From mboxrd@z Thu Jan 1 00:00:00 1970 From: Leandro Motta Barros Subject: Freeing boot memory Date: Tue, 16 Sep 2003 18:17:37 -0300 MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200309161817.37802.lmb@exatas.unisinos.br> Sender: owner-linux-mm@kvack.org Return-Path: To: linux-mm@kvack.org Cc: sisopiii-l@cscience.org List-ID: Hello, I and a colleague are studying the VM subsystem (actually this is the first time we are examining the Linux source code more closely) and have a question or two. Well, the questions concern the boot memory allocator. To be more precise, We're interested in the memory deallocation routines. We have seen that it is only possible to free full pages. So, theoretically, if we make several allocations smaller than one page, we will not be able to actually free this memory. I just don't know of this kind of situation happens in real life. Do we currently have some pages of memory "wasted" because the boot memory allocator was not able to free small allocations? Is there any estimate (or benchmark or whatever) on the number of pages that could be freed but are not? We have interest in hacking a little bit in the VM, and we thought that trying to find out ways to avoid this problem (if this is really a problem) could be nice. Do you have any thoughts about this? Thanks a lot, LMB -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: aart@kvack.org