From mboxrd@z Thu Jan 1 00:00:00 1970 From: Edward Shishkin Subject: Reiser4 for Linux-3.5 Date: Sat, 08 Sep 2012 22:51:26 +0200 Message-ID: <504BAFCE.7@gmail.com> Mime-Version: 1.0 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Sender: reiserfs-devel-owner@vger.kernel.org List-ID: Content-Type: text/plain; charset="windows-1252"; format="flowed" To: ReiserFS Development List Cc: Ivan Shapovalov , =?UTF-8?B?TWFyY2luIEJhY3p5xYRz?= =?UTF-8?B?a2k=?= Hello everyone. Please, find at http://sourceforge.net/projects/reiser4/files/reiser4-for-linux-3.x/ Changes since 2.6.39: Marcin Baczy=C5=84ski : =2E Port reiser4-for-2.6.39 to 3.2 Ivan Shapovalov : =2E Adjust reiser4-for-3.2 to 3.5; =2E Use umode_t as file mode type; =2E show_options() takes dentry; =2E Use d_make_root() instead of d_alloc_root(); =2E Use clear_inode() instead of end_writeback(); =2E Adjust to changes in encode_fh() API; =2E lock ->i_mutex in reiser4_sync_file_common(); =2E Introduce INODE_{INC,DROP,CLEAR,SET}_NLINK macros, which use library function for i_nlink manipulations; =2E Do not make reiser4_permission_common() bail out on rcu-walk; =2E Don't clear nlink in reiser4's do_create_vfs_child(). Edward Shishkin : =2E Reserve space in reiser4_write_begin() properly. Fix deadlock (reiser4 lock ordering violation): release page lock in reiser4_dirty_inode() before acquiring a longterm lock; Current status The bug in the balancing primitive INSERT_FLOW (which led to inconsistent key ordering) is not yet fixed. So the following features can lead to data corruption: =2E tail packing in the default file plugin; =2E transparent compression. =46sck will fix everything, however, it can be rather annoying to check production systems frequently.. Thanks, Edward. -- To unsubscribe from this list: send the line "unsubscribe reiserfs-deve= l" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html