Re: [PATCH V2] vfs: re-implement writeback_inodes_sb(_nr)_if_idle() and rename them

2013-01-11 Thread Fengguang Wu
On Fri, Jan 11, 2013 at 09:12:11PM +0100, Jan Kara wrote: > On Thu 10-01-13 13:47:57, Miao Xie wrote: > > writeback_inodes_sb(_nr)_if_idle() is re-implemented by replacing > > down_read() > > with down_read_trylock() because > > - If ->s_umount is write locked, then the sb is not idle. That is > >

Re: [PATCH V2] vfs: re-implement writeback_inodes_sb(_nr)_if_idle() and rename them

2013-01-11 Thread Jan Kara
On Thu 10-01-13 13:47:57, Miao Xie wrote: > writeback_inodes_sb(_nr)_if_idle() is re-implemented by replacing down_read() > with down_read_trylock() because > - If ->s_umount is write locked, then the sb is not idle. That is > writeback_inodes_sb(_nr)_if_idle() needn't wait for the lock. > - writ

[PATCH V2] vfs: re-implement writeback_inodes_sb(_nr)_if_idle() and rename them

2013-01-09 Thread Miao Xie
writeback_inodes_sb(_nr)_if_idle() is re-implemented by replacing down_read() with down_read_trylock() because - If ->s_umount is write locked, then the sb is not idle. That is writeback_inodes_sb(_nr)_if_idle() needn't wait for the lock. - writeback_inodes_sb(_nr)_if_idle() grabs s_umount lock w