From: Yang Erkun <yangerkun@huawei.com> mainline inclusion from mainline-v6.13-rc1 commit be8f982c369c965faffa198b46060f8853e0f1f0 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/IBEAMQ CVE: CVE-2024-56558 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?i... -------------------------------- The function `e_show` was called with protection from RCU. This only ensures that `exp` will not be freed. Therefore, the reference count for `exp` can drop to zero, which will trigger a refcount use-after-free warning when `exp_get` is called. To resolve this issue, use `cache_get_rcu` to ensure that `exp` remains active. ------------[ cut here ]------------ refcount_t: addition on 0; use-after-free. WARNING: CPU: 3 PID: 819 at lib/refcount.c:25 refcount_warn_saturate+0xb1/0x120 CPU: 3 UID: 0 PID: 819 Comm: cat Not tainted 6.12.0-rc3+ #1 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.1-2.fc37 04/01/2014 RIP: 0010:refcount_warn_saturate+0xb1/0x120 ... Call Trace: <TASK> e_show+0x20b/0x230 [nfsd] seq_read_iter+0x589/0x770 seq_read+0x1e5/0x270 vfs_read+0x125/0x530 ksys_read+0xc1/0x160 do_syscall_64+0x5f/0x170 entry_SYSCALL_64_after_hwframe+0x76/0x7e Fixes: bf18f163e89c ("NFSD: Using exp_get for export getting") Cc: stable@vger.kernel.org # 4.20+ Signed-off-by: Yang Erkun <yangerkun@huawei.com> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Li Lingfeng <lilingfeng@huaweicloud.com> --- fs/nfsd/export.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/fs/nfsd/export.c b/fs/nfsd/export.c index 21e404e7cb68..265f00d3ed25 100644 --- a/fs/nfsd/export.c +++ b/fs/nfsd/export.c @@ -1246,9 +1246,12 @@ static int e_show(struct seq_file *m, void *p) return 0; } - exp_get(exp); + if (!cache_get_rcu(&exp->h)) + return 0; + if (cache_check(cd, &exp->h, NULL)) return 0; + exp_put(exp); return svc_export_show(m, cd, cp); } -- 2.39.2

From: Yang Erkun <yangerkun@huawei.com> mainline inclusion from mainline-v6.13-rc1 commit f8c989a0c89a75d30f899a7cabdc14d72522bb8d category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/IBEAER CVE: CVE-2024-53216 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?i... -------------------------------- The last reference for `cache_head` can be reduced to zero in `c_show` and `e_show`(using `rcu_read_lock` and `rcu_read_unlock`). Consequently, `svc_export_put` and `expkey_put` will be invoked, leading to two issues: 1. The `svc_export_put` will directly free ex_uuid. However, `e_show`/`c_show` will access `ex_uuid` after `cache_put`, which can trigger a use-after-free issue, shown below. ================================================================== BUG: KASAN: slab-use-after-free in svc_export_show+0x362/0x430 [nfsd] Read of size 1 at addr ff11000010fdc120 by task cat/870 CPU: 1 UID: 0 PID: 870 Comm: cat Not tainted 6.12.0-rc3+ #1 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.1-2.fc37 04/01/2014 Call Trace: <TASK> dump_stack_lvl+0x53/0x70 print_address_description.constprop.0+0x2c/0x3a0 print_report+0xb9/0x280 kasan_report+0xae/0xe0 svc_export_show+0x362/0x430 [nfsd] c_show+0x161/0x390 [sunrpc] seq_read_iter+0x589/0x770 seq_read+0x1e5/0x270 proc_reg_read+0xe1/0x140 vfs_read+0x125/0x530 ksys_read+0xc1/0x160 do_syscall_64+0x5f/0x170 entry_SYSCALL_64_after_hwframe+0x76/0x7e Allocated by task 830: kasan_save_stack+0x20/0x40 kasan_save_track+0x14/0x30 __kasan_kmalloc+0x8f/0xa0 __kmalloc_node_track_caller_noprof+0x1bc/0x400 kmemdup_noprof+0x22/0x50 svc_export_parse+0x8a9/0xb80 [nfsd] cache_do_downcall+0x71/0xa0 [sunrpc] cache_write_procfs+0x8e/0xd0 [sunrpc] proc_reg_write+0xe1/0x140 vfs_write+0x1a5/0x6d0 ksys_write+0xc1/0x160 do_syscall_64+0x5f/0x170 entry_SYSCALL_64_after_hwframe+0x76/0x7e Freed by task 868: kasan_save_stack+0x20/0x40 kasan_save_track+0x14/0x30 kasan_save_free_info+0x3b/0x60 __kasan_slab_free+0x37/0x50 kfree+0xf3/0x3e0 svc_export_put+0x87/0xb0 [nfsd] cache_purge+0x17f/0x1f0 [sunrpc] nfsd_destroy_serv+0x226/0x2d0 [nfsd] nfsd_svc+0x125/0x1e0 [nfsd] write_threads+0x16a/0x2a0 [nfsd] nfsctl_transaction_write+0x74/0xa0 [nfsd] vfs_write+0x1a5/0x6d0 ksys_write+0xc1/0x160 do_syscall_64+0x5f/0x170 entry_SYSCALL_64_after_hwframe+0x76/0x7e 2. We cannot sleep while using `rcu_read_lock`/`rcu_read_unlock`. However, `svc_export_put`/`expkey_put` will call path_put, which subsequently triggers a sleeping operation due to the following `dput`. ============================= WARNING: suspicious RCU usage 5.10.0-dirty #141 Not tainted ----------------------------- ... Call Trace: dump_stack+0x9a/0xd0 ___might_sleep+0x231/0x240 dput+0x39/0x600 path_put+0x1b/0x30 svc_export_put+0x17/0x80 e_show+0x1c9/0x200 seq_read_iter+0x63f/0x7c0 seq_read+0x226/0x2d0 vfs_read+0x113/0x2c0 ksys_read+0xc9/0x170 do_syscall_64+0x33/0x40 entry_SYSCALL_64_after_hwframe+0x67/0xd1 Fix these issues by using `rcu_work` to help release `svc_expkey`/`svc_export`. This approach allows for an asynchronous context to invoke `path_put` and also facilitates the freeing of `uuid/exp/key` after an RCU grace period. Fixes: 9ceddd9da134 ("knfsd: Allow lockless lookups of the exports") Signed-off-by: Yang Erkun <yangerkun@huawei.com> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Conflicts: fs/nfsd/export.c [Commit 20ad856e4732 ("nfsd: report per-export stats") add definition of export_stats_destroy before svc_export_put and add ex_stats in struct svc_export.] Signed-off-by: Li Lingfeng <lilingfeng@huaweicloud.com> --- fs/nfsd/export.c | 31 +++++++++++++++++++++++++------ fs/nfsd/export.h | 4 ++-- 2 files changed, 27 insertions(+), 8 deletions(-) diff --git a/fs/nfsd/export.c b/fs/nfsd/export.c index 265f00d3ed25..de29a472e39b 100644 --- a/fs/nfsd/export.c +++ b/fs/nfsd/export.c @@ -40,15 +40,24 @@ #define EXPKEY_HASHMAX (1 << EXPKEY_HASHBITS) #define EXPKEY_HASHMASK (EXPKEY_HASHMAX -1) -static void expkey_put(struct kref *ref) +static void expkey_put_work(struct work_struct *work) { - struct svc_expkey *key = container_of(ref, struct svc_expkey, h.ref); + struct svc_expkey *key = + container_of(to_rcu_work(work), struct svc_expkey, ek_rcu_work); if (test_bit(CACHE_VALID, &key->h.flags) && !test_bit(CACHE_NEGATIVE, &key->h.flags)) path_put(&key->ek_path); auth_domain_put(key->ek_client); - kfree_rcu(key, ek_rcu); + kfree(key); +} + +static void expkey_put(struct kref *ref) +{ + struct svc_expkey *key = container_of(ref, struct svc_expkey, h.ref); + + INIT_RCU_WORK(&key->ek_rcu_work, expkey_put_work); + queue_rcu_work(system_wq, &key->ek_rcu_work); } static int expkey_upcall(struct cache_detail *cd, struct cache_head *h) @@ -331,14 +340,24 @@ static void nfsd4_fslocs_free(struct nfsd4_fs_locations *fsloc) fsloc->locations = NULL; } -static void svc_export_put(struct kref *ref) +static void svc_export_put_work(struct work_struct *work) { - struct svc_export *exp = container_of(ref, struct svc_export, h.ref); + struct svc_export *exp = + container_of(to_rcu_work(work), struct svc_export, ex_rcu_work); + path_put(&exp->ex_path); auth_domain_put(exp->ex_client); nfsd4_fslocs_free(&exp->ex_fslocs); kfree(exp->ex_uuid); - kfree_rcu(exp, ex_rcu); + kfree(exp); +} + +static void svc_export_put(struct kref *ref) +{ + struct svc_export *exp = container_of(ref, struct svc_export, h.ref); + + INIT_RCU_WORK(&exp->ex_rcu_work, svc_export_put_work); + queue_rcu_work(system_wq, &exp->ex_rcu_work); } static int svc_export_upcall(struct cache_detail *cd, struct cache_head *h) diff --git a/fs/nfsd/export.h b/fs/nfsd/export.h index e7daa1f246f0..ec2823d17d98 100644 --- a/fs/nfsd/export.h +++ b/fs/nfsd/export.h @@ -61,7 +61,7 @@ struct svc_export { u32 ex_layout_types; struct nfsd4_deviceid_map *ex_devid_map; struct cache_detail *cd; - struct rcu_head ex_rcu; + struct rcu_work ex_rcu_work; }; /* an "export key" (expkey) maps a filehandlefragement to an @@ -76,7 +76,7 @@ struct svc_expkey { u32 ek_fsid[6]; struct path ek_path; - struct rcu_head ek_rcu; + struct rcu_work ek_rcu_work; }; #define EX_ISSYNC(exp) (!((exp)->ex_flags & NFSEXP_ASYNC)) -- 2.39.2

From: Yang Erkun <yangerkun@huawei.com> mainline inclusion from mainline-v6.14-rc1 commit 2f55dbe4e2072c9e99298c6f37473778a98c9107 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/IBEAER CVE: CVE-2024-53216 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?i... -------------------------------- This is a prepare patch to add cache_check_rcu, will use it with follow patch. Suggested-by: NeilBrown <neilb@suse.de> Signed-off-by: Yang Erkun <yangerkun@huawei.com> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Li Lingfeng <lilingfeng@huaweicloud.com> --- include/linux/sunrpc/cache.h | 2 ++ net/sunrpc/cache.c | 41 +++++++++++++++++++++++------------- 2 files changed, 28 insertions(+), 15 deletions(-) diff --git a/include/linux/sunrpc/cache.h b/include/linux/sunrpc/cache.h index 5298d6dd61db..e55969b0cf87 100644 --- a/include/linux/sunrpc/cache.h +++ b/include/linux/sunrpc/cache.h @@ -222,6 +222,8 @@ static inline bool cache_is_expired(struct cache_detail *detail, struct cache_he return detail->flush_time >= h->last_refresh; } +extern int cache_check_rcu(struct cache_detail *detail, + struct cache_head *h, struct cache_req *rqstp); extern int cache_check(struct cache_detail *detail, struct cache_head *h, struct cache_req *rqstp); extern void cache_flush(void); diff --git a/net/sunrpc/cache.c b/net/sunrpc/cache.c index 522e43f66ecd..904c66b9a668 100644 --- a/net/sunrpc/cache.c +++ b/net/sunrpc/cache.c @@ -279,21 +279,7 @@ static int try_to_negate_entry(struct cache_detail *detail, struct cache_head *h return rv; } -/* - * This is the generic cache management routine for all - * the authentication caches. - * It checks the currency of a cache item and will (later) - * initiate an upcall to fill it if needed. - * - * - * Returns 0 if the cache_head can be used, or cache_puts it and returns - * -EAGAIN if upcall is pending and request has been queued - * -ETIMEDOUT if upcall failed or request could not be queue or - * upcall completed but item is still invalid (implying that - * the cache item has been replaced with a newer one). - * -ENOENT if cache entry was negative - */ -int cache_check(struct cache_detail *detail, +int cache_check_rcu(struct cache_detail *detail, struct cache_head *h, struct cache_req *rqstp) { int rv; @@ -334,6 +320,31 @@ int cache_check(struct cache_detail *detail, rv = -ETIMEDOUT; } } + + return rv; +} +EXPORT_SYMBOL_GPL(cache_check_rcu); + +/* + * This is the generic cache management routine for all + * the authentication caches. + * It checks the currency of a cache item and will (later) + * initiate an upcall to fill it if needed. + * + * + * Returns 0 if the cache_head can be used, or cache_puts it and returns + * -EAGAIN if upcall is pending and request has been queued + * -ETIMEDOUT if upcall failed or request could not be queue or + * upcall completed but item is still invalid (implying that + * the cache item has been replaced with a newer one). + * -ENOENT if cache entry was negative + */ +int cache_check(struct cache_detail *detail, + struct cache_head *h, struct cache_req *rqstp) +{ + int rv; + + rv = cache_check_rcu(detail, h, rqstp); if (rv) cache_put(h, detail); return rv; -- 2.39.2

From: Yang Erkun <yangerkun@huawei.com> mainline inclusion from mainline-v6.14-rc1 commit 1b10f0b603c066d81327c163a23c19f01e112366 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/IBEAER CVE: CVE-2024-53216 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?i... -------------------------------- rcu_read_lock/rcu_read_unlock has already provide protection for the pointer we will reference when we call c_show. Therefore, there is no need to obtain a cache reference to help protect cache_head. Additionally, the .put such as expkey_put/svc_export_put will invoke dput, which can sleep and break rcu. Stop get cache reference to fix them all. Fixes: ae74136b4bb6 ("SUNRPC: Allow cache lookups to use RCU protection rather than the r/w spinlock") Suggested-by: NeilBrown <neilb@suse.de> Signed-off-by: Yang Erkun <yangerkun@huawei.com> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Li Lingfeng <lilingfeng@huaweicloud.com> --- net/sunrpc/cache.c | 12 +++--------- 1 file changed, 3 insertions(+), 9 deletions(-) diff --git a/net/sunrpc/cache.c b/net/sunrpc/cache.c index 904c66b9a668..4b4ad9e52e69 100644 --- a/net/sunrpc/cache.c +++ b/net/sunrpc/cache.c @@ -1445,17 +1445,11 @@ static int c_show(struct seq_file *m, void *p) seq_printf(m, "# expiry=%lld refcnt=%d flags=%lx\n", convert_to_wallclock(cp->expiry_time), kref_read(&cp->ref), cp->flags); - if (!cache_get_rcu(cp)) - return 0; - if (cache_check(cd, cp, NULL)) - /* cache_check does a cache_put on failure */ + if (cache_check_rcu(cd, cp, NULL)) + seq_puts(m, "# "); + else if (cache_is_expired(cd, cp)) seq_puts(m, "# "); - else { - if (cache_is_expired(cd, cp)) - seq_puts(m, "# "); - cache_put(cp, cd); - } return cd->cache_show(m, cd, cp); } -- 2.39.2