Memory limit / xen for Conectiva

Memory limit / xen for Conectiva

[Flavio Leitner]

Using xm dmesg I found:

(XEN) Initialized 1983MB memory (507888 pages) on a 1983MB machine.

Well, this machine have 8GBytes of RAM and I already enable highmem on 
xen0 and xenU, but I suspect that this change must be on xen itself.


BTW, 
 The xen rpm package is now avaliable on Conectiva Linux Snapshot. :)


-- 
Flávio Bruno Leitner <fbl@conectiva.com.br>
[ E74B 0BD0 5E05 C385 239E  531C BC17 D670 7FF0 A9E0 ]


-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php

Re: Memory limit / xen for Conectiva

[Ian Pratt]

> Using xm dmesg I found:
> 
> (XEN) Initialized 1983MB memory (507888 pages) on a 1983MB machine.
> 
> Well, this machine have 8GBytes of RAM and I already enable highmem on 
> xen0 and xenU, but I suspect that this change must be on xen itself.

Please can you send us the output of a native linux 2.4 or 2.6
kernel booting. Your machine must have a memory hole, and we need
to be smarter about parsing memory tables. You should be able to
get 4GB working, but we have no plans to support PAE36. We're
concentrating on x86_64.

Thanks,
Ian


-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php

Re: Memory limit / xen for Conectiva

[Flavio Leitner]

On Fri, Sep 24, 2004 at 06:47:19PM +0100, Ian Pratt wrote:
> > (XEN) Initialized 1983MB memory (507888 pages) on a 1983MB machine.
> 
> Please can you send us the output of a native linux 2.4 or 2.6
> kernel booting. Your machine must have a memory hole, and we need
> to be smarter about parsing memory tables. You should be able to
> get 4GB working, but we have no plans to support PAE36. We're
> concentrating on x86_64.


Linux version 2.6.9-67090clenterprise (root@mapi8.distro.conectiva) (gcc version 3.4.1) #1 SMP Mon Aug 30 11:29:19 BRT 2004
BIOS-provided physical RAM map:
 BIOS-e820: 0000000000000000 - 000000000009e000 (usable)
 BIOS-e820: 000000000009e000 - 00000000000a0000 (reserved)
 BIOS-e820: 00000000000e9800 - 0000000000100000 (reserved)
 BIOS-e820: 0000000000100000 - 000000007bff0000 (usable)
 BIOS-e820: 000000007bff0000 - 000000007bfffc00 (ACPI data)
 BIOS-e820: 000000007bfffc00 - 000000007c000000 (ACPI NVS)
 BIOS-e820: 00000000fec00000 - 00000000fec10000 (reserved)
 BIOS-e820: 00000000fee00000 - 00000000fee01000 (reserved)
 BIOS-e820: 00000000ffc00000 - 0000000100000000 (reserved)
 BIOS-e820: 0000000100000000 - 0000000280000000 (usable)
9344MB HIGHMEM available.
896MB LOWMEM available.
found SMP MP-table at 000f5d00
On node 0 totalpages: 2621440
  DMA zone: 4096 pages, LIFO batch:1
  Normal zone: 225280 pages, LIFO batch:16
  HighMem zone: 2392064 pages, LIFO batch:16
DMI 2.3 present.
ACPI disabled because your bios is from 2000 and too old
You can enable it with acpi=force
Intel MultiProcessor Specification v1.4
    Virtual Wire compatibility mode.
OEM ID: INTEL    Product ID: SKA40        APIC at: 0xFEE00000
Processor #3 6:10 APIC version 17
Processor #0 6:10 APIC version 17
Processor #1 6:10 APIC version 17
Processor #2 6:10 APIC version 17
I/O APIC #4 Version 17 at 0xFEC00000.
I/O APIC #5 Version 17 at 0xFEC01000.
Enabling APIC mode:  Flat.  Using 2 I/O APICs
Processors: 4
Built 1 zonelists
Kernel command line: root=/dev/sda1 vga=0x314 
Initializing CPU#0
PID hash table entries: 4096 (order 12: 32768 bytes)
Detected 698.163 MHz processor.
Using tsc for high-res timesource
Console: colour dummy device 80x25
Dentry cache hash table entries: 131072 (order: 7, 524288 bytes)
Inode-cache hash table entries: 65536 (order: 6, 262144 bytes)
Memory: 8234364k/10485760k available (2089k kernel code, 87504k reserved, 884k data, 352k init, 7405504k highmem)
Checking if this processor honours the WP bit even in supervisor mode... Ok.
Calibrating delay loop... 1376.25 BogoMIPS
Security Scaffold v1.0.0 initialized
Mount-cache hash table entries: 512 (order: 0, 4096 bytes)
CPU: After generic identify, caps: 0383fbff 00000000 00000000 00000000
CPU: After vendor identify, caps:  0383fbff 00000000 00000000 00000000
CPU: L1 I cache: 16K, L1 D cache: 16K
CPU: L2 cache: 2048K
CPU: After all inits, caps:        0383fbff 00000000 00000000 00000040
Intel machine check architecture supported.
Intel machine check reporting enabled on CPU#0.
Enabling fast FPU save and restore... done.
Enabling unmasked SIMD FPU exception support... done.
Checking 'hlt' instruction... OK.
ACPI: IRQ9 SCI: Edge set to Level Trigger.
CPU0: Intel Pentium III (Cascades) stepping 01
per-CPU timeslice cutoff: 5845.92 usecs.
task migration cache decay timeout: 6 msecs.
Booting processor 1/0 eip 3000
Initializing CPU#1
Calibrating delay loop... 1392.64 BogoMIPS
CPU: After generic identify, caps: 0383fbff 00000000 00000000 00000000
CPU: After vendor identify, caps:  0383fbff 00000000 00000000 00000000
CPU: L1 I cache: 16K, L1 D cache: 16K
CPU: L2 cache: 2048K
CPU: After all inits, caps:        0383fbff 00000000 00000000 00000040
Intel machine check architecture supported.
Intel machine check reporting enabled on CPU#1.
CPU1: Intel Pentium III (Cascades) stepping 01
Booting processor 2/1 eip 3000
Initializing CPU#2
Calibrating delay loop... 1392.64 BogoMIPS
CPU: After generic identify, caps: 0383fbff 00000000 00000000 00000000
CPU: After vendor identify, caps:  0383fbff 00000000 00000000 00000000
CPU: L1 I cache: 16K, L1 D cache: 16K
CPU: L2 cache: 2048K
CPU: After all inits, caps:        0383fbff 00000000 00000000 00000040
Intel machine check architecture supported.
Intel machine check reporting enabled on CPU#2.
CPU2: Intel Pentium III (Cascades) stepping 01
Booting processor 3/2 eip 3000
Initializing CPU#3
Calibrating delay loop... 1392.64 BogoMIPS
CPU: After generic identify, caps: 0383fbff 00000000 00000000 00000000
CPU: After vendor identify, caps:  0383fbff 00000000 00000000 00000000
CPU: L1 I cache: 16K, L1 D cache: 16K
CPU: L2 cache: 2048K
CPU: After all inits, caps:        0383fbff 00000000 00000000 00000040
Intel machine check architecture supported.
Intel machine check reporting enabled on CPU#3.
CPU3: Intel Pentium III (Cascades) stepping 01
Total of 4 processors activated (5554.17 BogoMIPS).
ENABLING IO-APIC IRQs
..TIMER: vector=0x31 pin1=-1 pin2=0
...trying to set up timer (IRQ0) through the 8259A ... 
..... (found pin 0) ...works.
checking TSC synchronization across 4 CPUs: passed.
Brought up 4 CPUs
CPU0:  online
 domain 0: span 01
  groups: 01
  domain 1: span 0f
   groups: 01 02 04 08
CPU1:  online
 domain 0: span 02
  groups: 02
  domain 1: span 0f
   groups: 02 04 08 01
CPU2:  online
 domain 0: span 04
  groups: 04
  domain 1: span 0f
   groups: 04 08 01 02
CPU3:  online
 domain 0: span 08
  groups: 08
  domain 1: span 0f
   groups: 08 01 02 04
checking if image is initramfs...it isn't (no cpio magic); looks like an initrd
Freeing initrd memory: 416k freed
NET: Registered protocol family 16
EISA bus registered
PCI: PCI BIOS revision 2.10 entry at 0xfd976, last bus=2
PCI: Using configuration type 1
mtrr: v2.0 (20020519)
ACPI: Subsystem revision 20040715
ACPI: Interpreter disabled.
Linux Plug and Play Support v0.97 (c) Adam Belay
usbcore: registered new driver usbfs
usbcore: registered new driver hub
PCI: Probing PCI hardware
PCI: Probing PCI hardware (bus 00)
PCI: Discovered peer bus 01
PCI: Discovered peer bus 02
PCI->APIC IRQ transform: (B0,I1,P0) -> 169
PCI->APIC IRQ transform: (B0,I3,P0) -> 153
PCI->APIC IRQ transform: (B0,I4,P0) -> 169
PCI->APIC IRQ transform: (B0,I15,P0) -> 105
PCI->APIC IRQ transform: (B1,I5,P0) -> 161
PCI->APIC IRQ transform: (B1,I6,P0) -> 145
PCI->APIC IRQ transform: (B1,I6,P1) -> 145
PCI->APIC IRQ transform: (B2,I7,P0) -> 161
get_random_bytes called before random driver initialization
TC classifier action (bugs to netdev@oss.sgi.com cc hadi@cyberus.ca)
vesafb: framebuffer at 0xfd000000, mapped to 0xf8807000, size 1875k
vesafb: mode is 800x600x16, linelength=1600, pages=1
vesafb: protected mode interface info at c000:4796
vesafb: scrolling: redraw
vesafb: Truecolor: size=0:5:6:5, shift=0:11:5:0
mtrr: type mismatch for fd000000,100000 old: uncachable new: write-combining
mtrr: type mismatch for fd000000,80000 old: uncachable new: write-combining
mtrr: type mismatch for fd000000,40000 old: uncachable new: write-combining
mtrr: type mismatch for fd000000,20000 old: uncachable new: write-combining
mtrr: type mismatch for fd000000,10000 old: uncachable new: write-combining
mtrr: type mismatch for fd000000,8000 old: uncachable new: write-combining
mtrr: type mismatch for fd000000,4000 old: uncachable new: write-combining
mtrr: type mismatch for fd000000,2000 old: uncachable new: write-combining
mtrr: type mismatch for fd000000,1000 old: uncachable new: write-combining
mtrr: size and base must be multiples of 4 kiB
mtrr: size: 0x800  base: 0xfd000000
mtrr: size and base must be multiples of 4 kiB
mtrr: size: 0x400  base: 0xfd000000
mtrr: size and base must be multiples of 4 kiB
mtrr: size: 0x200  base: 0xfd000000
mtrr: size and base must be multiples of 4 kiB
mtrr: size: 0x100  base: 0xfd000000
mtrr: size and base must be multiples of 4 kiB
mtrr: size: 0x80  base: 0xfd000000
mtrr: size and base must be multiples of 4 kiB
mtrr: size: 0x40  base: 0xfd000000
mtrr: size and base must be multiples of 4 kiB
mtrr: size: 0x20  base: 0xfd000000
mtrr: size and base must be multiples of 4 kiB
mtrr: size: 0x10  base: 0xfd000000
mtrr: size and base must be multiples of 4 kiB
mtrr: size: 0x8  base: 0xfd000000
mtrr: size and base must be multiples of 4 kiB
mtrr: size: 0x4  base: 0xfd000000
mtrr: size and base must be multiples of 4 kiB
mtrr: size: 0x2  base: 0xfd000000
mtrr: size and base must be multiples of 4 kiB
mtrr: size: 0x1  base: 0xfd000000
fb0: VESA VGA frame buffer device
Machine check exception polling timer started.
Starting balanced_irq
audit: initializing netlink socket (disabled)
audit(1096036866.247:0): initialized
highmem bounce pool size: 64 pages
Total HugeTLB memory allocated, 0
VFS: Disk quotas dquot_6.5.1
Dquot-cache hash table entries: 1024 (order 0, 4096 bytes)
Initializing Cryptographic API
isapnp: Scanning for PnP cards...
isapnp: No Plug & Play device found
bootsplash 3.1.4-2004/02/19: looking for picture... no good signature found.
Console: switching to colour frame buffer device 100x37
Serial: 8250/16550 driver $Revision: 1.90 $ 54 ports, IRQ sharing enabled
ttyS0 at I/O 0x3f8 (irq = 4) is a 16550A
ttyS1 at I/O 0x2f8 (irq = 3) is a 16550A
Using anticipatory io scheduler
Floppy drive(s): fd0 is 1.44M
FDC 0 is a National Semiconductor PC87306
RAMDISK driver initialized: 16 RAM disks of 8192K size 1024 blocksize
Uniform Multi-Platform E-IDE driver Revision: 7.00alpha2
ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
SvrWks OSB4: IDE controller at PCI slot 0000:00:0f.1
SvrWks OSB4: chipset revision 0
SvrWks OSB4: not 100% native mode: will probe irqs later
    ide0: BM-DMA at 0x1880-0x1887, BIOS settings: hda:pio, hdb:pio
    ide1: BM-DMA at 0x1888-0x188f, BIOS settings: hdc:pio, hdd:pio
ohci_hcd: 2004 Feb 02 USB 1.1 'Open' Host Controller (OHCI) Driver (PCI)
ohci_hcd: block sizes: ed 64 td 64
ohci_hcd 0000:00:0f.2: ServerWorks OSB4/CSB5 OHCI USB Controller
ohci_hcd 0000:00:0f.2: irq 105, pci mem f89ea000
ohci_hcd 0000:00:0f.2: new USB bus registered, assigned bus number 1
hub 1-0:1.0: USB hub found
hub 1-0:1.0: 2 ports detected
USB Universal Host Controller Interface driver v2.2
usbcore: registered new driver hiddev
usbcore: registered new driver usbhid
drivers/usb/input/hid-core.c: v2.0:USB HID core driver
mice: PS/2 mouse device common for all mice
serio: i8042 AUX port at 0x60,0x64 irq 12
serio: i8042 KBD port at 0x60,0x64 irq 1
input: AT Translated Set 2 keyboard on isa0060/serio0
md: md driver 0.90.0 MAX_MD_DEVS=256, MD_SB_DISKS=27
EISA: Probing bus 0 at eisa0
Cannot allocate resource for EISA slot 1
Cannot allocate resource for EISA slot 2
EISA: Detected 0 cards.
NET: Registered protocol family 2
IP: routing cache hash table of 131072 buckets, 1024Kbytes
TCP: Hash tables configured (established 524288 bind 65536)
NET: Registered protocol family 1
NET: Registered protocol family 17
NET: Registered protocol family 8
NET: Registered protocol family 20
Software Suspend has malfunctioning SMP support. Disabled :(
PM: Reading pmdisk image.
PM: Resume from disk failed.
md: Autodetecting RAID arrays.
md: autorun ...
md: ... autorun DONE.
RAMDISK: Compressed image found at block 0
VFS: Mounted root (minix filesystem).
SCSI subsystem initialized
(scsi0) <Adaptec AIC-7899 Ultra 160/m SCSI host adapter> found at PCI 1/6/0
(scsi0) Wide Channel A, SCSI ID=7, 32/255 SCBs
(scsi0) Downloading sequencer code... 396 instructions downloaded
(scsi1) <Adaptec AIC-7899 Ultra 160/m SCSI host adapter> found at PCI 1/6/1
(scsi1) Wide Channel B, SCSI ID=8, 32/255 SCBs
(scsi1) Downloading sequencer code... 396 instructions downloaded
(scsi2) <Adaptec AIC-7880 Ultra SCSI host adapter> found at PCI 0/1/0
(scsi2) Wide Channel, SCSI ID=7, 16/255 SCBs
(scsi2) Downloading sequencer code... 436 instructions downloaded
scsi0 : Adaptec AHA274x/284x/294x (EISA/VLB/PCI-Fast SCSI) 5.2.6/5.2.0
       <Adaptec AIC-7899 Ultra 160/m SCSI host adapter>
scsi1 : Adaptec AHA274x/284x/294x (EISA/VLB/PCI-Fast SCSI) 5.2.6/5.2.0
       <Adaptec AIC-7899 Ultra 160/m SCSI host adapter>
(scsi1:0:1:0) message buffer busy, unable to abort.
(scsi1:0:1:0) Device reset, Message buffer in use
ohci_hcd 0000:00:0f.2: wakeup
  Vendor: SEAGATE   Model: ST318404LC        Rev: 0006
  Type:   Direct-Access                      ANSI SCSI revision: 03
(scsi1:0:1:0) Synchronous at 160.0 Mbyte/sec, offset 63.
SCSI device sda: 35843670 512-byte hdwr sectors (18352 MB)
SCSI device sda: drive cache: write back
 sda: sda1 sda2 < sda5 sda6 sda7 sda8 >
Attached scsi disk sda at scsi1, channel 0, id 1, lun 0
  Vendor: ESG-SHV   Model: SCA HSBP M9       Rev: 0.10
  Type:   Processor                          ANSI SCSI revision: 02
scsi2 : Adaptec AHA274x/284x/294x (EISA/VLB/PCI-Fast SCSI) 5.2.6/5.2.0
       <Adaptec AIC-7880 Ultra SCSI host adapter>
  Vendor: TOSHIBA   Model: CD-ROM XM-6401TA  Rev: 1001
  Type:   CD-ROM                             ANSI SCSI revision: 02
kjournald starting.  Commit interval 5 seconds
EXT3-fs: mounted filesystem with ordered data mode.
Freeing unused kernel memory: 352k freed
Real Time Clock Driver v1.12
EXT3 FS on sda1, internal journal
input: ImPS/2 Logitech Wheel Mouse on isa0060/serio1
input: PC Speaker
ts: Compaq touchscreen protocol output
Supermount version 2.0.4 for kernel 2.6
Linux agpgart interface v0.100 (c) Dave Jones
agpgart: Maximum main memory to use for agp memory: 9968M
agpgart: unable to determine aperture size.
agpgart: agp_backend_initialize() failed.
agpgart-serverworks: probe of 0000:00:00.0 failed with error -22
agpgart: Maximum main memory to use for agp memory: 9968M
agpgart: unable to determine aperture size.
agpgart: agp_backend_initialize() failed.
agpgart-serverworks: probe of 0000:00:00.1 failed with error -22
agpgart: Detected ServerWorks CNB20HE chipset: No AGP present.
agpgart: Detected ServerWorks CNB20HE chipset: No AGP present.
e100: Intel(R) PRO/100 Network Driver, 3.0.27-k2-NAPI
e100: Copyright(c) 1999-2004 Intel Corporation
e100: eth0: e100_probe: addr 0xfc201000, irq 153, MAC addr 00:03:47:1D:24:6A
e100: eth1: e100_probe: addr 0xfc202000, irq 169, MAC addr 00:D0:B7:A7:02:39
e100: eth0: e100_watchdog: link up, 10Mbps, half-duplex

--
Flávio Bruno Leitner <fbl@conectiva.com.br>
[ E74B 0BD0 5E05 C385 239E  531C BC17 D670 7FF0 A9E0 ]


-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php

Re: Memory limit / xen for Conectiva

[Ian Pratt]

> On Fri, Sep 24, 2004 at 06:47:19PM +0100, Ian Pratt wrote:
> > > (XEN) Initialized 1983MB memory (507888 pages) on a 1983MB machine.
> > 
> > Please can you send us the output of a native linux 2.4 or 2.6
> > kernel booting. Your machine must have a memory hole, and we need
> > to be smarter about parsing memory tables. You should be able to
> > get 4GB working, but we have no plans to support PAE36. We're
> > concentrating on x86_64.

>  BIOS-e820: 0000000000100000 - 000000007bff0000 (usable)
>  BIOS-e820: 0000000100000000 - 0000000280000000 (usable)

Yep, we're probably missing the second of these.  Although its
perfectly entitled to, its slightly baroque that your BIOS is
placing the IO hole at just below 2GB rather than 4GB.

Porting some more Linux table parsing code shouldn't be hard. I
guess we were going to see a machine like this sooner or later....

Ian


-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php

Re: Memory limit / xen for Conectiva

[Keir Fraser]

> > On Fri, Sep 24, 2004 at 06:47:19PM +0100, Ian Pratt wrote:
> > > > (XEN) Initialized 1983MB memory (507888 pages) on a 1983MB machine.
> > > 
> > > Please can you send us the output of a native linux 2.4 or 2.6
> > > kernel booting. Your machine must have a memory hole, and we need
> > > to be smarter about parsing memory tables. You should be able to
> > > get 4GB working, but we have no plans to support PAE36. We're
> > > concentrating on x86_64.
> 
> >  BIOS-e820: 0000000000100000 - 000000007bff0000 (usable)
> >  BIOS-e820: 0000000100000000 - 0000000280000000 (usable)
> 
> Yep, we're probably missing the second of these.  Although its
> perfectly entitled to, its slightly baroque that your BIOS is
> placing the IO hole at just below 2GB rather than 4GB.
> 
> Porting some more Linux table parsing code shouldn't be hard. I
> guess we were going to see a machine like this sooner or later....

Unfortunately this is not the case. You've got a great big memory hole
between 2GB and 4GB. So the remaining 6GB of your memory are all above
the 4GB boundary. That memory is inaccessible when not using PAE-36
mode, which we do not support.

It may be possible to tweak BIOS settings to get more of your memory
into the low 4GB. Probably it is arranged as it is so that large-ish
I/O regions such as the frame buffer can be allocated below the 4GB
boundary without overlapping with any RAM (which would then become
inaccessible). It's worth a look...

 -- Keir


-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php

Re: Memory limit / xen for Conectiva

[Flavio Leitner]

On Fri, Sep 24, 2004 at 09:04:38PM +0100, Keir Fraser wrote:
> Unfortunately this is not the case. You've got a great big memory hole
> between 2GB and 4GB. So the remaining 6GB of your memory are all above
> the 4GB boundary. That memory is inaccessible when not using PAE-36
> mode, which we do not support.

Ouch! 


> It may be possible to tweak BIOS settings to get more of your memory
> into the low 4GB. Probably it is arranged as it is so that large-ish
> I/O regions such as the frame buffer can be allocated below the 4GB
> boundary without overlapping with any RAM (which would then become
> inaccessible). It's worth a look...

There are only 2 options related with memory but no one can't fix that 
hole. I will try another machine to test xen again.

Thanks! 

-- 
Flávio Bruno Leitner <fbl@conectiva.com.br>
[ E74B 0BD0 5E05 C385 239E  531C BC17 D670 7FF0 A9E0 ]


-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php

RE: Memory limit / xen for Conectiva

[James Harper]

Is there no way to make use of this high memory? Not even as a ramdisk
or something? Or does any access at all require PAE36?

James

> -----Original Message-----
> From: xen-devel-admin@lists.sourceforge.net [mailto:xen-devel-
> admin@lists.sourceforge.net] On Behalf Of Ian Pratt
> Sent: Saturday, 25 September 2004 04:27
> To: Flavio Leitner
> Cc: Ian Pratt; xen-devel@lists.sourceforge.net; Ian.Pratt@cl.cam.ac.uk
> Subject: Re: [Xen-devel] Memory limit / xen for Conectiva
> 
> > On Fri, Sep 24, 2004 at 06:47:19PM +0100, Ian Pratt wrote:
> > > > (XEN) Initialized 1983MB memory (507888 pages) on a 1983MB
machine.
> > >
> > > Please can you send us the output of a native linux 2.4 or 2.6
> > > kernel booting. Your machine must have a memory hole, and we need
> > > to be smarter about parsing memory tables. You should be able to
> > > get 4GB working, but we have no plans to support PAE36. We're
> > > concentrating on x86_64.
> 
> >  BIOS-e820: 0000000000100000 - 000000007bff0000 (usable)
> >  BIOS-e820: 0000000100000000 - 0000000280000000 (usable)
> 
> Yep, we're probably missing the second of these.  Although its
> perfectly entitled to, its slightly baroque that your BIOS is
> placing the IO hole at just below 2GB rather than 4GB.
> 
> Porting some more Linux table parsing code shouldn't be hard. I
> guess we were going to see a machine like this sooner or later....
> 
> Ian
> 
> 
> -------------------------------------------------------
> This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
> Project Admins to receive an Apple iPod Mini FREE for your judgement
on
> who ports your project to Linux PPC the best. Sponsored by IBM.
> Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/xen-devel



-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php

Re: Memory limit / xen for Conectiva

[Keir Fraser]

> Is there no way to make use of this high memory? Not even as a ramdisk
> or something? Or does any access at all require PAE36?
> 
> James

Apart from PAE36, there is no way to make the CPU drive the four most
significant address lines. 

It's maybe something to think about in the future, but the work
required is non-trivial. We expect that people with big memory
requirements will move to x86-64 so we will concentrate on that
first. 

 -- Keir


-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php