Date: Tue, 1 Sep 1998 23:29:42 -0400 (EDT) From: Garrett Wollman <wollman@khavrinen.lcs.mit.edu> To: Stefan Bethke <stb@hanse.de> Cc: net@FreeBSD.ORG Subject: Re: Problems with MGET(m, M_WAIT, *) [was: Semantics of ...] Message-ID: <199809020329.XAA02805@khavrinen.lcs.mit.edu> In-Reply-To: <Pine.BSF.3.96.980901232001.28484I-100000@transit.hanse.de> References: <Pine.BSF.3.96.980825233251.25049C-100000@transit.hanse.de> <Pine.BSF.3.96.980901232001.28484I-100000@transit.hanse.de>
next in thread | previous in thread | raw e-mail | index | archive | help
<<On Wed, 2 Sep 1998 00:36:28 +0200 (CEST), Stefan Bethke <stb@hanse.de> said: > As Garrett said, the current mbuf allocator is going to go away. Is > it? Yes -- but it will take some time. The eventual goal is for interfaces to manage their own memory allocation -- some interfaces may have unusual requirements for memory which is being used as a buffer. (Something like the current mbuf allocator will probably remain to service those interfaces which don't have any special requirements. Even for those interfaces, however, it would be more memory-efficient to only allocate 1536 bytes per buffer (assuming Ethernet) than a full 2K cluster -- the current scheme wastes 25% on Ethernet-sized packets.) > Will the semantics of the new mbuf allocator be the same (M_WAIT may > return 0)? I guess so, but the one(s) working on the new allocator might > want to confirm that. This seems likely, although we ought to be able to do a bit better than the fixed-size map we have now. -GAWollman -- Garrett A. Wollman | O Siem / We are all family / O Siem / We're all the same wollman@lcs.mit.edu | O Siem / The fires of freedom Opinions not those of| Dance in the burning flame MIT, LCS, CRS, or NSA| - Susan Aglukark and Chad Irschick To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199809020329.XAA02805>