iscsi_tcp, ib_iser, cxgb*, be2iscsi and bnx2i do not use
the host lock and do not take the session lock against
a irq, so this patch drops the DEF_SCSI_QCMD use. Instead
we just take the session lock and disable bhs.
Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
This adds a more informative error code and message
for the iscsi scsi eh session drop paths. This allows
you to distinguish if the session was dropped due to
a connection failure vs the iscsi layer dropping
the session due to scsi eh failure processing.
Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
This patch adds a handful of missing CDBs defs that are used by TCM
persistent reservation logic in the SPC-4 defined CDB exclusion table for
registrations and reservations.
This includes a number of missing MI_* and MO_* prefixed service actions defs
for MAINTENANCE_IN and MAINTENANCE_OUT that are mentioned wrt to persistent
registration and reservation status for the SCSI Initiator Port.
Signed-off-by: Nicholas A. Bellinger <nab@linux-iscsi.org>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
The statistics for InputMegabytes and OutputMegabytes are
misnamed. They're accumulating bytes, not megabytes.
The statistic returned via /sys must be in megabytes, however,
which is what the HBA-API wants. The FCP code needs to accumulate
it in bytes and then divide by 1,000,000 (not 2^20) before it
presented via sysfs.
This affects fcoe.ko only, not fnic. The fnic driver
correctly by accumulating bytes and then converts to megabytes.
I checked that libhbalinux is using the /sys file directly without
conversion.
BTW, qla2xxx does divide by 2^20, which I'm not fixing here.
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>
Switches using multiple-FCFs may reject FLOGI in order to
balance the load between multiple FCFs. Even though the FCF
was available, it may have more load at the point we actually
send the FLOGI.
If the FLOGI fails, select a different FCF
if possible, among those with the same priority. If no other
FCF is available, just deliver the reject to libfc for retry.
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>
We can easily remove the tgt_flags from fc_fcp_pkt struct
and use rpriv->tgt_flags directly where needed.
Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
The patch fixes the following situations where NOP-Out pkt is called for:
- local unsolicited NOP-Out requests (requesting no NOP-In response)
- local NOP-Out responses to unsolicited NOP-In requests
kernel panic is observed due to double session spin_lock requests; one in the
bnx2i_process_nopin_local_cmpl routine in bnx2i_hwi.c and the other in the
iscsi_put_task routine in libiscsi.c
The proposed fix is to export the currently static __iscsi_put_task() routine
and have bnx2i call it directly instead of the iscsi_put_task() routine which
holds the session spin lock.
Signed-off-by: Eddie Wai <eddie.wai@broadcom.com>
Reviewed-by: Michael Chan <mchan@broadcom.com>
Reviewed-by: Anil Veerabhadrappa <anilgv@broadcom.com>
Acked-by: Benjamin Li <benli@broadcom.com>
Reviewed-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
To date libsas has only looked at the attached sas address when
determining the formation of wide ports. The specification and some
hardware expects that phys with different addresses will not form a wide
port unless the local peer phys also match each other. Introduce a flag
to select stricter behavior at sas_register_ha() time. The flag can be
dropped once it is known that all libsas users expect the same behavior.
Current drivers just initialize this field to zero and get the
traditional behavior.
Reported-by: Patrick Thomson <patrick.s.thomson@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Move the mid-layer's ->queuecommand() invocation from being locked
with the host lock to being unlocked to facilitate speeding up the
critical path for drivers who don't need this lock taken anyway.
The patch below presents a simple SCSI host lock push-down as an
equivalent transformation. No locking or other behavior should change
with this patch. All existing bugs and locking orders are preserved.
Additionally, add one parameter to queuecommand,
struct Scsi_Host *
and remove one parameter from queuecommand,
void (*done)(struct scsi_cmnd *)
Scsi_Host* is a convenient pointer that most host drivers need anyway,
and 'done' is redundant to struct scsi_cmnd->scsi_done.
Minimal code disturbance was attempted with this change. Most drivers
needed only two one-line modifications for their host lock push-down.
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
Acked-by: James Bottomley <James.Bottomley@suse.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-misc-2.6: (70 commits)
[SCSI] pmcraid: add support for set timestamp command and other fixes
[SCSI] pmcraid: remove duplicate struct member
[SCSI] qla4xxx: Fix cmd check in qla4xxx_cmd_wait
[SCSI] megaraid_sas: Version and documentation update
[SCSI] megaraid_sas: Add three times Online controller reset
[SCSI] megaraid_sas: Add input parameter for max_sectors
[SCSI] megaraid_sas: support devices update flag
[SCSI] libosd: write/read_sg_kern API
[SCSI] libosd: Support for scatter gather write/read commands
[SCSI] libosd: Free resources in reverse order of allocation
[SCSI] libosd: Fix bug in attr_page handling
[SCSI] lpfc 8.3.18: Update lpfc driver version to 8.3.18
[SCSI] lpfc 8.3.18: Add new WQE support
[SCSI] lpfc 8.3.18: Fix critical errors
[SCSI] lpfc 8.3.18: Adapter Shutdown and Unregistration cleanup
[SCSI] lpfc 8.3.18: Add logic to detect last devloss timeout
[SCSI] lpfc 8.3.18: Add support of received ELS commands
[SCSI] lpfc 8.3.18: FC/FCoE Discovery fixes
[SCSI] ipr: add definitions for a new adapter
[SCSI] bfa: fix comments for c files
...
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband: (63 commits)
IB/qib: clean up properly if pci_set_consistent_dma_mask() fails
IB/qib: Allow driver to load if PCIe AER fails
IB/qib: Fix uninitialized pointer if CONFIG_PCI_MSI not set
IB/qib: Fix extra log level in qib_early_err()
RDMA/cxgb4: Remove unnecessary KERN_<level> use
RDMA/cxgb3: Remove unnecessary KERN_<level> use
IB/core: Add link layer type information to sysfs
IB/mlx4: Add VLAN support for IBoE
IB/core: Add VLAN support for IBoE
IB/mlx4: Add support for IBoE
mlx4_en: Change multicast promiscuous mode to support IBoE
mlx4_core: Update data structures and constants for IBoE
mlx4_core: Allow protocol drivers to find corresponding interfaces
IB/uverbs: Return link layer type to userspace for query port operation
IB/srp: Sync buffer before posting send
IB/srp: Use list_first_entry()
IB/srp: Reduce number of BUSY conditions
IB/srp: Eliminate two forward declarations
IB/mlx4: Signal node desc changes to SM by using FW to generate trap 144
IB: Replace EXTRA_CFLAGS with ccflags-y
...
This is a trivial addition to the SG API that can receive kernel
pointers. It is only used by the out-of-tree test module. So
it's immediate need is questionable. For maintenance ease it might
just get in, as it's very small.
John.
do you need this in the Kernel, or is it only for osd_ktest.ko?
Signed-off-by: John A. Chandy <john.chandy@uconn.edu>
Signed-off-by: Boaz Harrosh <bharrosh@panasas.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
This patch adds the Scatter-Gather (sg) API to libosd.
Scatter-gather enables a write/read of multiple none-contiguous
areas of an object, in a single call. The extents may overlap
and/or be in any order.
The Scatter-Gather list is sent to the target in what is called
a "cdb continuation segment". This is yet another possible segment
in the osd-out-buffer. It is unlike all other segments in that it
sits before the actual "data" segment (which until now was always
first), and that it is signed by itself and not part of the data
buffer. This is because the cdb-continuation-segment is considered
a spill-over of the CDB data, and is therefor signed under
OSD_SEC_CAPKEY and higher.
TODO: A new osd_finalize_request_ex version should be supplied so
the @caps received on the network also contains a size parameter
and can be spilled over into the "cdb continuation segment".
Thanks to John Chandy <john.chandy@uconn.edu> for the original
code, and investigations. And the implementation of SG support
in the osd-target.
Original-coded-by: John Chandy <john.chandy@uconn.edu>
Signed-off-by: Boaz Harrosh <bharrosh@panasas.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
When number of NPIV ports created are greater than the xids
allocated per pool -- for eg., creating 255 NPIV ports on a
system with nr_cpu_ids of 32, with each pool containing 128
xids -- and then generating a link event - for eg.,
shutdown/no shutdown -- on the switch port causes the hang
with the following stack trace.
Call Trace:
schedule_timeout+0x19d/0x230
wait_for_common+0xc0/0x170
__cancel_work_timer+0xcf/0x1b0
fc_disc_stop+0x16/0x30 [libfc]
fc_lport_reset_locked+0x47/0x90 [libfc]
fc_lport_enter_reset+0x67/0xe0 [libfc]
fc_lport_disc_callback+0xbc/0xe0 [libfc]
fc_disc_done+0xa8/0xf0 [libfc]
fc_disc_timeout+0x29/0x40 [libfc]
run_workqueue+0xb8/0x140
worker_thread+0x96/0x110
kthread+0x96/0xa0
child_rip+0xa/0x20
Fix is to not cancel the disc_work if discovery is already
stopped, thus allowing lport state machine to restart and try
discovery again.
Signed-off-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com>
Acked-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>
This patch adds support for SRP_CRED_REQ to avoid a lockup by targets
that use that mechanism to return credits to the initiator. This
prevents a lockup observed in the field where we would never add the
credits from the SRP_CRED_REQ to our current count, and would therefore
never send another command to the target.
Minimal support for SRP_AER_REQ is also added, as these messages can
also be used to convey additional credits to the initiator.
Based upon extensive debugging and code by Bart Van Assche and a bug
report by Chris Worley.
Signed-off-by: David Dillow <dillowda@ornl.gov>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
* git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb-2.6: (141 commits)
USB: mct_u232: fix broken close
USB: gadget: amd5536udc.c: fix error path
USB: imx21-hcd - fix off by one resource size calculation
usb: gadget: fix Kconfig warning
usb: r8a66597-udc: Add processing when USB was removed.
mxc_udc: add workaround for ENGcm09152 for i.MX35
USB: ftdi_sio: add device ids for ScienceScope
USB: musb: AM35x: Workaround for fifo read issue
USB: musb: add musb support for AM35x
USB: AM35x: Add musb support
usb: Fix linker errors with CONFIG_PM=n
USB: ohci-sh - use resource_size instead of defining its own resource_len macro
USB: isp1362-hcd - use resource_size instead of defining its own resource_len macro
USB: isp116x-hcd - use resource_size instead of defining its own resource_len macro
USB: xhci: Fix compile error when CONFIG_PM=n
USB: accept some invalid ep0-maxpacket values
USB: xHCI: PCI power management implementation
USB: xHCI: bus power management implementation
USB: xHCI: port remote wakeup implementation
USB: xHCI: port power management implementation
...
Manually fix up (non-data) conflict: the SCSI merge gad renamed the
'hw_sector_size' member to 'physical_block_size', and the USB tree
brought a new use of it.
* 'for-2.6.37/barrier' of git://git.kernel.dk/linux-2.6-block: (46 commits)
xen-blkfront: disable barrier/flush write support
Added blk-lib.c and blk-barrier.c was renamed to blk-flush.c
block: remove BLKDEV_IFL_WAIT
aic7xxx_old: removed unused 'req' variable
block: remove the BH_Eopnotsupp flag
block: remove the BLKDEV_IFL_BARRIER flag
block: remove the WRITE_BARRIER flag
swap: do not send discards as barriers
fat: do not send discards as barriers
ext4: do not send discards as barriers
jbd2: replace barriers with explicit flush / FUA usage
jbd2: Modify ASYNC_COMMIT code to not rely on queue draining on barrier
jbd: replace barriers with explicit flush / FUA usage
nilfs2: replace barriers with explicit flush / FUA usage
reiserfs: replace barriers with explicit flush / FUA usage
gfs2: replace barriers with explicit flush / FUA usage
btrfs: replace barriers with explicit flush / FUA usage
xfs: replace barriers with explicit flush / FUA usage
block: pass gfp_mask and flags to sb_issue_discard
dm: convey that all flushes are processed as empty
...
This commit changes storage_common.h, file_storage.c and
f_mass_storage.c to use definitions of SCSI commands from
scsi/scsi.h file instead of redefining the commands in
storage_common.c.
scsi/scsi.h header file was missing READ_FORMAT_CAPACITIES and
READ_HEADER so this commit also add those to the header.
Signed-off-by: Michal Nazarewicz <mina86@mina86.com>
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: James Bottomley <James.Bottomley@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
I seem to have a knack for digging up buggy usb devices which don't work
with Linux, and I'm crazy enough to try to make them work. So this time a
friend of mine asked me to get an mp4 player (an mp3 player which can play
videos on a small screen) to work with Linux.
It is based on the well known rockbox chipset for which we already have an
unusual devs entries to work around some of its bugs. But this model
comes with an additional twist.
This model chokes on read_capacity_16 calls. Now normally we don't make
those calls, but this model comes with an sdcard slot and when there is no
card in there (and shipped from the factory there is none), it reports a
size of 0. However this time the programmers actually got the
read_capacity_10 response right! So they substract one from the size as
stored internally in the mp3 player before reporting it back, resulting in
an answer of ... 0xffffffff sectors, causing sd.c to try a
read_capacity_16, on which the device crashes.
This patch adds a flag to scsi_device to indicate that a a device cannot
handle read_capacity_16, and when this flag is set if a device reports an
lba of 0xffffffff as answer to a read_capacity_10, assumes it tries to
report a size of 0.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Cc: James Bottomley <James.Bottomley@HansenPartnership.com>
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: Matthew Dharm <mdharm-usb@one-eyed-alien.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Some USB devices emulate a usb-mass-storage attached (scsi) cdrom device,
usually this fake cdrom contains the windows software for the device.
While working on supporting Appotech ax3003 based photoframes, which do
this I discovered that they will go of into lala land when ever they see a
READ_DISC_INFO scsi command.
Thus this patch adds a scsi_device flag (which can then be set by the
usb-storage driver through an unsual-devs entry), to indicate this, and
makes the sr driver honor this flag.
I know this sucks, but as discussed on linux-scsi list there is no other
way to make this device work properly.
Looking at usb traces made under windows, windows never sends a
READ_DISC_INFO during normal interactions with a usb cdrom device. So as
this cdrom emulation thingie becomes more common we might see more of this
problem.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Cc: James Bottomley <James.Bottomley@HansenPartnership.com>
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: Matthew Dharm <mdharm-usb@one-eyed-alien.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
sd will get hung up issuing commands to flush write cache if a SAS
device behind the expander is unplugged without warning. Change libsas
to reject commands to domain devices that have already gone away.
[maciej.trela@intel.com: removed setting ->gone in sas_deform_port() to
permit sync cache commands at module removal]
Signed-off-by: Darrick J. Wong <djwong@us.ibm.com>
Tested-by: Haipao Fan <haipao.fan@intel.com>
Signed-off-by: Maciej Trela <maciej.trela@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
This adds a fc host dev loss sysfs file. Instead of
calling into the driver using the get_host_def_dev_loss_tmo
callback, we allow drivers to init the dev loss like is done
for other fc host params, and then the fc class will handle
updating the value if the user writes to the new sysfs file.
Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Some controllers have a hardware limit on the number of protection
information scatter-gather list segments they can handle.
Introduce a max_integrity_segments limit in the block layer and provide
a new scsi_host_template setting that allows HBA drivers to provide a
value suitable for the hardware.
Add support for honoring the integrity segment limit when merging both
bios and requests.
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Jens Axboe <axboe@carl.home.kernel.dk>
REQ_HARDBARRIER is deprecated. Remove spurious uses in the following
users. Please note that other than osdblk, all other uses were
already spurious before deprecation.
* osdblk: osdblk_rq_fn() won't receive any request with
REQ_HARDBARRIER set. Remove the test for it.
* pktcdvd: use of REQ_HARDBARRIER in pkt_generic_packet() doesn't mean
anything. Removed.
* aic7xxx_old: Setting MSG_ORDERED_Q_TAG on REQ_HARDBARRIER is
spurious. Removed.
* sas_scsi_host: Setting TASK_ATTR_ORDERED on REQ_HARDBARRIER is
spurious. Removed.
* scsi_tcq: The ordered tag path wasn't being used anyway. Removed.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Boaz Harrosh <bharrosh@panasas.com>
Cc: James Bottomley <James.Bottomley@suse.de>
Cc: Peter Osterlund <petero2@telia.com>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
This patch adds a fc_host setting to store the
default dev_loss_tmo. It is used if the driver
has a callack to get the value from the LLD. If
the callback is not set, then we use the fc class
module default value.
Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Commit 9e4f5e29 ("FC Pass Thru support") exported a number of header files
in include/scsi to user space, but didn't change the uX types to the
userspace-compatible __uX types. Without that you'll get compile errors
when including them - E.G.:
include/scsi/scsi.h:145: error: expected specifier-qualifier-list before `u8'
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Cc: Boaz Harrosh <bharrosh@panasas.com>
Cc: James Smart <james.smart@emulex.com>
Cc: James Bottomley <James.Bottomley@HansenPartnership.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This patch (as1398b) adds runtime PM support to the SCSI layer. Only
the machanism is provided; use of it is up to the various high-level
drivers, and the patch doesn't change any of them. Except for sg --
the patch expicitly prevents a device from being runtime-suspended
while its sg device file is open.
The implementation is simplistic. In general, hosts and targets are
automatically suspended when all their children are asleep, but for
them the runtime-suspend code doesn't actually do anything. (A host's
runtime PM status is propagated up the device tree, though, so a
runtime-PM-aware lower-level driver could power down the host adapter
hardware at the appropriate times.) There are comments indicating
where a transport class might be notified or some other hooks added.
LUNs are runtime-suspended by calling the drivers' existing suspend
handlers (and likewise for runtime-resume). Somewhat arbitrarily, the
implementation delays for 100 ms before suspending an eligible LUN.
This is because there typically are occasions during bootup when the
same device file is opened and closed several times in quick
succession.
The way this all works is that the SCSI core increments a device's
PM-usage count when it is registered. If a high-level driver does
nothing then the device will not be eligible for runtime-suspend
because of the elevated usage count. If a high-level driver wants to
use runtime PM then it can call scsi_autopm_put_device() in its probe
routine to decrement the usage count and scsi_autopm_get_device() in
its remove routine to restore the original count.
Hosts, targets, and LUNs are not suspended while they are being probed
or removed, or while the error handler is running. In fact, a fairly
large part of the patch consists of code to make sure that things
aren't suspended at such times.
[jejb: fix up compile issues in PM config variations]
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
We have two separate definitions for identical constants with nearly the
same name. One comes from the generic headers in scsi.h; the other is
an enum in libsas.h ... it's causing confusion about which one is
correct (fortunately they both are).
Fix this by eliminating the libsas.h duplicate
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
wait for session to come online in eh_device_reset_handler
and eh_target_reset_handler
Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: Vikas Chaudhary <vikas.chaudhary@qlogic.com>
Signed-off-by: Ravi Anand <ravi.anand@qlogic.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Incoming requests shouldn't require a local exchange if we're
just going to reply with one or two frames and don't expect
anything further. Don't allocate exchanges for such requests
until requested by the upper-layer protocol.
The sequence is always NULL for new requests, so remove
that as an argument to request handlers.
Also change the first argument to lport->tt.seq_els_rsp_send
from the sequence pointer to the received frame pointer, to
supply the exchange IDs and destination ID info.
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>
For incoming ELS and FCP requests, we often don't require an
exchange and sequence, however, sometimes we do. For those cases,
(primarily FCP requests for targets) add a function to set up
the exchange and sequence.
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>
Add functions to fill in an FC header given a request header.
These reduces code lines in fc_lport and fc_rport and works
without an exchange/sequence assigned.
fc_fill_reply_hdr() fills a header for a final reply frame.
fc_fill_hdr() which is similar but allows specifying the
f_ctl parameter.
Add defines for F_CTL values FC_FCTL_REQ and FC_FCTL_RESP.
These can be used for most request and response sequences.
v2 of patch adds a line to copy the frame encapsulation
info from the received frame.
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>
To pave the way for eliminating exchanges from incoming requests,
add simple inline fc_frame_sid() and fc_frame_did() functions
which get the FC_IDs from the frame header. This can be almost
as efficient as getting them from the sequence/exchange.
Move ntohll, htonll, ntoh24 and hton24 to <scsi/fc_frame.h>
since we need them there and that's included by <scsi/libfc.h>
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>
The LOGO state hasn't been used in a while, except in a brief
transition to DELETE state while holding the rport mutex.
All port LOGO responses have been ignored as well as any timeout
if we don't get a response.
So this patch just removes LOGO state and simplifies the response handler.
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>
The FC-BB-6 committee is proposing a new FIP usage model called
VN_port to VN_port mode. It allows VN_ports to discover each other
over a loss-free L2 Ethernet without any FCF or Fibre-channel fabric
services. This is point-to-multipoint. There is also a variant
of this called point-to-point which provides for making sure there
is just one pair of ports operating over the Ethernet fabric.
We add these new states: VNMP_START, _PROBE1, _PROBE2, _CLAIM, and _UP.
These usually go quickly in that sequence. After waiting a random
amount of time up to 100 ms in START, we select a pseudo-random
proposed locally-unique port ID and send out probes in states PROBE1
and PROBE2, 100 ms apart. If no probe responses are heard, we
proceed to CLAIM state 400 ms later and send a claim notification.
We wait another 400 ms to receive claim responses, which give us
a list of the other nodes on the network, including their FC-4
capabilities. After another 400 ms we go to VNMP_UP state and
should start interoperating with any of the nodes for whic we
receivec claim responses. More details are in the spec.j
Add the new mode as FIP_MODE_VN2VN. The driver must specify
explicitly that it wants to operate in this mode. There is
no automatic detection between point-to-multipoint and fabric
mode, and the local port initialization is affected, so it isn't
anticipated that there will ever be any such automatic switchover.
It may eventually be possible to have both fabric and VN2VN
modes on the same L2 network, which may be done by two separate
local VN_ports (lports).
When in VN2VN mode, FIP replaces libfc's fabric-oriented discovery
module with its own simple code that adds remote ports as they
are discovered from incoming claim notifications and responses.
These hooks are placed by fcoe_disc_init().
A linear list of discovered vn_ports is maintained under the
fcoe_ctlr struct. It is expected to be short for now, and
accessed infrequently. It is kept under RCU for lock-ordering
reasons. The lport and/or rport mutexes may be held when we
need to lookup a fcoe_vnport during an ELS send.
Change fcoe_ctlr_encaps() to lookup the destination vn_port in
the list of peers for the destination MAC address of the
FIP-encapsulated frame.
Add a new function fcoe_disc_init() to initialize just the
discovery portion of libfcoe for VN2VN mode.
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>
The FC-BB-6 committee is proposing a new FIP usage model called
VN_port to VN_port mode. It allows VN_ports to discover each other
over a loss-free L2 Ethernet without any FCF or Fibre-channel fabric
services. This is point-to-multipoint. There is also a variant
of this called point-to-point which provides for making sure there
is just one pair of ports operating over the Ethernet fabric.
This patch defines the new message type and subtypes as well as
one new descriptor type used by VN2VN mode.
These are all still at the proposed stage and subject to change.
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>
When an exchange is received with a FIP encapsulation, we need
to know that the response must be sent via FIP and what the original
ELS opcode was. This becomes important for VN2VN mode, where we may
receive FLOGI or LOGO from several peer VN_ports, and the LS_ACC or
LS_RJT must be sent FIP-encapsulated with the correct sub-type.
Add a field to the struct fc_frame, fr_encaps, to indicate the
encapsulation values. That term is chosen to be neutral and
LLD-agnostic in case non-FCoE/FIP LLDs might find it useful.
The frame fr_encaps is transferred from the ingress frame to the
exchange by fc_exch_recv_req(), and back to the outgoing frame
by fc_seq_send().
This is taking the last byte in the skb->cb array. If needed,
we could combine the info in sof, eof, flags, and encaps
together into one field, but it'd be better to do that if
and when its needed.
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>
The FIP proposal for VN_port to VN_port point-to-multipoint
operation requires a FLOGI be sent to each remote port.
The FLOGI is sent with the assigned S_ID and D_IDs of the
local and remote ports. This and the response get
FIP-encapsulated for Ethernet.
Add FLOGI state to the remote port state machine.
This will be skipped if not in point-to-multipoint mode.
To reduce a little duplication between PLOGI and FLOGI
response handling, added fc_rport_login_complete(), which
handles the parameters for the rdata struct.
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>
For VN_port to VN_port mode, the transport sets the port_id and
there's no lport FLOGI. This is similar to FC loop mode.
Add a point_to_multipoint flag that indicates the local port is in
point-to-multipoint mode. This skips FLOGI and discovery.
It also skips resetting the port_id on resets other than link down.
Add function fc_lport_set_local_id() that sets the local port_id.
This is called by libfcoe on behalf of the low-level driver
to set the port_id when the link comes up.
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>
There are three modes that libfcoe currently supports, and a new one
is coming. Change the fcoe_ctlr_init() interface to add the mode
desired. This should not change any functionality.
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>
For VN_port to VN_port mode, FIP will do discovery and needs a
way to find its state from the local port or discovery structure.
It seems that any other LLD that implements its own discovery
would also need something like this.
Replace disc->lport with disc->priv, and use container_of to
find the lport. We could use disc->priv for that, but
container_of is smaller and faster.
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>
It turns out most of the FIP work is now done from worker threads
or process context now, so there's no need to use a spin lock.
Change to use mutex instead of spin lock and delayed_work instead
of a timer.
This will make it nicer for the VN_port to VN_port feature that
will interact more with the libfc layers requiring that
spinlocks not be held.
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>
Add pre-zeroed space after the allocation for fc_rport_priv
for use by the lower-level driver.
This is primarily for VN2VN FIP mode, but could be used in
other ways someday.
The space required is specified in lport->rport_priv_size.
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>
To allow LLD to do lookups on rports without grabbing a mutex,
make them RCU-safe. The caller of lport->tt.rport_lookup will
have the choice of holding disc_mutex or the rcu_read_lock().
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>
This is per FC-BB-5 Annex-D recommendation and per that
if address checking fails then drop the frame.
FIP code paths are already doing this so only needed for fcoe
frames.
The src address checking is limited to only fip mode since
this might break non-fip mode used in p2p due to used OUI
based addressing in some p2p code paths, going forward FIP
will be the only mode, therefore limited this to only FIP
mode so that it won't break non-fip p2p mode for now.
-v2
Removes FCOE packet type checking since fcoe_rcv is
registered to receive only FCoE type packets from netdev
and it is already checked by netdev.
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>
Analyzing fcoe with sparse currently fails. This is because struct
fcoe_rcv_info contains two enum members that have been declared with
__attribute__((packed)). Apparently gcc honors this attribute while sparse
ignores it. The result is that sizeof(struct fcoe_rcv_info)
== sizeof(struct sk_buff::cb) == 48 on a 64-bit system according to gcc, but
not according to sparse. The patch below modifies the definition of
struct fcoe_rcv_info such that gcc and sparse interpret this structure
definition in the same way. The current sparse output is as follows:
$ cd linux-2.6.34
$ make C=2 M=drivers/scsi/fcoe modules
CHECK drivers/scsi/fcoe/fcoe.c
include/scsi/fc_frame.h:81:9: error: invalid bitfield width, -1.
CC [M] drivers/scsi/fcoe/fcoe.o
CHECK drivers/scsi/fcoe/libfcoe.c
include/scsi/fc_frame.h:81:9: error: invalid bitfield width, -1.
drivers/scsi/fcoe/libfcoe.c:56:37: error: invalid initializer
Signed-off-by: Bart Van Assche <bart.vanassche@gmail.com>
Cc: jeykholt@cisco.com
Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Remote ports were restarting indefinitely after getting
rejects in PRLI.
Fix by adding a counter of restarts and limiting that with
the port login retry limit as well.
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>
This patch somewhat combines two fixes to remote port handing in libfc.
The first problem was that rport work could be queued on a deleted
and freed rport. This is handled by not resetting rdata->event
ton NONE if the rdata is about to be deleted.
However, that fix led to the second problem, described by
Bhanu Gollapudi, as follows:
> Here is the sequence of events. T1 is first LOGO receive thread, T2 is
> fc_rport_work() scheduled by T1 and T3 is second LOGO receive thread and
> T4 is fc_rport_work scheduled by T3.
>
> 1. (T1)Received 1st LOGO in state Ready
> 2. (T1)Delete port & enter to RESTART state.
> 3. (T1)schdule event_work, since event is RPORT_EV_NONE.
> 4. (T1)set event = RPORT_EV_LOGO
> 5. (T1)Enter RESTART state as disc_id is set.
> 6. (T2)remember to PLOGI, and set event = RPORT_EV_NONE
> 6. (T3)Received 2nd LOGO
> 7. (T3)Delete Port & enter to RESTART state.
> 8. (T3)schedule event_work, since event is RPORT_EV_NONE.
> 9. (T3)Enter RESTART state as disc_id is set.
> 9. (T3)set event = RPORT_EV_LOGO
> 10.(T2)work restart, enter PLOGI state and issues PLOGI
> 11.(T4)Since state is not RESTART anymore, restart is not set, and the
> event is not reset to RPORT_EV_NONE. (current event is RPORT_EV_LOGO).
> 12. Now, PLOGI succeeds and fc_rport_enter_ready() will not schedule
> event_work, and hence the rport will never be created, eventually losing
> the target after dev_loss_tmo.
So, the problem here is that we were tracking the desire for
the rport be restarted by state RESTART, which was otherwise
equivalent to DELETE. A contributing factor is that we dropped
the lock between steps 6 and 10 in thread T2, which allows the
state to change, and we didn't completely re-evaluate then.
This is hopefully corrected by the following minor redesign:
Simplify the rport restart logic by making the decision to
restart after deleting the transport rport. That decision
is based on a new STARTED flag that indicates fc_rport_login()
has been called and fc_rport_logoff() has not been called
since then. This replaces the need for the RESTART state.
Only restart if the rdata is still in DELETED state
and only if it still has the STARTED flag set.
Also now, since we clear the event code much later in the
work thread, allow for the possibility that the rport may
have become READY again via incoming PLOGI, and if so,
queue another event to handle that.
In the problem scenario, the second LOGO received will
cause the LOGO event to occur again.
Reported-by: Bhanu Gollapudi <bprakash@broadcom.com>
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>
Mike Christie