On Tue, Jul 2, 2013 at 1:55 AM, Jeff Davis <pgsql@j-davis.com> wrote:
> On Sun, 2013-06-30 at 18:55 -0400, Greg Smith wrote:
>> This makes platform level testing a lot easier, thanks. Attached is an
>> updated copy of that program with some error checking. If the files it
>> creates already existed, the code didn't notice, and a series of write
>> errors happened. If you set the test up right it's not a problem, but
>> it's better if a bad setup is caught. I wrapped the whole test with a
>> shell script, also attached, which insures the right test sequence and
>> checks.
>
> Thank you.
>
>> 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.
>
> So your theory is that it may be slower because there are twice as many
> syscalls (one per 4K page rather than one per 8K page)? Interesting
> observation.
>
>> This is what I plan to benchmark specifically next.
>
> In the interest of keeping this patch moving forward, do you have an
> estimate for when this testing will be complete?
>
>> If the
>> posix_fallocate approach is actually slower than what's done now when
>> it's not getting kernel acceleration, which is the case on RHEL5 era
>> kernels, we might need to make the configure time test more complicated.
>> Whether posix_fallocate is defined isn't sensitive enough; on Linux it
>> may be the case that this only is usable when fallocate() is also there.
>
> I'd say that if posix_fallocate is slower than the existing code on
> pretty much any platform, we shouldn't commit the patch at all.
Even in that case, if a user can easily know which platform posix_fallocate
should be used in, we can commit the patch with the configurable GUC
parameter.
Regards,
--
Fujii Masao
