- 27 Jul, 2011 38 commits
-
-
Dan Rosenberg authored
There's a code path in pmcraid that can be reached via device ioctl that causes all sorts of ugliness, including heap corruption or triggering the OOM killer due to consecutive allocation of large numbers of pages. First, the user can call pmcraid_chr_ioctl(), with a type PMCRAID_PASSTHROUGH_IOCTL. This calls through to pmcraid_ioctl_passthrough(). Next, a pmcraid_passthrough_ioctl_buffer is copied in, and the request_size variable is set to buffer->ioarcb.data_transfer_length, which is an arbitrary 32-bit signed value provided by the user. If a negative value is provided here, bad things can happen. For example, pmcraid_build_passthrough_ioadls() is called with this request_size, which immediately calls pmcraid_alloc_sglist() with a negative size. The resulting math on allocating a scatter list can result in an overflow in the kzalloc() call (if num_elem is 0, the sglist will be smaller than expected), or if num_elem is unexpectedly large the subsequent loop will call alloc_pages() repeatedly, a high number of pages will be allocated and the OOM killer might be invoked. It looks like preventing this value from being negative in pmcraid_ioctl_passthrough() would be sufficient. Signed-off-by:
Dan Rosenberg <drosenberg@vsecurity.com> Cc: <stable@kernel.org> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
James Bottomley <JBottomley@Parallels.com>
-
Luben Tuikov authored
If expander discovery fails (sas_discover_expander()), remove the expander from the port device list (sas_ex_discover_expander()), before freeing it. Else the list is corrupted and, e.g., when we attempt to send SMP commands to other devices, the kernel oopses. Signed-off-by:
Luben Tuikov <ltuikov@yahoo.com> Cc: stable@kernel.org Reviewed-by:
Jack Wang <jack_wang@usish.com> Signed-off-by:
-
adam radford authored
Signed-off-by:
Adam Radford <aradford@gmail.com> Signed-off-by:
-
adam radford authored
Signed-off-by:
-
adam radford authored
Signed-off-by:
-
Bhanu Prakash Gollapudi authored
Based on PRLI response, identify if the target is FCP-2 (seq level error recovery) capable, and appropriately set the corresponding CONF, REC flags when offloading the session. Signed-off-by:
Bhanu Prakash Gollapudi <bprakash@broadcom.com> Signed-off-by:
-
Bhanu Prakash Gollapudi authored
Currently, bnx2fc has a hba structure that can work with only a single vlan interface. When there is a change in vlan id, it does not have the capability to switch to different vlan interface. To solve this problem, a new structure called 'interface' has been introduced, and each hba can now have multiple interfaces, one per vlan id. Most of the patch is a moving the interface specific fields from hba to the interface structure, and appropriately modifying the dereferences. A list of interfaces (if_list) is maintained along with adapter list. During a create call, the interface structure is allocated and added to if_list and deleted & freed on a destroy call. Link events are propagated to all interfaces belonging to the hba. Signed-off-by:
-
Stephen M. Cameron authored
In a shared SAS setup, target devices may be reset by one of several hosts, and outstanding commands on that device will be completed to corresponding hosts with status of UNSOLICITED_ABORT. Such commands should be retried instead of being treated as i/o errors. Also fixed a nearby spelling error. Signed-off-by:
Stephen M. Cameron <scameron@beardog.cce.hp.com> Signed-off-by:
-
Stephen M. Cameron authored
Most smartarrays tolerate it, but a few new ones don't. Without this change some newer Smart Arrays will lock up and i/o will grind to a halt. Signed-off-by:
-
James Smart authored
Change driver version to 8.3.25 Signed-off-by:
Alex Iannicelli <alex.iannicelli@emulex.com> Signed-off-by:
James Smart <james.smart@emulex.com> Signed-off-by:
-
James Smart authored
This patch implements a new FCF failover policy for the lpfc driver. It allows the driver to choose which FCF to failover to based on the FCF priority. This patch also introduces a new sysfs parameter (fcf_failover_policy) to allow the user to choose which FCF failover policy to use. Signed-off-by:
-
James Smart authored
Enhancements to Debug infrastructure - debugfs additions for new hardware. - Correct stack overflow in lpfc_debugfs_dumpHBASlim_data() - Correct warning on uninitialized reg_val in lpfc_idiag_drbacc_write() - Separated the iDiag command for capturing mailbox commands for generic issue mailbox command entry point and for BSG multi-buffer handling. - Added capturing dumping capabiliy of mailbox command and external buffer for the completion of the mailbox command so that the outcome can be examined. - Changed all the iDiag command structure data array indexing introduced so far with properly defined macros. - Added SLI4 device PCI BAR memory mapped register read/browse, write-by- value, set-bit, and clear-bit methods for both interface type 0 and interface type 2. - Corrected warnings on mbxstatus being uninitialized in error paths in lpfc_bsg.c Signed-off-by:
-
James Smart authored
PCI and SR-IOV Fixes - Call pci_save_state after the pci_restore_state completes. - After calling pci_enable_pcie_error_reporting() and checking the return value for logging messages from rc, reset rc to 0 to it will not later be interpreted for error. - Read PCI config space SR-IOV capability to get the number of VFs supported. - Check for the PF's supported number of VFs before invoking PCI enable sriov API call and log error message that user requested number of VFs is beyond the PF capability if such request is passed in. - Added check for Physical function with Virtual Functions attached. If so, first disable all the VFs before proceeding to device reset. Signed-off-by:
-
James Smart authored
Fabric and Target Discovery Fixes - Clear FC_VPORT_NEEDS_INIT_VPI flag during completion of REG_VFI mailbox command. - Prevent SLI3 Code from unregistering the physical VPI. - Add an else clause to the code that checks and sets sp->cmn.request_multiple_Nport to clear the bit. - Remove a redundant mbox free. - Modified lpfc_sli4_async_fip_evt to pass in physical VPI toi lpfc_find_vport_by_vpid function. - Modified lpfc_find_vport_by_vpid to translate physical VPI to logical VPI before comparing with vport VPI. Signed-off-by:
-
James Smart authored
Adapter Interface fixes and changes - Modify the macro field from lpfc_init_vpi_vpi to lpfc_init_vfi_vpi - Add the new CQE_CODE_RECEIVE_V1 CQE Code, add code in the driver to handle the new Code the same as the CQE_CODE_RECEIVE code except that there are two new checks for this code that will cause the driver to use the new V1 macros for rq_id and fcf_id. - Fix a bug in lpfc_prep_seq() where the size out of the first CQE was ONLY being used, even though multiple dmabufs make up the sequence, each have their own CQE with potentially different sizes. - Fix bug in lpfc_bsg_ct_unsol_event() where the ulpContext and ulpWord[3] fields of the XMIT_SEQUENCE64_CX IOCB were being calculated incorrectly. - Do physical to logical translation before indexing into the active XRI array. - Populate physical vpi in the iocb data structure. - Put the current accumulated total in each IOCB in the chain as we are walking thru then. The last IOCB in the chain should have the total length of the sequence. Signed-off-by:
-
James Smart authored
Miscellaneous Bug fixes and code cleanup - Fix 16G link speed reporting by adding check for 16G check. - Change the check and enforcement of MAILBOX_EXT_SIZE (2048B) to the check and enforcement of BSG_MBOX_SIZE - sizeof(MAILBOX_t) (3840B). - Instead of waiting for a fixed amount of time after performing firmware reset, the driver shall wait for the Lancer SLIPORT_STATUS register for the readiness of the firmware for bring up. - Add logging to indicate when dynamic parameters are changed. - Add revision and date to the firmware image format. - Use revision instead of rev_name to check firmware image version. - Update temporary offset after memcopy is complete for firmware update. - Consolidated the use of the macros to get rid of duplicated register offset definitions. - Removed the unused second parameter in routine lpfc_bsg_diag_mode_enter() - Enable debugfs when debugfs is enabled. - Update function comments for lpfc_sli4_alloc_xri and lpfc_sli4_init_rpi_hdrs. Signed-off-by:
-
James Smart authored
T10 DIF Fixes - Fix the case where the SCSI Host supplies the CRC and driver to controller protection is on. - Only support T10 DIF type 1. LBA always goes in ref tag and app tag is not checked. - Change the format of the sense data passed up to the SCSI layer to match the Descriptor Format Sense Data found in SPC-4 sections 4.5.2.1 and 4.5.2.2. - Fix Slip PDE implementation. - Remove BUG() in else casein lpfc_sc_to_bg_opcodes. Signed-off-by:
-
Krishna Gudipati authored
Signed-off-by:
Krishna Gudipati <kgudipat@brocade.com> Signed-off-by:
-
Krishna Gudipati authored
- Added support to enable / disable lunmasking on Brocade adapter ports. - Added support to query / clear lunmasking configuration. Signed-off-by:
-
Krishna Gudipati authored
- Added support to enable initiator based lun masking. - Initiator based Lun masking works similar to zoning where initiator port is allowed to see only those LUNs which are configured to be seen. Signed-off-by:
-
Krishna Gudipati authored
- Added dconf (Driver Config) BFA sub-module. - The dconf sub-module provides interfaces and manages flash writes to the flash DRV parition. - dconf sub-module also ensures that the whole 64K DRV partition is updated on a flash write. Signed-off-by:
-
Krishna Gudipati authored
Signed-off-by:
-
Krishna Gudipati authored
- Added support to configure QOS on Brocade adapter ports. - Added support to collect / reset QOS statistics. Signed-off-by:
-
Krishna Gudipati authored
- Added logic to enable / disable trunking on Brocade adapter ports. - Added logic to query trunking info. Signed-off-by:
-
Krishna Gudipati authored
Signed-off-by:
-
Krishna Gudipati authored
- Added support to collect and reset fcport stats. - Modified design to allow multiple requests for fcport stats. - fcport will remember the stats request in its stats_pending queue and service each of the queued requests after receiving a firmware response for the prior request. Signed-off-by: -
Krishna Gudipati authored
- Made changes to support IO profiling. - Added support to configure and query IO profiling info. Signed-off-by:
-
Krishna Gudipati authored
Extended BSG to support stats, port log and trace reset and to support adapter, port SET operations. Signed-off-by:
-
Krishna Gudipati authored
Added logic to check the loopback test input speed based on port mode. Signed-off-by:
-
Krishna Gudipati authored
- Made changes to always acknowledge RME interrupt and update consumer index (CI) when RME interrupt is generated. - Made changes to have ASIC specific hw_rspq_ack() handler. Signed-off-by:
-
Krishna Gudipati authored
Signed-off-by:
-
Krishna Gudipati authored
- Added support to post vendor unique events on fc_host. - Supports adapter, port, ioc, flash and remote port based AEN events. Signed-off-by:
-
Chandra Seetharaman authored
rdac hardware handler uses "Subsystem Identifier" from C4 inquiry page to uniquely identify a storage. The problem with that is that if any any of the bytes are non-ascii, subsys_id will all be spaces (hex 0x20). This creates lot of problems especially when there are multiple rdac storages are connected to the server. Use "Storage Array Unique Identifier" from C8 inquiry page, which is the world wide unique identifier for the storage array, to uniquely identify the storage. Signed-off-by:
Chandra Seetharaman <sekharan@us.ibm.com> Signed-off-by:
-
Kashyap, Desai authored
When READ_6 command is issued, the setting of SILI Bit in CDB is confirmed and if SILI bit is off, the processing of relavent Errata is executed. Earlier we did not have check for SILI bit in READ_6 CDB. As described in "ssc-r22", Now Driver is checking SILI bit for READ_6. Signed-off-by:
Kashyap Desai <kashyap.desai@lsi.com> Signed-off-by:
-
Saurav Kashyap authored
Signed-off-by:
Saurav Kashyap <saurav.kashyap@qlogic.com> Signed-off-by:
Giridhar Malavali <giridhar.malavali@qlogic.com> Signed-off-by:
-
Saurav Kashyap authored
The code is changed to support the new dynamic logging infrastructure. Following are the levels added. Default is 0 - no logging. 0x40000000 - Module Init & Probe. 0x20000000 - Mailbox Cmnds. 0x10000000 - Device Discovery. 0x08000000 - IO tracing. 0x04000000 - DPC Thread. 0x02000000 - Async events. 0x01000000 - Timer routines. 0x00800000 - User space. 0x00400000 - Task Management. 0x00200000 - AER/EEH. 0x00100000 - Multi Q. 0x00080000 - P3P Specific. 0x00040000 - Virtual Port. 0x00020000 - Buffer Dump. 0x00010000 - Misc. 0x7fffffff - For enabling all logs, can be too many logs. Setting ql2xextended_error_logging module parameter to any of the above value, will enable the debug for that particular level. Do LOGICAL OR of the value to enable more than one level. Signed-off-by:
Andrew Vasquez <andrew.vasquez@qlogic.com> Signed-off-by:
Joe Carnuccio <joe.carnuccio@qlogic.com> Signed-off-by:
Chad Dupuis <chad.dupuis@qlogic.com> Signed-off-by:
Madhuranath Iyengar <Madhu.Iyengar@qlogic.com> Signed-off-by:
-
Saurav Kashyap authored
This patch adds the dynamic logging framework to the qla2xxx driver. The user will be able to change the logging levels on the fly i.e. without load/unload of the driver. This also enables logging to be enabled for a particular section of the driver such as initialization, device discovery etc. Signed-off-by:
-
James Smart authored
During cable pull tests on our 16G FC adapter, we are seeing errors, typically reads to close targets, which fail due to CRC or framing errors caused by the cable being pull (return status DID_ERROR). The adapter detects the error on one of the first frames received, marks the FC exchange as dead (further frames go to bit bucket) and signals the host of the error. This action is so quick, and coupled with fast host CPUs, creates a scenario in which the midlayer sees the failure and retries the io almost immediately. We've seen link traces with the retry on the link while the original i/o is still being processed by the target. We're also seeing the time window for the "link to pull-apart" and the physical interface to report disconnected to be in the few millisecond range. Which means, we're encountering scenarios where the full retry count is exhausted (all with error) by the midlayer before the link disconnect state is detected. We looked at 8G FC behavior and occasionally see the same behavior, but as the link was slower, it rarely could exhaust all retries before the link reported disconnect. What is needed is a slight delay between io retries due to DID_ERROR to cover this error. It is inappropriate to put this delay in the driver, as the error is indistinguishable from other link-related errors, nor does the driver track whether the io is a retry or not. This is also easier than tracking between-io-error bursts that are seen in this scenario. The patch below updates the retry path so that it inserts a delay as if the target was busy. The busy delay is on the order of 6ms. This delay is sufficient to ensure the link down condition is reported before the retry count is exhausted (at most 1 retry is seen). Signed-off-by:
-
- 26 Jul, 2011 2 commits
-
-
Kashyap, Desai authored
Issue: This issue is seen on LSI H/W WarpDrive SSS6200 When filed direct I/O is tried as volume I/O the scmd field in internal lookup table get cleared and because of that the retried volume I/O never gets reported as completed to SML. Result: I/O timeout and Error handling thread will kicking off Fix: Setting back the scmd in the lookup table before retrying the failed direct i/o Signed-off-by:
-
Bhanu Prakash Gollapudi authored
Signed-off-by:
-
