diff options
author | Jens Axboe <axboe@kernel.dk> | 2025-09-18 14:16:53 -0600 |
---|---|---|
committer | Greg Kroah-Hartman <gregkh@linuxfoundation.org> | 2025-09-25 11:13:48 +0200 |
commit | 93e9d0293d3b4c247873c19ffe9426154bb82b9e (patch) | |
tree | f9b6c1db6045dcc5d79a309f4b1920d10236e610 | |
parent | e6b2b4a0ffd856b5f0291b14ccd2f83319c15ef8 (diff) |
io_uring/msg_ring: kill alloc_cache for io_kiocb allocations
Commit df8922afc37aa2111ca79a216653a629146763ad upstream.
A recent commit:
fc582cd26e88 ("io_uring/msg_ring: ensure io_kiocb freeing is deferred for RCU")
fixed an issue with not deferring freeing of io_kiocb structs that
msg_ring allocates to after the current RCU grace period. But this only
covers requests that don't end up in the allocation cache. If a request
goes into the alloc cache, it can get reused before it is sane to do so.
A recent syzbot report would seem to indicate that there's something
there, however it may very well just be because of the KASAN poisoning
that the alloc_cache handles manually.
Rather than attempt to make the alloc_cache sane for that use case, just
drop the usage of the alloc_cache for msg_ring request payload data.
Fixes: 50cf5f3842af ("io_uring/msg_ring: add an alloc cache for io_kiocb entries")
Link: https://lore.kernel.org/io-uring/68cc2687.050a0220.139b6.0005.GAE@google.com/
Reported-by: syzbot+baa2e0f4e02df602583e@syzkaller.appspotmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-rw-r--r-- | include/linux/io_uring_types.h | 3 | ||||
-rw-r--r-- | io_uring/io_uring.c | 5 | ||||
-rw-r--r-- | io_uring/msg_ring.c | 24 |
3 files changed, 2 insertions, 30 deletions
diff --git a/include/linux/io_uring_types.h b/include/linux/io_uring_types.h index 42fab0b501f2..f18f50d39598 100644 --- a/include/linux/io_uring_types.h +++ b/include/linux/io_uring_types.h @@ -400,9 +400,6 @@ struct io_ring_ctx { struct callback_head poll_wq_task_work; struct list_head defer_list; - struct io_alloc_cache msg_cache; - spinlock_t msg_lock; - #ifdef CONFIG_NET_RX_BUSY_POLL struct list_head napi_list; /* track busy poll napi_id */ spinlock_t napi_lock; /* napi_list lock */ diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c index cbca97d9d74b..68439eb0dc8f 100644 --- a/io_uring/io_uring.c +++ b/io_uring/io_uring.c @@ -316,9 +316,6 @@ static __cold struct io_ring_ctx *io_ring_ctx_alloc(struct io_uring_params *p) sizeof(struct io_async_rw)); ret |= io_alloc_cache_init(&ctx->uring_cache, IO_ALLOC_CACHE_MAX, sizeof(struct uring_cache)); - spin_lock_init(&ctx->msg_lock); - ret |= io_alloc_cache_init(&ctx->msg_cache, IO_ALLOC_CACHE_MAX, - sizeof(struct io_kiocb)); ret |= io_futex_cache_init(ctx); if (ret) goto free_ref; @@ -358,7 +355,6 @@ err: io_alloc_cache_free(&ctx->netmsg_cache, io_netmsg_cache_free); io_alloc_cache_free(&ctx->rw_cache, io_rw_cache_free); io_alloc_cache_free(&ctx->uring_cache, kfree); - io_alloc_cache_free(&ctx->msg_cache, io_msg_cache_free); io_futex_cache_free(ctx); kfree(ctx->cancel_table.hbs); kfree(ctx->cancel_table_locked.hbs); @@ -2743,7 +2739,6 @@ static __cold void io_ring_ctx_free(struct io_ring_ctx *ctx) io_alloc_cache_free(&ctx->netmsg_cache, io_netmsg_cache_free); io_alloc_cache_free(&ctx->rw_cache, io_rw_cache_free); io_alloc_cache_free(&ctx->uring_cache, kfree); - io_alloc_cache_free(&ctx->msg_cache, io_msg_cache_free); io_futex_cache_free(ctx); io_destroy_buffers(ctx); mutex_unlock(&ctx->uring_lock); diff --git a/io_uring/msg_ring.c b/io_uring/msg_ring.c index b68e009bce21..97708e5132bc 100644 --- a/io_uring/msg_ring.c +++ b/io_uring/msg_ring.c @@ -11,7 +11,6 @@ #include "io_uring.h" #include "rsrc.h" #include "filetable.h" -#include "alloc_cache.h" #include "msg_ring.h" /* All valid masks for MSG_RING */ @@ -76,13 +75,7 @@ static void io_msg_tw_complete(struct io_kiocb *req, struct io_tw_state *ts) struct io_ring_ctx *ctx = req->ctx; io_add_aux_cqe(ctx, req->cqe.user_data, req->cqe.res, req->cqe.flags); - if (spin_trylock(&ctx->msg_lock)) { - if (io_alloc_cache_put(&ctx->msg_cache, req)) - req = NULL; - spin_unlock(&ctx->msg_lock); - } - if (req) - kfree_rcu(req, rcu_head); + kfree_rcu(req, rcu_head); percpu_ref_put(&ctx->refs); } @@ -104,19 +97,6 @@ static int io_msg_remote_post(struct io_ring_ctx *ctx, struct io_kiocb *req, return 0; } -static struct io_kiocb *io_msg_get_kiocb(struct io_ring_ctx *ctx) -{ - struct io_kiocb *req = NULL; - - if (spin_trylock(&ctx->msg_lock)) { - req = io_alloc_cache_get(&ctx->msg_cache); - spin_unlock(&ctx->msg_lock); - if (req) - return req; - } - return kmem_cache_alloc(req_cachep, GFP_KERNEL | __GFP_NOWARN | __GFP_ZERO); -} - static int io_msg_data_remote(struct io_kiocb *req) { struct io_ring_ctx *target_ctx = req->file->private_data; @@ -124,7 +104,7 @@ static int io_msg_data_remote(struct io_kiocb *req) struct io_kiocb *target; u32 flags = 0; - target = io_msg_get_kiocb(req->ctx); + target = kmem_cache_alloc(req_cachep, GFP_KERNEL | __GFP_NOWARN | __GFP_ZERO); if (unlikely(!target)) return -ENOMEM; |