Memory recognition problem on Acer DAA75L - aParker motherboard (x1470)

English Hardware
1

Hello -

I recently installed opensuse 12.2 RC2 on a quad-AMD 3.0GHz (F1 socket) system with 16GB of RAM (4GBx4) on an Acer DAA75L-aParker motherboard (which was pulled from a X1470 tower).

The POST recognizes the 16GB of RAM installed. However, when I boot to OpenSuSE 12.2 RC2, it only sees 4GB of RAM. The same goes for pretty much any recently released Linux distribution.

I went to Acer’s web site and found a bios update that they called “Bios for Linux,” but it’s for a slightly different motherboard (when I attempt to update the bios using this bios file, it returns “ROM file ROMID not compatible…” While I could override the check, I’m afraid of bricking my motherboard.

I’ve tried over-riding the kernel parameter by adding “mem=16384M” on the boot command-line, but it didn’t help.

Any help would be deeply appreciated.

Andrew

2

On 09/23/2012 09:36 PM, thermalconductor wrote:
>
> Hello -
>
> I recently installed opensuse 12.2 RC2

-=WELCOME=- new poster…

first, why RC2? the 12.2 is available here: http://software.opensuse.org/

suggest you install it, or at least download and run the 12.2 Live CD
for (your choice) KDE or Gnome…

> on a quad-AMD 3.0GHz (F1 socket)
> system with 16GB of RAM (4GBx4) on an Acer DAA75L-aParker motherboard
> (which was pulled from a X1470 tower).

so, i guess (because i’m not going to google it–you should tell us)
that is a 64 bit capable board, and you downloaded a 64 bit system?

and, did the 12.2 install a “default” kernel or a “desktop” kernel?

> The POST recognizes the 16GB of RAM installed.

hmmmmmm, in what way does the POST do that? (does it give a beep code
indicating 16 GB? *)

or, are you saying when you boot up you see a BIOS message indicating
16GB installed? OR, are you interrupting the boot process very early and
going into the BIOS setup screens and seeing 16 GB there?

> However, when I boot to
> OpenSuSE 12.2 RC2, it only sees 4GB of RAM. The same goes for pretty
> much any recently released Linux distribution.

does “pretty much any” mean that some see 16GB? but most don’t?

did the board have 16GB installed when it was in the X1470 tower, or did
you add them afterwards?

>
> I went to Acer’s web site and found a bios update that they called
> “Bios for Linux,” but it’s for a slightly different motherboard (when I
> attempt to update the bios using this bios file, it returns “ROM file
> ROMID not compatible…” While I could override the check, I’m afraid
> of bricking my motherboard.

i’d say it is a good plan to NOT run in that BIOS update because in my
book “slightly different” is FAR FAR too different…

as is the “slightly different” RAM available out there–which is why i
wonder if maybe you have 4GB of the right ram (that was in the X1470)
and then bought 12GB of “slightly different” RAM and loaded it in…so,
what does the boards user manual say is the required RAM, and are all
sticks of that?

>
> I’ve tried over-riding the kernel parameter by adding “mem=16384M” on
> the boot command-line, but it didn’t help.

when i look at that boot option in the doc which came with my kernel
(automatically installed at
file:///usr/src/linux/Documentation/kernel-parameters.txt if you fetch
the kernel source and docs) it says to

[quote]
Use together with memmap= to avoid physical address space collisions.
Without memmap= PCI devices could be placed at addresses belonging to
unused RAM.

[quote]

and, then it offers four different memmap= options and four comments on
what all to set–did you also add memmap as you added mem= ??

> Any help would be deeply appreciated.

finally, when you say “when I boot to OpenSuSE 12.2 RC2, it only sees
4GB of RAM” please don’t just say that, instead copy and paste the
output of the following terminal command back to this thread using the
instructions here: http://goo.gl/i3wnr


free -m
cat /proc/meminfo | grep -i total

and, again: welcome!
since you are new here please tell us more about your linux background…

–
dd http://tinyurl.com/DD-Caveat
*

3

Thank you for the welcome. I appreciate that. But, I must confess, after reading your entire response, I didn’t feel very welcome. I do appreciate you taking the time to help, but, please, I can do without the snark. I’m an experienced programmer, not a kernel developer. Please understand that any omissions are because I spend most of my time cutting code that is not system related.

To start, I installed RC2 because that was the latest version available. Yes, I should have installed the newest version. I didn’t do it for various reasons, but I will do it as soon as time allows. As soon as I do that, I’ll report back. That will likely be a week from now.

Yes, this is a 64-bit installation – OpenSuSE 12.2 RC2 64-bit.

I have the desktop kernel installed, according to Yast.

During the POST, it displays 16GB (and counts through it), and the bios setup displays 16GB.

I bought 4 sticks of matched RAM. I never, ever mix RAM. It’s new RAM and it fits the specs the motherboard calls for (1333MHz). I know the repercussions of mixing RAM, so I can assure you it’s properly spec’d.

I have not tried the memmap option, but will try it after I move to the 12.2 release. I will also post the output from both commands at that time, too. I work away from home, which is why I can’t do things rapidly. And when I am home, my time is sparse. Again, I will post all of this sometime around Sept. 29 or 30.

Thank you for your help.
Andrew

4

On 09/23/2012 10:16 PM, thermalconductor wrote:
>
> Thank you for the welcome. I appreciate that. But, I must confess,
> after reading your entire response, I didn’t feel very welcome. I do
> appreciate you taking the time to help, but, please, I can do without
> the snark. I’m an experienced programmer, not a kernel developer.
> Please understand that any omissions are because I spend most of my time
> cutting code that is not system related.

You may feel unwelcome, but the volunteers here work with all sorts of
experience. When a particular post does not state if the system is 32- or
64-bit, and they use an older pre-release version after the system is released,
we must assume a beginner. For all we knew, you were using a 32-bit system that
would not be able to handle as much as 16 GB.

> To start, I installed RC2 because that was the latest version
> available. Yes, I should have installed the newest version. I didn’t
> do it for various reasons, but I will do it as soon as time allows. As
> soon as I do that, I’ll report back. That will likely be a week from
> now.
>
> Yes, this is a 64-bit installation – OpenSuSE 12.2 RC2 64-bit.
>
> I have the desktop kernel installed, according to Yast.
>
> During the POST, it displays 16GB (and counts through it), and the bios
> setup displays 16GB.
>
> I bought 4 sticks of matched RAM. I never, ever mix RAM. It’s new RAM
> and it fits the specs the motherboard calls for (1333MHz). I know the
> repercussions of mixing RAM, so I can assure you it’s properly spec’d.
>
> I have not tried the memmap option, but will try it after I move to the
> 12.2 release. I will also post the output from both commands at that
> time, too. I work away from home, which is why I can’t do things
> rapidly. And when I am home, my time is sparse. Again, I will post all
> of this sometime around Sept. 29 or 30.

When you post the info from your kernel, also post the e820 lines from the
output of dmesg.

5

> I can do without the snark.

though you read such, there were none…

not one bit…

every question/comment was leading down a technical trail to a potential
solution and none were intended to discount your technical expertise…

for example: i cannot see that you are experienced and know how to
match RAM–but, if you wander around these fora you will find many
examples of folks who do not know to and my question would lead to a
quick solution for them…i can’t not ask about that nor whatever you
read as ‘snark’…and, for all i know RC2 contained a know bug when
operating with your motherboard and over 4GB RAM…

admitted: maybe “why are you running 12.2 RC2?” sounds like snark while
“Upgrade from the known buggy testing software to the released and
supported software and post if the situation persists” would not??

again, welcome!!

–
dd

6

Thank you for your reply. It appears that you are from Germany. Perhaps it’s a cultural difference in how we communicate that led me to believe you were being snarky? If so, I apologize for making a snap judgment.

I have a German friend of many years, and his mother always seems curt. The fact is, she’s a sweet, wonderful woman once you get to know her! My friend understands this and always tells me this joke: “There once was a German man that loved his wife so much that he actually told her.” This isn’t a rip on Germans, as I have nothing but respect for them (I think Germans make the best cars in the world!). But I think our exchange somewhat illustrates our cultural differences.

FYI, I’m an old unix hand. I’ve used pyramid, HPUX, AIX, SCO, IRIX, SunOS, Solaris and more. I absolutely love openSuSE and would never consider any other OS. It’s my OS of choice for work and home. Most of my department at work is on Windows and iMacs. In fact, my partner and I are the only two in the entire 80-person IT department that use Linux distros. He uses Ubuntu, I use openSuSE. Between us, we maintain the back-end code for a web site that brings in about $5 million per week in sales.

In closing, this problem is related to a computer that I pieced together using all new parts. I originally suspected the RAM, but somewhat ruled that out based on the POST/bios setup. I’m confident we’ll get to the root of this problem. All I ask is your patience in allowing me to get that information to you later this week.

Thank you to all those that responded.

7

On 09/24/2012 09:06 PM, thermalconductor wrote:
> Thank you for your reply. It appears that you are from Germany.
> Perhaps it’s a cultural difference in how we communicate that led me to
> believe you were being snarky? If so, I apologize for making a snap
> judgment.

heh. apology not needed…(but, accepted).

however, imho it is not usually beneficial to make generalized
assumptions about an individual’s intentions based on things like
race, color, religion, national origin, age, disability, political
affiliation, belief, citizenship, gender, sexual orientation and etc
(those kinda things which some countries hold lawfully off limits as the
basis for discrimination)…

i don’t know if we have a cultural difference or not…all i know is my
prose seemed snarky/curt to you…so, therefore it was that for you…

perhaps instead of simply trying to explain that was not my intention, i
should have apologized…

Sorry.

though i don’t know why i need to apologize because you mistook my
‘style’ as being ‘snarky’. (but, you are not the first, and probably not
the last to read “standard military directness” as curt." so, sorry.)

you decide if there is a cultural thing: i was born in Tennessee, vote
in Texas and am a resident of Denmark…since you are still working i am
probably older than you and by 1991 had done enough active duty military
service to earn a full pension (plus)…

–
dd http://goo.gl/PUjnL
http://tinyurl.com/DD-Caveat

8

Thanks for the reply. No need to apologize, my friend. It’s easy to misread posts when one doesn’t have the advantage of seeing the other person’s face. I’m grateful that you’re taking time to help me.

Please note that I didn’t boot the kernel with the mem or the memmap options because I thought I’d post the requested system output before attempting any stunts with memory. I’m hoping that there is something obvious to someone out there reading this thread, and that they can point out what further steps to take.

Anyhow, here is the output that you requested:


clyde:/home/andy # free -m
             total       used       free     shared    buffers     cached
Mem:          3665       1454       2210          0         32        797
-/+ buffers/cache:        625       3040
Swap:         2053          0       2053


clyde:/home/andy # cat /proc/meminfo | grep -i total
MemTotal:        3753592 kB
SwapTotal:       2103292 kB
VmallocTotal:   34359738367 kB
HugePages_Total:       0


clyde:/home/andy # dmesg
(snip)
    0.000000] BIOS-provided physical RAM map:
    0.000000]  BIOS-e820: 0000000000000000 - 000000000009d800 (usable)
    0.000000]  BIOS-e820: 000000000009d800 - 00000000000a0000 (reserved)
    0.000000]  BIOS-e820: 00000000000e0000 - 0000000000100000 (reserved)
    0.000000]  BIOS-e820: 0000000000100000 - 00000000be550000 (usable)
    0.000000]  BIOS-e820: 00000000be550000 - 00000000be59a000 (ACPI NVS)
    0.000000]  BIOS-e820: 00000000be59a000 - 00000000be5a3000 (ACPI data)
    0.000000]  BIOS-e820: 00000000be5a3000 - 00000000be62f000 (reserved)
    0.000000]  BIOS-e820: 00000000be62f000 - 00000000be632000 (ACPI NVS)
    0.000000]  BIOS-e820: 00000000be632000 - 00000000be633000 (usable)
    0.000000]  BIOS-e820: 00000000be633000 - 00000000be86e000 (reserved)
    0.000000]  BIOS-e820: 00000000be86e000 - 00000000bf86f000 (ACPI NVS)
    0.000000]  BIOS-e820: 00000000bf86f000 - 00000000bf87f000 (reserved)
    0.000000]  BIOS-e820: 00000000bf87f000 - 00000000bf880000 (ACPI NVS)
    0.000000]  BIOS-e820: 00000000bf880000 - 00000000bf891000 (reserved)
    0.000000]  BIOS-e820: 00000000bf891000 - 00000000bf898000 (ACPI NVS)
    0.000000]  BIOS-e820: 00000000bf898000 - 00000000bf8c0000 (reserved)
    0.000000]  BIOS-e820: 00000000bf8c0000 - 00000000bfac3000 (ACPI NVS)
    0.000000]  BIOS-e820: 00000000bfac3000 - 00000000bff00000 (reserved)
    0.000000]  BIOS-e820: 00000000e0000000 - 00000000f0000000 (reserved)
    0.000000]  BIOS-e820: 00000000fec00000 - 00000000fec01000 (reserved)
    0.000000]  BIOS-e820: 00000000fec10000 - 00000000fec11000 (reserved)
    0.000000]  BIOS-e820: 00000000fed00000 - 00000000fed01000 (reserved)
    0.000000]  BIOS-e820: 00000000fed40000 - 00000000fed45000 (reserved)
    0.000000]  BIOS-e820: 00000000fed61000 - 00000000fed71000 (reserved)
    0.000000]  BIOS-e820: 00000000fed80000 - 00000000fed90000 (reserved)
    0.000000]  BIOS-e820: 00000000ff000000 - 0000000100000000 (reserved)
    0.000000]  BIOS-e820: 0000000100001000 - 000000012f000000 (usable)
    0.000000] NX (Execute Disable) protection: active
    0.000000] DMI 2.7 present.
    0.000000] DMI: Acer Aspire X1470/Aspire X1470, BIOS P01-A1 08/05/2011
    0.000000] e820 update range: 0000000000000000 - 0000000000010000 (usable) ==> (reserved)
    0.000000] e820 remove range: 00000000000a0000 - 0000000000100000 (usable)
    0.000000] No AGP bridge found
    0.000000] last_pfn = 0x12f000 max_arch_pfn = 0x400000000
    0.000000] MTRR default type: uncachable
    0.000000] MTRR fixed ranges enabled:
    0.000000]   00000-9FFFF write-back
    0.000000]   A0000-BFFFF write-through
    0.000000]   C0000-CEFFF write-protect
    0.000000]   CF000-E7FFF uncachable
    0.000000]   E8000-FFFFF write-protect
    0.000000] MTRR variable ranges enabled:
    0.000000]   0 base 0000000000 mask FF00000000 write-back
    0.000000]   1 base 00BFF00000 mask FFFFF00000 uncachable
    0.000000]   2 base 00C0000000 mask FFC0000000 uncachable
    0.000000]   3 disabled
    0.000000]   4 disabled
    0.000000]   5 disabled
    0.000000]   6 disabled
    0.000000]   7 disabled
    0.000000] TOM2: 000000013f000000 aka 5104M
    0.000000] x86 PAT enabled: cpu 0, old 0x7040600070406, new 0x7010600070106
    0.000000] e820 update range: 00000000bff00000 - 0000000100000000 (usable) ==> (reserved)
    0.000000] last_pfn = 0xbe633 max_arch_pfn = 0x400000000
    0.000000] found SMP MP-table at [ffff8800000fcf80] fcf80
    0.000000] initial memory mapped : 0 - 20000000
    0.000000] Base memory trampoline at [ffff880000098000] 98000 size 20480
    0.000000] Using GB pages for direct mapping
    0.000000] init_memory_mapping: 0000000000000000-00000000be633000
    0.000000]  0000000000 - 0080000000 page 1G
    0.000000]  0080000000 - 00be600000 page 2M
    0.000000]  00be600000 - 00be633000 page 4k
    0.000000] kernel direct mapping tables up to be633000 @ 1fffd000-20000000
    0.000000] init_memory_mapping: 0000000100000000-000000012f000000
    0.000000]  0100000000 - 012f000000 page 2M
    0.000000] kernel direct mapping tables up to 12f000000 @ be54e000-be550000
    0.000000] RAMDISK: 36026000 - 3700b000
    0.000000] ACPI: RSDP 00000000000f0450 00024 (v02 ACRSYS)
    0.000000] ACPI: XSDT 00000000be59a070 0005C (v01 ACRSYS ACRPRDCT 01072009 AMI  00010013)
    0.000000] ACPI: FACP 00000000be5a0608 000F4 (v04 ACRSYS ACRPRDCT 01072009 AMI  00010013)
    0.000000] ACPI Warning: Optional field Pm2ControlBlock has zero address or length: 0x0000000000000000/0x1 - not using it (20120320/tbfadt-579)
    0.000000] ACPI: DSDT 00000000be59a158 064A9 (v02 ACRSYS ACRPRDCT 00000000 INTL 20051117)
    0.000000] ACPI: FACS 00000000bf892f80 00040
    0.000000] ACPI: APIC 00000000be5a0700 00072 (v03 ACRSYS ACRPRDCT 01072009 AMI  00010013)
    0.000000] ACPI: MCFG 00000000be5a0778 0003C (v01 A M I  GMCH945. 01072009 MSFT 00000097)
    0.000000] ACPI: SLIC 00000000be5a07b8 00176 (v01 ACRSYS ACRPRDCT 01072009 AMI  00010013)
    0.000000] ACPI: HPET 00000000be5a0930 00038 (v01 ACRSYS ACRPRDCT 01072009 AMI  00000004)
    0.000000] ACPI: SSDT 00000000be5a0968 00FD8 (v01 AMD    POWERNOW 00000001 AMD  00000001)
    0.000000] ACPI: SSDT 00000000be5a1940 00695 (v02    AMD     ALIB 00000001 MSFT 04000000)
    0.000000] ACPI: Local APIC address 0xfee00000
    0.000000] No NUMA configuration found
    0.000000] Faking a node at 0000000000000000-000000012f000000
    0.000000] Initmem setup node 0 0000000000000000-000000012f000000
    0.000000]   NODE_DATA [000000012efec000 - 000000012effffff]
    0.000000]  [ffffea0000000000-ffffea00043fffff] PMD -> [ffff88012aa00000-ffff88012dffffff] on node 0
(snip)
9

On 09/30/2012 02:46 AM, thermalconductor wrote:
> clyde:/home/andy # dmesg

i can’t help with the dmesg output…wait for Larry/others to view that
before proceeding…

my foggy and unpolished crystal ball says you suffer from a hardware or
BIOS problem…(i can give the reasons if you wish…)

i can’t help further (except with maybe hardware test methods…)
good luck.

–
dd

10

On 09/30/2012 01:06 AM, dd@home.dk wrote:
> On 09/30/2012 02:46 AM, thermalconductor wrote:
>> clyde:/home/andy # dmesg
>
> i can’t help with the dmesg output…wait for Larry/others to view that before
> proceeding…
>
> my foggy and unpolished crystal ball says you suffer from a hardware or BIOS
> problem…(i can give the reasons if you wish…)
>
> i can’t help further (except with maybe hardware test methods…)
> good luck.

The highest address line listed in your e820 output is

0.000000]  BIOS-e820: 0000000100001000 - 000000012f000000 (usable)

Converting the hexadecimal to decimal yields 4 GiB to 4.73 GiB. That is all your
BIOS is reporting to the Linux kernel. To me, this certainly looks like a BIOS bug.