On Sun, Jun 30, 2013 at 5:55 PM, Greg Smith <greg@2ndquadrant.com> wrote:
>
>
> pwrite(4, "\0", 1, 16769023) = 1
> pwrite(4, "\0", 1, 16773119) = 1
> pwrite(4, "\0", 1, 16777215) = 1
>
> That's glibc helpfully converting your call to posix_fallocate into small
> writes, because the OS doesn't provide a better way in that kernel. It's
> not hard to imagine this being slower than what the WAL code is doing right
> now. I'm not worried about correctness issues anymore, but my gut paranoia
> about this not working as expected on older systems was justified. Everyone
> who thought I was just whining owes me a cookie.
I had noted in the very early part of the thread that glibc emulates
posix_fallocate when the (Linux-specific) 'fallocate' systemcall
fails. In this case, it's writing 4 bytes of zeros and then
essentially seeking forward 4092 (4096-4) bytes. This prevents files
with holes in them because the holes have to be at least 4kiB in size,
if I recall properly. It's *not* writing out 16MiB in 4 byte
increments.
--
Jon