- 08 May, 2007 16 commits
-
-
Heiko Carstens authored
Must clear adapter failed flag if an fsf request times out. This is necessary because on link down situations the failed flags gets set but the QDIO queues are still up. Since an adapter reopen will be skipped if the failed flag is set an adapter_reopen that is issued on fsf request timeout has no effect if the local link is down. Might lead to locked up system if the SCSI stack is waiting for abort completion. Signed-off-by:
Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by:
Swen Schillig <swen@vnet.ibm.com> Signed-off-by:
James Bottomley <James.Bottomley@SteelEye.com>
-
Heiko Carstens authored
Simplify request ID management and make sure that frequently used functions are inlined. Also fix a memory leak in zfcp_adapter_enqueue() which only gets hit in error handling. Signed-off-by:
-
Christof Schmitt authored
The SCSI stack requires low level drivers to register and unregister devices. For zfcp this leads to the situation where zfcp calls the SCSI stack, the SCSI tries to scan the new device and the scan SCSI command fails. This would require the zfcp erp, but the erp thread is already blocked in the register call. The fix is to make sure that the calls from the ERP thread to the SCSI stack do not block the ERP thread. In detail: 1) Use a workqueue to avoid blocking of the scsi_scan_target calls. 2) When removing a unit make sure that no scsi_scan_target call is pending. 3) Replace scsi_flush_work with scsi_target_unblock. This avoids blocking and has the same result. Signed-off-by:
Christof Schmitt <christof.schmitt@de.ibm.com> Signed-off-by:
-
Christof Schmitt authored
There is a possible race condition while generating the unique request ids and sequence numbers. Both might be read at the same time and have the same value. Fix this by serializing the access through the queue lock of the adapter: First call zfcp_fsf_req_sbal_get that acquires the lock, then read and increment the unique ids. Signed-off-by:
-
Christof Schmitt authored
S_ID and D_ID are defined in the FCP spec as 3 byte fields. Change the output in zfcp print statements accordingly to print them with only 3 bytes. Signed-off-by:
-
Brian King authored
Use a newly added PCI API to issue a PCI Fundamental reset (warm reset) to a new ipr PCI-E adapter. Typically, the ipr adapter uses the start BIST bit in config space to reset an adapter. Issuing start BIST on this particular adapter results in the PCI-E logic on the card losing sync, which causes PCI-E errors, making the card unusable. The only reset mechanism that exists on this hardware that does not have this problem is PCI Fundamental reset (warm reset). Signed-off-by:
Brian King <brking@linux.vnet.ibm.com> Signed-off-by:
-
Andrew Vasquez authored
Signed-off-by:
Andrew Vasquez <andrew.vasquez@qlogic.com> Signed-off-by:
-
Andrew Vasquez authored
Supported ISP types include ISP2422 and revision-2 type ISP2432 chips. Signed-off-by:
-
Andrew Vasquez authored
Signed-off-by:
-
Andrew Vasquez authored
Signed-off-by:
-
Andrew Vasquez authored
Hardcoding the qlport_down_retry module-parameter effectively disallowed any user-defined NVRAM setting to go into effect. Signed-off-by:
-
Andrew Vasquez authored
Remove a stale check against ha->device_flags (DFLG_NO_CABLE) as topology scanning is performed within the DPC-thread context. Signed-off-by:
-
Christof Schmitt authored
For each request that is sent to the FCP adapter, zfcp allocates memory. Status information and data that is being read from the device is written to this memory by the hardware. After that, the hardware signals this via the response queue and zfcp continues processing. Now, if zfcp detects that there is a signal for an incoming response from the hardware, but there is no outstanding request for that request id, then some memory that can be in use anywhere in the system has just been overwritten. This should never happen, but if it does, stop the system with a panic. Signed-off-by:
-
Olaf Hering authored
_ convert void* to struct mesh_state* - remove unused irq argument from mesh_interrupt() Signed-off-by:
Olaf Hering <olaf@aepfle.de> Cc: James Bottomley <James.Bottomley@steeleye.com> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
-
Amol Lad authored
For this driver cond_resched() seems to be a better alternative Signed-off-by:
Amol Lad <amol@verismonetworks.com> Signed-off-by:
-
walter harms authored
drivers/scsi/megaraid.c: In function 'megaraid_probe_one': drivers/scsi/megaraid.c:4893: warning: implicit declaration of function 'mega_create_proc_entry' drivers/scsi/megaraid.c: In function 'megaraid_remove_one': drivers/scsi/megaraid.c:4968: warning: unused variable 'buf' Fix by adding #defines Signed-off-by:
walter harms <wharms@bfs.de> Signed-off-by:
Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by:
-
- 06 May, 2007 24 commits
-
-
Salyzyn, Mark authored
Correct SUN products in aacraid documentation, preliminary names were changed from internal project to customer product prior to release. Signed-off-by:
Mark Salyzyn <aacraid@adaptec.com> Signed-off-by:
-
Salyzyn, Mark authored
The kexec patch introduced a superfluous (and otherwise inert) reset of some adapters. The register can have a hardware default value that has zeros for the undefined interrupts. This patch refines the test of the interrupt enable register to focus on only the interrupts that affect the driver in order to detect if an incomplete shutdown of the Adapter had occurred (kdump). Signed-off-by:
-
Salyzyn, Mark authored
Another layer on this onion also discovered by Duane, the interrupt enable handler also needed to be set ... The interrupt enable was called from within the synchronous command handler. Signed-off-by:
-
Guennadi Liakhovetski authored
Dynamically map the buffer for PIO for the residue byte. Signed-off-by:
G. Liakhovetski <g.liakhovetski@gmx.de> Signed-off-by:
-
Guennadi Liakhovetski authored
cpu_to_le32 endianness conversions in tmscsim.c, followed by arithmetic operations don't look correct. Besides, {in,out}[wl] already perform the necessary conversions. Further, bus addresses of request buffers are guaranteed to be (mapped) under 4G by current scsi- and block-layer defaults. This could be explicitly enforced by using blk_queue_bounce_limit(), which, however, doesn't seem to be the common practice among SCSI drivers. Signed-off-by: -
Guennadi Liakhovetski authored
DMA_INT code is disabled since 1998, remove it to prepare for further cleanup. Signed-off-by:
-
Brian King authored
If an ipr adapter encounters an adapter error requiring an adapter reset to recover from prior to driver load time, the error will be ignored and recovery will not happen until the initial timeout occurs waiting for the firmware to come ready, which means a five minute timeout. Fix is to read the interrupt register before clearing any of the interrupts at probe time. Signed-off-by:
-
Christoph Hellwig authored
Now that we have the much better esp_scsi driver and low level drivers are easy to port over deprecate the old NCR53C9x driver. I've Cc'ed the m68k and mips lists because all but one bus glues are for these platforms. Chances stand bad for the remaining driver, mca_53c9x which hasn't gotten any non-trivial update since it was merge in late 2.1.x and whos maintainers mail address bounces. Signed-off-by:
Christoph Hellwig <hch@lst.de> Signed-off-by:
-
James Smart authored
Per the comment in the change - it's not always prudent to immediately remove the rport upon first notice of a disconnect. Make all rports wait dev_loss_tmo before being deleted (and each could have a separate dev_loss_tmo value). The original post was: http://marc.info/?l=linux-scsi&m=117392196006703&w=2 The repost contains the following changes: - Bug fix in fc_starget_delete(). Dev_loss_tmo_callbk() was called prior to tearing down the target. The callback is to be the last thing called, as it tells the LLDD that the rport is completely finished and can be torn down. Rework so that terminate_rport_io() is called to terminate the outstanding io. Isolated work so it's is simply "starget" work. - Fix holes in original patch. There were code paths that did not expect the dev_loss_tmo timer to be running for the non-fcp rports. - Bug Fix: the transport wasn't protecting against a LLDD calling fc_remote_port_delete() back-to-back. Thus, the dev_loss_tmo timer could be restarted such that it fires after the rport had been deleted. Validate rport state before starting the timer. Signed-off-by:
James Smart <James.Smart@emulex.com> Signed-off-by:
-
Brian King authored
Enables multi-initiator support on ipr RAID adapters that support it. Signed-off-by:
-
Brian King authored
Formats ipr dual adapter errors so that they are more compact. Signed-off-by:
-
Brian King authored
Increases the adapter operational timeout for some adapters that support dual controller configurations, since they may take longer to come ready. Signed-off-by:
-
Brian King authored
In ipr dual adapter configurations, the ipr adapter firmware may require an adapter reset for various reasons. The reset is requested by the adapter firmware logging an error with an IOASC of 0x02048000. Add support to log this error, and reset the adapter. Signed-off-by:
-
Brian King authored
On newer levels of microcode for ipr RAID adapters supporting multi-initiator configurations, the disk array, or VSET, resources are capable of generating a check condition. This patch prevents ipr from generating sense data in this scenario and retrieving it from the logical device instead. Signed-off-by:
-
Brian King authored
On certain ipr RAID adapters, which are capable of multi-initiator configurations, the disk array, or VSET, resources will be in a POR Unit Attention state following an adapter reset. In order for the midlayer to handle the UA, I must report a bus reset to the logical disk array bus at the end of an adapter reset. Signed-off-by:
-
Brian King authored
Since driver_data for pci ids in the ipr driver is now just flags, we can allow these to be passed in from userspace for dynamic ids. Signed-off-by:
-
Brian King authored
This patch fixes some scenarios where an ipr adapter could get reset overlapped, which could cause very long timeouts to occur, or PCI bus errors. Signed-off-by:
-
Christoph Hellwig authored
Signed-off-by:
-
Adrian Bunk authored
- make needlessly global code static - #if 0 unused functions Signed-off-by:
Adrian Bunk <bunk@stusta.de> Acked-by:
Ravi Anand <ravi.anand@qlogic.com> Signed-off-by:
-
James Smart authored
Change version number to 8.1.12 Signed-off-by:
-
James Smart authored
Update copyright year to 2007 Signed-off-by:
-
James Smart authored
Added support for 8G speed and new HBAs. Signed-off-by:
-
James Smart authored
Add support for async scanning Notes: This is the async scan patch to our driver from Matthew Wilcox. The async scan logic is still subject to errors in insmod/rmmod, as the async scan threads don't get shutdown when the module unloads underneath them. See http://marc.info/?l=linux-scsi&m=117551999925582&w=2Signed-off-by:
-
James Smart authored
Don't process ERATT interrupts when issuing KILL_BOARD mbx command Signed-off-by:
-
