Hello. Could you please try the attached patch? This patch is against reiser4-for-2.6.36. Also please make sure you have the latest one (I have updated the stuff 11-Dec-2010). Report, if any problems. Thanks, Edward. On Sun, Dec 19, 2010 at 11:33 PM, Edward Shishkin <edward.shishkin@xxxxxxxxx> wrote: > ÐÐÐÑÐÐ ÐÐÐÑÐÐÐÐ wrote: >> Wanted to try reiser4. I have compiled a kernel with reiser4, format >> the partition and began to copy the folder /home. >> After a short time the screen goes black and there was one console >> with the error output. After switching to other tty and back up >> continued. dmesg output copied to the pastebin >> (http://pastebin.com/XyDKWJ8J). >> This was repeated several times and the computer hangs perfectly still. >> I have ubuntu 10.10 on a laptop Acer Aspire 7720z. > > Yup, I see the bug: > instead of writing pages reiser4 kicks ext4 to do it, > and, I guess, doesn't clear journal_info (this is why > ext4 oopses). > > [ Â278.062271] Â[<c0318822>] ? jbd2_journal_start+0x12/0x20 > [ Â278.062322] Â[<c02f5858>] ? ext4_journal_start_sb+0xc8/0x100 > [ Â278.062377] Â[<c02e0957>] ? ext4_meta_trans_blocks+0x67/0xc0 > [ Â278.062431] Â[<c02e47e3>] ? ext4_da_writepages+0x203/0x560 > [ Â278.062484] Â[<c060fccd>] ? _raw_spin_lock+0xd/0x10 > [ Â278.062530] Â[<c0281ca2>] ? flush_some_atom+0x312/0x540 > [ Â278.062578] Â[<c02a4cdd>] ? writepages_cryptcompress+0x19d/0x220 > [ Â278.062636] Â[<c02913a8>] ? reiser4_writeout+0xa8/0x1a0 > [ Â278.062684] Â[<c022c186>] ? iput+0x36/0x270 > [ Â278.062724] Â[<c02367e4>] ? generic_writeback_sb_inodes+0xe4/0x150 > [ Â278.062782] Â[<c01dfaac>] ? do_writepages+0x1c/0x40 > [ Â278.062826] Â[<c0236125>] ? writeback_single_inode+0x85/0x250 > [ Â278.062877] Â[<c060fccd>] ? _raw_spin_lock+0xd/0x10 > [ Â278.062921] Â[<c028db64>] ? reiser4_writeback_inodes+0xa4/0x130 > [ Â278.062974] Â[<c02367b1>] ? generic_writeback_sb_inodes+0xb1/0x150 > [ Â278.063031] Â[<c0236c6b>] ? writeback_inodes_wb+0x10b/0x150 > > > here should be writeback_sb_inodes() instead of writeback_inodes_wb > > > [ Â278.063082] Â[<c0291d06>] ? entd_flush+0xa6/0xe0 > [ Â278.063127] Â[<c0291d40>] ? entd+0x0/0x240 > [ Â278.063166] Â[<c0291e51>] ? entd+0x111/0x240 > [ Â278.063208] Â[<c0165fd0>] ? autoremove_wake_function+0x0/0x50 > [ Â278.064008] Â[<c0291d40>] ? entd+0x0/0x240 > [ Â278.067417] Â[<c0165ba4>] ? kthread+0x74/0x80 > [ Â278.067417] Â[<c0165b30>] ? kthread+0x0/0x80 > [ Â278.067417] Â[<c010357e>] ? kernel_thread_helper+0x6/0x10 > > > It is interesting that the path > entd -> writeback_inodes_wb -> ext4_da_writepages > has never been triggered on my testing machine (I also > have reiser4 and ext4 there). Perhaps, this is because > of different system configurations. > > I'll try to fix it in the near future, so that you'll be able to > try reiser4. > > Thanks for your report, > Edward. >
Attachment:
reiser4-fix-entd_flush.patch
Description: Binary data
