 120 MB), please let me
know.

// core.txt <<

Fatal trap 12: page fault while in kernel mode
cpuid = 0; apic id = 00
fault virtual address = 0x34
fault code = supervisor write, page not present
instruction pointer = 0x20:0xc0dffdd6
stack pointer = 0x28:0xf0598720
frame pointer = 0x28:0xf0598720
code segment = base 0x0, limit 0xfffff, type 0x1b
= DPL 0, pres 1, def32 1, gran 1
processor eflags = interrupt enabled, resume, IOPL = 0
current process = 1033 (xfce4-terminal)
trap number = 12
panic: page fault
cpuid = 0
KDB: stack backtrace:
#0 0xc0b7be92 at kdb_backtrace+0x52
#1 0xc0b3c1db at vpanic+0x11b
#2 0xc0b3c0bb at panic+0x1b
#3 0xc10650cb at trap_fatal+0x30b
#4 0xc1065435 at trap_pfault+0x355
#5 0xc1064b04 at trap+0x674
#6 0xc104f96a at calltrap+0x6
#7 0xc105af1f at pmap_extract_and_hold+0x17f
#8 0xc0dec302 at vm_fault_quick_hold_pages+0xb2
#9 0xc0bfae75 at vn_io_fault1+0x285
#10 0xc0bfaaf3 at vn_rdwr+0x273
#11 0xc0bfb066 at vn_rdwr_inchunks+0xb6
#12 0xc0ad904d at elf32_coredump+0xabd
#13 0xc0b3fe92 at sigexit+0xbc2
#14 0xc0b4064e at postsig+0x2de
#15 0xc0b8fc28 at ast+0x548
#16 0xc10505b5 at doreti_ast+0x1b
Uptime: 1m20s
Physical memory: 1854 MB
Dumping 115 MB: 100 84 68 52 36 20 4

// core.txt >>

Any help in getting my laptop to work properly will be greatly
appreciated. Thank you.

Regards
Manish Jain

[Manish Jain]

Post by Manish Jain

Post by Manish Jain

Post by Polytropon

Post by Manish Jain
I reinstalled the whole system using xfce as the desktop. While things
have improved, the system is still not workable.
1) The installer 'forgot' to place the boot code on the MBR. I had to
use live CD + boot0cfg to fix that.

Use MBR only if needed. Today's preference is to use GPT for
http://www.wonkity.com/~wblock/docs/html/disksetup.html
Use MBR only if you have a valid reason as it probably has
reached the state of "not supported anymore".

Post by Manish Jain
2) The first time X started, xfce4-panel crashed
3) The second time, xfce-terminal crashed
4) The third time, trying to launch firefox rebooted the system.

This (and especially the last problem) leads to the impression
that there is a _massive_ problem, it's probably not related to
the FreeBSD OS, and surely not to installed applications. Are
you sure there isn't a hardware problem lurking in the background?
Like... starting a web browser reboots the system... this simply
sounds totally wrong.

Post by Manish Jain
Overall, 10.3 i386 seems to have gremlins in the belly. What do I do ?

I have a 10.2 installation on an older Dell laptop, running Gnome 2,
no problems so far (but I haven't tried updating it yet, as I'm a
big fan of "never touch a running system", a symptom of lazyness
and "fat fingers"). :-)

Post by Manish Jain
Try some linux distro ? This is exactly what I have been trying to
avoid.

Check out PC-BSD or try the Lxde desktop. Maybe you can also
settle for a less convoluted desktop, i.e. a window manager
that you like.

Post by Manish Jain
BTW, does anyone have any idea where can I get Windows XP drivers for
the NE56R laptop ?
From the manufacturer of course. :-)
If nothing else works, XP remains my fall-back option.

For offline use that's probably more or less okay. :-)

I strongly suspect some hardware problem, quite possible RAM : there
is a page fault at boot time and the report is in /var/crash. I am
running memtest86+ for 24 hours, and will report results tomorrow. If
memtest86+ detects no errors, I will paste the contents of the crash
report for some expert to analyse.
Regards
Manish Jain

I ran memtest86+ for about 30 hours, doing 65 passes. No errors
detected. So I rebooted, but I continue to get page fault and
launching xfce-terminal rebooted the system. I am pasting the top few
lines of the crash report here. If anybody would like the complete
report (something like 2500 lines), please email me separately. If
anybody would further like to have the core file itself (something
like 120 MB), please let me know.
// core.txt <<
Fatal trap 12: page fault while in kernel mode
cpuid = 0; apic id = 00
fault virtual address = 0x34
fault code = supervisor write, page not present
instruction pointer = 0x20:0xc0dffdd6
stack pointer = 0x28:0xf0598720
frame pointer = 0x28:0xf0598720
code segment = base 0x0, limit 0xfffff, type 0x1b
= DPL 0, pres 1, def32 1, gran 1
processor eflags = interrupt enabled, resume, IOPL = 0
current process = 1033 (xfce4-terminal)
trap number = 12
panic: page fault
cpuid = 0
#0 0xc0b7be92 at kdb_backtrace+0x52
#1 0xc0b3c1db at vpanic+0x11b
#2 0xc0b3c0bb at panic+0x1b
#3 0xc10650cb at trap_fatal+0x30b
#4 0xc1065435 at trap_pfault+0x355
#5 0xc1064b04 at trap+0x674
#6 0xc104f96a at calltrap+0x6
#7 0xc105af1f at pmap_extract_and_hold+0x17f
#8 0xc0dec302 at vm_fault_quick_hold_pages+0xb2
#9 0xc0bfae75 at vn_io_fault1+0x285
#10 0xc0bfaaf3 at vn_rdwr+0x273
#11 0xc0bfb066 at vn_rdwr_inchunks+0xb6
#12 0xc0ad904d at elf32_coredump+0xabd
#13 0xc0b3fe92 at sigexit+0xbc2
#14 0xc0b4064e at postsig+0x2de
#15 0xc0b8fc28 at ast+0x548
#16 0xc10505b5 at doreti_ast+0x1b
Uptime: 1m20s
Physical memory: 1854 MB
Dumping 115 MB: 100 84 68 52 36 20 4
// core.txt >>
Any help in getting my laptop to work properly will be greatly
appreciated. Thank you.
Regards
Manish Jain

I found what is the problem, with the help of Ultimate Boot CD. UBCD
says the video memory is corrupt. Does that mean I have to trash the
entire laptop, or is there some way to replace the video memory ?

Thanks for any help.


Regards
Manish Jain

[Polytropon]

Post by Manish Jain
I found what is the problem, with the help of Ultimate Boot CD. UBCD
says the video memory is corrupt.

Ha! As assumed, it looked too much like hardware error...

Post by Manish Jain
Does that mean I have to trash the
entire laptop, or is there some way to replace the video memory ?

Depends.

There are laptops where the GPU memory is allocated from the
installed memory modules that serve as RAM. Other models have
dedicated chips for the GPU which are independent. In such a
case, you probably cannot replace them. But if it's the first
case mentioned, exchange the RAM modules. The memtest results
usually are "false-negative" because the GPU memory is already
"cut off" from the cells available as regular RAM, so those do
not get tested.

[Manish Jain]

Post by Polytropon

Post by Manish Jain
I found what is the problem, with the help of Ultimate Boot CD. UBCD
says the video memory is corrupt.

Ha! As assumed, it looked too much like hardware error...

Post by Manish Jain
Does that mean I have to trash the
entire laptop, or is there some way to replace the video memory ?

Depends.
There are laptops where the GPU memory is allocated from the
installed memory modules that serve as RAM. Other models have
dedicated chips for the GPU which are independent. In such a
case, you probably cannot replace them. But if it's the first
case mentioned, exchange the RAM modules. The memtest results
usually are "false-negative" because the GPU memory is already
"cut off" from the cells available as regular RAM, so those do
not get tested.

In case the video RAM is part of a dedicated chip, then should it not be
possible to replace the chipset ? With PC's, this is possible, I know.
But laptops, I have no idea.

Regards
Manish Jain

[Polytropon]

Post by Manish Jain

Post by Polytropon

Post by Manish Jain
I found what is the problem, with the help of Ultimate Boot CD. UBCD
says the video memory is corrupt.

Ha! As assumed, it looked too much like hardware error...

Post by Manish Jain
Does that mean I have to trash the
entire laptop, or is there some way to replace the video memory ?

Depends.
There are laptops where the GPU memory is allocated from the
installed memory modules that serve as RAM. Other models have
dedicated chips for the GPU which are independent. In such a
case, you probably cannot replace them. But if it's the first
case mentioned, exchange the RAM modules. The memtest results
usually are "false-negative" because the GPU memory is already
"cut off" from the cells available as regular RAM, so those do
not get tested.

In case the video RAM is part of a dedicated chip, then should it not be
possible to replace the chipset ? With PC's, this is possible, I know.
But laptops, I have no idea.

I have not yet seen a laptop where the GPU was "a module" that could
be replaced easily. It's usually glued+soldered onto the mainboard,
as it is not intended to be replaced. That kind of memory also doesn't
come in sockets, so it's probably impossible to replace. The common
solution is to replace the mainboard, which implies to replace the
whole laptop. :-)

Still it might be possible that such "modular laptop mainboards" do
exist. It's just that _I_ haven't seen one...

[Warren Block]

Post by Manish Jain

Post by Polytropon

Post by Manish Jain
I found what is the problem, with the help of Ultimate Boot CD. UBCD
says the video memory is corrupt.

Ha! As assumed, it looked too much like hardware error...

Post by Manish Jain
Does that mean I have to trash the
entire laptop, or is there some way to replace the video memory ?

Depends.
There are laptops where the GPU memory is allocated from the
installed memory modules that serve as RAM. Other models have
dedicated chips for the GPU which are independent. In such a
case, you probably cannot replace them. But if it's the first
case mentioned, exchange the RAM modules. The memtest results
usually are "false-negative" because the GPU memory is already
"cut off" from the cells available as regular RAM, so those do
not get tested.

In case the video RAM is part of a dedicated chip, then should it not be
possible to replace the chipset ? With PC's, this is possible, I know. But
laptops, I have no idea.

Possible, yes, but in a notebook it will be soldered to the motherboard,
so replacing it would require hot-air rework equipment and experience.
Replacing the motherboard is simpler, although sometimes not simple,
even if you have done it before.

Neither method is likely to be practical. The term is "BER": beyond
economic repair. Replacement of the system will be cheaper.

[Manish Jain]

Post by Warren Block

Post by Manish Jain

Post by Polytropon

Post by Manish Jain
I found what is the problem, with the help of Ultimate Boot CD. UBCD
says the video memory is corrupt.

Ha! As assumed, it looked too much like hardware error...

Post by Manish Jain
Does that mean I have to trash the
entire laptop, or is there some way to replace the video memory ?

Depends.
There are laptops where the GPU memory is allocated from the
installed memory modules that serve as RAM. Other models have
dedicated chips for the GPU which are independent. In such a
case, you probably cannot replace them. But if it's the first
case mentioned, exchange the RAM modules. The memtest results
usually are "false-negative" because the GPU memory is already
"cut off" from the cells available as regular RAM, so those do
not get tested.

In case the video RAM is part of a dedicated chip, then should it not
be possible to replace the chipset ? With PC's, this is possible, I
know. But laptops, I have no idea.

Possible, yes, but in a notebook it will be soldered to the
motherboard, so replacing it would require hot-air rework equipment
and experience. Replacing the motherboard is simpler, although
sometimes not simple, even if you have done it before.
Neither method is likely to be practical. The term is "BER": beyond
economic repair. Replacement of the system will be cheaper.

Is there any way I can find out whether the video RAM is part of the
main memory or is it wired into a chip ?

Regards
Manish Jain

[Polytropon]

Post by Manish Jain
Is there any way I can find out whether the video RAM is part of the
main memory or is it wired into a chip ?

Check the documentation (technical handbook) of the laptop.
In worst case, visual inspection of the mainboard might help.

[Solène Rapenne]

Post by Polytropon

Post by Manish Jain

Post by Polytropon

Post by Manish Jain
I found what is the problem, with the help of Ultimate Boot CD. UBCD
says the video memory is corrupt.

Ha! As assumed, it looked too much like hardware error...

Post by Manish Jain
Does that mean I have to trash the
entire laptop, or is there some way to replace the video memory ?

Depends.
There are laptops where the GPU memory is allocated from the
installed memory modules that serve as RAM. Other models have
dedicated chips for the GPU which are independent. In such a
case, you probably cannot replace them. But if it's the first
case mentioned, exchange the RAM modules. The memtest results
usually are "false-negative" because the GPU memory is already
"cut off" from the cells available as regular RAM, so those do
not get tested.

In case the video RAM is part of a dedicated chip, then should it not
be
possible to replace the chipset ? With PC's, this is possible, I know.
But laptops, I have no idea.

I have not yet seen a laptop where the GPU was "a module" that could
be replaced easily. It's usually glued+soldered onto the mainboard,
as it is not intended to be replaced. That kind of memory also doesn't
come in sockets, so it's probably impossible to replace. The common
solution is to replace the mainboard, which implies to replace the
whole laptop. :-)
Still it might be possible that such "modular laptop mainboards" do
exist. It's just that _I_ haven't seen one...

Some laptop have dedicated cards, they call it MXM :
https://en.wikipedia.org/wiki/Mobile_PCI_Express_Module

I have one laptop with a "discrete" nvidia MXM card (optimus technology)
that I can remove or change. But usually, you find those cards on laptop
with good video performance, sold as gamer laptop.

[Warren Block]

Is there any way I can find out whether the video RAM is part of the main
memory or is it wired into a chip ?

If it has an Intel GPU only, then video RAM is from system RAM. If it
has an additional Nvidia or ATI/AMD GPU... it still probably uses the
main RAM.

If it has an additional GPU that is truly discrete, that could have its
own RAM.

So "does it have a second GPU?" is the first question. Posting the
exact model number might help answer that.

[Manish Jain]

Post by Manish Jain
Is there any way I can find out whether the video RAM is part of the
main memory or is it wired into a chip ?

If it has an Intel GPU only, then video RAM is from system RAM. If it
has an additional Nvidia or ATI/AMD GPU... it still probably uses the
main RAM.
If it has an additional GPU that is truly discrete, that could have
its own RAM.
So "does it have a second GPU?" is the first question. Posting the
exact model number might help answer that.

The laptop is a Gateway NE56R. Looks to like it has dedicated chip for
video RAM. You can look up this URL :
https://panam.gateway.com/s/notebook/2012/Gateway/NE/NE56R/NE56Rsp2.shtml

Thanks for all the help.

Regards
Manish Jain

[Warren Block]

Is there any way I can find out whether the video RAM is part of the main
memory or is it wired into a chip ?

If it has an Intel GPU only, then video RAM is from system RAM. If it has
an additional Nvidia or ATI/AMD GPU... it still probably uses the main RAM.
If it has an additional GPU that is truly discrete, that could have its own
RAM.
So "does it have a second GPU?" is the first question. Posting the exact
model number might help answer that.

The laptop is a Gateway NE56R. Looks to like it has dedicated chip for video
https://panam.gateway.com/s/notebook/2012/Gateway/NE/NE56R/NE56Rsp2.shtml

But that link shows how many variations there are of that one model.
The exact model number should be printed on the system itself, probably
on the bottom or in the battery compartment. Like a lot of other
manufacturers, Gateway used to use an additional code after the main
model number to identify the specific version. A quick search shows
there are NE56R10u, NE56R20u, NE56R27u, and so on.

[Manish Jain]

Post by Warren Block

Post by Manish Jain

Post by Manish Jain
Is there any way I can find out whether the video RAM is part of
the main memory or is it wired into a chip ?

If it has an Intel GPU only, then video RAM is from system RAM. If
it has an additional Nvidia or ATI/AMD GPU... it still probably uses
the main RAM.
If it has an additional GPU that is truly discrete, that could have
its own RAM.
So "does it have a second GPU?" is the first question. Posting the
exact model number might help answer that.

The laptop is a Gateway NE56R. Looks to like it has dedicated chip
https://panam.gateway.com/s/notebook/2012/Gateway/NE/NE56R/NE56Rsp2.shtml

But that link shows how many variations there are of that one model.
The exact model number should be printed on the system itself,
probably on the bottom or in the battery compartment. Like a lot of
other manufacturers, Gateway used to use an additional code after the
main model number to identify the specific version. A quick search
shows there are NE56R10u, NE56R20u, NE56R27u, and so on.

What is printed underneath is NE56R14l-B9502G32Mnks

I can't be sure whether the l is actually l or I or 1

Regards
Manish Jain

[Warren Block]

Post by Manish Jain
What is printed underneath is NE56R14l-B9502G32Mnks
I can't be sure whether the l is actually l or I or 1

Looks like this one:
http://in.gateway.com/gw/en/IN/content/model/NX.Y1USI.005

Which does not have a separate GPU, and means that the bad memory would
be in one of the removable DIMMs.

[Manish Jain]

Post by Warren Block

Post by Manish Jain
What is printed underneath is NE56R14l-B9502G32Mnks
I can't be sure whether the l is actually l or I or 1

http://in.gateway.com/gw/en/IN/content/model/NX.Y1USI.005
Which does not have a separate GPU, and means that the bad memory
would be in one of the removable DIMMs.

Something to cheer about, at last. But then why does not memtest86+ pick
up any errors in 65 passes ?

Regards
Manish Jain

[Polytropon]

Post by Manish Jain

Post by Warren Block

Post by Manish Jain
What is printed underneath is NE56R14l-B9502G32Mnks
I can't be sure whether the l is actually l or I or 1

http://in.gateway.com/gw/en/IN/content/model/NX.Y1USI.005
Which does not have a separate GPU, and means that the bad memory
would be in one of the removable DIMMs.

Something to cheer about, at last. But then why does not memtest86+ pick
up any errors in 65 passes ?

As I mentioned earlier, this is a "false-negative" result. The portions
of the RAM dedicated for the GPU are "cut away" quite eary, so memtest
can only test what's left for "normal RAM". And those portions don't seem
to have the error. The "GPU RAM" portions cannot be tested with memtest
because it cannot access them: To the memtest access routine, the situation
is as follows: Let's say there are 4 GB RAM, 1 GB dedicated for GPU, so
for memtest there's 3 GB RAM to check. _Where_ this is located - well,
that depends on how the system BIOS/EFI decides on where to place the
GPU RAM segment ("internal addressing"). Let's assume you have two chips
of 2 GB each, this _could_ be the layout:

+--------------------+
Bank 0: |VVVFVVVVVV | 2 GB
+----------^---------+
RAM address 00000000 here

+--------------------+
Bank 1: | | 2 GB
+--------------------+

Then memtest will check from 00000000 to the end of the 3 GB RAM, and
the fault (F) is within the GPU RAM (V). It won't be tested.

Maybe you could just swap the modules an re-run the test. However, it's
hard to say _where_ the GPU RAM will end up, maybe at the beginning of
bank 1? Maybe a "random choice"? Or maybe it isn't even one contiguous
memory space?

[krad]

usually there is an option in the bios somewhere to alter the allocation of
ram to the video card. It can usually go from 64M-256M. Set it to the
lowest setting to present as much of the ram to the OS. Then run memtest
again. You may get different results.

Post by Polytropon

Post by Manish Jain

Post by Warren Block

Post by Manish Jain
What is printed underneath is NE56R14l-B9502G32Mnks
I can't be sure whether the l is actually l or I or 1

http://in.gateway.com/gw/en/IN/content/model/NX.Y1USI.005
Which does not have a separate GPU, and means that the bad memory
would be in one of the removable DIMMs.

Something to cheer about, at last. But then why does not memtest86+ pick
up any errors in 65 passes ?

As I mentioned earlier, this is a "false-negative" result. The portions
of the RAM dedicated for the GPU are "cut away" quite eary, so memtest
can only test what's left for "normal RAM". And those portions don't seem
to have the error. The "GPU RAM" portions cannot be tested with memtest
because it cannot access them: To the memtest access routine, the situation
is as follows: Let's say there are 4 GB RAM, 1 GB dedicated for GPU, so
for memtest there's 3 GB RAM to check. _Where_ this is located - well,
that depends on how the system BIOS/EFI decides on where to place the
GPU RAM segment ("internal addressing"). Let's assume you have two chips
+--------------------+
Bank 0: |VVVFVVVVVV | 2 GB
+----------^---------+
RAM address 00000000 here
+--------------------+
Bank 1: | | 2 GB
+--------------------+
Then memtest will check from 00000000 to the end of the 3 GB RAM, and
the fault (F) is within the GPU RAM (V). It won't be tested.
Maybe you could just swap the modules an re-run the test. However, it's
hard to say _where_ the GPU RAM will end up, maybe at the beginning of
bank 1? Maybe a "random choice"? Or maybe it isn't even one contiguous
memory space?
--
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
_______________________________________________
https://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "

[Solène Rapenne]

Post by Manish Jain

Post by Warren Block

Post by Manish Jain
What is printed underneath is NE56R14l-B9502G32Mnks
I can't be sure whether the l is actually l or I or 1

http://in.gateway.com/gw/en/IN/content/model/NX.Y1USI.005
Which does not have a separate GPU, and means that the bad memory
would be in one of the removable DIMMs.

Something to cheer about, at last. But then why does not memtest86+
pick up any errors in 65 passes ?
Regards
Manish Jain
_______________________________________________
https://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to

If you have 2 memory module, you can try each one separately. If you
encounter your problem with one and not the other one, you found the
problem.

[Warren Block]

Post by Warren Block

Post by Manish Jain
What is printed underneath is NE56R14l-B9502G32Mnks
I can't be sure whether the l is actually l or I or 1

http://in.gateway.com/gw/en/IN/content/model/NX.Y1USI.005
Which does not have a separate GPU, and means that the bad memory would be
in one of the removable DIMMs.

Something to cheer about, at last. But then why does not memtest86+ pick up
any errors in 65 passes ?

Possibly because they are hidden by being in an area of memory that
memtest86 can't test. Testing with only one DIMM could change that, or
changing the amount of RAM allocated to video in the BIOS. Only change
one thing at a time.