Date: Sat, 2 Feb 2008 21:54:16 -0800 From: "Kip Macy" <kip.macy@gmail.com> To: "Aaditya Rai" <filesync@yahoo.com> Cc: freebsd-net@freebsd.org Subject: Re: m_getjcl for contiguous jumbo allocations Message-ID: <b1fa29170802022154i415ca6afn5117d353e365e7b4@mail.gmail.com> In-Reply-To: <659520.27857.qm@web32211.mail.mud.yahoo.com> References: <659520.27857.qm@web32211.mail.mud.yahoo.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Feb 2, 2008 8:46 PM, Aaditya Rai <filesync@yahoo.com> wrote: > Hello, > > I'm using release 6.2. I need to do allocations upto > 8K which are physically contiguous. > > Is m_getjcl the right call for this? > > Also, if I use m_getjcl to do 4K allocations, after > two or three allocations I start getting cluster/mbuf > allocation failures; or weird memory corruptions. If I > change it to use MGETCL etc. and do 2K allocations, > the same code works fine. > > Is m_getjcl() stable in 6.2? Should I fall back on > implementing my own little backend contig-allocator? On 6.2 you'll have to fall back your own allocator if you are unable to chain 4k jumbo frames. In -CURRENT physically contiguous jumbo frames are the default. I'm not sure if that made it in to 7.0. -Kip
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?b1fa29170802022154i415ca6afn5117d353e365e7b4>