kernel/drivers/s390, branch linux-5.1.y

scsi: zfcp: fix request object use-after-free in send path causing wrong traces

2019-07-26T07:12:52Z

commit 106d45f350c7cac876844dc685845cba4ffdb70b upstream. When tracing instances where we open and close WKA ports, we also pass the request-ID of the respective FSF command. But after successfully sending the FSF command we must not use the request-object anymore, as this might result in an use-after-free (see "zfcp: fix request object use-after-free in send path causing seqno errors" ). To fix this add a new variable that caches the request-ID before sending the request. This won't change during the hand-off to the FCP channel, and so it's safe to trace this cached request-ID later, instead of using the request object. Signed-off-by: Benjamin Block Fixes: d27a7cb91960 ("zfcp: trace on request for open and close of WKA port") Cc: #2.6.38+ Reviewed-by: Steffen Maier Reviewed-by: Jens Remus Signed-off-by: Martin K. Petersen Signed-off-by: Greg Kroah-Hartman

scsi: zfcp: fix request object use-after-free in send path causing seqno errors

2019-07-26T07:12:52Z

commit b76becde2b84137faa29bbc9a3b20953b5980e48 upstream. With a recent change to our send path for FSF commands we introduced a possible use-after-free of request-objects, that might further lead to zfcp crafting bad requests, which the FCP channel correctly complains about with an error (FSF_PROT_SEQ_NUMB_ERROR). This error is then handled by an adapter-wide recovery. The following sequence illustrates the possible use-after-free: Send Path: int zfcp_fsf_open_port(struct zfcp_erp_action *erp_action) { struct zfcp_fsf_req *req; ... spin_lock_irq(&qdio->req_q_lock); // ^^^^^^^^^^^^^^^^ // protects QDIO queue during sending ... req = zfcp_fsf_req_create(qdio, FSF_QTCB_OPEN_PORT_WITH_DID, SBAL_SFLAGS0_TYPE_READ, qdio->adapter->pool.erp_req); // ^^^^^^^^^^^^^^^^^^^ // allocation of the request-object ... retval = zfcp_fsf_req_send(req); ... spin_unlock_irq(&qdio->req_q_lock); return retval; } static int zfcp_fsf_req_send(struct zfcp_fsf_req *req) { struct zfcp_adapter *adapter = req->adapter; struct zfcp_qdio *qdio = adapter->qdio; ... zfcp_reqlist_add(adapter->req_list, req); // ^^^^^^^^^^^^^^^^ // add request to our driver-internal hash-table for tracking // (protected by separate lock req_list->lock) ... if (zfcp_qdio_send(qdio, &req->qdio_req)) { // ^^^^^^^^^^^^^^ // hand-off the request to FCP channel; // the request can complete at any point now ... } /* Don't increase for unsolicited status */ if (!zfcp_fsf_req_is_status_read_buffer(req)) // ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ // possible use-after-free adapter->fsf_req_seq_no++; // ^^^^^^^^^^^^^^^^ // because of the use-after-free we might // miss this accounting, and as follow-up // this results in the FCP channel error // FSF_PROT_SEQ_NUMB_ERROR adapter->req_no++; return 0; } static inline bool zfcp_fsf_req_is_status_read_buffer(struct zfcp_fsf_req *req) { return req->qtcb == NULL; // ^^^^^^^^^ // possible use-after-free } Response Path: void zfcp_fsf_reqid_check(struct zfcp_qdio *qdio, int sbal_idx) { ... struct zfcp_fsf_req *fsf_req; ... for (idx = 0; idx < QDIO_MAX_ELEMENTS_PER_BUFFER; idx++) { ... fsf_req = zfcp_reqlist_find_rm(adapter->req_list, req_id); // ^^^^^^^^^^^^^^^^^^^^ // remove request from our driver-internal // hash-table (lock req_list->lock) ... zfcp_fsf_req_complete(fsf_req); } } static void zfcp_fsf_req_complete(struct zfcp_fsf_req *req) { ... if (likely(req->status & ZFCP_STATUS_FSFREQ_CLEANUP)) zfcp_fsf_req_free(req); // ^^^^^^^^^^^^^^^^^ // free memory for request-object else complete(&req->completion); // ^^^^^^^^ // completion notification for code-paths that wait // synchronous for the completion of the request; in // those the memory is freed separately } The result of the use-after-free only affects the send path, and can not lead to any data corruption. In case we miss the sequence-number accounting, because the memory was already re-purposed, the next FSF command will fail with said FCP channel error, and we will recover the whole adapter. This causes no additional errors, but it slows down traffic. There is a slight chance of the same thing happen again recursively after the adapter recovery, but so far this has not been seen. This was seen under z/VM, where the send path might run on a virtual CPU that gets scheduled away by z/VM, while the return path might still run, and so create the necessary timing. Running with KASAN can also slow down the kernel sufficiently to run into this user-after-free, and then see the report by KASAN. To fix this, simply pull the test for the sequence-number accounting in front of the hand-off to the FCP channel (this information doesn't change during hand-off), but leave the sequence-number accounting itself where it is. To make future regressions of the same kind less likely, add comments to all closely related code-paths. Signed-off-by: Benjamin Block Fixes: f9eca0227600 ("scsi: zfcp: drop duplicate fsf_command from zfcp_fsf_req which is also in QTCB header") Cc: #5.0+ Reviewed-by: Steffen Maier Reviewed-by: Jens Remus Signed-off-by: Martin K. Petersen Signed-off-by: Greg Kroah-Hartman

s390/qdio: handle PENDING state for QEBSM devices

2019-07-26T07:12:31Z

[ Upstream commit 04310324c6f482921c071444833e70fe861b73d9 ] When a CQ-enabled device uses QEBSM for SBAL state inspection, get_buf_states() can return the PENDING state for an Output Queue. get_outbound_buffer_frontier() isn't prepared for this, and any PENDING buffer will permanently stall all further completion processing on this Queue. This isn't a concern for non-QEBSM devices, as get_buf_states() for such devices will manually turn PENDING buffers into EMPTY ones. Fixes: 104ea556ee7f ("qdio: support asynchronous delivery of storage blocks") Signed-off-by: Julian Wiedmann Signed-off-by: Heiko Carstens Signed-off-by: Sasha Levin

s390/qdio: don't touch the dsci in tiqdio_add_input_queues()

2019-07-21T07:02:03Z

commit ac6639cd3db607d386616487902b4cc1850a7be5 upstream. Current code sets the dsci to 0x00000080. Which doesn't make any sense, as the indicator area is located in the _left-most_ byte. Worse: if the dsci is the _shared_ indicator, this potentially clears the indication of activity for a _different_ device. tiqdio_thinint_handler() will then have no reason to call that device's IRQ handler, and the device ends up stalling. Fixes: d0c9d4a89fff ("[S390] qdio: set correct bit in dsci") Cc: Signed-off-by: Julian Wiedmann Signed-off-by: Vasily Gorbik Signed-off-by: Greg Kroah-Hartman

s390/qdio: (re-)initialize tiqdio list entries

2019-07-21T07:02:03Z

commit e54e4785cb5cb4896cf4285964aeef2125612fb2 upstream. When tiqdio_remove_input_queues() removes a queue from the tiq_list as part of qdio_shutdown(), it doesn't re-initialize the queue's list entry and the prev/next pointers go stale. If a subsequent qdio_establish() fails while sending the ESTABLISH cmd, it calls qdio_shutdown() again in QDIO_IRQ_STATE_ERR state and tiqdio_remove_input_queues() will attempt to remove the queue entry a second time. This dereferences the stale pointers, and bad things ensue. Fix this by re-initializing the list entry after removing it from the list. For good practice also initialize the list entry when the queue is first allocated, and remove the quirky checks that papered over this omission. Note that prior to commit e521813468f7 ("s390/qdio: fix access to uninitialized qdio_q fields"), these checks were bogus anyway. setup_queues_misc() clears the whole queue struct, and thus needs to re-init the prev/next pointers as well. Fixes: 779e6e1c724d ("[S390] qdio: new qdio driver.") Cc: Signed-off-by: Julian Wiedmann Signed-off-by: Vasily Gorbik Signed-off-by: Greg Kroah-Hartman

s390/qeth: fix VLAN attribute in bridge_hostnotify udev event

2019-06-25T03:34:52Z

[ Upstream commit 335726195e460cb6b3f795b695bfd31f0ea70ef0 ] Enabling sysfs attribute bridge_hostnotify triggers a series of udev events for the MAC addresses of all currently connected peers. In case no VLAN is set for a peer, the device reports the corresponding MAC addresses with VLAN ID 4096. This currently results in attribute VLAN=4096 for all non-VLAN interfaces in the initial series of events after host-notify is enabled. Instead, no VLAN attribute should be reported in the udev event for non-VLAN interfaces. Only the initial events face this issue. For dynamic changes that are reported later, the device uses a validity flag. This also changes the code so that it now sets the VLAN attribute for MAC addresses with VID 0. On Linux, no qeth interface will ever be registered with VID 0: Linux kernel registers VID 0 on all network interfaces initially, but qeth will drop .ndo_vlan_rx_add_vid for VID 0. Peers with other OSs could register MACs with VID 0. Fixes: 9f48b9db9a22 ("qeth: bridgeport support - address notifications") Signed-off-by: Alexandra Winter Signed-off-by: Julian Wiedmann Signed-off-by: David S. Miller Signed-off-by: Sasha Levin

s390/qeth: check dst entry before use

2019-06-25T03:34:52Z

[ Upstream commit 0cd6783d3c7d40be165d1f3c811cedf0e3dfcdf1 ] While qeth_l3 uses netif_keep_dst() to hold onto the dst, a skb's dst may still have been obsoleted (via dst_dev_put()) by the time that we end up using it. The dst then points to the loopback interface, which means the neighbour lookup in qeth_l3_get_cast_type() determines a bogus cast type of RTN_BROADCAST. For IQD interfaces this causes us to place such skbs on the wrong HW queue, resulting in TX errors. Fix-up the various call sites to first validate the dst entry with dst_check(), and fall back accordingly. Signed-off-by: Julian Wiedmann Signed-off-by: David S. Miller Signed-off-by: Sasha Levin

s390/qeth: handle limited IPv4 broadcast in L3 TX path

2019-06-25T03:34:52Z

[ Upstream commit 72c87976c5abbf8a834ad85f10d03c0cd58b985c ] When selecting the cast type of a neighbourless IPv4 skb (eg. on a raw socket), qeth_l3 falls back to the packet's destination IP address. For this case we should classify traffic sent to 255.255.255.255 as broadcast. This fixes DHCP requests, which were misclassified as unicast (and for IQD interfaces thus ended up on the wrong HW queue). Signed-off-by: Julian Wiedmann Signed-off-by: David S. Miller Signed-off-by: Sasha Levin

s390/zcrypt: Fix wrong dispatching for control domain CPRBs

2019-06-22T06:09:12Z

[ Upstream commit 7379e652797c0b9b5f6caea1576f2dff9ce6a708 ] The zcrypt device driver does not handle CPRBs which address a control domain correctly. This fix introduces a workaround: The domain field of the request CPRB is checked if there is a valid domain value in there. If this is true and the value is a control only domain (a domain which is enabled in the crypto config ADM mask but disabled in the AQM mask) the CPRB is forwarded to the default usage domain. If there is no default domain, the request is rejected with an ENODEV. This fix is important for maintaining crypto adapters. For example one LPAR can use a crypto adapter domain ('Control and Usage') but another LPAR needs to be able to maintain this adapter domain ('Control'). Scenarios like this did not work properly and the patch enables this. Signed-off-by: Harald Freudenberger Signed-off-by: Heiko Carstens Signed-off-by: Sasha Levin

scsi: zfcp: fix to prevent port_remove with pure auto scan LUNs (only sdevs)

2019-06-09T07:16:09Z

commit ef4021fe5fd77ced0323cede27979d80a56211ca upstream. When the user tries to remove a zfcp port via sysfs, we only rejected it if there are zfcp unit children under the port. With purely automatically scanned LUNs there are no zfcp units but only SCSI devices. In such cases, the port_remove erroneously continued. We close the port and this implicitly closes all LUNs under the port. The SCSI devices survive with their private zfcp_scsi_dev still holding a reference to the "removed" zfcp_port (still allocated but invisible in sysfs) [zfcp_get_port_by_wwpn in zfcp_scsi_slave_alloc]. This is not a problem as long as the fc_rport stays blocked. Once (auto) port scan brings back the removed port, we unblock its fc_rport again by design. However, there is no mechanism that would recover (open) the LUNs under the port (no "ersfs_3" without zfcp_unit [zfcp_erp_strategy_followup_success]). Any pending or new I/O to such LUN leads to repeated: Done: NEEDS_RETRY Result: hostbyte=DID_IMM_RETRY driverbyte=DRIVER_OK See also v4.10 commit 6f2ce1c6af37 ("scsi: zfcp: fix rport unblock race with LUN recovery"). Even a manual LUN recovery (echo 0 > /sys/bus/scsi/devices/H:C:T:L/zfcp_failed) does not help, as the LUN links to the old "removed" port which remains to lack ZFCP_STATUS_COMMON_RUNNING [zfcp_erp_required_act]. The only workaround is to first ensure that the fc_rport is blocked (e.g. port_remove again in case it was re-discovered by (auto) port scan), then delete the SCSI devices, and finally re-discover by (auto) port scan. The port scan includes an fc_rport unblock, which in turn triggers a new scan on the scsi target to freshly get new pure auto scan LUNs. Fix this by rejecting port_remove also if there are SCSI devices (even without any zfcp_unit) under this port. Re-use mechanics from v3.7 commit d99b601b6338 ("[SCSI] zfcp: restore refcount check on port_remove"). However, we have to give up zfcp_sysfs_port_units_mutex earlier in unit_add to prevent a deadlock with scsi_host scan taking shost->scan_mutex first and then zfcp_sysfs_port_units_mutex now in our zfcp_scsi_slave_alloc(). Signed-off-by: Steffen Maier Fixes: b62a8d9b45b9 ("[SCSI] zfcp: Use SCSI device data zfcp scsi dev instead of zfcp unit") Fixes: f8210e34887e ("[SCSI] zfcp: Allow midlayer to scan for LUNs when running in NPIV mode") Cc: #2.6.37+ Reviewed-by: Benjamin Block Signed-off-by: Martin K. Petersen Signed-off-by: Greg Kroah-Hartman