On Sat, 29 Sep 2001, Justin Erenkrantz wrote: > in CPU usage. I also know that gstein has gone record on his > advogato diary that mmap may not be the performance win we > think it may be.
Dean's been telling us that for ages... well, at least when you compare writev to sendfile. =-) > It may be worth trying to compile without > mmap and see how we perform. Would be interesting to see, but because we're talking about parsed files, I'm willing to bet it would be a loss to disable MMAP on most platforms, since we're talking about the difference between MMAPing the file and getting a single bucket out or apr_file_read()'ing in the bucket in 8KB chunks. That's potentially a lot of extra system calls and a lot of extra buckets to deal with... but it's at least worth trying. > Also, I'm guessing this is on Solaris 8? Any chance we could get the > sendfilev patch on there? (I know it is part of 7/01 and Sol9 as > well.) Of course that would only apply to non-parsed files, but yes, it would be quite interesting to see it if it's not already on there... --Cliff -------------------------------------------------------------- Cliff Woolley [EMAIL PROTECTED] Charlottesville, VA
