[lkml]   [1998]   [Jun]   [25]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: Thread implementations...
: caddr_t buf = mmap(0, len, PROT_READ, MAP_FILE | MAP_SHARED, ifd, 0);
: write(ofd, buf, len);
: If we make the write work directly from user space (if we need
: to lock down the memory for DMA, we can put any process/thread
: that tries to unmap it to sleep) and get rid of copying for
: mmap, then we are down to the minimum number of copies.


: > mmap() does have significant overhead
: CPU overhead. The question is whether that is the bottleneck.

In the for what it is worth department, you are venturing into charted
waters. SGI has done all of these tricks and more for years and I think
we can learn from their experiences. A few points:

1) Your model above still does a copy. So the cold cache numbers can
never be faster than 1/4 of memory speed: DMA in, copy, DMA out.
The best numbers are 1/2 memory speed: DMA in, DMA out.

2) On SGI's, for server type of operations, the mmap() is the bottleneck.
You are setting up and tearing down a virtual mapping that you don't
need: the ``currency'' you are dealing in at both ends is physical
pages, not virtual pages. This starts to become a bottleneck for
files smaller than 8K (Linux) or 32K (most other operating systems).
Linux is better because it is lighter.

3) You can read my writeup of how I thought I/O ought to be done after
doing it at SGI for a while, it's in (also

It's not what I'd call detailed, but it is interesting reading and I'd
love to discuss the idea with you at length. Stephen Tweedie and I
started to talk about it at Linux Expo; maybe he has some thoughts
he cares to share?


To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to

 \ /
  Last update: 2005-03-22 13:43    [W:0.114 / U:0.432 seconds]
©2003-2017 Jasper Spaans. hosted at Digital OceanAdvertise on this site