Skip to contentSkip to conversation

flush_cache_page

DISCUSSION Jorgen Lundman 2 participants 3 messages Jan 13, 2008 – Jan 14, 2008 MIPS

lore
Share & tools
lore
#0 lore raw

Due to cache coherence bugs, Fuse has an extra call to work around it;

        flush_cache_page(vma, cs->addr, page_to_pfn(cs->pg));

But my kernel (2.6.15 for mips 4KEc Tangox board) does not have a flush_cache_page().

If I use kangox_flush_all() Fuse works rather well, but the performance is abysmal. Can I simulate this call using one of the calls I do have;

__flush_dcache_page
flush_data_cache_page
tangox_flush_cache_all
cache_flush
kc_flush_cache

Or alternatively, does anyone have the source for flush_cache_page() for said CPU?

Signature · 4 lines
Jorgen Lundman       | <[email protected]>
Unix Administrator   | +81 (0)3 -5456-2687 ext 1017 (work)
Shibuya-ku, Tokyo    | +81 (0)90-5578-8500          (cell)
Japan                | +81 (0)3 -3375-1767          (home)
#1 lore raw

Man, 2.6.15 is like 2-3 years old....

Quoted 10 lines · Jorgen Lundman

Due to cache coherence bugs, Fuse has an extra call to work around it;

       flush_cache_page(vma, cs->addr, page_to_pfn(cs->pg));

But my kernel (2.6.15 for mips 4KEc Tangox board) does not have a flush_cache_page().

If I use kangox_flush_all() Fuse works rather well, but the performance is abysmal. Can I simulate this call using one of the calls I do have;

Quoted 16 lines

__flush_dcache_page
flush_data_cache_page
tangox_flush_cache_all
cache_flush
kc_flush_cache

Or alternatively, does anyone have the source for flush_cache_page() for said CPU?

Signature · 4 lines
Jorgen Lundman       | <[email protected]>
Unix Administrator   | +81 (0)3 -5456-2687 ext 1017 (work)
Shibuya-ku, Tokyo    | +81 (0)90-5578-8500          (cell)
Japan                | +81 (0)3 -3375-1767          (home)

_______________________________________

Mr Markus Gothe
Software Engineer

Phone: +46 (0)13 21 81 20 (ext. 1046)
Fax: +46 (0)13 21 21 15
Mobile: +46 (0)73 718 72 80 Diskettgatan 11, SE-583 35 Linköping, Sweden www.27m.com

#2 lore raw
Quoted 1 line · Markus Gothe

Man, 2.6.15 is like 2-3 years old....

Thank you, that is very useful. And if you can let me know how I do the signature, and which AES keys I need to encrypt the kernel with to be able to boot it on the SMP8634 I could upgrade.

But for now, I can not change what the device runs, only try to work around its limitations.

Lund

Quoted 42 lines · Jorgen Lundman
Quoted 11 lines

Due to cache coherence bugs, Fuse has an extra call to work around it;

       flush_cache_page(vma, cs->addr, page_to_pfn(cs->pg));

But my kernel (2.6.15 for mips 4KEc Tangox board) does not have a flush_cache_page().

If I use kangox_flush_all() Fuse works rather well, but the performance is abysmal. Can I simulate this call using one of the calls I do have;

Quoted 16 lines

__flush_dcache_page
flush_data_cache_page
tangox_flush_cache_all
cache_flush
kc_flush_cache

Or alternatively, does anyone have the source for flush_cache_page() for said CPU?

Signature · 4 lines
Jorgen Lundman       | <[email protected] <mailto:[email protected]>>
Unix Administrator   | +81 (0)3 -5456-2687 ext 1017 (work)
Shibuya-ku, Tokyo    | +81 (0)90-5578-8500          (cell)
Japan                | +81 (0)3 -3375-1767          (home)

_______________________________________

Mr Markus Gothe
Software Engineer

Phone: +46 (0)13 21 81 20 (ext. 1046)
Fax: +46 (0)13 21 21 15
Mobile: +46 (0)73 718 72 80 Diskettgatan 11, SE-583 35 Linköping, Sweden www.27m.com <http://www.27m.com>

Signature · 4 lines
Jorgen Lundman       | <[email protected]>
Unix Administrator   | +81 (0)3 -5456-2687 ext 1017 (work)
Shibuya-ku, Tokyo    | +81 (0)90-5578-8500          (cell)
Japan                | +81 (0)3 -3375-1767          (home)