Date: Mon, 24 Mar 2014 16:03:04 +0400 From: Dmitry Sivachenko <trtrmitya@gmail.com> To: =?utf-8?Q?Trond_Endrest=C3=B8l?= <Trond.Endrestol@fagskolen.gjovik.no> Cc: hackers@freebsd.org Subject: Re: madvise() vs posix_fadvise() Message-ID: <C5489EF2-34D8-412C-88AC-476120D3F1F4@gmail.com> In-Reply-To: <alpine.BSF.2.00.1403211725140.56113@mail.fig.ol.no> References: <D6BD48AF-9522-495D-8D54-37854E53C272@gmail.com> <alpine.BSF.2.00.1403211725140.56113@mail.fig.ol.no>
next in thread | previous in thread | raw e-mail | index | archive | help
On 21 =D0=BC=D0=B0=D1=80=D1=82=D0=B0 2014 =D0=B3., at 20:27, Trond = Endrest=C3=B8l <Trond.Endrestol@fagskolen.gjovik.no> wrote: > On Fri, 21 Mar 2014 18:56+0400, Dmitry Sivachenko wrote: >=20 >> Hello! >>=20 >> I have a program which uses large data files (read-only, via mmap()). >>=20 >> These machines have a bit more RAM that these files occupy, so it is=20= >> possible to have all these data in memory. >>=20 >> What techniques should I use to promote this data not to be purged=20 >> from RAM: >>=20 >> -- madvise(MADV_WILLNEED) >> -- posix_fadvise(POSIX_FADV_WILLNEED) >> -- both? >=20 > Although a bit dangerous, mlock(2) might be your ticket. That system=20= > call prevents your memory region from being swapped/paged away from=20 > physical memory. >=20 I know about mlock(2), it is a bit overkill. Can someone please explain the difference between madvise(MADV_WILLNEED) = and posix_fadvise(POSIX_FADV_WILLNEED)?
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?C5489EF2-34D8-412C-88AC-476120D3F1F4>