From mboxrd@z Thu Jan 1 00:00:00 1970 From: Theodore Ts'o Subject: [PATCH 0/5] Break up "fall back to vmalloc() for large allocations" Date: Mon, 1 Aug 2011 09:13:13 -0400 Message-ID: <1312204398-12460-1-git-send-email-tytso@mit.edu> References: Cc: Theodore Ts'o To: Ext4 Developers List Return-path: Received: from li9-11.members.linode.com ([67.18.176.11]:40451 "EHLO test.thunk.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751841Ab1HANN0 (ORCPT ); Mon, 1 Aug 2011 09:13:26 -0400 In-Reply-To: Sender: linux-ext4-owner@vger.kernel.org List-ID: This patch series breaks up Andreas and Yu Jian's patch "fall back to vmalloc() for large allocations" into 5 patches, with the following changes: 1) I created new helper functions, ext4_kvmalloc(), ext4_kzalloc, and ext4_kvfree() to simplify the code. 2) Fixed the patch so that online resize would work correctly if s_group_info and s_group_desc were allocated using vmalloc(). (Yeah, there are other reasons why online resize won't work for such large allocations, but let's not add to the problems; also, there's always the possibility that kmalloc might fail for even for a small allocation.) 3) Fixed many more places in mballoc.c where ext4_msg() would be better than just printk(). I was waiting for Andreas and/or Yu Jian to resubmit, but since they didn't, and I wanted these fixes, I decided to break up the patch myself. - Ted Theodore Ts'o (3): ext4: introduce ext4_kvmalloc(), ext4_kzalloc(), and ext4_kvfree() ext4: use ext4_kvzalloc()/ext4_kvmalloc() for s_group_desc and s_group_info ext4: use ext4_msg() instead of printk in mballoc Yu Jian (2): ext4: use EXT4_BAD_INO for buddy cache to avoid colliding with valid inode # ext4: prevent memory leaks from ext4_mb_init_backend() on error path fs/ext4/ext4.h | 3 ++ fs/ext4/mballoc.c | 103 +++++++++++++++++++++++++++++------------------------ fs/ext4/resize.c | 13 ++++--- fs/ext4/super.c | 63 +++++++++++++++++++++----------- 4 files changed, 107 insertions(+), 75 deletions(-) -- 1.7.4.1.22.gec8e1.dirty