> On Mon, Oct 25, 2010 at 08:52:43AM +0100, Matthias Scheler wrote: > > On Sun, Oct 24, 2010 at 10:56:40PM +0000, David Holland wrote: > > > Anyway, ISTM that writing from the mmap buffer in say 64K chunks would > > > retain nearly all the advantages and get rid of the latency problem. > > > > The way the code is currently written it only uses mmap(2) for > > files smaller than 8MB anyway. Your suggested change would require > > more work than reducing the size of the mapped memory. > > Forget that, I misunderstood what you said. We can still use one > mmap(2) but should write out that memory with multile write(2) > system calls.
FWIW, bozohttpd defaults to mapping upto 64MB regions and writing out up to 64KB at a time. .mrg.
