From: Jeff Layton <jlayton@kernel.org> mainline inclusion from mainline-v7.3-rc1 commit f307b4f7ed174a3252d2f292904fd276c78c77de category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/18939 CVE: CVE-2026-89676 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?i... -------------------------------- struct nfsd4_copy served two roles: as &u->copy it is a transient per-COMPOUND argument in the request buffer; as the heap async_copy it is a durable object (worker kthread, reaper linkage, CB_OFFLOAD callback, IDR stateid) that outlives the COMPOUND, with dup_copy_fields() shuttling state between them. That dual identity was the root of the recent lifetime bugs. Introduce struct nfsd4_async_copy for the durable object. It embeds a struct nfsd4_copy (cp_copy) for the operation parameters/result and adds the durable-only fields: async_copies linkage, task_struct, refcount, reaper TTL, copy stateid, and CB_OFFLOAD callback. The durable object therefore never points into the request buffer. cp_clp stays in nfsd4_copy -- it is a request property read by the sync-copy tracepoints on the transient object. Mechanical split, no intended behavioral change; a step toward folding the copy stateids into the common nfs4_stid infrastructure. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Jeff Layton <jlayton@kernel.org> Link: https://patch.msgid.link/20260710-nfsd-testing-v3-8-a0ff7db6aa3e@kernel.org Signed-off-by: Chuck Lever <cel@kernel.org> Conflicts: fs/nfsd/nfs4proc.c fs/nfsd/nfs4state.c fs/nfsd/state.h fs/nfsd/xdr4.h [1)OLK-5.10 uses the older 5.10 async-copy model: cp_src is an embedded struct, stop state is a bool, and a standalone nfsd4_copy (cb_copy) is still allocated for CB_OFFLOAD; the durable worker fields were split out while refcount stayed in nfsd4_copy for cb_copy, a new nfs4_put_cb_copy() releases it, and the reaper/has_active/cancel_by_sb helper hunks do not exist in this baseline; 2)nfs4_free_copy_state() now takes nfsd4_async_copy and reaches cp_clp through cp_copy.cp_clp; 3)Prototypes for nfs4_init_copy_state()/nfs4_free_copy_state(), nfs4_put_copy() and find_async_copy() repointed to struct nfsd4_async_copy; 4)Only copies/copy_task/stopped/cp_stateid moved; refcount stays in nfsd4_copy for the standalone CB_OFFLOAD object; cp_nn, ss_mnt, cp_cb, cp_intra and cp_synchronous remain in nfsd4_copy] Signed-off-by: Lu Jialin <lujialin4@huawei.com> --- fs/nfsd/nfs4proc.c | 81 +++++++++++++++++++++++++++------------------ fs/nfsd/nfs4state.c | 6 ++-- fs/nfsd/state.h | 9 ++--- fs/nfsd/xdr4.h | 31 ++++++++++++++--- 4 files changed, 82 insertions(+), 45 deletions(-) diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index d9a2941086e3..67c9d0dec1ba 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -1110,31 +1110,42 @@ nfsd4_clone(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, return status; } -void nfs4_put_copy(struct nfsd4_copy *copy) +void nfs4_put_copy(struct nfsd4_async_copy *copy) { if (!refcount_dec_and_test(©->refcount)) return; /* Drop the task_struct pinned in nfsd4_copy(); NULL on error paths. */ if (copy->copy_task) put_task_struct(copy->copy_task); - atomic_dec(©->cp_nn->pending_async_copies); + atomic_dec(©->cp_copy.cp_nn->pending_async_copies); + kfree(copy); +} + +/* + * OLK-5.10 still allocates a standalone nfsd4_copy for the CB_OFFLOAD + * callback; it has no kthread or pending-copy accounting of its own. + */ +static void nfs4_put_cb_copy(struct nfsd4_copy *copy) +{ + if (!refcount_dec_and_test(©->refcount)) + return; kfree(copy); } static bool -check_and_set_stop_copy(struct nfsd4_copy *copy) +check_and_set_stop_copy(struct nfsd4_async_copy *copy) { bool value; - spin_lock(©->cp_clp->async_lock); + spin_lock(©->cp_copy.cp_clp->async_lock); value = copy->stopped; if (!copy->stopped) copy->stopped = true; - spin_unlock(©->cp_clp->async_lock); + spin_unlock(©->cp_copy.cp_clp->async_lock); return value; } -static void nfsd4_stop_copy(struct nfsd4_copy *copy) +static void nfsd4_stop_copy(struct nfsd4_async_copy *copy) { /* only 1 thread should stop the copy */ if (!check_and_set_stop_copy(copy)) @@ -1142,14 +1153,14 @@ static void nfsd4_stop_copy(struct nfsd4_copy *copy) nfs4_put_copy(copy); } -static struct nfsd4_copy *nfsd4_get_copy(struct nfs4_client *clp) +static struct nfsd4_async_copy *nfsd4_get_copy(struct nfs4_client *clp) { - struct nfsd4_copy *copy = NULL; + struct nfsd4_async_copy *copy = NULL; spin_lock(&clp->async_lock); if (!list_empty(&clp->async_copies)) { - copy = list_first_entry(&clp->async_copies, struct nfsd4_copy, - copies); + copy = list_first_entry(&clp->async_copies, + struct nfsd4_async_copy, copies); refcount_inc(©->refcount); } spin_unlock(&clp->async_lock); @@ -1158,7 +1169,7 @@ static struct nfsd4_copy *nfsd4_get_copy(struct nfs4_client *clp) void nfsd4_shutdown_copy(struct nfs4_client *clp) { - struct nfsd4_copy *copy; + struct nfsd4_async_copy *copy; while ((copy = nfsd4_get_copy(clp)) != NULL) nfsd4_stop_copy(copy); @@ -1356,7 +1367,7 @@ static void nfsd4_cb_offload_release(struct nfsd4_callback *cb) { struct nfsd4_copy *copy = container_of(cb, struct nfsd4_copy, cp_cb); - nfs4_put_copy(copy); + nfs4_put_cb_copy(copy); } static int nfsd4_cb_offload_done(struct nfsd4_callback *cb, @@ -1440,28 +1451,28 @@ static void dup_copy_fields(struct nfsd4_copy *src, struct nfsd4_copy *dst) if (src->cp_intra) /* for inter, file_src doesn't exist yet */ dst->nf_src = nfsd_file_get(src->nf_src); - memcpy(&dst->cp_stateid, &src->cp_stateid, sizeof(src->cp_stateid)); memcpy(&dst->cp_src, &src->cp_src, sizeof(struct nl4_server)); memcpy(&dst->stateid, &src->stateid, sizeof(src->stateid)); memcpy(&dst->c_fh, &src->c_fh, sizeof(src->c_fh)); dst->ss_mnt = src->ss_mnt; } -static void cleanup_async_copy(struct nfsd4_copy *copy) +static void cleanup_async_copy(struct nfsd4_async_copy *copy) { nfs4_free_copy_state(copy); - nfsd_file_put(copy->nf_dst); - if (copy->cp_intra) - nfsd_file_put(copy->nf_src); - spin_lock(©->cp_clp->async_lock); + nfsd_file_put(copy->cp_copy.nf_dst); + if (copy->cp_copy.cp_intra) + nfsd_file_put(copy->cp_copy.nf_src); + spin_lock(©->cp_copy.cp_clp->async_lock); list_del(©->copies); - spin_unlock(©->cp_clp->async_lock); + spin_unlock(©->cp_copy.cp_clp->async_lock); nfs4_put_copy(copy); } static int nfsd4_do_async_copy(void *data) { - struct nfsd4_copy *copy = (struct nfsd4_copy *)data; + struct nfsd4_async_copy *async = data; + struct nfsd4_copy *copy = &async->cp_copy; struct nfsd4_copy *cb_copy; if (!copy->cp_intra) { /* Inter server SSC */ @@ -1496,7 +1507,7 @@ static int nfsd4_do_async_copy(void *data) out: if (!copy->cp_intra) kfree(copy->nf_src); - cleanup_async_copy(copy); + cleanup_async_copy(async); return 0; } @@ -1506,7 +1517,7 @@ nfsd4_copy(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, { struct nfsd4_copy *copy = &u->copy; __be32 status; - struct nfsd4_copy *async_copy = NULL; + struct nfsd4_async_copy *async_copy = NULL; if (!copy->cp_intra) { /* Inter server SSC */ if (!inter_copy_offload_enable || copy->cp_synchronous) { @@ -1536,21 +1547,25 @@ nfsd4_copy(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, struct nfsd_net *nn = net_generic(SVC_NET(rqstp), nfsd_net_id); struct task_struct *task; - async_copy = kzalloc(sizeof(struct nfsd4_copy), GFP_KERNEL); + async_copy = kzalloc(sizeof(struct nfsd4_async_copy), GFP_KERNEL); if (!async_copy) goto out_err; - async_copy->cp_nn = nn; + async_copy->cp_copy.cp_nn = nn; refcount_set(&async_copy->refcount, 1); /* Arbitrary cap on number of pending async copy operations */ if (atomic_inc_return(&nn->pending_async_copies) > (int)rqstp->rq_pool->sp_nrthreads) goto out_err; - dup_copy_fields(copy, async_copy); if (!nfs4_init_copy_state(nn, async_copy)) goto out_err; memcpy(©->cp_res.cb_stateid, &async_copy->cp_stateid.stid, sizeof(copy->cp_res.cb_stateid)); + /* + * dup after writing cb_stateid; duplicating first would leave + * the callback stateid zeroed. + */ + dup_copy_fields(copy, &async_copy->cp_copy); task = kthread_create(nfsd4_do_async_copy, async_copy, "%s", "copy thread"); if (IS_ERR(task)) @@ -1561,10 +1576,10 @@ nfsd4_copy(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, */ get_task_struct(task); async_copy->copy_task = task; - spin_lock(&async_copy->cp_clp->async_lock); + spin_lock(&async_copy->cp_copy.cp_clp->async_lock); list_add(&async_copy->copies, - &async_copy->cp_clp->async_copies); - spin_unlock(&async_copy->cp_clp->async_lock); + &async_copy->cp_copy.cp_clp->async_copies); + spin_unlock(&async_copy->cp_copy.cp_clp->async_lock); wake_up_process(async_copy->copy_task); status = nfs_ok; } else { @@ -1583,10 +1598,10 @@ nfsd4_copy(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, goto out; } -struct nfsd4_copy * +struct nfsd4_async_copy * find_async_copy(struct nfs4_client *clp, stateid_t *stateid) { - struct nfsd4_copy *copy; + struct nfsd4_async_copy *copy; spin_lock(&clp->async_lock); list_for_each_entry(copy, &clp->async_copies, copies) { @@ -1606,7 +1621,7 @@ nfsd4_offload_cancel(struct svc_rqst *rqstp, union nfsd4_op_u *u) { struct nfsd4_offload_status *os = &u->offload_status; - struct nfsd4_copy *copy; + struct nfsd4_async_copy *copy; struct nfs4_client *clp = cstate->clp; copy = find_async_copy(clp, &os->stateid); @@ -1695,12 +1710,12 @@ nfsd4_offload_status(struct svc_rqst *rqstp, { struct nfsd4_offload_status *os = &u->offload_status; __be32 status = 0; - struct nfsd4_copy *copy; + struct nfsd4_async_copy *copy; struct nfs4_client *clp = cstate->clp; copy = find_async_copy(clp, &os->stateid); if (copy) { - os->count = copy->cp_res.wr_bytes_written; + os->count = copy->cp_copy.cp_res.wr_bytes_written; nfs4_put_copy(copy); } else status = nfserr_bad_stateid; diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 3e1aeb08ea6b..f97dbb2c093b 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -820,7 +820,7 @@ static int nfs4_init_cp_state(struct nfsd_net *nn, copy_stateid_t *stid, return 1; } -int nfs4_init_copy_state(struct nfsd_net *nn, struct nfsd4_copy *copy) +int nfs4_init_copy_state(struct nfsd_net *nn, struct nfsd4_async_copy *copy) { return nfs4_init_cp_state(nn, ©->cp_stateid, NFS4_COPY_STID); } @@ -854,12 +854,12 @@ struct nfs4_cpntf_state *nfs4_alloc_init_cpntf_state(struct nfsd_net *nn, return NULL; } -void nfs4_free_copy_state(struct nfsd4_copy *copy) +void nfs4_free_copy_state(struct nfsd4_async_copy *copy) { struct nfsd_net *nn; WARN_ON_ONCE(copy->cp_stateid.sc_type != NFS4_COPY_STID); - nn = net_generic(copy->cp_clp->net, nfsd_net_id); + nn = net_generic(copy->cp_copy.cp_clp->net, nfsd_net_id); spin_lock(&nn->s2s_cp_lock); idr_remove(&nn->s2s_cp_stateids, copy->cp_stateid.stid.si_opaque.so_id); diff --git a/fs/nfsd/state.h b/fs/nfsd/state.h index a16e3d6fcbc0..6c080c63e880 100644 --- a/fs/nfsd/state.h +++ b/fs/nfsd/state.h @@ -636,6 +636,7 @@ struct nfsd4_blocked_lock { struct nfsd4_compound_state; struct nfsd_net; struct nfsd4_copy; +struct nfsd4_async_copy; extern __be32 nfs4_preprocess_stateid_op(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, struct svc_fh *fhp, @@ -646,8 +647,8 @@ __be32 nfsd4_lookup_stateid(struct nfsd4_compound_state *cstate, struct nfs4_stid **s, struct nfsd_net *nn); struct nfs4_stid *nfs4_alloc_stid(struct nfs4_client *cl, struct kmem_cache *slab, void (*sc_free)(struct nfs4_stid *)); -int nfs4_init_copy_state(struct nfsd_net *nn, struct nfsd4_copy *copy); -void nfs4_free_copy_state(struct nfsd4_copy *copy); +int nfs4_init_copy_state(struct nfsd_net *nn, struct nfsd4_async_copy *copy); +void nfs4_free_copy_state(struct nfsd4_async_copy *copy); struct nfs4_cpntf_state *nfs4_alloc_init_cpntf_state(struct nfsd_net *nn, struct nfs4_stid *p_stid); void nfs4_unhash_stid(struct nfs4_stid *s); @@ -676,8 +677,8 @@ extern bool nfs4_has_reclaimed_state(struct xdr_netobj name, struct nfsd_net *nn struct nfs4_file *find_file(struct knfsd_fh *fh); void put_nfs4_file(struct nfs4_file *fi); -extern void nfs4_put_copy(struct nfsd4_copy *copy); -extern struct nfsd4_copy * +extern void nfs4_put_copy(struct nfsd4_async_copy *copy); +extern struct nfsd4_async_copy * find_async_copy(struct nfs4_client *clp, stateid_t *staetid); extern void nfs4_put_cpntf_state(struct nfsd_net *nn, struct nfs4_cpntf_state *cps); diff --git a/fs/nfsd/xdr4.h b/fs/nfsd/xdr4.h index 1ced060704a3..03607083945a 100644 --- a/fs/nfsd/xdr4.h +++ b/fs/nfsd/xdr4.h @@ -569,18 +569,39 @@ struct nfsd4_copy { struct nfsd_file *nf_src; struct nfsd_file *nf_dst; - copy_stateid_t cp_stateid; - - struct list_head copies; - struct task_struct *copy_task; + /* + * Kept in nfsd4_copy because OLK-5.10 still allocates a separate + * nfsd4_copy for the CB_OFFLOAD callback (cb_copy in + * nfsd4_do_async_copy()) and that object is released through this + * refcount. + */ refcount_t refcount; - bool stopped; struct vfsmount *ss_mnt; struct nfs_fh c_fh; nfs4_stateid stateid; struct nfsd_net *cp_nn; }; + +/* + * Durable state for an async (background) server-side COPY. + * + * struct nfsd4_copy is transient: it lives in the COMPOUND argument buffer + * and is reused once the op returns. An async COPY outlives the COMPOUND + * (worker kthread, async_copies linkage), so its params and result are + * snapshotted into the embedded cp_copy and it never points into the request + * buffer. + */ +struct nfsd4_async_copy { + struct nfsd4_copy cp_copy; /* operation params + result */ + + struct list_head copies; /* nfs4_client.async_copies */ + struct task_struct *copy_task; + refcount_t refcount; + bool stopped; + copy_stateid_t cp_stateid; /* s2s_cp_stateids IDR entry */ +}; + extern bool inter_copy_offload_enable; struct nfsd4_seek { -- 2.34.1