| Message ID | 20201118111531.455814-1-yukuai3@huawei.com (mailing list archive) |
|---|---|
| State | Accepted, archived |
| Headers | show |
| Series | xfs: return corresponding errcode if xfs_initialize_perag() fail | expand |
On Wed, Nov 18, 2020 at 07:15:31PM +0800, Yu Kuai wrote: > In xfs_initialize_perag(), if kmem_zalloc(), xfs_buf_hash_init(), or > radix_tree_preload() failed, the returned value 'error' is not set > accordingly. > > Fixes: commit 8b26c5825e02 ("xfs: handle ENOMEM correctly during initialisation of perag structures") /me suspects you really want Fixes: 9b2471797942 ("xfs: cache unlinked pointers in an rhashtable") but otherwise this is legit. Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com> --D > Reported-by: Hulk Robot <hulkci@huawei.com> > Signed-off-by: Yu Kuai <yukuai3@huawei.com> > --- > fs/xfs/xfs_mount.c | 11 ++++++++--- > 1 file changed, 8 insertions(+), 3 deletions(-) > > diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c > index 150ee5cb8645..7110507a2b6b 100644 > --- a/fs/xfs/xfs_mount.c > +++ b/fs/xfs/xfs_mount.c > @@ -194,20 +194,25 @@ xfs_initialize_perag( > } > > pag = kmem_zalloc(sizeof(*pag), KM_MAYFAIL); > - if (!pag) > + if (!pag) { > + error = -ENOMEM; > goto out_unwind_new_pags; > + } > pag->pag_agno = index; > pag->pag_mount = mp; > spin_lock_init(&pag->pag_ici_lock); > INIT_RADIX_TREE(&pag->pag_ici_root, GFP_ATOMIC); > - if (xfs_buf_hash_init(pag)) > + > + error = xfs_buf_hash_init(pag); > + if (error) > goto out_free_pag; > init_waitqueue_head(&pag->pagb_wait); > spin_lock_init(&pag->pagb_lock); > pag->pagb_count = 0; > pag->pagb_tree = RB_ROOT; > > - if (radix_tree_preload(GFP_NOFS)) > + error = radix_tree_preload(GFP_NOFS); > + if (error) > goto out_hash_destroy; > > spin_lock(&mp->m_perag_lock); > -- > 2.25.4 >
diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index 150ee5cb8645..7110507a2b6b 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -194,20 +194,25 @@ xfs_initialize_perag( } pag = kmem_zalloc(sizeof(*pag), KM_MAYFAIL); - if (!pag) + if (!pag) { + error = -ENOMEM; goto out_unwind_new_pags; + } pag->pag_agno = index; pag->pag_mount = mp; spin_lock_init(&pag->pag_ici_lock); INIT_RADIX_TREE(&pag->pag_ici_root, GFP_ATOMIC); - if (xfs_buf_hash_init(pag)) + + error = xfs_buf_hash_init(pag); + if (error) goto out_free_pag; init_waitqueue_head(&pag->pagb_wait); spin_lock_init(&pag->pagb_lock); pag->pagb_count = 0; pag->pagb_tree = RB_ROOT; - if (radix_tree_preload(GFP_NOFS)) + error = radix_tree_preload(GFP_NOFS); + if (error) goto out_hash_destroy; spin_lock(&mp->m_perag_lock);
In xfs_initialize_perag(), if kmem_zalloc(), xfs_buf_hash_init(), or radix_tree_preload() failed, the returned value 'error' is not set accordingly. Fixes: commit 8b26c5825e02 ("xfs: handle ENOMEM correctly during initialisation of perag structures") Reported-by: Hulk Robot <hulkci@huawei.com> Signed-off-by: Yu Kuai <yukuai3@huawei.com> --- fs/xfs/xfs_mount.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-)