Hi Vikas,
ARM 11 supports the following pages
• 16MB super sections
• 1MB sections
• 64KB large pages
• 4KB small pages.
while in ARM9
1MB (sections),
64KB (large pages),
4KB (small pages), and
1KB (tiny pages)
Regards,
Zaheer.
Vikash Kumar wrote:
Thanks Sandeep.. thanks Mulyadi.. thanks Lin.. This clears
the confusion for x86.
However I have one more question.. I have seen PAGE_SIZE being changed
on PPC.. sizes of 16K, 32k, 64k, 128k can be achieved.. but can it be
done on ARM architecture?
Because in our case target can be ARM as well.
Regards,
-Vikash
On Tue, May 12, 2009 at 7:47 AM, Pei Lin <telent997@xxxxxxxxx>
wrote:
yeah,
i think the page size refer to hardware mechanism.
reference as "Intel® 64 and IA-32 Architectures
Software Developer’s Manual
Volume 3A:
System Programming Guide, Part 1"
"When paging is used, the processor divides the linear address space
into fixed-size pages (of 4 KBytes, 2 MBytes, or 4 MBytes in length)
that can be mapped into physical
memory and/or disk storage. When a program (or task) references a
logical address in memory, the processor translates the address into a
linear address and then uses its paging mechanism to translate the
linear address into a corresponding physical address."
"The PSE flag enables large page sizes: 4-MByte pages or 2-MByte pages
(when the PAE flag is set). When the PSE flag is clear, the more
common page length of 4 KBytes is used. See Section 3.7.2, “Linear
Address Translation (4-MByte Pages)”, Section 3.8.3, “Linear Address
Translation With PAE Enabled (2-MByte Pages)”, and Section 3.9,
“36-Bit Physical Addressing Using the PSE-36 Paging Mechanism” for
more information about the use of the PSE flag."
"The TLBs store the most recently used page-directory and page-table
entries. They speed up memory accesses when paging is enabled by
reducing the number of memory accesses that are required to read the
page tables stored in system memory. The TLBs are divided into four
groups: instruction TLBs for 4-KByte pages, data TLBs for 4-KByte
pages; instruction TLBs for large pages (2-MByte or 4-MByte pages),
and data TLBs for large pages. The TLBs are normally active only in
protected mode with paging enabled. When paging is disabled or the
processor is in real-address mode, the TLBs maintain their contents
until explicitly or implicitly flushed"
i hope this will be helpful.
BRs,
Lin
2009/5/11 Mulyadi Santosa <mulyadi.santosa@xxxxxxxxx>:
> On Mon, May 11, 2009 at 9:53 PM, Sandeep K
Sinha
> < sandeepksinha@xxxxxxxxx>
wrote:
>> Does x86 support 8K pages???
>>
>> IMO, No.
>>
>> On Mon, May 11, 2009 at 8:18 PM, Vikash Kumar < vikashskumar@xxxxxxxxx>
>> wrote:
>>>
>>> Dear all,
>>>
>>> I am trying to increase the default page size (by default
4k) for x86. As
>>> I understand I need to modify the PAGE_SHIFT macro in
>>> "include/asm-x86/page.h" to achieve this. I changed the
default value of 12
>>> to 13 (to make the page size 8K). But when I build the
kernel and try to
>>> boot using it, I get crash
>
> Adding what Sandeep has said, 4K is default page size. Other than
> that, you're left with 2MB or 4MB page size ...but AFAIK you need
to
> turn on PSE (page size extension). And probably hugetlb too.
>
> regards,
>
> Mulyadi.
>
> --
> To unsubscribe from this list: send an email with
> "unsubscribe kernelnewbies" to ecartis@xxxxxxxxxxxx
> Please read the FAQ at http://kernelnewbies.org/FAQ
>
>
The information contained in this electronic message and any attachments to this message are intended for the exclusive
use of the addressee(s) and may contain proprietary, confidential or privileged information. If you are not the intended
recipient, you should not disseminate, distribute or copy this e-mail. Please notify the sender immediately and destroy
all copies of this message and any attachments contained in it.
|
