*PAE* Kernel Still Does Not Detect All RAM

[Mordac]

Hey. I've actually been having this problem for a while, but it was never that big of a deal so I just was never prompted to do much about it. I guess I'm just bored now so I decided to sign up and ask about it.

A while ago I upgraded to 4GB of RAM on an x86 system, up from 2GB. However, even with the PAE kernel installed ("2.6.40-4.fc15.i686.PAE", at the moment), I can't actually access the full 4GB, only 2.9GB as if I was running a non-PAE kernel. Both the BIOS and lshw can see the 4GB installed, and my CPU does support PAE, so I really don't know what could be causing this. Anyone got any ideas?

For what it's worth, I'll post the blurs lshw gives about my CPU and RAM:

Originally Posted by lshw

*-cpu:0
description: CPU
product: Intel(R) Core(TM) Duo CPU T2450 @ 2.00GHz
vendor: Intel Corp.
physical id: 4
bus info: cpu@0
version: 6.14.12
serial: 0000-06EC-0000-0000-0000-0000
slot: U2E1
size: 800MHz
capacity: 2048MHz
width: 32 bits
capabilities: boot fpu fpu_exception wp vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe nx constant_tsc arch_perfmon bts aperfmperf pni monitor est tm2 xtpr pdcm cpufreq
configuration: id=0


*-memory
description: System Memory
physical id: d
slot: System board or motherboard
size: 4GiB
*-bank:0
description: SODIMM DDR Synchronous
physical id: 0
slot: M1
size: 2GiB
width: 32 bits
*-bank:1
description: SODIMM DDR Synchronous
physical id: 1
slot: M2
size: 2GiB
width: 32 bits

Cheers!

---------- Post added at 07:43 PM ---------- Previous post was at 07:19 PM ----------

Should this have been in the Hardware board instead? If so, could a mod please move it? Thanks!

[bodhi.zazen]

I see all 4 Gb listed in that output. Is there some other output you are concerned with ?

size: 4GiB

[Mordac]

Originally Posted by bodhi.zazen

I see all 4 Gb listed in that output. Is there some other output you are concerned with ?

Oh, yeah, I guess I forgot to list a piece of the puzzle! I did say that lshw sees the 4GB of hardware, but the OS cannot actually address all 4. See, for instance:

$ free -m
total used free shared buffers cached
Mem: 3017 1836 1180 0 98 1323
-/+ buffers/cache: 414 2602
Swap: 4031 0 4031

Or if I open Gnome System Monitor, it also shows only 2.9 GB being available.

[mikee]

Just to cover all the bases..
Did you reboot after installing this kernel?
Show us "uname -a" output so we can see which kernel is running.

[Mordac]

Originally Posted by mikee

Just to cover all the bases..
Did you reboot after installing this kernel?
Show us "uname -a" output so we can see which kernel is running.

I've upgraded the hardware months ago, so yes, I have rebooted since then. I just put off dealing with this until now for general laziness reasons.

Anyway, it's no bother at all, so here's as requested:

$ uname -a
Linux [...] 2.6.40-4.fc15.i686.PAE #1 SMP Fri Jul 29 18:47:58 UTC 2011 i686 i686 i386 GNU/Linux

[nirik]

This is likely a hardware limitation.
Is this a dell box from a few years back? There were some that they shipped with a 32bit north bridge, so at most you get 3.3ish on a 4GB system.

You can look in 'dmesg' for more info on the memory it sees and uses... look toward the top of the boot process.

is your BIOS up to date? my computer is about 5 years old, and last BIOS update was in 2008 and my PAE kernel see's 3.9 Gigs Ram even though i have 4Gigs ( 32bit F15 )

[Mordac]

Originally Posted by nirik

This is likely a hardware limitation.
Is this a dell box from a few years back? There were some that they shipped with a 32bit north bridge, so at most you get 3.3ish on a 4GB system.

You can look in 'dmesg' for more info on the memory it sees and uses... look toward the top of the boot process.

It's a Toshiba, actually, though it is from a few years back, so it might have the same issues? Toshiba says is supports 4GB, but...

Here's the dmesg output, at least until it seems relevant, as I'm not going to post the whole monstrosity. If something's missing...

[ 0.000000] Initializing cgroup subsys cpuset
[ 0.000000] Initializing cgroup subsys cpu
[ 0.000000] Linux version 2.6.40-4.fc15.i686.PAE (mockbuild@x86-18.phx2.fedoraproject.org) (gcc version 4.6.0 20110530 (Red Hat 4.6.0-9)
(GCC) ) #1 SMP Fri Jul 29 18:47:58 UTC 2011
[ 0.000000] Disabled fast string operations
[ 0.000000] BIOS-provided physical RAM map:
[ 0.000000] BIOS-e820: 0000000000000000 - 000000000009f800 (usable)
[ 0.000000] BIOS-e820: 000000000009f800 - 00000000000a0000 (reserved)
[ 0.000000] BIOS-e820: 00000000000dc000 - 0000000000100000 (reserved)
[ 0.000000] BIOS-e820: 0000000000100000 - 00000000bf680000 (usable)
[ 0.000000] BIOS-e820: 00000000bf680000 - 00000000bf700000 (ACPI NVS)
[ 0.000000] BIOS-e820: 00000000bf700000 - 00000000c0000000 (reserved)
[ 0.000000] BIOS-e820: 00000000e0000000 - 00000000f0000000 (reserved)
[ 0.000000] BIOS-e820: 00000000fec00000 - 00000000fec10000 (reserved)
[ 0.000000] BIOS-e820: 00000000fed00000 - 00000000fed00400 (reserved)
[ 0.000000] BIOS-e820: 00000000fed14000 - 00000000fed1a000 (reserved)
[ 0.000000] BIOS-e820: 00000000fed1c000 - 00000000fed90000 (reserved)
[ 0.000000] BIOS-e820: 00000000fee00000 - 00000000fee01000 (reserved)
[ 0.000000] BIOS-e820: 00000000ff000000 - 0000000100000000 (reserved)
[ 0.000000] NX (Execute Disable) protection: active
[ 0.000000] DMI present.
[ 0.000000] DMI: TOSHIBA Satellite A200/ISKAE, BIOS V1.80 08/21/2007
[ 0.000000] e820 update range: 0000000000000000 - 0000000000010000 (usable) ==> (reserved)
[ 0.000000] e820 remove range: 00000000000a0000 - 0000000000100000 (usable)
[ 0.000000] last_pfn = 0xbf680 max_arch_pfn = 0x1000000
[ 0.000000] MTRR default type: uncachable
[ 0.000000] MTRR fixed ranges enabled:
[ 0.000000] 00000-9FFFF write-back
[ 0.000000] A0000-BFFFF uncachable
[ 0.000000] C0000-CFFFF write-protect
[ 0.000000] D0000-DFFFF uncachable
[ 0.000000] E0000-FFFFF write-protect
[ 0.000000] MTRR variable ranges enabled:
[ 0.000000] 0 base 000000000 mask F80000000 write-back
[ 0.000000] 1 base 080000000 mask FC0000000 write-back
[ 0.000000] 2 base 0BF700000 mask FFFF00000 uncachable
[ 0.000000] 3 base 0BF800000 mask FFF800000 uncachable
[ 0.000000] 4 disabled
[ 0.000000] 5 disabled
[ 0.000000] 6 disabled
[ 0.000000] 7 disabled
[ 0.000000] PAT not supported by CPU.
[ 0.000000] original variable MTRRs
[ 0.000000] reg 0, base: 0GB, range: 2GB, type WB
[ 0.000000] reg 1, base: 2GB, range: 1GB, type WB
[ 0.000000] reg 2, base: 3063MB, range: 1MB, type UC
[ 0.000000] reg 3, base: 3064MB, range: 8MB, type UC
[ 0.000000] total RAM covered: 3063M
[ 0.000000] Found optimal setting for mtrr clean up
[ 0.000000] gran_size: 64K chunk_size: 16M num_reg: 4 lose cover RAM: 0G
[ 0.000000] New variable MTRRs
[ 0.000000] reg 0, base: 0GB, range: 2GB, type WB
[ 0.000000] reg 1, base: 2GB, range: 1GB, type WB
[ 0.000000] reg 2, base: 3063MB, range: 1MB, type UC
[ 0.000000] reg 3, base: 3064MB, range: 8MB, type UC
[ 0.000000] found SMP MP-table at [c00f7670] f7670
[ 0.000000] initial memory mapped : 0 - 01200000
[ 0.000000] Base memory trampoline at [c009b000] 9b000 size 16384
[ 0.000000] init_memory_mapping: 0000000000000000-00000000371fe000
[ 0.000000] 0000000000 - 0000200000 page 4k
[ 0.000000] 0000200000 - 0037000000 page 2M
[ 0.000000] 0037000000 - 00371fe000 page 4k
[ 0.000000] kernel direct mapping tables up to 371fe000 @ 11f6000-1200000
[ 0.000000] RAMDISK: 37219000 - 37ff0000
[ 0.000000] Allocated new RAMDISK: 36427000 - 371fd735
[ 0.000000] Move RAMDISK from 0000000037219000 - 0000000037fef734 to 36427000 - 371fd734
[ 0.000000] ACPI: RSDP 000f75e0 00014 (v00 TOSCPL)
[ 0.000000] ACPI: RSDT bf687740 00054 (v01 TOSCPL TOSCPL00 06040000 LTP 00000000)
[ 0.000000] ACPI: FACP bf68fc78 00074 (v01 TOSCPL CALISTGA 06040000 LOHR 0000005A)
[ 0.000000] ACPI: DSDT bf6894ed 0678B (v01 TOSCPL CALISTGA 06040000 INTL 20060608)
[ 0.000000] ACPI: FACS bf690fc0 00040
[ 0.000000] ACPI: SLIC bf68fcec 00176 (v01 TOSCPL TOSCPL00 06040000 LOHR 00000000)
[ 0.000000] ACPI: APIC bf68fe62 00068 (v01 INTEL CALISTGA 06040000 LOHR 0000005A)
[ 0.000000] ACPI: HPET bf68feca 00038 (v01 INTEL CALISTGA 06040000 LOHR 0000005A)
[ 0.000000] ACPI: MCFG bf68ff02 0003C (v01 INTEL CALISTGA 06040000 LOHR 0000005A)
[ 0.000000] ACPI: BOOT bf68ffd8 00028 (v01 PTLTD $SBFTBL$ 06040000 LTP 00000001)
[ 0.000000] ACPI: APIC bf68ff70 00068 (v01 PTLTD ? APIC 06040000 LTP 00000000)
[ 0.000000] ACPI: SSDT bf688e9a 0064F (v01 SataRe SataPri 00001000 INTL 20050624)
[ 0.000000] ACPI: SSDT bf688808 00692 (v01 SataRe SataSec 00001000 INTL 20050624)
[ 0.000000] ACPI: SSDT bf687d20 0025F (v01 PmRef Cpu0Tst 00003000 INTL 20050624)
[ 0.000000] ACPI: SSDT bf687c7a 000A6 (v01 PmRef Cpu1Tst 00003000 INTL 20050624)
[ 0.000000] ACPI: SSDT bf687794 004E6 (v01 PmRef CpuPm 00003000 INTL 20050624)
[ 0.000000] ACPI: BIOS bug: multiple APIC/MADT found, using 0
[ 0.000000] ACPI: If "acpi_apic_instance=2" works better, notify linux-acpi@vger.kernel.org
[ 0.000000] ACPI: Local APIC address 0xfee00000
[ 0.000000] 2180MB HIGHMEM available.
[ 0.000000] 881MB LOWMEM available.
[ 0.000000] mapped low ram: 0 - 371fe000
[ 0.000000] low ram: 0 - 371fe000
[ 0.000000] Zone PFN ranges:
[ 0.000000] DMA 0x00000010 -> 0x00001000
[ 0.000000] Normal 0x00001000 -> 0x000371fe
[ 0.000000] HighMem 0x000371fe -> 0x000bf680
[ 0.000000] Movable zone start PFN for each node
[ 0.000000] early_node_map[2] active PFN ranges
[ 0.000000] 0: 0x00000010 -> 0x0000009f
[ 0.000000] 0: 0x00000100 -> 0x000bf680
[ 0.000000] On node 0 totalpages: 783887
[ 0.000000] free_area_init_node: node 0, pgdat c0ac46c0, node_mem_map f4c36200
[ 0.000000] DMA zone: 32 pages used for memmap
[ 0.000000] DMA zone: 0 pages reserved
[ 0.000000] DMA zone: 3951 pages, LIFO batch:0
[ 0.000000] Normal zone: 1732 pages used for memmap
[ 0.000000] Normal zone: 219962 pages, LIFO batch:31
[ 0.000000] HighMem zone: 4362 pages used for memmap
[ 0.000000] HighMem zone: 553848 pages, LIFO batch:31
[ 0.000000] Using APIC driver default
[.......]
[ 0.000000] ACPI: HPET id: 0x8086a201 base: 0xfed00000
[ 0.000000] SMP: Allowing 2 CPUs, 0 hotplug CPUs
[ 0.000000] nr_irqs_gsi: 40
[ 0.000000] PM: Registered nosave memory: 000000000009f000 - 00000000000a0000
[ 0.000000] PM: Registered nosave memory: 00000000000a0000 - 00000000000dc000
[ 0.000000] PM: Registered nosave memory: 00000000000dc000 - 0000000000100000
[ 0.000000] Allocating PCI resources starting at c0000000 (gap: c0000000:20000000)
[ 0.000000] Booting paravirtualized kernel on bare hardware
[ 0.000000] setup_percpu: NR_CPUS:64 nr_cpumask_bits:64 nr_cpu_ids:2 nr_node_ids:1
[ 0.000000] PERCPU: Embedded 13 pages/cpu @f4a00000 s28928 r0 d24320 u1048576
[ 0.000000] pcpu-alloc: s28928 r0 d24320 u1048576 alloc=1*2097152
[ 0.000000] pcpu-alloc: [0] 0 1
[ 0.000000] Built 1 zonelists in Zone order, mobility grouping on. Total pages: 777761
[ 0.000000] PID hash table entries: 4096 (order: 2, 16384 bytes)
[ 0.000000] Dentry cache hash table entries: 131072 (order: 7, 524288 bytes)
[ 0.000000] Inode-cache hash table entries: 65536 (order: 6, 262144 bytes)
[ 0.000000] Initializing CPU#0
[ 0.000000] allocated 12543744 bytes of page_cgroup
[ 0.000000] please try 'cgroup_disable=memory' option if you don't want memory cgroups
[ 0.000000] Initializing HighMem for node 0 (000371fe:000bf680)
[ 0.000000] Memory: 3075024k/3136000k available (4279k kernel code, 60524k reserved, 2718k data, 612k init, 2232840k highmem)
[ 0.000000] virtual kernel memory layout:
[ 0.000000] fixmap : 0xff575000 - 0xfffff000 (10792 kB)
[ 0.000000] pkmap : 0xff200000 - 0xff400000 (2048 kB)
[ 0.000000] vmalloc : 0xf79fe000 - 0xff1fe000 ( 120 MB)
[ 0.000000] lowmem : 0xc0000000 - 0xf71fe000 ( 881 MB)
[ 0.000000] .init : 0xc0ad6000 - 0xc0b6f000 ( 612 kB)
[ 0.000000] .data : 0xc082dedc - 0xc0ad5940 (2718 kB)
[ 0.000000] .text : 0xc0400000 - 0xc082dedc (4279 kB)

I think this is it as far as memory related stuff goes, but if there's something that's still lemme know and I'll pluck it from the rest. My skills really don't go as far as interpreting most of this guff, I'm afraid. :\

Originally Posted by detox

is your BIOS up to date? my computer is about 5 years old, and last BIOS update was in 2008 and my PAE kernel see's 3.9 Gigs Ram even though i have 4Gigs ( 32bit F15 )

I have the most recent BIOS, though that's from 2007. :P However, when I go into the BIOS setup at boot time, it can see all 4GB, at least at a hardware level, though if it allocates them as available or not I have no idea. I can tell you that the previous version of the BIOS did not even do that, only saw 3GB, but it started seeing 4GB after I flashed the new version.

[nirik]

I fear this is the 32bit northbridge issue... it's around the right time for it.

You could pull down a 64 bit live media and boot that. If it sees the same memory then that is the issue. ;(

[Mordac]

Originally Posted by nirik

I fear this is the 32bit northbridge issue... it's around the right time for it.

You could pull down a 64 bit live media and boot that. If it sees the same memory then that is the issue. ;(

I couldn't, as it's not an x64 box! =) Fair enough though, I guess there's really no way around this. It sucks that I blew money on that extra 2GB RAM stick, but oh well.

Cheers guys.

Originally Posted by Mordac

I couldn't, as it's not an x64 box! =) Fair enough though, I guess there's really no way around this. It sucks that I blew money on that extra 2GB RAM stick, but oh well.

Cheers guys.

you could always pull one stick out or are they 2x 2Gig sticks?

[Mordac]

Originally Posted by detox

you could always pull one stick out or are they 2x 2Gig sticks?

Two 2GB sticks I'm afraid. I could have a 1 GB stick from the ones that were in there before with a 2GB one, but I understand combining different RAM sticks is inadvisable (especially as they're at a different speed).

[jpollard]

How much memory does the video hardware take?

[Mordac]

256MB. It's not dedicated though, it's an Intel integrated graphics card.

[jpollard]

That memory isn't available to the system though - which cuts memory to 3.75G.

I know the kernel, by default, takes 1G for itself to cover the I/O hole, and itself, which leaves about 2.75 for user space. There could be a bit more due to my VERY rough estimates. But it comes close to what you actually see.

stevea may have a better understanding of the physical limits, but it looks like your system is seeing all of available memory.