Re: [00/17] Large Blocksize Support V3

Andy Whitcroft wrote:
To my knowledge, moving to a higher base page size has its advantages in
TLB reach, but brings with it some pretty serious downsides. Especially
in caching small files. Internal fragmentation in the page cache
significantly affecting system performance. So much so that development
is ongoing to see if supporting sub-base-page objects in the buffer
cache could be beneficial.

On Fri, Apr 27, 2007 at 01:08:17AM +1000, Nick Piggin wrote:
I think 16K would be pretty reasonable (ia64 tends to use it). I guess
powerpc went to 64k either because that's what databases want or because
their TLB refills are too slow, so the internal fragmentation bites
them a lot harder.
But that was more of a side comment, because I still think io controllers
should be easily capable of operation on 4K pages. Graphics cards are,
aren't they?

Base page size need only be a power-of-two multiple of the MMU pagesize.
ISTR the 64KB PAGE_SIZE in ppc64 being motivated by TLB considerations.

There is not going to be a way to improve how well IO controllers deal
with 4K pages. The scatter/gather lists for smaller fragments will
require more device communication overhead no matter what, and
furthermore, the more discontiguity, the more seekiness.


-- wli
-
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

  • cant start v440 after upgrading 512 Mb to 1 Gb RAM
    ... 0>DMMU TLB DATA RAM Access ... 0>Init mmu regs ... 0>Setup L2 Cache ... 1>Quick Block Mem Test ...
    (comp.unix.solaris)
  • Re: -mmX 4G patches feedback [numbers: how much performance impact]
    ... most of the first chart supports my measurements. ... the cache size of the l2 is 512k, that's the point where slowing down walking ... > effect of the TLB flush has to be on the order of 100-200 usecs. ... with page layout effects you mean different page coloring I assume, ...
    (Linux-Kernel)
  • Re: -mmX 4G patches feedback [numbers: how much performance impact]
    ... i assumed a 1MB cache. ... > note that it's not only the tlb flush having the cost, ... this is what i called the 'secondary' cost of the TLB flush. ... the pagetables, necessiating 128 cachemisses. ...
    (Linux-Kernel)
  • Re: RISC outside of SPECCPU
    ... >Transactional workloads have ememse code and data footprints. ... >The TLBs for these RISC machines were rather large--typically 64 ... and software reloaded the TLB from the hash table ... >with data in the cache, and the TLB reload software contended with code ...
    (comp.arch)
  • Re: Can context switches be faster?
    ... But the bulk of the mapping ... lookups will actually occur in a cache of the virtual memory mappings ... called the translation lookaside buffer (TLB). ... with address space switches originate. ...
    (Linux-Kernel)