openwrt/staging/blogic.git
14 years ago[SCSI] be2iscsi: Adding Ring Mode Wrb's V3
Jayamohan Kallickal [Fri, 23 Oct 2009 06:23:49 +0000 (11:53 +0530)]
[SCSI] be2iscsi: Adding Ring Mode Wrb's V3

This patch adds support for ring based wrbs

Signed-off-by: Jayamohan Kallickal <jayamohank@serverengines.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] be2iscsi: Adding msix and mcc_rings V3
Jayamohan Kallickal [Fri, 23 Oct 2009 06:22:33 +0000 (11:52 +0530)]
[SCSI] be2iscsi: Adding msix and mcc_rings V3

This patch enables msix for be2iscsi. It also enables use
of mcc_rings for fw commands. Since the mcc eq creation is
dependent on msix I am sending as one patch

Signed-off-by: Jayamohan Kallickal <jayamohank@serverengines.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] libfc: fix free of fc_rport_priv with timer pending
Joe Eykholt [Wed, 21 Oct 2009 23:28:30 +0000 (16:28 -0700)]
[SCSI] libfc: fix free of fc_rport_priv with timer pending

Timer crashes were caused by freeing a struct fc_rport_priv
with a timer pending, causing the timer facility list to be
corrupted.  This was during FC uplink flap tests with a lot
of targets.

After discovery, we were doing an PLOGI on an rdata that was
in DELETE state but not yet removed from the lookup list.
This moved the rdata from DELETE state to PLOGI state.
If the PLOGI exchange allocation failed and needed to be
retried, the timer scheduling could race with the free
being done by fc_rport_work().

When fc_rport_login() is called on a rport in DELETE state,
move it to a new state RESTART.  In fc_rport_work, when
handling a LOGO, STOPPED or FAILED event, look for restart
state.  In the RESTART case, don't take the rdata off the
list and after the transport remote port is deleted and
exchanges are reset, re-login to the remote port.

Note that the new RESTART state also corrects a problem we
had when re-discovering a port that had moved to DELETE state.
In that case, a new rdata was created, but the old rdata
would do an exchange manager reset affecting the FC_ID
for both the new rdata and old rdata.  With the new state,
the new port isn't logged into until after any old exchanges
are reset.

Signed-off-by: Joe Eykholt <jeykholt@cisco.com>
Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] fnic: Pad the unused bytes of CDB to 0s
Abhijeet Joglekar [Wed, 21 Oct 2009 23:28:25 +0000 (16:28 -0700)]
[SCSI] fnic: Pad the unused bytes of CDB to 0s

Signed-off-by: Abhijeet Joglekar <abjoglek@cisco.com>
Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] fnic: Set max_cmd_len to driver supported CDB length
Abhijeet Joglekar [Wed, 21 Oct 2009 23:28:19 +0000 (16:28 -0700)]
[SCSI] fnic: Set max_cmd_len to driver supported CDB length

Signed-off-by: Abhijeet Joglekar <abjoglek@cisco.com>
Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] fnic: Process all cq entries per ISR
Abhijeet Joglekar [Wed, 21 Oct 2009 23:28:14 +0000 (16:28 -0700)]
[SCSI] fnic: Process all cq entries per ISR

Driver was processing a fixed max number of cq descriptors per ISR. For
instance, for the SCSI IO queue, number of IOs processed per ISR were 8.
If hardware writes 9 cq descriptors to the cq and generates an interrupt,
driver would process only 8 descriptors and decrement the outstanding
credit count by 8. Unless another interrupt event happens, the hw does
not generate any additional interrupt. This results in the cq descriptor
sitting in the queue without being procesed and can cause IO timeouts
and aborts.

Modify all ISR functions to process all queued cq descriptors in one shot.
Since bulk of ELS frame processing is done in thread context and bulk
of SCSI IO processing is done in soft ISR deferred context, the cycles
spent in the ISR per cq descriptor is small.

Signed-off-by: Herman Lee <hermlee@cisco.com>
Signed-off-by: Abhijeet Joglekar <abjoglek@cisco.com>
Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] libfc: fix memory corruption caused by double frees and bad error handling
Chris Leech [Wed, 21 Oct 2009 23:28:09 +0000 (16:28 -0700)]
[SCSI] libfc: fix memory corruption caused by double frees and bad error handling

I was running into several different panics under stress, which I traced down
to a few different possible slab corruption issues in error handling paths.
I have not yet looked into why these exchange sends fail, but with these
fixes my test system is much more stable under stress than before.

fc_elsct_send() could fail and either leave the passed in frame intact
(failure in fc_ct/els_fill) or the frame could have been freed if the
failure was is fc_exch_seq_send().  The caller had no way of knowing, and
there was a potential double free in the error handling in fc_fcp_rec().

Make fc_elsct_send() always free the frame before returning, and remove the
fc_frame_free() call in fc_fcp_rec().

While fc_exch_seq_send() did always consume the frame, there were double free
bugs in the error handling of fc_fcp_cmd_send() and fc_fcp_srr() as well.

Numerous calls to error handling routines (fc_disc_error(),
fc_lport_error(), fc_rport_error_retry() ) were passing in a frame pointer that
had already been freed in the case of an error.  I have changed the call
sites to pass in a NULL pointer, but there may be more appropriate error
codes to use.

Question:  Why do these error routines take a frame pointer anyway?  I
understand passing in a pointer encoded error to the response handlers, but
the error routines take no action on a valid pointer and should never be
called that way.

Signed-off-by: Chris Leech <christopher.leech@intel.com>
Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] fcoe: Call ndo_fcoe_enable/disable to turn FCoE feature on/off in LLD
Yi Zou [Wed, 21 Oct 2009 23:28:03 +0000 (16:28 -0700)]
[SCSI] fcoe: Call ndo_fcoe_enable/disable to turn FCoE feature on/off in LLD

Calls ndo_fcoe_enabled() of the associated netdev upon creating the FCoE
instance to make sure LLD has all necessary resources allocated and setup
properly before passing FCoE traffic. Similarly, calls ndo_fcoe_disable()
upon destroying the FCoE instance on the associated netdev to allow the LLD
to release all allocated resources for FCoE.

Signed-off-by: Yi Zou <yi.zou@intel.com>
Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] libfc: Fix frags in frame exceeding SKB_MAX_FRAGS in fc_fcp_send_data
Yi Zou [Wed, 21 Oct 2009 23:27:58 +0000 (16:27 -0700)]
[SCSI] libfc: Fix frags in frame exceeding SKB_MAX_FRAGS in fc_fcp_send_data

In case of sequence offload, in fc_fcp_send_data(), the skb_fill_page_info()
called may end up adding more frags to the skb_shinfo(fp_skb(fp))->frags[],
exceeding SKB_MAX_FRAGS, this eventually corrupts the memory. I am adding the
FR_FRAME_SG_LEN back, but as SKB_MAX_FRAGS -1, leaving 1 for our fcoe_eof_crc
page. And send will be broken into multiple large sends if the frame already
contains more frags than skb handle.

Signed-off-by: Yi Zou <yi.zou@intel.com>
Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] fcoe: Use NETIF_F_FCOE_MTU flag to set up max frame size (lport->mfs)
Yi Zou [Wed, 21 Oct 2009 23:27:52 +0000 (16:27 -0700)]
[SCSI] fcoe: Use NETIF_F_FCOE_MTU flag to set up max frame size (lport->mfs)

Add a define of FCOE_MTU as 2158 bytes and use FCOE_MTU when the LLD is found
to support NETIF_F_FCOE_MTU. The lport->mfs is then calculated out of the
2158 FCOE_MTU. Otherwise, we stick with the netdev->mtu, i.e., LAN MTU. Also,
change the notification on NETDEV_CHANGEMTU event to bypass changing mfs when
LAN MTU is changed if NETIF_F_FCOE_MTU is supported.

Signed-off-by: Yi Zou <yi.zou@intel.com>
Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] fcoe: initialize return value in fcoe_destroy
Mike Christie [Wed, 21 Oct 2009 23:27:44 +0000 (16:27 -0700)]
[SCSI] fcoe: initialize return value in fcoe_destroy

When doing echo ethX > /sys..../destroy I am getting
errors when the tear down succeeds. It looks like the
reason for this is because the rc var is not getting set
when the destruction works. This just sets it to zero.

Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] libfc: removes unused disc_work and ex_list
Vasu Dev [Wed, 21 Oct 2009 23:27:39 +0000 (16:27 -0700)]
[SCSI] libfc: removes unused disc_work and ex_list

Reported-by: Alex Lyakas <alexl@mellanox.co.il>
Signed-off-by: Vasu Dev <vasu.dev@intel.com>
Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] libfc: adds missing exch release for accepted RRQ
Vasu Dev [Wed, 21 Oct 2009 23:27:33 +0000 (16:27 -0700)]
[SCSI] libfc: adds missing exch release for accepted RRQ

Adds missing exch release when RRQ is accepted by calling
fc_seq_ls_acc. Adds common exch release for fc_exch_els_rrq
by use of out label.

Reported-by: Alex Lyakas <alexl@mellanox.co.il>
Signed-off-by: Vasu Dev <vasu.dev@intel.com>
Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] libfc: removes initializing fc_cpu_order and fc_cpu_mask per lport
Vasu Dev [Wed, 21 Oct 2009 23:27:28 +0000 (16:27 -0700)]
[SCSI] libfc: removes initializing fc_cpu_order and fc_cpu_mask per lport

Initializing these libfc globals per lport could mess up exch
allocation/free for existing lport.

So this patch moves their initialization to fc_setup_exch_mgr
so that these globals gets initialized only once for libfc.

Reported-by: Alex Lyakas <alexl@mellanox.co.il>
Signed-off-by: Vasu Dev <vasu.dev@intel.com>
Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] libfc: don't WARN_ON in lport_timeout for RESET state
Joe Eykholt [Wed, 21 Oct 2009 23:27:22 +0000 (16:27 -0700)]
[SCSI] libfc: don't WARN_ON in lport_timeout for RESET state

It's possible and harmless to get FLOGI timeouts
while in RESET state.  Don't do a WARN_ON in that case.

Also, split out the other WARN_ONs in fc_lport_timeout, so
we can tell which one is hit by its line number.

Signed-off-by: Joe Eykholt <jeykholt@cisco.com>
Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] libfc: lport: fix minor documentation errors
Joe Eykholt [Wed, 21 Oct 2009 23:27:17 +0000 (16:27 -0700)]
[SCSI] libfc: lport: fix minor documentation errors

Fix minor errors.
A debug message said an RLIR was received instead of ECHO.
"Expected" was misspelled in several places.
Fix a type cast from u32 to __be32.

Rob, Some of these may have been also taken care of in your
other doc cleanup patch.  Feel free to fold them in.

Signed-off-by: Joe Eykholt <jeykholt@cisco.com>
Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] libfc: Fix wrong scsi return status under FC_DATA_UNDRUN
Yi Zou [Wed, 21 Oct 2009 23:27:12 +0000 (16:27 -0700)]
[SCSI] libfc: Fix wrong scsi return status under FC_DATA_UNDRUN

This bug is exposed when there is a link flap in LLD. Particularly, when it
happens right after a SCSI write command is sent out, no FCP_DATA is sent,
causing fsp->status_code to be set as FC_DATA_UNDRUN in fc_fcp_complete_locked
even no SCSI status is received. Consequently, fc_io_compl treats this as DID_OK.
This results in SCSI returning successful to the initial I/O request even
there is no DATA actually sent. Particularly, if you run an I/O tool w/ data
verification on, the read back for verification is gonna fail.

This is fixed here by checking when FC_DATA_UNDRUN happens, SCSI status is
received w/ FC_SRB_RCV_STATUS set in fsp->state.

Signed-off-by: Yi Zou <yi.zou@intel.com>
Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] libfc: Remove unused fc_lport pointer from fc_fcp_pkt_abort
Robert Love [Wed, 21 Oct 2009 23:27:06 +0000 (16:27 -0700)]
[SCSI] libfc: Remove unused fc_lport pointer from fc_fcp_pkt_abort

This argument isn't used, let's not pass it into the routine.

Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] libfc, fcoe: Don't EXPORT_SYMBOLS unnecessarily
Robert Love [Wed, 21 Oct 2009 23:27:01 +0000 (16:27 -0700)]
[SCSI] libfc, fcoe: Don't EXPORT_SYMBOLS unnecessarily

These are a few functions that were not used by other
modules. They did not need to be exported so this patch
removes the EXPORT_SYMBOLS call for each.

Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] fcoe: remove redundant checking of netdev->netdev_ops
Yi Zou [Wed, 21 Oct 2009 23:26:55 +0000 (16:26 -0700)]
[SCSI] fcoe: remove redundant checking of netdev->netdev_ops

Remove the redundant checking of netdev->netdev_ops as it will never be NULL.

Signed-off-by: Yi Zou <yi.zou@intel.com>
Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] libfc: fix ddp in fc_fcp for 0 xid
Yi Zou [Wed, 21 Oct 2009 23:26:50 +0000 (16:26 -0700)]
[SCSI] libfc: fix ddp in fc_fcp for 0 xid

xid 0 was used as an indication of invalid xid before but now xid 0
can be used as a valid exchange i. This patch fixes the ddp completion
in fcp layer, i.e., in fc_fcp.c:fc_fcp_ddp_done() function, to make sure it
does not use xid 0 for indication of an invalid xid, instead, it now
uses use FC_XID_UNKNOWN for such indication.

Signed-off-by: Yi Zou <yi.zou@intel.com>
Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] libfc: fix typo in retry check on received PRLI
Joe Eykholt [Wed, 21 Oct 2009 23:26:45 +0000 (16:26 -0700)]
[SCSI] libfc: fix typo in retry check on received PRLI

A received Fibre Channel ELS PRLI request contains a bit that
indicates whether the remote port supports certain retry processing
sequences.  The test for this bit was somehow coded to use multiply
instead of AND!

This case would apply only for target mode operation, and it is
unlikely to be noticed as an initiator.

Signed-off-by: Joe Eykholt <jeykholt@cisco.com>
Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] ibmvfc: Driver version 1.0.7
Brian King [Mon, 19 Oct 2009 20:07:55 +0000 (15:07 -0500)]
[SCSI] ibmvfc: Driver version 1.0.7

Bump driver version to 1.0.7.

Signed-off-by: Brian King <brking@linux.vnet.ibm.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] ibmvfc: Add FC Passthru support
Brian King [Mon, 19 Oct 2009 20:07:54 +0000 (15:07 -0500)]
[SCSI] ibmvfc: Add FC Passthru support

Adds support for FC passthru via BSG.

Signed-off-by: Brian King <brking@linux.vnet.ibm.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] ibmvfc: Fix adapter cancel flags for terminate_rport_io
Brian King [Mon, 19 Oct 2009 20:07:53 +0000 (15:07 -0500)]
[SCSI] ibmvfc: Fix adapter cancel flags for terminate_rport_io

When issuing a Cancel to the virtual fibre channel adapter,
the interface specifies a flags field for the client to indicate
what kind of error recovery is being performed. Fix up these
flags for terminate_rport_io to indicate an abort task set
rather than a target reset.

Signed-off-by: Brian King <brking@linux.vnet.ibm.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] ibmvfc: Remove unnecessary parameter to ibmvfc_init_host
Brian King [Mon, 19 Oct 2009 20:07:49 +0000 (15:07 -0500)]
[SCSI] ibmvfc: Remove unnecessary parameter to ibmvfc_init_host

Remove a parameter to ibmvfc_init_host which is always set to
zero by all callers.

Signed-off-by: Brian King <brking@linux.vnet.ibm.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] ibmvfc: Fix locking in ibmvfc_remove
Brian King [Mon, 19 Oct 2009 20:07:48 +0000 (15:07 -0500)]
[SCSI] ibmvfc: Fix locking in ibmvfc_remove

Need to grab the host lock around the call to ibmvfc_link_down.

Signed-off-by: Brian King <brking@linux.vnet.ibm.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] ibmvfc: Fixup TMF response handling
Brian King [Mon, 19 Oct 2009 20:07:47 +0000 (15:07 -0500)]
[SCSI] ibmvfc: Fixup TMF response handling

When processing the response to either a LUN reset,
target reset, or an abort task set, the ibmvfc driver needs to
treat as success receiving a response with a non-zero
status in the response IU along with a general transport
error with the FCP response code being zero. The VIOS
currently guarantees this cannot happen, but a future version
of VIOS may allow this to be returned, so ensure we handle
this response combination correctly for TMFs, as we already
do for SCSI commands.

Signed-off-by: Brian King <brking@linux.vnet.ibm.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] scsi_debug: Thin provisioning support
Martin K. Petersen [Thu, 15 Oct 2009 18:45:27 +0000 (14:45 -0400)]
[SCSI] scsi_debug: Thin provisioning support

This version fixes 64-bit modulo on 32-bit as well as inadvertent map
updates when TP was disabled.

Implement support for thin provisioning in scsi_debug.  No actual memory
de-allocation is taking place.  The intent is to emulate a thinly
provisioned storage device, not to be one.

There are four new module options:

 - unmap_granularity specifies the granularity at which to track mapped
   blocks (specified in number of logical blocks).  2048 (1 MB) is a
   realistic value for disk arrays although some may have a finer
   granularity.

 - unmap_alignment specifies the first LBA which is naturally aligned on
   an unmap_granularity boundary.

 - unmap_max_desc specifies the maximum number of ranges that can be
   unmapped using one UNMAP command.  If this is 0, only WRITE SAME is
   supported and UNMAP will cause a check condition.

 - unmap_max_blocks specifies the maximum number of blocks that can be
   unmapped using a single UNMAP command.  Default is 0xffffffff.

These parameters are reported in the new and extended block limits VPD.

If unmap_granularity is specified the device is tagged as thin
provisioning capable in READ CAPACITY(16).  A bitmap is allocated to
track whether blocks are mapped or not.  A WRITE request will cause a
block to be mapped.  So will WRITE SAME unless the UNMAP bit is set.

Blocks can be unmapped using either WRITE SAME or UNMAP.  No accounting
is done to track partial blocks.  This means that only whole blocks will
be marked free.  This is how the array people tell me their firmwares
work.

GET LBA STATUS is also supported.  This command reports whether a block
is mapped or not, and how long the adjoining mapped/unmapped extent is.

The block allocation bitmap can also be viewed from user space via:

/sys/bus/pseudo/drivers/scsi_debug/map

Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Acked-by: Douglas Gilbert <dgilbert@interlog.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] scsi: Add missing command definitions
Martin K. Petersen [Thu, 15 Oct 2009 18:43:23 +0000 (14:43 -0400)]
[SCSI] scsi: Add missing command definitions

Add definitions for UNMAP, WRITE SAME{16,32} and GET LBA STATUS
commands.

Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] qla2xxx: Update version number to 8.03.01-k7
Giridhar Malavali [Tue, 13 Oct 2009 22:16:53 +0000 (15:16 -0700)]
[SCSI] qla2xxx: Update version number to 8.03.01-k7

Signed-off-by: Giridhar Malavali <giridhar.malavali@qlogic.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] qla2xxx: Properly handle UNDERRUN completion statuses.
Lalit Chandivade [Tue, 13 Oct 2009 22:16:52 +0000 (15:16 -0700)]
[SCSI] qla2xxx: Properly handle UNDERRUN completion statuses.

Correct issues where the lower scsi-status would be improperly
cleared, instead, allow the midlayer to process the status after
the proper residual-count checks are performed.  Finally,
validate firmware status flags prior to assigning values from the
FCP_RSP frame.

Signed-off-by: Lalit Chandivade <lalit.chandivade@qlogic.com>
Signed-off-by: Michael Hernandez <michael.hernandez@qlogic.com>
Signed-off-by: Ravi Anand <ravi.anand@qlogic.com>
Signed-off-by: Andrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: Giridhar Malavali <giridhar.malavali@qlogic.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] qla2xxx: Properly re-register FC4/FDMI after physical and logical link disrupt...
Andrew Vasquez [Tue, 13 Oct 2009 22:16:51 +0000 (15:16 -0700)]
[SCSI] qla2xxx: Properly re-register FC4/FDMI after physical and logical link disruptions.

Original code would not register FC4 nor FDMI information after a
logical tear-down of an VFC link.  Code now triggers registration
date during processing of a 'Report ID Acquisition IOCB', which
is submitted after a FLOGI or FDISC completes.

Signed-off-by: Andrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: Giridhar Malavali <giridhar.malavali@qlogic.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] qla2xxx: Properly check FCP_RSP response-info field after TMF completion.
Andrew Vasquez [Tue, 13 Oct 2009 22:16:50 +0000 (15:16 -0700)]
[SCSI] qla2xxx: Properly check FCP_RSP response-info field after TMF completion.

Original code discarded response-info field information and
assumed the command completed successfully without verifying the
target's status within the FCP_RSP packet.

Signed-off-by: Andrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: Giridhar Malavali <giridhar.malavali@qlogic.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] qla2xxx: Retrieve firmware's maximum number of supported FCFs.
Andrew Vasquez [Tue, 13 Oct 2009 22:16:49 +0000 (15:16 -0700)]
[SCSI] qla2xxx: Retrieve firmware's maximum number of supported FCFs.

Signed-off-by: Andrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: Giridhar Malavali <giridhar.malavali@qlogic.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] qla2xxx: Set the size of the host buffer used to fetch DCBX and XGMAC paramete...
Giridhar Malavali [Tue, 13 Oct 2009 22:16:48 +0000 (15:16 -0700)]
[SCSI] qla2xxx: Set the size of the host buffer used to fetch DCBX and XGMAC parameters to 4K.

Signed-off-by: Giridhar Malavali <giridhar.malavali@qlogic.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] qla2xxx: Reread firmware versions information after an ISP abort.
Lalit Chandivade [Tue, 13 Oct 2009 22:16:47 +0000 (15:16 -0700)]
[SCSI] qla2xxx: Reread firmware versions information after an ISP abort.

In some case, the MPI and PHY versions when retrieved after the
Execute-FW mailbox-command are incorrect (255.255.255.255).
Instead, query the information after the check for firmware ready
is done in the abort ISP path.

Signed-off-by: Lalit Chandivade <lalit.chandivade@qlogic.com>
Signed-off-by: Andrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: Giridhar Malavali <giridhar.malavali@qlogic.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] qla2xxx: Display additional mailbox registers during AEN handling.
Andrew Vasquez [Tue, 13 Oct 2009 22:16:46 +0000 (15:16 -0700)]
[SCSI] qla2xxx: Display additional mailbox registers during AEN handling.

The mailbox register values may assist in debugging efforts.

Signed-off-by: Andrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: Giridhar Malavali <giridhar.malavali@qlogic.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] qla2xxx: Add firmware-dump kobject uevent notification.
Andrew Vasquez [Tue, 13 Oct 2009 22:16:45 +0000 (15:16 -0700)]
[SCSI] qla2xxx: Add firmware-dump kobject uevent notification.

Signed-off-by: Andrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: Giridhar Malavali <giridhar.malavali@qlogic.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] bfa: fixed checkpatch errors for bfad files
Jing Huang [Fri, 25 Sep 2009 19:29:54 +0000 (12:29 -0700)]
[SCSI] bfa: fixed checkpatch errors for bfad files

This patch fixes checkpatch errors/warnings in bfad files.

Signed-off-by: Jing Huang <huangj@brocade.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] scsi_transport_fc: remove invalid BUG_ON
Michael Reed [Fri, 9 Oct 2009 19:15:59 +0000 (14:15 -0500)]
[SCSI] scsi_transport_fc: remove invalid BUG_ON

I was doing some large lun count testing with 2.6.31 and hit
a BUG_ON() in fc_timeout_deleted_rport(), and it seems like it
should have been just a matter of time before someone did.

It seems invalid to set port_state under lock, then expect it to
remain set after releasing the lock.  Another thread called
fc_remote_port_add() when the lock was released, changing the
port_state.

This patch removes the BUG_ON and moves the test of the
port_state to inside the host_lock.  It's been running for
several weeks now with no ill effect.

Signed-off-by: Michael Reed <mdr@sgi.com>
Acked-by: James Smart <james.smart@emulex.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] fix propogation of integrity errors
Mike Christie [Mon, 28 Sep 2009 20:50:52 +0000 (15:50 -0500)]
[SCSI] fix propogation of integrity errors

When the Integrity check is done in scsi_io_completion it will
set error to -EILSEQ. However, at this point error is no longer
used, and blk_end_request_err has -EIO hardcoded.

It looks like there was just porting mistake with this patch
http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=3e695f89c5debb735e4ff051e9e58d8fb4e95110
and we meant to send error upwards, so this patch changes the hard
coded EIO to the error variable.

I have only boot tested this patch.

Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Acked-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] stex: update version to 4.6.0000.4
Ed Lin [Tue, 29 Sep 2009 06:58:36 +0000 (22:58 -0800)]
[SCSI] stex: update version to 4.6.0000.4

Update version to 4.6.0000.4.

Signed-off-by: Ed Lin <ed.lin@promise.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] stex: add support for reset request from firmware
Ed Lin [Tue, 29 Sep 2009 06:58:33 +0000 (22:58 -0800)]
[SCSI] stex: add support for reset request from firmware

Add support for reset request from firmware for controllers
of st_shasta and st_yel type. Code adjustments necessary
for this change are also included.

Signed-off-by: Ed Lin <ed.lin@promise.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] stex: add small dma buffer support
Ed Lin [Tue, 29 Sep 2009 06:58:17 +0000 (22:58 -0800)]
[SCSI] stex: add small dma buffer support

The controllers of st_seq and st_vsc type can work
if only small dma buffer is available, with a reduced
firmware feature set. Add support for this case.

Signed-off-by: Ed Lin <ed.lin@promise.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] mptfusion: Bump version to 3.04.13
Kashyap, Desai [Wed, 7 Oct 2009 05:58:50 +0000 (11:28 +0530)]
[SCSI] mptfusion: Bump version to 3.04.13

Bump version 3.04.13.

Signed-off-by: Kashyap Desai <kashyap.desai@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] mptspi: Fix for incorrect data underrun errata
Kashyap, Desai [Wed, 7 Oct 2009 05:57:40 +0000 (11:27 +0530)]
[SCSI] mptspi: Fix for incorrect data underrun errata

Errata:
Certain conditions on the scsi bus may casue the 53C1030 to incorrectly signal
a SCSI_DATA_UNDERRUN to the host.

Workaround 1:
For an Errata on LSI53C1030 When the length of request data
and transfer data are different with result of command (READ or VERIFY),
DID_SOFT_ERROR is set.

Workaround 2:
For potential trouble on LSI53C1030. It is checked whether the length of
request data is equal to the length of transfer and residual.
MEDIUM_ERROR is set by incorrect data.

Signed-off-by: Kashyap Desai <kashyap.desai@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] mptctl : Remove printk which floods unnecessary messages to var/log/message
Kashyap, Desai [Wed, 7 Oct 2009 05:56:54 +0000 (11:26 +0530)]
[SCSI] mptctl : Remove printk which floods unnecessary messages to var/log/message

Signed-off-by: Kashyap Desai <kashyap.desai@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] megaraid_sas: Update version number and documentation
Yang, Bo [Tue, 6 Oct 2009 20:55:09 +0000 (14:55 -0600)]
[SCSI] megaraid_sas: Update version number and documentation

Signed-off-by Bo Yang<bo.yang@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] megaraid_sas: use the firmware boot timeout when waiting for commands
Yang, Bo [Tue, 6 Oct 2009 20:52:20 +0000 (14:52 -0600)]
[SCSI] megaraid_sas: use the firmware boot timeout when waiting for commands

use the constant MEGASAS_RESET_WAIT_TIME when waiting for firmware
commands to complete (currently 3 minutes).

Signed-off-by Bo Yang<bo.yang@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] megaraid_sas: fix 64 bit sense pointer truncation
Yang, Bo [Tue, 6 Oct 2009 20:52:20 +0000 (14:52 -0600)]
[SCSI] megaraid_sas: fix 64 bit sense pointer truncation

The current sense pointer is cast to a u32 pointer, which can truncate
on 64 bits.  Fix by using unsigned long instead.

Signed-off-by Bo Yang<bo.yang@lsi.com>
Cc: stable@kernel.org
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] megaraid_sas: Add the support for updating the OS after adding/removing the...
Yang, Bo [Tue, 6 Oct 2009 20:50:17 +0000 (14:50 -0600)]
[SCSI] megaraid_sas: Add the support for updating the OS after adding/removing the devices from FW

Driver will update the OS devices after adding and deleting the device
from FW.  When driver receive add or delete AEN from FW, driver will
send the DCMD cmd to get the System PD list from FW.  Then driver will
check if this device already in the OS: If add event and OS don't have
the device (but it is in the list), driver add the device to OS,
otherwise driver will not add.  If remove event, driver will check the
list, if is not in the list, but OS have the device, driver will
remove the device.

Signed-off-by Bo Yang<bo.yang@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] megaraid_sas: Fix the fix for fw hang caused by megaraid sas application
Yang, Bo [Tue, 6 Oct 2009 20:47:35 +0000 (14:47 -0600)]
[SCSI] megaraid_sas: Fix the fix for fw hang caused by megaraid sas application

Add a lock to the skinny firmware initialisation sequence to prevent
the two stage write being non atomic if multiple instances use it.

Add a flag to the driver shutdown sequence to prevent aen ioctls being
called after shutdown begins.

Signed-off-by Bo Yang<bo.yang@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] megaraid_sas: add the IEEE SGE support to SAS2 controller
Yang, Bo [Tue, 6 Oct 2009 20:43:28 +0000 (14:43 -0600)]
[SCSI] megaraid_sas: add the IEEE SGE support to SAS2 controller

To increase the performance, megaraid sas driver added the IEEE SGE
support to support SAS2 controller.

Signed-off-by Bo Yang<bo.yang@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] megaraid_sas: allocate the application cmds to sas2 controller
Yang, Bo [Tue, 6 Oct 2009 20:40:58 +0000 (14:40 -0600)]
[SCSI] megaraid_sas: allocate the application cmds to sas2 controller

MegaRAID SAS2 controller ioctl can't use 32 cmd for applications.
Driver need to divide different number of cmds to IO and application.

Signed-off-by Bo Yang<bo.yang@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] megaraid_sas: report system PDs to OS
Yang, Bo [Tue, 6 Oct 2009 20:33:06 +0000 (14:33 -0600)]
[SCSI] megaraid_sas: report system PDs to OS

When OS issue inquiry, it will check driver's internal pd_list.

Signed-off-by Bo Yang<bo.yang@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] megaraid_sas: infrastructure to get PDs from FW
Yang, Bo [Tue, 6 Oct 2009 20:27:54 +0000 (14:27 -0600)]
[SCSI] megaraid_sas: infrastructure to get PDs from FW

Add system PDs to OS.  Driver implemented the get_pd_list function to
get the system PD from FW.

Signed-off-by Bo Yang<bo.yang@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] megaraid_sas: Add new megaraid SAS 2 controller support to the driver
Yang, Bo [Tue, 6 Oct 2009 20:31:54 +0000 (14:31 -0600)]
[SCSI] megaraid_sas: Add new megaraid SAS 2 controller support to the driver

Add the new megaraid sas 2 controller to the driver.  megaraid sas2 is
LSI next generation SAS products.  driver add the interface to support
this product.

Signed-off-by Bo Yang<bo.yang@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] megaraid_sas: add sysfs for AEN polling
Yang, Bo [Tue, 6 Oct 2009 20:20:59 +0000 (14:20 -0600)]
[SCSI] megaraid_sas: add sysfs for AEN polling

update the sysfs parameter to tell application driver support AEN poll

Signed-off-by Bo Yang<bo.yang@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] megaraid_sas: Add poll mechanism to megaraid sas driver
Yang, Bo [Tue, 6 Oct 2009 20:18:02 +0000 (14:18 -0600)]
[SCSI] megaraid_sas: Add poll mechanism to megaraid sas driver

Add Poll_wait mechanism to SAS-2 MegaRAID SAS Linux driver. Driver
will wakeup poll after the driver get event from MegaRAID SAS FW.

Signed-off-by Bo Yang<bo.yang@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] megaraid_sas: tape drive support fix
Yang, Bo [Tue, 6 Oct 2009 20:12:21 +0000 (14:12 -0600)]
[SCSI] megaraid_sas: tape drive support fix

Add the Tape drive fix to the megaraid_sas driver: If the command is
for the tape device, set the FW pthru timeout to the os layer timeout
value.

Signed-off-by Bo Yang<bo.yang@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] mpt2sas: Bump version 03.100.03.00
Kashyap, Desai [Wed, 23 Sep 2009 12:10:05 +0000 (17:40 +0530)]
[SCSI] mpt2sas: Bump version 03.100.03.00

Bump version to 03.100.03.00

Signed-off-by: Kashyap Desai <kashyap.desai@lsi.com>
Signed-off-by: Eric Moore <Eric.moore@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] mpt2sas: No link rate change, do not call update links nor unblock device
Kashyap, Desai [Wed, 23 Sep 2009 12:06:52 +0000 (17:36 +0530)]
[SCSI] mpt2sas: No link rate change, do not call update links nor unblock device

(1) target resets are sending link change rate events with no link rate
change -> thus said the driver was modified so when there is no link rate
change, we don't need to call mpt2sas_transport_update_links nor
_scsih_ublock_io_device.
(2) There were changes made in _scsih_sas_topology_change_event_debug to
change the debug strings so they are more clear.  Also the link rate change
information was added to display the new and previous link rate.
for the MPI2_EVENT_SAS_TOPOLOGY_CHANGE_LIST event when the ExpStatus is
set to zero, display "responding" instead of "unknown status".

Signed-off-by: Kashyap Desai <kashyap.desai@lsi.com>
Signed-off-by: Eric Moore <Eric.moore@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] mpt2sas : Add support for RAID Action System Shutdown Initiated at OS shutdown
Kashyap, Desai [Mon, 5 Oct 2009 10:26:56 +0000 (15:56 +0530)]
[SCSI] mpt2sas : Add support for RAID Action System Shutdown Initiated at OS shutdown

(1) Added new function _scsih_ir_shutdown.  This function will issue the
MPI2_RAID_ACTION_SYSTEM_SHUTDOWN_INITIATED request via
MPI2_FUNCTION_RAID_ACTION. The function will wait 10 seconds for reply
message frame, then print out the ioc status and loginfo.  This function is
only called when there are raid volumes present.

(2) Add shutdown callback in the struct pci_driver object scsih_driver. This
will be called only when the system is shutting down. From this function, we
will call _scsih_ir_shutdown mentioned above.

(3) Add support in _scsih_remove to call _scsih_ir_shutdown. The function
_scsih_remove will be called when the driver is unloaded (and system is
still running).

scsih internal command contex is added to send internal message frames
from mpt2sas_scsih.c.

Signed-off-by: Kashyap Desai <kashyap.desai@lsi.com>
Signed-off-by: Eric Moore <Eric.moore@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] mpt2sas: Freeze the sdev IO queue when firmware sends internal dev reset
Kashyap, Desai [Wed, 23 Sep 2009 12:05:41 +0000 (17:35 +0530)]
[SCSI] mpt2sas: Freeze the sdev IO queue when firmware sends internal dev reset

When receiving the MPI2_EVENT_SAS_DEV_STAT_RC_INTERNAL_DEVICE_RESET event,
the driver will set the tm_busy flag in the sdev private host data, When
tm_busy flag is set, the driver will return SCSI_MLQUEUE_DEVICE_BUSY,
effectly freezing the IO to the device. The tm_busy flag is cleared with the
MPI2_EVENT_SAS_DEV_STAT_RC_CMP_INTERNAL_DEV_RESET event.

Signed-off-by: Kashyap Desai <kashyap.desai@lsi.com>
Signed-off-by: Eric Moore <Eric.moore@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] mpt2sas : PPC (power pc) endian bug fix's
Kashyap, Desai [Mon, 5 Oct 2009 10:23:06 +0000 (15:53 +0530)]
[SCSI] mpt2sas : PPC (power pc) endian bug fix's

(1) EEDP(End to End data protection) was not working. This was due to not
setting EEDP BlockSize and Flags to little endian format in the message
frame.
(2) Some expander sysfs attributes were not getting set properly.  The sas
format was not getting set due to endian issues with sas_format field in the
struct rep_manu_reply. Since sas_format was not set properly, the
component_vendor_id, component_revision_id, and component_id were not set.
(3) In _transport_smp_handler: we don't need to convert the smid from little
endian to cpu prior to calling mpt2sas_base_free_smid, because its allready
in cpu format. (4) Some loginfos and ioc status were not xonverted from
little endian to cpu.

Signed-off-by: Kashyap Desai <kashyap.desai@lsi.com>
Signed-off-by: Eric Moore <Eric.moore@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] mpt2sas: mpt2sas_base_get_sense_buffer_dma should be returning little endian
Kashyap, Desai [Wed, 23 Sep 2009 12:04:13 +0000 (17:34 +0530)]
[SCSI] mpt2sas: mpt2sas_base_get_sense_buffer_dma should be returning little endian

cpu_to_le64 when calculating the physical dma address. This will properly
handle endianess on big endian systems.  The return value of this function
was changed from dma_addr_t to __le64. Remove the typecasting of u32 when
setting the SenseBufferLowAddress, since its already in __le32 format.

Signed-off-by: Kashyap Desai <kashyap.desai@lsi.com>
Signed-off-by: Eric Moore <Eric.moore@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] mpt2sas: Return DID_TRANSPORT_DISRUPTED in nexus loss,SCSI_MLQUEUE_DEVICE_BUSY...
Kashyap, Desai [Wed, 23 Sep 2009 12:03:14 +0000 (17:33 +0530)]
[SCSI] mpt2sas: Return DID_TRANSPORT_DISRUPTED in nexus loss,SCSI_MLQUEUE_DEVICE_BUSY if device is busy

1 Its observed that the OS was sending request to the driver after it had been
put into blocking state, so the driver was modified to return
SCSI_MLQUEUE_DEVICE_BUSY.
2. Driver will return DID_TRANSPORT_DISRUPTED when sdev is haivng nexus loss.
This occurrs when sdev is blocked, between the
MPI2_EVENT_SAS_TOPO_RC_DELAY_NOT_RESPONDING and
MPI2_EVENT_SAS_TOPO_RC_TARG_NOT_RESPONDING events.

Signed-off-by: Kashyap Desai <kashyap.desai@lsi.com>
Signed-off-by: Eric Moore <Eric.moore@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] mpt2sas: Retrieve the ioc facts prior to putting the controller into READY...
Kashyap, Desai [Wed, 23 Sep 2009 12:02:06 +0000 (17:32 +0530)]
[SCSI] mpt2sas: Retrieve the ioc facts prior to putting the controller into READY state

The driver needs to retrieve the ioc facts prior to putting the controller
into READY state. The current design is calling ioc facts after putting the
controller into READY state, which means the driver is sending a diag reset
instead of message unit reset becuase the capability information is not yet
available.

Signed-off-by: Kashyap Desai <kashyap.desai@lsi.com>
Signed-off-by: Eric Moore <Eric.moore@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] mpt2sas: Added new info messages for IR and Expander events.
Kashyap, Desai [Wed, 23 Sep 2009 12:01:01 +0000 (17:31 +0530)]
[SCSI] mpt2sas: Added new info messages for IR and Expander events.

(1) for the MPI2_EVENT_IR_OPERATION_STATUS event, add support to print
"background init" or "make data consistent" for debugging purposes.  If the
RAIDOperation is set to a value not defined, then don't print anything

(2) for the MPI2_EVENT_SAS_DEVICE_STATUS_CHANGE event, add support to print
"expander reduced functionality" and "expander reduced functionality
complete", which are new events.

Signed-off-by: Kashyap Desai <kashyap.desai@lsi.com>
Signed-off-by: Eric Moore <Eric.moore@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] mpt2sas: Limit the max_depth to 32 for SATA devices which are not part of...
Kashyap, Desai [Wed, 23 Sep 2009 12:00:22 +0000 (17:30 +0530)]
[SCSI] mpt2sas: Limit the max_depth to 32 for SATA devices which are not part of volume

Added sanity check in _scsih_change_queue_depth to limit the max_depth to 32
for SATA devices. This is only for physical devices not part of a volume.

Signed-off-by: Kashyap Desai <kashyap.desai@lsi.com>
Signed-off-by: Eric Moore <Eric.moore@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] mpt2sas: Added support to set the TimeStamp when sending ioc_init
Kashyap, Desai [Wed, 23 Sep 2009 11:59:29 +0000 (17:29 +0530)]
[SCSI] mpt2sas: Added support to set the TimeStamp when sending ioc_init

Added support to set the TimeStamp when sending ioc_init.
Signed-off-by: Kashyap Desai <kashyap.desai@lsi.com>
Signed-off-by: Eric Moore <Eric.moore@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] mpt2sas: Add Extended Type for Diagnostic Buffer support
Kashyap, Desai [Wed, 23 Sep 2009 11:58:59 +0000 (17:28 +0530)]
[SCSI] mpt2sas: Add Extended Type for Diagnostic Buffer support

Added tests for registry entries of EXBuffSize, EXImmed, and EXType to
support the new Extended diag buffer type.  Modified code where necessary to
handle the new ExtendedType field in the F/W diagnostic Post and Release
messages.

Signed-off-by: Kashyap Desai <kashyap.desai@lsi.com>
Signed-off-by: Eric Moore <Eric.moore@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] mpt2sas: Added command line option diag_buffer_enable.
Kashyap, Desai [Wed, 23 Sep 2009 11:58:09 +0000 (17:28 +0530)]
[SCSI] mpt2sas: Added command line option diag_buffer_enable.

Added command line option diag_buffer_enable. When the command line option is
set, the driver will automatically post diag buffers at driver load time.
The command line option diag_buffer_enable is bitwise, so it's possible to
enable both and/or snapshot + trace buffers.  For trace, the driver will
allocate 1MB buffer, whereas for snapshot its 2MB. The purpose for this is
so the enduser doesn't have to manually use an application to setup diag
buffers for debugging firmware related issues.

Here is some examples
trace:
# insmod mpt2sas.ko diag_buffer_enable=1

snapshot:
# insmod mpt2sas.ko diag_buffer_enable=2

both trace and snapshot:
# insmod mpt2sas.ko diag_buffer_enable=3

Signed-off-by: Kashyap Desai <kashyap.desai@lsi.com>
Signed-off-by: Eric Moore <Eric.moore@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] mpt2sas: Fixed some of the comment
Kashyap, Desai [Wed, 23 Sep 2009 11:57:41 +0000 (17:27 +0530)]
[SCSI] mpt2sas: Fixed some of the comment

Fixed some of the comments sections for some of the function so "@ioc:
pointer to scsi command object" was changed to "@ioc: per adapter object"

Signed-off-by: Kashyap Desai <kashyap.desai@lsi.com>
Signed-off-by: Eric Moore <Eric.moore@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] mpt2sas: Support for stopping driver when Firmware encounters
Kashyap, Desai [Wed, 23 Sep 2009 11:56:58 +0000 (17:26 +0530)]
[SCSI] mpt2sas: Support for stopping driver when Firmware encounters

Added command line option and shost sysfs attribute called
mpt2sas_fwfault_debug. When enduser writes a "1" to this parameter, this
will enable support in the driver for debugging firmware timeout related
issues.  This handling was added in three areas (a) scsi error handling
callback called task_abort, (b) IOCTL interface, and (c) other timeouts that
result in diag resets, such as manufacturing config pages.  When this
support is enabled, the driver will provide dump_stack to console, halt
controller firmware, and panic driver. The end user probably would want to
setup serial console redirection so the dump stack can be seen.

Here are the three methods for enable this support:

(a) # insmod mpt2sas.ko mpt2sas_fwfault_debug=1
(b) # echo 1 > /sys/module/mpt2sas/parameters/mpt2sas_fwfault_debug
(c) # echo 1 > /sys/class/scsi_host/host#/fwfault_debug  (where # is
the host number)

Signed-off-by: Kashyap Desai <kashyap.desai@lsi.com>
Signed-off-by: Eric Moore <Eric.moore@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] mpt2sas: Adding MPI Headers - revision L
Kashyap, Desai [Wed, 23 Sep 2009 11:56:20 +0000 (17:26 +0530)]
[SCSI] mpt2sas: Adding MPI Headers - revision L

The new headers contain the following changes:
(1) Added IO Unit Page 7.
(2) Added new device ids for SAS2208.
(3) Added SAS IO Unit Page 5.
(4) Added partial and slumber power management capable flags to SAS Device
    Page 0 Flags field.
(5) Added PhyInfo defines for power condition.
(6) Added Ethernet configuration pages.

Signed-off-by: Kashyap Desai <kashyap.desai@lsi.com>
Signed-off-by: Eric Moore <Eric.moore@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] mpt2sas: New device SAS2208 support is added
Kashyap, Desai [Wed, 23 Sep 2009 11:54:27 +0000 (17:24 +0530)]
[SCSI] mpt2sas: New device SAS2208 support is added

Added device ids range for { 0x80 - 87 } , modified mpi/mpi2_cnfg.h containing
MPI2_MFGPAGE_DEVID_SAS2208_X.

Signed-off-by: Kashyap Desai <kashyap.desai@lsi.com>
Signed-off-by: Eric Moore <Eric.moore@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] mpt2sas: Add support in the driver to check for valid response info
Kashyap, Desai [Wed, 23 Sep 2009 11:53:07 +0000 (17:23 +0530)]
[SCSI] mpt2sas: Add support in the driver to check for valid response info

Add support in the driver to check for valid response info in the scsi
state, then check to see if the response code is
MPI2_SCSITASKMGMT_RSP_INVALID_FRAME; when this condition occurrs, the driver
will return DID_SOFT_ERROR.  A return code of DID_SOFT_ERROR will result in
a retry at the scsi-mid layer level.  An additional change added to obtain
the response code from the 1st byte of the response info instead of last.

Signed-off-by: Kashyap Desai <kashyap.desai@lsi.com>
Signed-off-by: Eric Moore <Eric.moore@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] mpt2sas: Expander remove fails when it is processing another expander add.
Kashyap, Desai [Wed, 23 Sep 2009 11:52:37 +0000 (17:22 +0530)]
[SCSI] mpt2sas: Expander remove fails when it is processing another expander add.

This handles the case where driver receives a expander removal event while
it is in the middle of processing an expander add event. The existing
implementation will stop processing futher device adds when a expander
delete arrives on top of add expander add. Due to a sanity check in the
driver, the devices there were not added, were never handshaked to firmware
with the device removal handshake protocal. Since the driver didnt' do the
handshake, the controller never provide further add events.  To fix this
issue, the sanity check was removed so the driver will always do the device
removal handshake protocal.

Signed-off-by: Kashyap Desai <kashyap.desai@lsi.com>
Signed-off-by: Eric Moore <Eric.moore@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years ago[SCSI] mpt2sas: Driver will use sas address instead of handle as a lookup
Kashyap, Desai [Wed, 23 Sep 2009 11:51:29 +0000 (17:21 +0530)]
[SCSI] mpt2sas: Driver will use sas address instead of handle as a lookup

The device driver was not handling updating device handles in all cases
across diag resets. To fix this issue, the driver is converted to using sas
address instead of handle as a lookup reference to the parent expander or
sas_host. Also, for both expanders and sas host, the phy handle will be one
unique handle. In the sas host case, the phy handle can be different for
every phy, so the change is to set the handle to the handle of the first
phy; every phy will be one single sas address(phy 0) instead of a different
sas address for every phy(previous implementation). So making one consistent
sas address for all the direct attachedports to the sas host, will make it
better user experience when using udev /dev/disk/by-path dev nodes

Signed-off-by: Kashyap Desai <kashyap.desai@lsi.com>
Signed-off-by: Eric Moore <Eric.moore@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
15 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/ide-2.6
Linus Torvalds [Thu, 29 Oct 2009 16:22:34 +0000 (09:22 -0700)]
Merge git://git./linux/kernel/git/davem/ide-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/ide-2.6:
  ide: Serialize CMD643 and CMD646 to fix a hardware bug with SSD

15 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6
Linus Torvalds [Thu, 29 Oct 2009 16:22:08 +0000 (09:22 -0700)]
Merge git://git./linux/kernel/git/davem/net-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6: (43 commits)
  net: Fix 'Re: PACKET_TX_RING: packet size is too long'
  netdev: usb: dm9601.c can drive a device not supported yet, add support for it
  qlge: Fix firmware mailbox command timeout.
  qlge: Fix EEH handling.
  AF_RAW: Augment raw_send_hdrinc to expand skb to fit iphdr->ihl (v2)
  bonding: fix a race condition in calls to slave MII ioctls
  virtio-net: fix data corruption with OOM
  sfc: Set ip_summed correctly for page buffers passed to GRO
  cnic: Fix L2CTX_STATUSB_NUM offset in context memory.
  MAINTAINERS: rt2x00 list is moderated
  airo: Reorder tests, check bounds before element
  mac80211: fix for incorrect sequence number on hostapd injected frames
  libertas spi: fix sparse errors
  mac80211: trivial: fix spelling in mesh_hwmp
  cfg80211: sme: deauthenticate on assoc failure
  mac80211: keep auth state when assoc fails
  mac80211: fix ibss joining
  b43: add 'struct b43_wl' missing declaration
  b43: Fix Bugzilla #14181 and the bug from the previous 'fix'
  rt2x00: Fix crypto in TX frame for rt2800usb
  ...

15 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/percpu
Linus Torvalds [Thu, 29 Oct 2009 16:19:29 +0000 (09:19 -0700)]
Merge branch 'for-linus' of git://git./linux/kernel/git/tj/percpu

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/percpu:
  sched: move rq_weight data array out of .percpu
  percpu: allow pcpu_alloc() to be called with IRQs off

15 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-param-fixes
Linus Torvalds [Thu, 29 Oct 2009 16:18:20 +0000 (09:18 -0700)]
Merge git://git./linux/kernel/git/rusty/linux-2.6-param-fixes

* git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-param-fixes:
  param: fix setting arrays of bool
  param: fix NULL comparison on oom
  param: fix lots of bugs with writing charp params from sysfs, by leaking mem.

15 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-for-linus
Linus Torvalds [Thu, 29 Oct 2009 16:17:59 +0000 (09:17 -0700)]
Merge git://git./linux/kernel/git/rusty/linux-2.6-for-linus

* git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-for-linus:
  virtio: order used ring after used index read
  virtio-pci: fix per-vq MSI-X request logic

15 years agoMerge branch 'for-linus' of git://git.kernel.dk/linux-2.6-block
Linus Torvalds [Thu, 29 Oct 2009 16:17:19 +0000 (09:17 -0700)]
Merge branch 'for-linus' of git://git.kernel.dk/linux-2.6-block

* 'for-linus' of git://git.kernel.dk/linux-2.6-block:
  backing-dev: ensure that a removed bdi no longer has super_block referencing it
  block: use after free bug in __blkdev_get
  block: silently error unsupported empty barriers too

15 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jbarnes...
Linus Torvalds [Thu, 29 Oct 2009 16:17:02 +0000 (09:17 -0700)]
Merge branch 'for-linus' of git://git./linux/kernel/git/jbarnes/pci-2.6

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jbarnes/pci-2.6:
  Revert "PCI: get larger bridge ranges when space is available"

15 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland...
Linus Torvalds [Thu, 29 Oct 2009 16:16:23 +0000 (09:16 -0700)]
Merge branch 'for-linus' of git://git./linux/kernel/git/roland/infiniband

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband:
  mlx4_core: Add a new supported 40 GigE device ID

15 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-rc-fixes-2.6
Linus Torvalds [Thu, 29 Oct 2009 16:16:01 +0000 (09:16 -0700)]
Merge git://git./linux/kernel/git/jejb/scsi-rc-fixes-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-rc-fixes-2.6:
  [SCSI] zfcp: Flush SCSI registration work when adding unit
  [SCSI] zfcp: Fix timer initialization for ct and els requests
  [SCSI] zfcp: Warn about storage devices with broken PLOGI data
  [SCSI] zfcp: Handle WWPN mismatch in PLOGI payload
  [SCSI] zfcp: fix kfree handling in zfcp_init_device_setup
  [SCSI] fix memory leak in initialization

15 years agoMerge branch 'drm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/airlied...
Linus Torvalds [Thu, 29 Oct 2009 16:15:02 +0000 (09:15 -0700)]
Merge branch 'drm-fixes' of git://git./linux/kernel/git/airlied/drm-2.6

* 'drm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/airlied/drm-2.6:
  drm/kms: fix kms/fbdev colormap support properly.
  drm: Add the basic check for the detailed timing in EDID
  drm/radeon/kms: ignore vga arbiter return.

15 years agoMerge branch 'sh/for-2.6.32' of git://git.kernel.org/pub/scm/linux/kernel/git/lethal...
Linus Torvalds [Thu, 29 Oct 2009 16:07:15 +0000 (09:07 -0700)]
Merge branch 'sh/for-2.6.32' of git://git./linux/kernel/git/lethal/sh-2.6

* 'sh/for-2.6.32' of git://git.kernel.org/pub/scm/linux/kernel/git/lethal/sh-2.6:
  sh: Fix hugetlbfs dependencies for SH-3 && MMU configurations.
  sh: Document uImage.bin target in archhelp.
  sh: add uImage.bin target
  sh: rsk7203 CONFIG_MTD=n fix
  sh: Check for return_to_handler when unwinding the stack
  sh: Build fix: define more __movmem* symbols
  sh: __irq_entry annotate do_IRQ().

Fix up sh/powerpc conflicts in fs/Kconfig

15 years agoMerge branch 'bugfixes' of git://git.linux-nfs.org/projects/trondmy/nfs-2.6
Linus Torvalds [Thu, 29 Oct 2009 16:02:24 +0000 (09:02 -0700)]
Merge branch 'bugfixes' of git://git.linux-nfs.org/projects/trondmy/nfs-2.6

* 'bugfixes' of git://git.linux-nfs.org/projects/trondmy/nfs-2.6:
  NFSv4: The link() operation should return any delegation on the file
  NFSv4: Fix two unbalanced put_rpccred() issues.
  NFSv4: Fix a bug when the server returns NFS4ERR_RESOURCE
  nfs: Panic when commit fails

15 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/sfrench/cifs-2.6
Linus Torvalds [Thu, 29 Oct 2009 16:02:01 +0000 (09:02 -0700)]
Merge git://git./linux/kernel/git/sfrench/cifs-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/sfrench/cifs-2.6:
  [CIFS] Fixing to avoid invalid kfree() in cifs_get_tcp_session()

15 years agoray_cs: Fix copy_from_user handling
Alan Cox [Tue, 27 Oct 2009 15:35:55 +0000 (15:35 +0000)]
ray_cs: Fix copy_from_user handling

I've not touched the other stuff here but the word "locking" comes to mind.

Signed-off-by: Alan Cox <alan@linux.intel.com>
Cc: stable@kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
15 years agodpt_i2o: Fix up copy*user
Alan Cox [Tue, 27 Oct 2009 15:35:35 +0000 (15:35 +0000)]
dpt_i2o: Fix up copy*user

Signed-off-by: Alan Cox <alan@linux.intel.com>
Cc: stable@kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
15 years agoMerge branch 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc
Linus Torvalds [Thu, 29 Oct 2009 15:59:06 +0000 (08:59 -0700)]
Merge branch 'merge' of git://git./linux/kernel/git/benh/powerpc

* 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc:
  powerpc/ppc64: Use preempt_schedule_irq instead of preempt_schedule
  powerpc: Minor cleanup to lib/Kconfig.debug
  powerpc: Minor cleanup to sound/ppc/Kconfig
  powerpc: Minor cleanup to init/Kconfig
  powerpc: Limit memory hotplug support to PPC64 Book-3S machines
  powerpc: Limit hugetlbfs support to PPC64 Book-3S machines
  powerpc: Fix compile errors found by new ppc64e_defconfig
  powerpc: Add a Book-3E 64-bit defconfig
  powerpc/booke: Fix xmon single step on PowerPC Book-E
  powerpc: Align vDSO base address
  powerpc: Fix segment mapping in vdso32
  powerpc/iseries: Remove compiler version dependent hack
  powerpc/perf_events: Fix priority of MSR HV vs PR bits
  powerpc/5200: Update defconfigs
  drivers/serial/mpc52xx_uart.c: Use UPIO_MEM rather than SERIAL_IO_MEM
  powerpc/boot/dts: drop obsolete 'fsl5200-clocking'
  of: Remove nested function
  mpc5200: support for the MAN mpc5200 based board mucmc52
  mpc5200: support for the MAN mpc5200 based board uc101

15 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sameo/mfd-2.6
Linus Torvalds [Thu, 29 Oct 2009 15:44:04 +0000 (08:44 -0700)]
Merge branch 'for-linus' of git://git./linux/kernel/git/sameo/mfd-2.6

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sameo/mfd-2.6:
  mfd: Do not dereference null pointer in twl4030 error path
  mfd: Always initialise WM831x IRQ mutex

15 years agoMerge branch 'hwpoison-2.6.32' of git://git.kernel.org/pub/scm/linux/kernel/git/ak...
Linus Torvalds [Thu, 29 Oct 2009 15:20:00 +0000 (08:20 -0700)]
Merge branch 'hwpoison-2.6.32' of git://git./linux/kernel/git/ak/linux-mce-2.6

* 'hwpoison-2.6.32' of git://git.kernel.org/pub/scm/linux/kernel/git/ak/linux-mce-2.6:
  HWPOISON: fix invalid page count in printk output
  HWPOISON: Allow schedule_on_each_cpu() from keventd
  HWPOISON: fix/proc/meminfo alignment
  HWPOISON: fix oops on ksm pages
  HWPOISON: Fix page count leak in hwpoison late kill in do_swap_page
  HWPOISON: return early on non-LRU pages
  HWPOISON: Add brief hwpoison description to Documentation
  HWPOISON: Clean up PR_MCE_KILL interface

15 years agoMerge branch 'for-linus' of git://oss.sgi.com/xfs/xfs
Linus Torvalds [Thu, 29 Oct 2009 15:18:25 +0000 (08:18 -0700)]
Merge branch 'for-linus' of git://oss.sgi.com/xfs/xfs

* 'for-linus' of git://oss.sgi.com/xfs/xfs:
  xfs: fix double IRELE in xfs_dqrele_inode