This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: Seeking consensus on BZ 16734
- From: OndÅej BÃlka <neleai at seznam dot cz>
- To: Andreas Schwab <schwab at suse dot de>
- Cc: Rich Felker <dalias at libc dot org>, Paul Pluzhnikov <ppluzhnikov at google dot com>, "H.J. Lu" <hjl dot tools at gmail dot com>, GLIBC Devel <libc-alpha at sourceware dot org>, Daniel Colascione <dancol at dancol dot org>, Andrew Pinski <pinskia at gmail dot com>
- Date: Mon, 2 Feb 2015 13:16:16 +0100
- Subject: Re: Seeking consensus on BZ 16734
- Authentication-results: sourceware.org; auth=none
- References: <CALoOobP_7jpdZUqSFmKCTFds6t8TTdnxfOfg2jCTr_TjvU+t2w at mail dot gmail dot com> <CAMe9rOrp6jCuPe4ZX-kdHdO_4_k-Dpf7ha-PxtCJmJLnL3K0-A at mail dot gmail dot com> <CALoOobMZFx7c+i0GCFRg1-1Z=2H3xDDH8+td-D=0k9muAFvPAA at mail dot gmail dot com> <20150202051410 dot GG23507 at brightrain dot aerifal dot cx> <mvma90wipim dot fsf at hawking dot suse dot de>
On Mon, Feb 02, 2015 at 11:20:01AM +0100, Andreas Schwab wrote:
> Rich Felker <dalias@libc.org> writes:
>
> > The original idea seems to be some misguided idea that read/write
> > should perform better with a page-aligned buffer. I can't make any
> > sense of this believe except in the case of the O_DIRECT silliness
> > Linux supports; normally IO is going to be memcpy to/from fs cache
> > buffers and there's no reason to expect page alignment to make that
> > faster.
>
> Page aligned copies from the page cache should be faster.
>
That is false, processor cache is divided into cache lines which are
generaly 64 byte long, not pages so aligning does not give speedup.
It could be harmful for some processors, if they use hash function that
conflicts when cache lines are multiple of page appart causing
performance loss.