Date: Wed, 10 May 2006 09:30:27 +0930 From: "Daniel O'Connor" <doconnor@gsoft.com.au> To: freebsd-hackers@freebsd.org Cc: Bharma Ji <bharmaji@gmail.com> Subject: Re: differences between M_CACHE, M_DEVBUF, M_TEMP Message-ID: <200605100930.28030.doconnor@gsoft.com.au> In-Reply-To: <67beabb0605081559j3904ada3vb5a952cc3195ae44@mail.gmail.com> References: <67beabb0605081559j3904ada3vb5a952cc3195ae44@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
--nextPart1261166.GfGITZO878 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline On Tuesday 09 May 2006 08:29, Bharma Ji wrote: > I am trying to understand the difference in these three different memorie= s. > Code comments in kern_malloc.c says that M_DEVBUF should be used for devi= ce > driver. I didn't however see any major difference between the three > memories. Can a device theoretically take up memory from M_TEMP (or > M_CACHE) segment? Similary can device drivers using M_DEVBUF also allocate > from M_CACHE / M_TEMP if needed. The M_* stuff is just a categorisation for memory allocations. Run vmstat -= m=20 to see what I mean. AFAIK the actual take has no bearing on the performance of the allocation=20 itself. =2D-=20 Daniel O'Connor software and network engineer for Genesis Software - http://www.gsoft.com.au "The nice thing about standards is that there are so many of them to choose from." -- Andrew Tanenbaum GPG Fingerprint - 5596 B766 97C0 0E94 4347 295E E593 DC20 7B3F CE8C --nextPart1261166.GfGITZO878 Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2.2 (FreeBSD) iD8DBQBEYS0c5ZPcIHs/zowRAgqvAKCScKFU0oVNRhKz4vXyjYmwTMoq1ACggD+6 Ce75GuHX/jjxAsvdoB09OXI= =xlfU -----END PGP SIGNATURE----- --nextPart1261166.GfGITZO878--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200605100930.28030.doconnor>