From e1654f413fe08ffbc3292d8d2b8958b2cc5cb5e8 Mon Sep 17 00:00:00 2001 From: Luis Chamberlain Date: Mon, 27 Sep 2021 14:59:58 -0700 Subject: nbd: add error handling support for add_disk() We never checked for errors on add_disk() as this function returned void. Now that this is fixed, use the shiny new error handling. Reviewed-by: Christoph Hellwig Reviewed-by: Hannes Reinecke Signed-off-by: Luis Chamberlain Signed-off-by: Jens Axboe --- drivers/block/nbd.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'drivers/block/nbd.c') diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c index 1183f7872b71..f4101d3efe7d 100644 --- a/drivers/block/nbd.c +++ b/drivers/block/nbd.c @@ -1762,7 +1762,9 @@ static struct nbd_device *nbd_dev_add(int index, unsigned int refs) disk->fops = &nbd_fops; disk->private_data = nbd; sprintf(disk->disk_name, "nbd%d", index); - add_disk(disk); + err = add_disk(disk); + if (err) + goto out_err_disk; /* * Now publish the device. @@ -1771,6 +1773,8 @@ static struct nbd_device *nbd_dev_add(int index, unsigned int refs) nbd_total_devices++; return nbd; +out_err_disk: + blk_cleanup_disk(disk); out_free_idr: mutex_lock(&nbd_index_mutex); idr_remove(&nbd_index_idr, index); -- cgit v1.2.3 From 4e6eef5dc25b528e08ac5b5f64f6ca9d9987241d Mon Sep 17 00:00:00 2001 From: Yu Kuai Date: Thu, 16 Sep 2021 17:33:44 +0800 Subject: nbd: don't handle response without a corresponding request message While handling a response message from server, nbd_read_stat() will try to get request by tag, and then complete the request. However, this is problematic if nbd haven't sent a corresponding request message: t1 t2 submit_bio nbd_queue_rq blk_mq_start_request recv_work nbd_read_stat blk_mq_tag_to_rq blk_mq_complete_request nbd_send_cmd Thus add a new cmd flag 'NBD_CMD_INFLIGHT', it will be set in nbd_send_cmd() and checked in nbd_read_stat(). Noted that this patch can't fix that blk_mq_tag_to_rq() might return a freed request, and this will be fixed in following patches. Signed-off-by: Yu Kuai Reviewed-by: Ming Lei Reviewed-by: Josef Bacik Link: https://lore.kernel.org/r/20210916093350.1410403-2-yukuai3@huawei.com Signed-off-by: Jens Axboe --- drivers/block/nbd.c | 22 +++++++++++++++++++++- 1 file changed, 21 insertions(+), 1 deletion(-) (limited to 'drivers/block/nbd.c') diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c index f4101d3efe7d..d18ba557a69d 100644 --- a/drivers/block/nbd.c +++ b/drivers/block/nbd.c @@ -131,6 +131,12 @@ struct nbd_device { }; #define NBD_CMD_REQUEUED 1 +/* + * This flag will be set if nbd_queue_rq() succeed, and will be checked and + * cleared in completion. Both setting and clearing of the flag are protected + * by cmd->lock. + */ +#define NBD_CMD_INFLIGHT 2 struct nbd_cmd { struct nbd_device *nbd; @@ -405,6 +411,7 @@ static enum blk_eh_timer_return nbd_xmit_timeout(struct request *req, if (!mutex_trylock(&cmd->lock)) return BLK_EH_RESET_TIMER; + __clear_bit(NBD_CMD_INFLIGHT, &cmd->flags); if (!refcount_inc_not_zero(&nbd->config_refs)) { cmd->status = BLK_STS_TIMEOUT; mutex_unlock(&cmd->lock); @@ -734,6 +741,12 @@ static struct nbd_cmd *nbd_read_stat(struct nbd_device *nbd, int index) cmd = blk_mq_rq_to_pdu(req); mutex_lock(&cmd->lock); + if (!__test_and_clear_bit(NBD_CMD_INFLIGHT, &cmd->flags)) { + dev_err(disk_to_dev(nbd->disk), "Suspicious reply %d (status %u flags %lu)", + tag, cmd->status, cmd->flags); + ret = -ENOENT; + goto out; + } if (cmd->cmd_cookie != nbd_handle_to_cookie(handle)) { dev_err(disk_to_dev(nbd->disk), "Double reply on req %p, cmd_cookie %u, handle cookie %u\n", req, cmd->cmd_cookie, nbd_handle_to_cookie(handle)); @@ -833,6 +846,7 @@ static bool nbd_clear_req(struct request *req, void *data, bool reserved) return true; mutex_lock(&cmd->lock); + __clear_bit(NBD_CMD_INFLIGHT, &cmd->flags); cmd->status = BLK_STS_IOERR; mutex_unlock(&cmd->lock); @@ -969,7 +983,13 @@ again: * returns EAGAIN can be retried on a different socket. */ ret = nbd_send_cmd(nbd, cmd, index); - if (ret == -EAGAIN) { + /* + * Access to this flag is protected by cmd->lock, thus it's safe to set + * the flag after nbd_send_cmd() succeed to send request to server. + */ + if (!ret) + __set_bit(NBD_CMD_INFLIGHT, &cmd->flags); + else if (ret == -EAGAIN) { dev_err_ratelimited(disk_to_dev(nbd->disk), "Request send failed, requeueing\n"); nbd_mark_nsock_dead(nbd, nsock, 1); -- cgit v1.2.3 From 07175cb1baf4c51051b1fbd391097e349f9a02a9 Mon Sep 17 00:00:00 2001 From: Yu Kuai Date: Thu, 16 Sep 2021 17:33:45 +0800 Subject: nbd: make sure request completion won't concurrent commit cddce0116058 ("nbd: Aovid double completion of a request") try to fix that nbd_clear_que() and recv_work() can complete a request concurrently. However, the problem still exists: t1 t2 t3 nbd_disconnect_and_put flush_workqueue recv_work blk_mq_complete_request blk_mq_complete_request_remote -> this is true WRITE_ONCE(rq->state, MQ_RQ_COMPLETE) blk_mq_raise_softirq blk_done_softirq blk_complete_reqs nbd_complete_rq blk_mq_end_request blk_mq_free_request WRITE_ONCE(rq->state, MQ_RQ_IDLE) nbd_clear_que blk_mq_tagset_busy_iter nbd_clear_req __blk_mq_free_request blk_mq_put_tag blk_mq_complete_request -> complete again There are three places where request can be completed in nbd: recv_work(), nbd_clear_que() and nbd_xmit_timeout(). Since they all hold cmd->lock before completing the request, it's easy to avoid the problem by setting and checking a cmd flag. Signed-off-by: Yu Kuai Reviewed-by: Ming Lei Reviewed-by: Josef Bacik Link: https://lore.kernel.org/r/20210916093350.1410403-3-yukuai3@huawei.com Signed-off-by: Jens Axboe --- drivers/block/nbd.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) (limited to 'drivers/block/nbd.c') diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c index d18ba557a69d..0bb3c1e2d575 100644 --- a/drivers/block/nbd.c +++ b/drivers/block/nbd.c @@ -411,7 +411,11 @@ static enum blk_eh_timer_return nbd_xmit_timeout(struct request *req, if (!mutex_trylock(&cmd->lock)) return BLK_EH_RESET_TIMER; - __clear_bit(NBD_CMD_INFLIGHT, &cmd->flags); + if (!__test_and_clear_bit(NBD_CMD_INFLIGHT, &cmd->flags)) { + mutex_unlock(&cmd->lock); + return BLK_EH_DONE; + } + if (!refcount_inc_not_zero(&nbd->config_refs)) { cmd->status = BLK_STS_TIMEOUT; mutex_unlock(&cmd->lock); @@ -846,7 +850,10 @@ static bool nbd_clear_req(struct request *req, void *data, bool reserved) return true; mutex_lock(&cmd->lock); - __clear_bit(NBD_CMD_INFLIGHT, &cmd->flags); + if (!__test_and_clear_bit(NBD_CMD_INFLIGHT, &cmd->flags)) { + mutex_unlock(&cmd->lock); + return true; + } cmd->status = BLK_STS_IOERR; mutex_unlock(&cmd->lock); -- cgit v1.2.3 From fcf3d633d8e101e84a0e072ab79957b938ac7e06 Mon Sep 17 00:00:00 2001 From: Yu Kuai Date: Thu, 16 Sep 2021 17:33:46 +0800 Subject: nbd: check sock index in nbd_read_stat() The sock that clent send request in nbd_send_cmd() and receive reply in nbd_read_stat() should be the same. Signed-off-by: Yu Kuai Reviewed-by: Ming Lei Reviewed-by: Josef Bacik Link: https://lore.kernel.org/r/20210916093350.1410403-4-yukuai3@huawei.com Signed-off-by: Jens Axboe --- drivers/block/nbd.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'drivers/block/nbd.c') diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c index 0bb3c1e2d575..49c501cf8a8b 100644 --- a/drivers/block/nbd.c +++ b/drivers/block/nbd.c @@ -751,6 +751,10 @@ static struct nbd_cmd *nbd_read_stat(struct nbd_device *nbd, int index) ret = -ENOENT; goto out; } + if (cmd->index != index) { + dev_err(disk_to_dev(nbd->disk), "Unexpected reply %d from different sock %d (expected %d)", + tag, index, cmd->index); + } if (cmd->cmd_cookie != nbd_handle_to_cookie(handle)) { dev_err(disk_to_dev(nbd->disk), "Double reply on req %p, cmd_cookie %u, handle cookie %u\n", req, cmd->cmd_cookie, nbd_handle_to_cookie(handle)); -- cgit v1.2.3 From 0de2b7a4dd08eeb82b8cc3fe5d6b8ba49a32f7bd Mon Sep 17 00:00:00 2001 From: Yu Kuai Date: Thu, 16 Sep 2021 17:33:47 +0800 Subject: nbd: don't start request if nbd_queue_rq() failed commit 6a468d5990ec ("nbd: don't start req until after the dead connection logic") move blk_mq_start_request() from nbd_queue_rq() to nbd_handle_cmd() to skip starting request if the connection is dead. However, request is still started in other error paths. Currently, blk_mq_end_request() will be called immediately if nbd_queue_rq() failed, thus start request in such situation is useless. So remove blk_mq_start_request() from error paths in nbd_handle_cmd(). Signed-off-by: Yu Kuai Reviewed-by: Ming Lei Reviewed-by: Josef Bacik Link: https://lore.kernel.org/r/20210916093350.1410403-5-yukuai3@huawei.com Signed-off-by: Jens Axboe --- drivers/block/nbd.c | 3 --- 1 file changed, 3 deletions(-) (limited to 'drivers/block/nbd.c') diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c index 49c501cf8a8b..87066cad5711 100644 --- a/drivers/block/nbd.c +++ b/drivers/block/nbd.c @@ -939,7 +939,6 @@ static int nbd_handle_cmd(struct nbd_cmd *cmd, int index) if (!refcount_inc_not_zero(&nbd->config_refs)) { dev_err_ratelimited(disk_to_dev(nbd->disk), "Socks array is empty\n"); - blk_mq_start_request(req); return -EINVAL; } config = nbd->config; @@ -948,7 +947,6 @@ static int nbd_handle_cmd(struct nbd_cmd *cmd, int index) dev_err_ratelimited(disk_to_dev(nbd->disk), "Attempted send on invalid socket\n"); nbd_config_put(nbd); - blk_mq_start_request(req); return -EINVAL; } cmd->status = BLK_STS_OK; @@ -972,7 +970,6 @@ again: */ sock_shutdown(nbd); nbd_config_put(nbd); - blk_mq_start_request(req); return -EIO; } goto again; -- cgit v1.2.3 From f52c0e08237e7864a44311fc78bc9bf2e045611b Mon Sep 17 00:00:00 2001 From: Yu Kuai Date: Thu, 16 Sep 2021 17:33:48 +0800 Subject: nbd: clean up return value checking of sock_xmit() Check if sock_xmit() return 0 is useless because it'll never return 0, comment it and remove such checkings. Signed-off-by: Yu Kuai Reviewed-by: Ming Lei Reviewed-by: Josef Bacik Link: https://lore.kernel.org/r/20210916093350.1410403-6-yukuai3@huawei.com Signed-off-by: Jens Axboe --- drivers/block/nbd.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) (limited to 'drivers/block/nbd.c') diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c index 87066cad5711..1d34560d8c1e 100644 --- a/drivers/block/nbd.c +++ b/drivers/block/nbd.c @@ -495,7 +495,8 @@ done: } /* - * Send or receive packet. + * Send or receive packet. Return a positive value on success and + * negtive value on failue, and never return 0. */ static int sock_xmit(struct nbd_device *nbd, int index, int send, struct iov_iter *iter, int msg_flags, int *sent) @@ -621,7 +622,7 @@ static int nbd_send_cmd(struct nbd_device *nbd, struct nbd_cmd *cmd, int index) result = sock_xmit(nbd, index, 1, &from, (type == NBD_CMD_WRITE) ? MSG_MORE : 0, &sent); trace_nbd_header_sent(req, handle); - if (result <= 0) { + if (result < 0) { if (was_interrupted(result)) { /* If we havne't sent anything we can just return BUSY, * however if we have sent something we need to make @@ -665,7 +666,7 @@ send_pages: skip = 0; } result = sock_xmit(nbd, index, 1, &from, flags, &sent); - if (result <= 0) { + if (result < 0) { if (was_interrupted(result)) { /* We've already sent the header, we * have no choice but to set pending and @@ -717,7 +718,7 @@ static struct nbd_cmd *nbd_read_stat(struct nbd_device *nbd, int index) reply.magic = 0; iov_iter_kvec(&to, READ, &iov, 1, sizeof(reply)); result = sock_xmit(nbd, index, 0, &to, MSG_WAITALL, NULL); - if (result <= 0) { + if (result < 0) { if (!nbd_disconnected(config)) dev_err(disk_to_dev(nbd->disk), "Receive control failed (result %d)\n", result); @@ -788,7 +789,7 @@ static struct nbd_cmd *nbd_read_stat(struct nbd_device *nbd, int index) rq_for_each_segment(bvec, req, iter) { iov_iter_bvec(&to, READ, &bvec, 1, bvec.bv_len); result = sock_xmit(nbd, index, 0, &to, MSG_WAITALL, NULL); - if (result <= 0) { + if (result < 0) { dev_err(disk_to_dev(nbd->disk), "Receive data failed (result %d)\n", result); /* @@ -1234,7 +1235,7 @@ static void send_disconnects(struct nbd_device *nbd) iov_iter_kvec(&from, WRITE, &iov, 1, sizeof(request)); mutex_lock(&nsock->tx_lock); ret = sock_xmit(nbd, i, 1, &from, 0, NULL); - if (ret <= 0) + if (ret < 0) dev_err(disk_to_dev(nbd->disk), "Send disconnect failed %d\n", ret); mutex_unlock(&nsock->tx_lock); -- cgit v1.2.3 From 3fe1db626a56cdf259c348404f2c5429e2f065a1 Mon Sep 17 00:00:00 2001 From: Yu Kuai Date: Thu, 16 Sep 2021 17:33:49 +0800 Subject: nbd: partition nbd_read_stat() into nbd_read_reply() and nbd_handle_reply() Prepare to fix uaf in nbd_read_stat(), no functional changes. Signed-off-by: Yu Kuai Reviewed-by: Ming Lei Reviewed-by: Josef Bacik Link: https://lore.kernel.org/r/20210916093350.1410403-7-yukuai3@huawei.com Signed-off-by: Jens Axboe --- drivers/block/nbd.c | 74 +++++++++++++++++++++++++++++++---------------------- 1 file changed, 44 insertions(+), 30 deletions(-) (limited to 'drivers/block/nbd.c') diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c index 1d34560d8c1e..06e292f0b0ae 100644 --- a/drivers/block/nbd.c +++ b/drivers/block/nbd.c @@ -700,38 +700,45 @@ out: return 0; } -/* NULL returned = something went wrong, inform userspace */ -static struct nbd_cmd *nbd_read_stat(struct nbd_device *nbd, int index) +static int nbd_read_reply(struct nbd_device *nbd, int index, + struct nbd_reply *reply) { - struct nbd_config *config = nbd->config; - int result; - struct nbd_reply reply; - struct nbd_cmd *cmd; - struct request *req = NULL; - u64 handle; - u16 hwq; - u32 tag; - struct kvec iov = {.iov_base = &reply, .iov_len = sizeof(reply)}; + struct kvec iov = {.iov_base = reply, .iov_len = sizeof(*reply)}; struct iov_iter to; - int ret = 0; + int result; - reply.magic = 0; - iov_iter_kvec(&to, READ, &iov, 1, sizeof(reply)); + reply->magic = 0; + iov_iter_kvec(&to, READ, &iov, 1, sizeof(*reply)); result = sock_xmit(nbd, index, 0, &to, MSG_WAITALL, NULL); if (result < 0) { - if (!nbd_disconnected(config)) + if (!nbd_disconnected(nbd->config)) dev_err(disk_to_dev(nbd->disk), "Receive control failed (result %d)\n", result); - return ERR_PTR(result); + return result; } - if (ntohl(reply.magic) != NBD_REPLY_MAGIC) { + if (ntohl(reply->magic) != NBD_REPLY_MAGIC) { dev_err(disk_to_dev(nbd->disk), "Wrong magic (0x%lx)\n", - (unsigned long)ntohl(reply.magic)); - return ERR_PTR(-EPROTO); + (unsigned long)ntohl(reply->magic)); + return -EPROTO; } - memcpy(&handle, reply.handle, sizeof(handle)); + return 0; +} + +/* NULL returned = something went wrong, inform userspace */ +static struct nbd_cmd *nbd_handle_reply(struct nbd_device *nbd, int index, + struct nbd_reply *reply) +{ + int result; + struct nbd_cmd *cmd; + struct request *req = NULL; + u64 handle; + u16 hwq; + u32 tag; + int ret = 0; + + memcpy(&handle, reply->handle, sizeof(handle)); tag = nbd_handle_to_tag(handle); hwq = blk_mq_unique_tag_to_hwq(tag); if (hwq < nbd->tag_set.nr_hw_queues) @@ -774,9 +781,9 @@ static struct nbd_cmd *nbd_read_stat(struct nbd_device *nbd, int index) ret = -ENOENT; goto out; } - if (ntohl(reply.error)) { + if (ntohl(reply->error)) { dev_err(disk_to_dev(nbd->disk), "Other side returned error (%d)\n", - ntohl(reply.error)); + ntohl(reply->error)); cmd->status = BLK_STS_IOERR; goto out; } @@ -785,6 +792,7 @@ static struct nbd_cmd *nbd_read_stat(struct nbd_device *nbd, int index) if (rq_data_dir(req) != WRITE) { struct req_iterator iter; struct bio_vec bvec; + struct iov_iter to; rq_for_each_segment(bvec, req, iter) { iov_iter_bvec(&to, READ, &bvec, 1, bvec.bv_len); @@ -798,7 +806,7 @@ static struct nbd_cmd *nbd_read_stat(struct nbd_device *nbd, int index) * and let the timeout stuff handle resubmitting * this request onto another connection. */ - if (nbd_disconnected(config)) { + if (nbd_disconnected(nbd->config)) { cmd->status = BLK_STS_IOERR; goto out; } @@ -822,24 +830,30 @@ static void recv_work(struct work_struct *work) work); struct nbd_device *nbd = args->nbd; struct nbd_config *config = nbd->config; + struct nbd_sock *nsock; struct nbd_cmd *cmd; struct request *rq; while (1) { - cmd = nbd_read_stat(nbd, args->index); - if (IS_ERR(cmd)) { - struct nbd_sock *nsock = config->socks[args->index]; + struct nbd_reply reply; - mutex_lock(&nsock->tx_lock); - nbd_mark_nsock_dead(nbd, nsock, 1); - mutex_unlock(&nsock->tx_lock); + if (nbd_read_reply(nbd, args->index, &reply)) + break; + + cmd = nbd_handle_reply(nbd, args->index, &reply); + if (IS_ERR(cmd)) break; - } rq = blk_mq_rq_from_pdu(cmd); if (likely(!blk_should_fake_timeout(rq->q))) blk_mq_complete_request(rq); } + + nsock = config->socks[args->index]; + mutex_lock(&nsock->tx_lock); + nbd_mark_nsock_dead(nbd, nsock, 1); + mutex_unlock(&nsock->tx_lock); + nbd_config_put(nbd); atomic_dec(&config->recv_threads); wake_up(&config->recv_wq); -- cgit v1.2.3 From 8663b210f8c169a49aaeed3af92471a147545477 Mon Sep 17 00:00:00 2001 From: Yu Kuai Date: Thu, 16 Sep 2021 22:18:10 +0800 Subject: nbd: fix uaf in nbd_handle_reply() There is a problem that nbd_handle_reply() might access freed request: 1) At first, a normal io is submitted and completed with scheduler: internel_tag = blk_mq_get_tag -> get tag from sched_tags blk_mq_rq_ctx_init sched_tags->rq[internel_tag] = sched_tag->static_rq[internel_tag] ... blk_mq_get_driver_tag __blk_mq_get_driver_tag -> get tag from tags tags->rq[tag] = sched_tag->static_rq[internel_tag] So, both tags->rq[tag] and sched_tags->rq[internel_tag] are pointing to the request: sched_tags->static_rq[internal_tag]. Even if the io is finished. 2) nbd server send a reply with random tag directly: recv_work nbd_handle_reply blk_mq_tag_to_rq(tags, tag) rq = tags->rq[tag] 3) if the sched_tags->static_rq is freed: blk_mq_sched_free_requests blk_mq_free_rqs(q->tag_set, hctx->sched_tags, i) -> step 2) access rq before clearing rq mapping blk_mq_clear_rq_mapping(set, tags, hctx_idx); __free_pages() -> rq is freed here 4) Then, nbd continue to use the freed request in nbd_handle_reply Fix the problem by get 'q_usage_counter' before blk_mq_tag_to_rq(), thus request is ensured not to be freed because 'q_usage_counter' is not zero. Signed-off-by: Yu Kuai Reviewed-by: Ming Lei Link: https://lore.kernel.org/r/20210916141810.2325276-1-yukuai3@huawei.com Signed-off-by: Jens Axboe --- drivers/block/nbd.c | 18 +++++++++++++++++- 1 file changed, 17 insertions(+), 1 deletion(-) (limited to 'drivers/block/nbd.c') diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c index 06e292f0b0ae..0d064fab6186 100644 --- a/drivers/block/nbd.c +++ b/drivers/block/nbd.c @@ -830,6 +830,7 @@ static void recv_work(struct work_struct *work) work); struct nbd_device *nbd = args->nbd; struct nbd_config *config = nbd->config; + struct request_queue *q = nbd->disk->queue; struct nbd_sock *nsock; struct nbd_cmd *cmd; struct request *rq; @@ -840,13 +841,28 @@ static void recv_work(struct work_struct *work) if (nbd_read_reply(nbd, args->index, &reply)) break; + /* + * Grab .q_usage_counter so request pool won't go away, then no + * request use-after-free is possible during nbd_handle_reply(). + * If queue is frozen, there won't be any inflight requests, we + * needn't to handle the incoming garbage message. + */ + if (!percpu_ref_tryget(&q->q_usage_counter)) { + dev_err(disk_to_dev(nbd->disk), "%s: no io inflight\n", + __func__); + break; + } + cmd = nbd_handle_reply(nbd, args->index, &reply); - if (IS_ERR(cmd)) + if (IS_ERR(cmd)) { + percpu_ref_put(&q->q_usage_counter); break; + } rq = blk_mq_rq_from_pdu(cmd); if (likely(!blk_should_fake_timeout(rq->q))) blk_mq_complete_request(rq); + percpu_ref_put(&q->q_usage_counter); } nsock = config->socks[args->index]; -- cgit v1.2.3 From 0c98057be9efa32de78dbc4685fc73da9d71faa1 Mon Sep 17 00:00:00 2001 From: Ye Bin Date: Wed, 20 Oct 2021 15:39:59 +0800 Subject: nbd: Fix use-after-free in pid_show I got issue as follows: [ 263.886511] BUG: KASAN: use-after-free in pid_show+0x11f/0x13f [ 263.888359] Read of size 4 at addr ffff8880bf0648c0 by task cat/746 [ 263.890479] CPU: 0 PID: 746 Comm: cat Not tainted 4.19.90-dirty #140 [ 263.893162] Call Trace: [ 263.893509] dump_stack+0x108/0x15f [ 263.893999] print_address_description+0xa5/0x372 [ 263.894641] kasan_report.cold+0x236/0x2a8 [ 263.895696] __asan_report_load4_noabort+0x25/0x30 [ 263.896365] pid_show+0x11f/0x13f [ 263.897422] dev_attr_show+0x48/0x90 [ 263.898361] sysfs_kf_seq_show+0x24d/0x4b0 [ 263.899479] kernfs_seq_show+0x14e/0x1b0 [ 263.900029] seq_read+0x43f/0x1150 [ 263.900499] kernfs_fop_read+0xc7/0x5a0 [ 263.903764] vfs_read+0x113/0x350 [ 263.904231] ksys_read+0x103/0x270 [ 263.905230] __x64_sys_read+0x77/0xc0 [ 263.906284] do_syscall_64+0x106/0x360 [ 263.906797] entry_SYSCALL_64_after_hwframe+0x44/0xa9 Reproduce this issue as follows: 1. nbd-server 8000 /tmp/disk 2. nbd-client localhost 8000 /dev/nbd1 3. cat /sys/block/nbd1/pid Then trigger use-after-free in pid_show. Reason is after do step '2', nbd-client progress is already exit. So it's task_struct already freed. To solve this issue, revert part of 6521d39a64b3's modify and remove useless 'recv_task' member of nbd_device. Fixes: 6521d39a64b3 ("nbd: Remove variable 'pid'") Signed-off-by: Ye Bin Reviewed-by: Josef Bacik Link: https://lore.kernel.org/r/20211020073959.2679255-1-yebin10@huawei.com Signed-off-by: Jens Axboe --- drivers/block/nbd.c | 18 +++++++++--------- 1 file changed, 9 insertions(+), 9 deletions(-) (limited to 'drivers/block/nbd.c') diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c index 0d064fab6186..0ee104fbb628 100644 --- a/drivers/block/nbd.c +++ b/drivers/block/nbd.c @@ -122,10 +122,10 @@ struct nbd_device { struct work_struct remove_work; struct list_head list; - struct task_struct *task_recv; struct task_struct *task_setup; unsigned long flags; + pid_t pid; /* pid of nbd-client, if attached */ char *backend; }; @@ -223,7 +223,7 @@ static ssize_t pid_show(struct device *dev, struct gendisk *disk = dev_to_disk(dev); struct nbd_device *nbd = (struct nbd_device *)disk->private_data; - return sprintf(buf, "%d\n", task_pid_nr(nbd->task_recv)); + return sprintf(buf, "%d\n", nbd->pid); } static const struct device_attribute pid_attr = { @@ -335,7 +335,7 @@ static int nbd_set_size(struct nbd_device *nbd, loff_t bytesize, nbd->config->bytesize = bytesize; nbd->config->blksize_bits = __ffs(blksize); - if (!nbd->task_recv) + if (!nbd->pid) return 0; if (nbd->config->flags & NBD_FLAG_SEND_TRIM) { @@ -1300,7 +1300,7 @@ static void nbd_config_put(struct nbd_device *nbd) if (test_and_clear_bit(NBD_RT_HAS_PID_FILE, &config->runtime_flags)) device_remove_file(disk_to_dev(nbd->disk), &pid_attr); - nbd->task_recv = NULL; + nbd->pid = 0; if (test_and_clear_bit(NBD_RT_HAS_BACKEND_FILE, &config->runtime_flags)) { device_remove_file(disk_to_dev(nbd->disk), &backend_attr); @@ -1341,7 +1341,7 @@ static int nbd_start_device(struct nbd_device *nbd) int num_connections = config->num_connections; int error = 0, i; - if (nbd->task_recv) + if (nbd->pid) return -EBUSY; if (!config->socks) return -EINVAL; @@ -1360,7 +1360,7 @@ static int nbd_start_device(struct nbd_device *nbd) } blk_mq_update_nr_hw_queues(&nbd->tag_set, config->num_connections); - nbd->task_recv = current; + nbd->pid = task_pid_nr(current); nbd_parse_flags(nbd); @@ -1616,8 +1616,8 @@ static int nbd_dbg_tasks_show(struct seq_file *s, void *unused) { struct nbd_device *nbd = s->private; - if (nbd->task_recv) - seq_printf(s, "recv: %d\n", task_pid_nr(nbd->task_recv)); + if (nbd->pid) + seq_printf(s, "recv: %d\n", nbd->pid); return 0; } @@ -2198,7 +2198,7 @@ static int nbd_genl_reconfigure(struct sk_buff *skb, struct genl_info *info) mutex_lock(&nbd->config_lock); config = nbd->config; if (!test_bit(NBD_RT_BOUND, &config->runtime_flags) || - !nbd->task_recv) { + !nbd->pid) { dev_err(nbd_to_dev(nbd), "not configured, cannot reconfigure\n"); ret = -EINVAL; -- cgit v1.2.3