Caching control

A feature that would be useful is one where the level of caching can be
controlled for a descriptor. An ioctl() call should be fine. The value
given would specify the maximum number of page-size units of caching to
keep for the descriptor. This would be a _maximum_ and the kernel would
be allowed to cache less than this amount (which would happen if it does
the actual physical I/O faster than the process calls write). But if the
process does call write() more often, the kernel would prevent more from
being written by blocking that write() call.

It would be OK by me if, when that blocking would happen, and when the
descriptor is set to non-blocking mode, that the write() returns with
the EAGAIN error status. But that would not actually be necessary if the
caching can be controlled as described above.

--
|WARNING: Due to extreme spam, googlegroups.com is blocked. Due to ignorance |
| by the abuse department, bellsouth.net is blocked. If you post to |
| Usenet from these places, find another Usenet provider ASAP. |
| Phil Howard KA9WGN (email for humans: first name in lower case at ipal.net) |
.

Relevant Pages

  • Re: Caching control
    ... controlled for a descriptor. ... given would specify the maximum number of page-size units of caching to ... process does call writemore often, the kernel would prevent more from ... There's kind of an unwritten rule that when you propose a new feature, ...
    (comp.os.linux.development.system)
  • Re: Clumsy interface of epoll
    ... with a file descriptor, I conclude that this thing is already ... But it already is duplicated in user space. ... returned while deleting the fd from the epoll pollset. ... Because there is no reason for kernel space to do something that can ...
    (comp.os.linux.development.apps)
  • Re: IIS6, ASP.NET and Kernel Caching
    ... If ASP.Net or any other handler is scriptmapped to handle the resource, ... declarative kernel response caching, but only if you ask for it in the ASPX ... on the originating server ...
    (microsoft.public.inetserver.iis)
  • Re: aio_read/write versus O_NONBLOCK
    ... data buffer in the kernel associated with a particular 'readable' file ... be only the process which created the descriptor. ... filesystem is designed to be a ressource shared among all applications ... The connection is in fact a shared resource with both ...
    (comp.unix.programmer)
  • Re: how to return from a blocked read() ?
    ... It doesn't matter. ... it's blocking, it's in the read. ... That thread can use that descriptor... ... on some of your posts. ...
    (comp.os.linux.development.apps)