It seems the original code doesn't pass the right arg gfp_t to decide how to allocate. Just applying this patch, fsstress will fail. So please ignore this patch, will resend later..
Thanks, Wang > From: Wang Shilong <wangsl-f...@cn.fujitsu.com> > > We have passed arg gfp_mask to tree_mod_alloc(), so > just use it rather than always use GFP_ATOMIC. > > Signed-off-by: Wang Shilong <wangsl-f...@cn.fujitsu.com> > --- > fs/btrfs/ctree.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c > index de6de8e..0e3514f 100644 > --- a/fs/btrfs/ctree.c > +++ b/fs/btrfs/ctree.c > @@ -553,7 +553,7 @@ static inline int tree_mod_alloc(struct btrfs_fs_info > *fs_info, gfp_t flags, > * once we switch from spin locks to something different, we should > * honor the flags parameter here. > */ > - tm = *tm_ret = kzalloc(sizeof(*tm), GFP_ATOMIC); > + tm = *tm_ret = kzalloc(sizeof(*tm), flags); > if (!tm) > return -ENOMEM; > > -- > 1.7.11.7 > -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
