> > What about a zero-copy read-interface? > > An ioctl (or something) which enables the kernel to do dma directly to > > the userspace. Of course this should be limited to the root-user or a > > user with special capabilities (rights) since if a drive screws up, data > > from a different sector (or so) might end up in the proces' memory. Of > > course copying a sector from kernel- to userspace can be done pretty > > fast but i.m.h.o. all possible speedimprovements should be made unless > > unclean. > Just use mmap(). Unlike your proposal, it cooperates with the page > cache. Doesn't that one also use copying? I've also heard that using mmap is expensive due to pagefaulting. I've found, for example, that copying a 1.3GB file using read/write instead of mmap & memcpy is seconds faster. Folkert van Heusden -- Auto te koop, zie: http://www.vanheusden.com/daihatsu.php ---------------------------------------------------------------------- Get your PGP/GPG key signed at www.biglumber.com! ---------------------------------------------------------------------- Phone: +31-6-41278122, PGP-key: 1F28D8AE, www.vanheusden.com
Attachment:
signature.asc
Description: Digital signature
- Follow-Ups:
- Re: zero-copy read() interface
- From: Andrew Morton <[email protected]>
- Re: zero-copy read() interface
- From: dean gaudet <[email protected]>
- Re: zero-copy read() interface
- From: Jörn Engel <[email protected]>
- Re: zero-copy read() interface
- References:
- zero-copy read() interface
- From: Folkert van Heusden <[email protected]>
- Re: zero-copy read() interface
- From: Jörn Engel <[email protected]>
- zero-copy read() interface
- Prev by Date: Re: FPU-intensive programs crashing with floating point exception on Cyrix MII
- Next by Date: Re: Flash erase groups and filesystems
- Previous by thread: Re: zero-copy read() interface
- Next by thread: Re: zero-copy read() interface
- Index(es):
