Re: [PATCH] x86: 46 bit PAE support




* Pavel Machek <pavel@xxxxxx> wrote:

On Wed 2009-05-06 08:20:59, Rik van Riel wrote:
H. Peter Anvin wrote:
Rik van Riel wrote:
Testing: booted it on an x86-64 system with 6GB RAM. Did you really think
I had access to a system with 64TB of RAM? :)

No, but it would be good if we could test it under Qemu or KVM with an
appropriately set up sparse memory map.

I don't have a system with 1TB either, which is how much space
the memmap[] would take...

Do we really have 1 byte overhead per 64 bytes of RAM?
Pavel

Yes, struct page is ~64 bytes, and 64*64 == 4096.

Alas, it's not a problem: my suggestion wasnt to simulate 64 TB of
RAM. My suggestion was to create a sparse physical memory map (in a
virtual machine) that spreads ~1GB of RAM all around the 64 TB
physical address space. That will test whether the kernel is able to
map and work with such physical addresses. (which will cover most of
the issues)

A good look at /debug/x86/dump_pagetables with such a system booted
up would be nice as well - to make sure every virtual memory range
is in its proper area, and that there's enough free space around
them.

Ingo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/



Relevant Pages

  • Re: After executing MDCreateThread data abort occurs
    ... This is almost always RAM configuration, memory map, or cache problems. ... target.The code under Startup and KernelStart routines are running fine ... So the Kernel is not letting this address 0xc2000000 to be programmed. ...
    (microsoft.public.windowsce.embedded)
  • Re: RAM Question
    ... It is rather obvious the OP looked at some kind reading to determine what he ... missed the fact that when he had 512mgs RAM, his system used ... Do yourself a big favor, and mainly for others seeking help here, don't ... > see your memory map. ...
    (microsoft.public.windowsxp.general)
  • Re: Linux for Asymmetric Multi Processing Systems.
    ... the ARM7 Memory Map through a _Shared RAM_. ... the Memory Map of ARM7 processors is visible to ARM9 ...
    (Linux-Kernel)
  • Re: Linux on ARM: bootloader & remapping
    ... MR> U-Boot already supports many arm926-based boards. ... My concern is how to define a memory map and consequently a valid TEXT_BASE ... NOR flash, RAM is mapped to 0x1000_0000. ...
    (comp.os.linux.development.system)
  • Re: [PATCH] x86: 46 bit PAE support
    ... my suggestion wasnt to simulate 64 TB of ... My suggestion was to create a sparse physical memory map (in a ... virtual machine) that spreads ~1GB of RAM all around the 64 TB ...
    (Linux-Kernel)