From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Fri, 14 Jun 2002 10:10:35 -0700 From: Tom Rini To: Dan Malek Cc: linuxppc-embedded@lists.linuxppc.org, Paul Mackerras , David Gibson Subject: Re: consistent_free() Message-ID: <20020614171035.GM13541@opus.bloom.county> References: <20020614042928.GK26146@zax> <20020614153954.GL13541@opus.bloom.county> <3D0A1D53.6040205@embeddededge.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <3D0A1D53.6040205@embeddededge.com> Sender: owner-linuxppc-embedded@lists.linuxppc.org List-Id: On Fri, Jun 14, 2002 at 12:44:03PM -0400, Dan Malek wrote: [snip] > The newer ARM version only allocates the number of pages necessary to > cover the request. The old version, and the one in PowerPC right now, > would just use __get_free_pages, which was a power-of-two allocator. > You could end up wasting lots of memory if you weren't allocating power of > two sized buffers. There were a couple of implementation iterations > to make this work, and I guess this is where ARM ended up. I think > one version really wanted that page struct, but now it doesn't really > matter. I ended up noticing that myself, and have things almost right, I think (patch shortly, if so..) > >.... Dan? Can we get some more insight > >into the workings of your mind? :) > > I've been watching the ARM updates and plan to move their changes into > the PowerPC sources. I don't know why they don't call vfree() directly, > I thought it did the proper clean up. I'm not sure either. __iounmap on ARM just is vfree() with the casts. -- Tom Rini (TR1265) http://gate.crashing.org/~trini/ ** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/