]> git.karo-electronics.de Git - karo-tx-linux.git/commit
include/linux/fs.h: disable preempt when acquire i_size_seqcount write lock
authorFan Du <fan.du@windriver.com>
Thu, 7 Feb 2013 01:27:18 +0000 (12:27 +1100)
committerStephen Rothwell <sfr@canb.auug.org.au>
Mon, 18 Feb 2013 05:46:59 +0000 (16:46 +1100)
commit3981a1cf257e9250299fb1329387e20dfc10edfb
tree6a1949db6302aa6a9da98cc8aeb684949abc825e
parentbe8c011a4813fb8cb8eb5a65c422aa910bb18908
include/linux/fs.h: disable preempt when acquire i_size_seqcount write lock

Two rt tasks bind to one CPU core.

The higher priority rt task A preempts a lower priority rt task B which
has already taken the write seq lock, and then the higher priority rt task
A try to acquire read seq lock, it's doomed to lockup.

rt task A with lower priority: call write
i_size_write                                        rt task B with higher priority: call sync, and preempt task A
  write_seqcount_begin(&inode->i_size_seqcount);    i_size_read
  inode->i_size = i_size;                             read_seqcount_begin <-- lockup here...

So disable preempt when acquiring every i_size_seqcount *write* lock will
cure the problem.

Signed-off-by: Fan Du <fan.du@windriver.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
include/linux/fs.h