From 9450918293b3c35f11883231a53da1aed2c78403 Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Wed, 29 Jul 2015 22:27:13 -0700 Subject: target: Perform RCU callback barrier before backend/fabric unload This patch addresses a v4.2-rc1 regression where backend driver module unload happening immediately after TBO->free_device() does internal call_rcu(), will currently result in IRQ context rcu_process_callbacks() use-after-free paging OOPsen. It adds the missing rcu_barrier() in target_backend_unregister() to perform an explicit RCU barrier waiting for all RCU callbacks to complete before releasing target_backend_ops memory, and allowing TBO->module exit to proceed. Also, do the same for fabric drivers in target_unregister_template() to ensure se_deve_entry->rcu_head -> kfree_rcu() callbacks have completed, before allowing target_core_fabric_ops->owner module exit to proceed. Acked-by: Paul E. McKenney Cc: Christoph Hellwig Cc: Hannes Reinecke Cc: Sagi Grimberg Signed-off-by: Nicholas Bellinger --- drivers/target/target_core_configfs.c | 9 ++++++++- drivers/target/target_core_hba.c | 10 +++++++++- 2 files changed, 17 insertions(+), 2 deletions(-) diff --git a/drivers/target/target_core_configfs.c b/drivers/target/target_core_configfs.c index c2e9fea90b4a..860e84046177 100644 --- a/drivers/target/target_core_configfs.c +++ b/drivers/target/target_core_configfs.c @@ -457,8 +457,15 @@ void target_unregister_template(const struct target_core_fabric_ops *fo) if (!strcmp(t->tf_ops->name, fo->name)) { BUG_ON(atomic_read(&t->tf_access_cnt)); list_del(&t->tf_list); + mutex_unlock(&g_tf_lock); + /* + * Wait for any outstanding fabric se_deve_entry->rcu_head + * callbacks to complete post kfree_rcu(), before allowing + * fabric driver unload of TFO->module to proceed. + */ + rcu_barrier(); kfree(t); - break; + return; } } mutex_unlock(&g_tf_lock); diff --git a/drivers/target/target_core_hba.c b/drivers/target/target_core_hba.c index 62ea4e8e70a8..be9cefc07407 100644 --- a/drivers/target/target_core_hba.c +++ b/drivers/target/target_core_hba.c @@ -84,8 +84,16 @@ void target_backend_unregister(const struct target_backend_ops *ops) list_for_each_entry(tb, &backend_list, list) { if (tb->ops == ops) { list_del(&tb->list); + mutex_unlock(&backend_mutex); + /* + * Wait for any outstanding backend driver ->rcu_head + * callbacks to complete post TBO->free_device() -> + * call_rcu(), before allowing backend driver module + * unload of target_backend_ops->owner to proceed. + */ + rcu_barrier(); kfree(tb); - break; + return; } } mutex_unlock(&backend_mutex); -- cgit v1.2.3 From 9547308bda296b6f69876c840a0291fcfbeddbb8 Mon Sep 17 00:00:00 2001 From: Alexei Potashnik Date: Tue, 21 Jul 2015 15:07:56 -0700 Subject: target/iscsi: Fix double free of a TUR followed by a solicited NOPOUT Make sure all non-READ SCSI commands get targ_xfer_tag initialized to 0xffffffff, not just WRITEs. Double-free of a TUR cmd object occurs under the following scenario: 1. TUR received (targ_xfer_tag is uninitialized and left at 0) 2. TUR status sent 3. First unsolicited NOPIN is sent to initiator (gets targ_xfer_tag of 0) 4. NOPOUT for NOPIN (with TTT=0) arrives - its ExpStatSN acks TUR status, TUR is queued for removal - LIO tries to find NOPIN with TTT=0, but finds the same TUR instead, TUR is queued for removal for the 2nd time (Drop unbalanced conditional bracket usage - nab) Signed-off-by: Alexei Potashnik Signed-off-by: Spencer Baugh Cc: # v3.1+ Signed-off-by: Nicholas Bellinger --- drivers/target/iscsi/iscsi_target.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/target/iscsi/iscsi_target.c b/drivers/target/iscsi/iscsi_target.c index cd77a064c772..fd092909a457 100644 --- a/drivers/target/iscsi/iscsi_target.c +++ b/drivers/target/iscsi/iscsi_target.c @@ -968,9 +968,9 @@ int iscsit_setup_scsi_cmd(struct iscsi_conn *conn, struct iscsi_cmd *cmd, cmd->cmd_flags |= ICF_NON_IMMEDIATE_UNSOLICITED_DATA; conn->sess->init_task_tag = cmd->init_task_tag = hdr->itt; - if (hdr->flags & ISCSI_FLAG_CMD_READ) { + if (hdr->flags & ISCSI_FLAG_CMD_READ) cmd->targ_xfer_tag = session_get_next_ttt(conn->sess); - } else if (hdr->flags & ISCSI_FLAG_CMD_WRITE) + else cmd->targ_xfer_tag = 0xFFFFFFFF; cmd->cmd_sn = be32_to_cpu(hdr->cmdsn); cmd->exp_stat_sn = be32_to_cpu(hdr->exp_statsn); -- cgit v1.2.3 From 9c395170a559d3b23dad100b01fc4a89d661c698 Mon Sep 17 00:00:00 2001 From: Roland Dreier Date: Fri, 24 Jul 2015 12:11:46 -0700 Subject: target: REPORT LUNS should return LUN 0 even for dynamic ACLs If an initiator doesn't have any real LUNs assigned, we should report LUN 0 and a LUN list length of 1. Some versions of Solaris at least go beserk if we report a LUN list length of 0. Signed-off-by: Roland Dreier Cc: # v3.1+ Signed-off-by: Nicholas Bellinger --- drivers/target/target_core_spc.c | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) diff --git a/drivers/target/target_core_spc.c b/drivers/target/target_core_spc.c index b5ba1ec3c354..556ea1b2cdd8 100644 --- a/drivers/target/target_core_spc.c +++ b/drivers/target/target_core_spc.c @@ -1221,11 +1221,9 @@ sense_reason_t spc_emulate_report_luns(struct se_cmd *cmd) * coming via a target_core_mod PASSTHROUGH op, and not through * a $FABRIC_MOD. In that case, report LUN=0 only. */ - if (!sess) { - int_to_scsilun(0, (struct scsi_lun *)&buf[offset]); - lun_count = 1; + if (!sess) goto done; - } + nacl = sess->se_node_acl; rcu_read_lock(); @@ -1248,6 +1246,14 @@ sense_reason_t spc_emulate_report_luns(struct se_cmd *cmd) * See SPC3 r07, page 159. */ done: + /* + * If no LUNs are accessible, report virtual LUN 0. + */ + if (lun_count == 0) { + int_to_scsilun(0, (struct scsi_lun *)&buf[offset]); + lun_count = 1; + } + lun_count *= 8; buf[0] = ((lun_count >> 24) & 0xff); buf[1] = ((lun_count >> 16) & 0xff); -- cgit v1.2.3 From 0f69a38737538c8a265f75d26996818f98c899e4 Mon Sep 17 00:00:00 2001 From: Roland Dreier Date: Thu, 13 Aug 2015 21:59:19 -0700 Subject: target: Fix handling of small allocation lengths in REPORT LUNS REPORT LUNS should not fail just because the allocation length is less than 16. The relevant section of SPC-4 is: 4.2.5.6 Allocation length The ALLOCATION LENGTH field specifies the maximum number of bytes or blocks that an application client has allocated in the Data-In Buffer. The ALLOCATION LENGTH field specifies bytes unless a different requirement is stated in the command definition. An allocation length of zero specifies that no data shall be transferred. This condition shall not be considered an error. So we should just truncate our response rather than return an error. Signed-off-by: Roland Dreier Signed-off-by: Spencer Baugh Signed-off-by: Nicholas Bellinger --- drivers/target/target_core_spc.c | 32 ++++++++++++++++---------------- 1 file changed, 16 insertions(+), 16 deletions(-) diff --git a/drivers/target/target_core_spc.c b/drivers/target/target_core_spc.c index 556ea1b2cdd8..f87d4cef6d39 100644 --- a/drivers/target/target_core_spc.c +++ b/drivers/target/target_core_spc.c @@ -1203,17 +1203,13 @@ sense_reason_t spc_emulate_report_luns(struct se_cmd *cmd) struct se_dev_entry *deve; struct se_session *sess = cmd->se_sess; struct se_node_acl *nacl; + struct scsi_lun slun; unsigned char *buf; u32 lun_count = 0, offset = 8; - - if (cmd->data_length < 16) { - pr_warn("REPORT LUNS allocation length %u too small\n", - cmd->data_length); - return TCM_INVALID_CDB_FIELD; - } + __be32 len; buf = transport_kmap_data_sg(cmd); - if (!buf) + if (cmd->data_length && !buf) return TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE; /* @@ -1234,10 +1230,12 @@ sense_reason_t spc_emulate_report_luns(struct se_cmd *cmd) * See SPC2-R20 7.19. */ lun_count++; - if ((offset + 8) > cmd->data_length) + if (offset >= cmd->data_length) continue; - int_to_scsilun(deve->mapped_lun, (struct scsi_lun *)&buf[offset]); + int_to_scsilun(deve->mapped_lun, &slun); + memcpy(buf + offset, &slun, + min(8u, cmd->data_length - offset)); offset += 8; } rcu_read_unlock(); @@ -1250,16 +1248,18 @@ done: * If no LUNs are accessible, report virtual LUN 0. */ if (lun_count == 0) { - int_to_scsilun(0, (struct scsi_lun *)&buf[offset]); + int_to_scsilun(0, &slun); + if (cmd->data_length > 8) + memcpy(buf + offset, &slun, + min(8u, cmd->data_length - offset)); lun_count = 1; } - lun_count *= 8; - buf[0] = ((lun_count >> 24) & 0xff); - buf[1] = ((lun_count >> 16) & 0xff); - buf[2] = ((lun_count >> 8) & 0xff); - buf[3] = (lun_count & 0xff); - transport_kunmap_data_sg(cmd); + if (buf) { + len = cpu_to_be32(lun_count * 8); + memcpy(buf, &len, min_t(int, sizeof len, cmd->data_length)); + transport_kunmap_data_sg(cmd); + } target_complete_cmd_with_length(cmd, GOOD, 8 + lun_count * 8); return 0; -- cgit v1.2.3