- 09 Jun, 2009 11 commits
-
-
Kashyap, Desai authored
1.) rewrite taskmanagement request and completion routines, making them single threaded and using the generic MPT_MGMT struct, deleting mptscsih_TMHandler, replacing with single request TM handler mptscsih_IssueTaskMgmt, and killing the watchdog timer functions. 2.) cleanup ioc_reset callback handlers, introducing wrappers for synchronizing error recovery (mpt_set_taskmgmt_in_progress_flag, mpt_clear_taskmgmt_in_progress_flag), as the fusion firmware only handles one task management request at a time Signed-off-by:
Kashyap Desai <kadesai@lsi.com> Signed-off-by:
James Bottomley <James.Bottomley@HansenPartnership.com>
-
Kashyap, Desai authored
Rewrite of all internal generated functions that issue commands to firmware, porting them to be single threaded using the generic MPT_MGMT struct. Implemented using completion Queue. Signed-off-by:
-
Kashyap, Desai authored
1) Previously we had mutliple #defines to use same values. Now those #defines are optimized. MPT_IOCTL_STATUS_* is removed and MPT_MGMT_STATUS_* are new #defines. 2.) config path is optimized. Instead of wait Queue and timer, using completion Q. 3.) mpt_timer_expired is not used. [jejb: elide patch to eliminate mpt_timer_expired] Signed-off-by:
-
Kashyap, Desai authored
SendEventNotification was handled through FIFO, now it is using doorbell to communicate with hardware. Added Sleep Flag as an extra argument to support Can-Sleep feature. Resending patch including compilation error fix reviewed by Grant Grundler. Signed-off-by:
-
Kashyap, Desai authored
sas_discovery_quiesce_io flag is used to control IO start/resume functionality. IO will be stoped while doing discovery of topology. Once discovery is completed It will resume IO. Resending patch including James review. Signed-off-by:
-
Kashyap, Desai authored
The reason for this change is there is a data corruption when four different physical memory regions in the 36GB to 37GB region are accessed. This is only affecting 1078. The solution is we need to use different addressing when filling in the scatter gather table for the effected memory regions. So instead of snooping on all four different memory holes, we treat any physical addresses in the 36GB address with the same algorithm. The fix is explained below 1) Ensure that the message frames are NOT located in the trouble region. There is no remapping available for message frames, they must be allocated outside the problem region. 2) Ensure that Sense buffers are NOT in the trouble region. There is no remapping available. 3) Walk through the SGE entries and if any are inside the trouble region then they need to be remapped as discussed below. 1) Set the Local Address bit in the SGE Flags field. MPI_SGE_FLAGS_LOCAL_ADDRESS 2) Ensure we are using 64-bit SGEs 3) Set MSb (Bit 63) of the 64-bit address, this will indicate buffer location is Host Memory. Signed-off-by:
-
Alan Cox authored
Closes-bug: http://bugzilla.kernel.org/show_bug.cgi?id=13438 Closes-bug: http://bugzilla.kernel.org/show_bug.cgi?id=13437Signed-off-by:
Alan Cox <alan@linux.intel.com> Signed-off-by:
-
Michael Chan authored
New iSCSI driver for Broadcom BNX2 devices. The driver interfaces with the CNIC driver to access the hardware. Signed-off-by:
Anil Veerabhadrappa <anilgv@broadcom.com> Signed-off-by:
Michael Chan <mchan@broadcom.com> Signed-off-by:
Mike Christie <michaelc@cs.wisc.edu> Signed-off-by:
-
Michael Chan authored
The CNIC driver controls BNX2 hardware rings and resources used by iSCSI. Most hardware resources for iSCSI are separate from those used for ethernet networking. iSCSI uses a separate MAC address and IP address. The CNIC driver creates a UIO interface to handle the non-offloaded packets such as ARP, etc in userspace. Signed-off-by:
David S. Miller <davem@davemloft.net> Signed-off-by:
-
Michael Chan authored
Add interface and functions to support a new CNIC driver to drive the Broadcom bnx2 hardware for iSCSI offload. Signed-off-by:
-
Michael Chan authored
Add ISCSI_NETLINK messages for iSCSI NICs to get information such as path from userspace. Original iscsid messages are now always sent as multicast to group 1. The new messages are sent to group 2. The multicast changes were made by Mike Christie. Signed-off-by:
Benjamin Li <benli@broadcom.com> Signed-off-by:
-
- 08 Jun, 2009 29 commits
-
-
Brian King authored
Add support to ibmvscsi for the capabilities MAD. This command gets sent to the Virtual I/O server prior to login in order to communicate client capabilities. Additionally it returns information regarding capabilities that the server supports. The two main capabilities communicated in this MAD are related to partition migration and client reserve. Client reserve allows for SCSI-2 reservations to be sent to virtual disks which are backed by physical LUNs and will result in the reservation being sent to the physical LUN. Signed-off-by:
Brian King <brking@linux.vnet.ibm.com> Signed-off-by:
-
Robert Jennings authored
A new mode of error reporting, fast fail, has been added to the VIOS which allows failover to happen more quickly. If this new fast fail mode is enabled on the VIOS and the vSCSI client supports the mode, the VIOS will not return MEDIUM error on path failures, but rather return VIOSRP_ADAPTER_FAIL in the crq response, which ibmvscsi will translate to DID_ERROR. This new mode can be enabled for single path configurations as well, so it is the new default error reporting mode. A module parameter is provided to disable this new behavior on the off chance it causes a problem on some old VIOS version. Signed-off-by:
-
Brian King authored
The ibmvscsi driver currently sends the SRP Login before sending the Adapter Info MAD, which can result in commands getting sent to the virtual adapter before we are ready for them. This results in a slight window where the target devices may not behave as expected. Change the order and close the window. Signed-off-by:
-
Robert Jennings authored
Previously we had one timeout that was used for all types of operations. This adds specific timeout values for different operations (init, login, adapter info MAD, abort task, and LUN reset). Signed-off-by:
-
Brian King authored
Adds support for 16 byte CDBs to the ibmvscsi driver. Signed-off-by:
-
Andrew Vasquez authored
Signed-off-by:
Andrew Vasquez <andrew.vasquez@qlogic.com> Signed-off-by:
-
Andrew Vasquez authored
Ensure MPS remains in synchronization across all NIC/FCoE functions after a reset. Signed-off-by:
-
Andrew Vasquez authored
Signed-off-by:
-
Andrew Vasquez authored
Queued work processing will now be serialized with its own lower-priority spinlock. This also simplifies the work-queue interface for future work-queue consumers. Signed-off-by:
-
Andrew Vasquez authored
As firmware will ultimately terminate (stop) and port states-cleared. Signed-off-by:
-
Andrew Vasquez authored
ISP24xx and above must query the host-status register, not HCCR. Signed-off-by:
-
Andrew Vasquez authored
ISP24xx and above ISPs perform a RISC reset in qla24xx_reset_chip(), which is called prior to qla24xx_chip_diag(). Signed-off-by:
-
Andrew Vasquez authored
Signed-off-by:
-
Andrew Vasquez authored
Signed-off-by:
-
Andrew Vasquez authored
Signed-off-by:
-
Andrew Vasquez authored
With RSCN states not being kept across qla2x00_configure_loop() invocations, loop-resync distruptions during fabric-discovery may cause ports to remain in a lost state. Force state renegotiation during a follow-on configure-loop iteration. Signed-off-by:
-
Andrew Vasquez authored
Unlike earlier ISPs, recent ISPs (ISP81xx) can in fact fail this mailbox command. Signed-off-by:
-
Anirban Chakraborty authored
This patch fixes a NULL pointer bug that occurs when IO is being carried out on a vport for which the cpu affinity mode is turned on. Signed-off-by:
Anirban Chakraborty <anirban.chakraborty@qlogic.com> Signed-off-by:
-
Andrew Vasquez authored
The SCSI-midlayer's fast-fail codes consider an DID_ERROR status as a driver-error and the failed I/O would then be retried in the midlayer without being fast-failed to dm-multipath. DID_BUS_BUSY status returns would induce unneeded path-failures events being propagated to the DM/MD. Signed-off-by:
-
Andrew Vasquez authored
In case the onboard firmware is unable to be read or loaded for operation, attempt to fallback to a limited-operational firmware image stored in a different flash region. This will allow a user to reflash and correct a board with proper operational firmware. Signed-off-by:
-
Anirban Chakraborty authored
Signed-off-by:
-
Andrew Vasquez authored
Signed-off-by:
-
Andrew Vasquez authored
Firmware currently provides PB and PGF TLVs. Signed-off-by:
-
Andrew Vasquez authored
Signed-off-by:
-
Andrew Vasquez authored
Signed-off-by:
-
Joe Eykholt authored
FIP is the FCoE Initialization Protocol and this patch adds the protocol ethertype to the kernel's list of ethertypes. Signed-off-by:
Joe Eykholt <jeykholt@cisco.com> Signed-off-by:
Robert Love <robert.w.love@intel.com> Acked-by:
-
Vasu Dev authored
This is not required as VLAN header is added by device interface driver, this was causing bad FC_CRC in FCoE pkts when using VLAN interface. Signed-off-by:
Vasu Dev <vasu.dev@intel.com> Signed-off-by:
-
Vasu Dev authored
Removes periodic fcoe_watchdog timer used across all fcoe interface maintained in fcoe_hostlist instead added new fcoe_queue_timer per fcoe interface. Added timer is armed only when some pending skb need to be flushed as oppose to periodic 1 second fcoe_watchdog, since now fcoe_queue_timer is used on demand thus set this to 2 jiffies. Now fcoe_queue_timer is much simple than fcoe_watchdog using lock to process all fcoe interface from fcoe_hostlist. I noticed +ve performance result with using 2 jiffies timer as this helps flushing fcoe_pending_queue quickly. Signed-off-by:
-
Vasu Dev authored
Currently fcoe_pending_queue.lock held twice for every new skb adding to this queue when already least one pkt is pending in this queue and that is not uncommon once skb pkts starts getting queued here upon fcoe_start_io => dev_queue_xmit failure. This patch moves most fcoe_pending_queue logic to fcoe_check_wait_queue function, this new logic grabs fcoe_pending_queue.lock only once to add a new skb instead twice as used to be. I think after this patch call flow around fcoe_check_wait_queue calling in fcoe_xmit is bit simplified with modified fcoe_check_wait_queue function taking care of adding and removing pending skb in one function. Signed-off-by:
-
