[CIFS] readdir (ffirst) enablement of accurate timestamps from legacy servers

Signed-off-by: Steve French <sfrench@us.ibm.com>

[CIFS] Fix compiler warning with previous patch

Signed-off-by: Steve French <sfrench@us.ibm.com>

[CIFS] Fix typo

Signed-off-by: Steve French <sfrench@us.ibm.com>

[CIFS] Allow for 15 minute TZs (e.g. Nepal) and be more explicit about
not setting time on close

Signed-off-by: Guenter Kukkukk <linux@kukkukk.com>
Signed-off-by: Steve French <sfrench@us.ibm.com>

[CIFS] Fix readdir of large directories for backlevel servers

(were not setting all of resume key)

Signed-off-by: Steve French <sfrench@us.ibm.com>

[CIFS] Allow LANMAN21 support even in both POSIX non-POSIX path

Signed-off-by: Guenter Kukkukk <linux@kukkukk.com>
Signed-off-by: Steve French <sfrench@us.ibm.com>

[CIFS] Make use of newer QFSInfo dependent on capability bit instead of
whether we negotiated legacy lanman dialect so we do not keep retrying
for mount to WindowsME

Signed-off-by: Steve French <sfrench@us.ibm.com>

[CIFS] Do not send newer QFSInfo to legacy servers which can not support it

Fix dialect negotiation to save off when we have negotiated lanman.
This allows us to avoid sending some somewhat newer requests that the server
can not handle and go directly to the older version (infolevel) of the same
call. Make sure we try to negotiate a level which allows us to get the
server OS (which we check so we can detect Win9x vs. other legacy servers
and eventually work around the Win9x DOS time bug (they reverse date/time
fields).

Signed-off-by: Steve French <sfrench@us.ibm.com>

[CIFS] Fix typo in name of new cifs_show_stats

Signed-off-by: Steve French <sfrench@us.ibm.com>

[CIFS] Rename server time zone field

Server time zone is not really a time zone, rather a time adjustement
in seconds.

CC: Guenter Kukkukk <linux@kukkukk.com>
Signed-off-by: Steve French <sfrench@us.ibm.com>

[CIFS] Handle legacy servers which return undefined time zone

Signed-off-by: Guenter Kukkukk <linux@kukkukk.com>
Signed-off-by: Steve French <sfrench@us.ibm.com>

[CIFS] CIFS support for /proc/<pid>/mountstats part 1

Signed-off-by: Steve French <sfrench@us.ibm.com>

[CIFS] Fix build break ifdef in wrong place

Signed-off-by: Steve French <sfrench@us.ibm.com>

[CIFS] More removing of unused functions

Signed-off-by: Steve French <sfrench@us.ibm.com>

[CIFS] Remove unused prototypes

Signed-off-by: Steve French <sfrench@us.ibm.com>

[CIFS] Fix build break

Signed-off-by: Steve French <sfrench@us.ibm.com>

[CIFS] Remove static and unused symbols

Most cases of the ones found by Shaggy by
"make namespacecheck"
could be removed or made static

Ack: Dave Kleikamp <shaggy@austin.ibm.com>
Signed-off-by: Steve French <sfrench@us.ibm.com>

[CIFS] Legacy time handling for Win9x and OS/2 part 1

Signed-off-by: Steve French <sfrench@us.ibm.com>

CIFS: Use SEEK_END instead of hardcoded value

Signed-off-by: Josef 'Jeff' Sipek <jeffpc@josefsipek.net>
Signed-off-by: Steve French <sfrench@us.ibm.com>

Merge git://git./linux/kernel/git/sfrench/cifs-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/sfrench/cifs-2.6:
  [CIFS] statfs for cifs unix extensions no longer experimental
  [CIFS] New POSIX locking code not setting rc properly to zero on successful
  [CIFS] Support deep tree mounts (e.g. mounts to //server/share/path)

Merge /pub/scm/linux/kernel/git/davej/agpgart

* master.kernel.org:/pub/scm/linux/kernel/git/davej/agpgart:
  [AGPGART] Rework AGPv3 modesetting fallback.
  [AGPGART] Add suspend callback for i965
  [AGPGART] Fix number of aperture sizes in 830 gart structs.
  [AGPGART] Intel 965 Express support.
  [AGPGART] agp.h: constify struct agp_bridge_data::version
  [AGPGART] const'ify VIA AGP PCI table.
  [AGPGART] CONFIG_PM=n slim: drivers/char/agp/intel-agp.c
  [AGPGART] CONFIG_PM=n slim: drivers/char/agp/efficeon-agp.c
  [AGPGART] Const'ify the agpgart driver version.
  [AGPGART] remove private page protection map

Merge /pub/scm/linux/kernel/git/davej/cpufreq

* master.kernel.org:/pub/scm/linux/kernel/git/davej/cpufreq:
  [CPUFREQ] sw_any_bug_dmi_table can be used on resume, so it isn't initdata
  [CPUFREQ] Fix some more CPU hotplug locking.
  [CPUFREQ] Workaround for BIOS bug in software coordination of frequency
  [CPUFREQ] Longhaul - Add voltage scaling to driver
  [CPUFREQ] Fix sparse warning in ondemand
  [CPUFREQ] make drivers/cpufreq/cpufreq_ondemand.c:powersave_bias_target() static
  [CPUFREQ] Longhaul - Add ignore_latency option
  [CPUFREQ] Longhaul - Disable arbiter
  [CPUFREQ][2/2] ondemand: updated add powersave_bias tunable
  [CPUFREQ][1/2] ondemand: updated tune for hardware coordination
  [CPUFREQ] Fix typo.

[PATCH] fallout from hcd-core patch

missing le16_to_cpu()

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[PATCH] fix the survivors of fbcon_vbl_handler() renaming

In

|Author: James Simmons <jsimmons@kozmo.(none)>
|Date:   Thu Mar 13 22:37:08 2003 -0800
|
|    [FBCON] Cursor handling clean up. I nuked several static variables.

we have

-static void fbcon_vbl_handler(int irq, void *dummy, struct pt_regs *fp)
+static void fb_vbl_handler(int irq, void *dev_id, struct pt_regs *fp)

and 3 years later a couple of instances missed back then still remains
there.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[PATCH] sun4: fix sbus_setup_iommu()

iommu_init() and iounit_init() are never called for sun4, but that's not
enough - these calls should be ifdefed out since the functions in question
simply do not exist for CONFIG_SUN4 kernel.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[PATCH] asm/backlight.h is ppc-only

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[PATCH] sanitize frv archclean

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[PATCH] aoa is pmac-only

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[PATCH] memcpy_fromio() missing in istallion

memcpy() from iomem is a bad thing...

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[PATCH] fix ancient breakage in ebus_init()

Back when pci_dev had base_address[], loop of form
base = &...->base_address[0];
for (.....) {
...
*base++ = addr;
}
was fine, but when that array got spread in ->resource[...].start
replacing the initialization with
base = &...->resource[0].start;
was not a sufficient modification.  IOW this code got broken for cases
when there had been more than one resource to fill.  All way back in
2.3.41-pre3...

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[PATCH] fix missing ifdefs in syscall classes hookup for generic targets

several targets have no ....at() family and m32r calls its only chown variant
chown32(), with __NR_chown being undefined.  creat(2) is also absent in some
targets.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[CPUFREQ] sw_any_bug_dmi_table can be used on resume, so it isn't initdata

sw_any_bug_dmi_table can be used on resume, so it isn't initdata.

Signed-off-by: Jeremy Fitzhardinge <jeremy@goop.org>
Signed-off-by: Dave Jones <davej@redhat.com>

[CPUFREQ] Fix some more CPU hotplug locking.

Lukewarm IQ detected in hotplug locking
BUG: warning at kernel/cpu.c:38/lock_cpu_hotplug()
[<b0134a42>] lock_cpu_hotplug+0x42/0x65
[<b02f8af1>] cpufreq_update_policy+0x25/0xad
[<b0358756>] kprobe_flush_task+0x18/0x40
[<b0355aab>] schedule+0x63f/0x68b
[<b01377c2>] __link_module+0x0/0x1f
[<b0119e7d>] __cond_resched+0x16/0x34
[<b03560bf>] cond_resched+0x26/0x31
[<b0355b0e>] wait_for_completion+0x17/0xb1
[<f965c547>] cpufreq_stat_cpu_callback+0x13/0x20 [cpufreq_stats]
[<f9670074>] cpufreq_stats_init+0x74/0x8b [cpufreq_stats]
[<b0137872>] sys_init_module+0x91/0x174
[<b0102c81>] sysenter_past_esp+0x56/0x79

As there are other places that call cpufreq_update_policy without
the hotplug lock, it seems better to keep the hotplug locking
at the lower level for the time being until this is revamped.

Signed-off-by: Dave Jones <davej@redhat.com>

Merge branch 'for-linus' of /linux/kernel/git/roland/infiniband

* 'for-linus' of master.kernel.org:/pub/scm/linux/kernel/git/roland/infiniband: (65 commits)
  IB: Fix typo in kerneldoc for ib_set_client_data()
  IPoIB: Add some likely/unlikely annotations in hot path
  IPoIB: Remove unused include of vmalloc.h
  IPoIB: Rejoin all multicast groups after a port event
  IPoIB: Create MCGs with all attributes required by RFC
  IB/sa: fix ib_sa_selector names
  IB/iser: INFINIBAND_ISER depends on INET
  IB/mthca: Simplify calls to mthca_cq_clean()
  RDMA/cma: Document rdma_accept() error handling
  IB/mthca: Recover from catastrophic errors
  RDMA/cma: Document rdma_destroy_id() function
  IB/cm: Do not track remote QPN in timewait state
  IB/sa: Require SA registration
  IPoIB: Refactor completion handling
  IB/iser: Do not use FMR for a single dma entry sg
  IB/iser: fix some debug prints
  IB/iser: make FMR "page size" be 4K and not PAGE_SIZE
  IB/iser: Limit the max size of a scsi command
  IB/iser: fix a check of SG alignment for RDMA
  RDMA/cma: Protect against adding device during destruction
  ...

Merge branch 'upstream-linus' of /linux/kernel/git/jgarzik/netdev-2.6

* 'upstream-linus' of master.kernel.org:/pub/scm/linux/kernel/git/jgarzik/netdev-2.6:
  [netdrvr] mv643xx_eth: fix obvious typo, which caused build breakage
  [netdrvr] lp486e: fix typo

IB: Fix typo in kerneldoc for ib_set_client_data()

Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IPoIB: Add some likely/unlikely annotations in hot path

Signed-off-by: Eli Cohen <eli@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IPoIB: Remove unused include of vmalloc.h

IPoIB doesn't use anything from <linux/vmalloc.h>, so don't include it.

Signed-off-by: Dotan Barak <dotanb@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IPoIB: Rejoin all multicast groups after a port event

When ipoib_ib_dev_flush() is called because of a port event, the
driver needs to rejoin all multicast groups, since the flush will call
ipoib_mcast_dev_flush() (via ipoib_ib_dev_down()).  Otherwise no
(non-broadcast) multicast groups will be rejoined until the networking
core calls ->set_multicast_list again, and so multicast reception will
be broken for potentially a long time.

Signed-off-by: Eli Cohen <eli@mellanox.co.il>
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IPoIB: Create MCGs with all attributes required by RFC

RFC 4391 ("Transmission of IP over InfiniBand (IPoIB)") says:

  If the IB multicast group does not already exist, one must be
  created first with the IPoIB link MTU.  The MGID MUST use the same
  P_Key, Q_Key, SL, MTU, and HopLimit as those used in the
  broadcast-GID.  The rest of attributes SHOULD follow the values used
  in the broadcast-GID as well.

However, the current IPoIB driver is only setting the attributes
required by the InfiniBand spec to create a multicast group, so in
particular the MTU and HopLimit are not being set.  Add these
attributes when creating MCGs, and also set the Rate attribute, since
IPoIB pays attention to that attribute as well.

Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/sa: fix ib_sa_selector names

Relevant SA queries are actually "greater than" / "less than", not
"greater than or equal" / "less than or equal" as the names imply.
(See IB spec 1.2 Vol 1, 15.2.5.16 PATHRECORD/Table 205 PathRecord)

Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/iser: INFINIBAND_ISER depends on INET

iSER won't build without CONFIG_INET enabled, so make Kconfig reflect that.

Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/mthca: Simplify calls to mthca_cq_clean()

If a QP has separate send and receive CQs, then the send CQ will never
have receive completions from that QP in it.  So when cleaning the
send CQ, there's no need to pass in an SRQ pointer, even if the QP is
attached to an SRQ.

Signed-off-by: Roland Dreier <rolandd@cisco.com>

RDMA/cma: Document rdma_accept() error handling

Document the reject sending and modifying QP to error done in rdma_accept().

Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com>
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/mthca: Recover from catastrophic errors

Trigger device remove and then add when a catastrophic error is
detected in hardware.  This, in turn, will cause a device reset, which
we hope will recover from the catastrophic condition.

Since this might interefere with debugging the root cause, add a
module option to suppress this behaviour.

Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

RDMA/cma: Document rdma_destroy_id() function

Clarify that rdma_destroy_id cancels outstanding asynchronous operations on the
Associated id.

Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com>
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/cm: Do not track remote QPN in timewait state

Do not track remote QPN in TimeWait state, since QP is not connected.

Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/sa: Require SA registration

Require users to register with SA module, to prevent the sa_query
module text from going away while an SA query callback is still
running.  Update all in-tree users for the new interface.

Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IPoIB: Refactor completion handling

Split up ipoib_ib_handle_wc() into ipoib_ib_handle_rx_wc() and
ipoib_ib_handle_tx_wc() to make the code easier to read.  This will
also help implement NAPI in the future.

Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/iser: Do not use FMR for a single dma entry sg

Fast Memory Registration (fmr) is used to register for rdma an sg whose
elements are not linearly sequential after dma mapping.

The IB verbs layer provides an "all dma memory MR (memory region)" which
can be used for RDMA-ing a dma linearly sequential buffer.

Change the code to use the dma mr instead of doing fmr when dma mapping
produces a single dma entry sg.

Signed-off-by: Erez Zilber <erezz@voltaire.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/iser: fix some debug prints

fix and add some debug prints related to iser
handling of memory for rdma.

Signed-off-by: Erez Zilber <erezz@voltaire.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/iser: make FMR "page size" be 4K and not PAGE_SIZE

As iser is able to use at most one rdma operation for the
execution of a scsi command, and registration of the sg
associated with scsi command has its restrictions, the code
checks if an sg is "aligned for rdma".

Alignment for rdma is measured in "fmr page" units whose
possible resolutions are different between HCAs and can be
smaller, equal or bigger to the system page size.

When the system page size is bigger than 4KB (eg the default
with ia64 kernels) there a bigger chance that an sg would be
aligned for rdma if the fmr page size is 4KB.

Change the code to create FMR whose pages are of size 4KB
and to take that into account when processing the sg.

Signed-off-by: Erez Zilber <erezz@voltaire.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/iser: Limit the max size of a scsi command

Currently, the data length of a command coming down from scsi-ml
is limited only by the size of its sg list (sg_tablesize). The
max data length may be different for different page size values.
By setting max_sectors, we limit the data length to
max_sectors*512 bytes.

Signed-off-by: Erez Zilber <erezz@voltaire.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/iser: fix a check of SG alignment for RDMA

dma mapping may include a "compaction" of the sg associated with scsi command.
Hence, the size of the maximal prefix of the SG which is aligned for rdma must be
compared against the length of the dma mapped sg (mem->dma_nents) and not against
the size of it before it was mapped (mem->size).

Signed-off-by: Erez Zilber <erezz@voltaire.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

RDMA/cma: Protect against adding device during destruction

Closes a window where address resolution can attach an rdma_cm_id to a
device during destruction of the rdma_cm_id.  This can result in the
rdma_cm_id remaining in the device list after its memory has been
freed.

Signed-off-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

RDMA/amso1100: Add driver for Ammasso 1100 RNIC

Add a driver for the Ammasso 1100 gigabit ethernet RNIC.

Signed-off-by: Tom Tucker <tom@opengridcomputing.com>
Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

RDMA: iWARP Core Changes.

Modifications to the existing rdma header files, core files, drivers,
and ulp files to support iWARP, including:
 - Hook iWARP CM into the build system and use it in rdma_cm.
 - Convert enum ib_node_type to enum rdma_node_type, which includes
   the possibility of RDMA_NODE_RNIC, and update everything for this.

Signed-off-by: Tom Tucker <tom@opengridcomputing.com>
Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

RDMA: iWARP Connection Manager.

Add an iWARP Connection Manager (CM), which abstracts connection
management for iWARP devices (RNICs).  It is a logical instance of the
xx_cm where xx is the transport type (ib or iw).  The symbols exported
are used by the transport independent rdma_cm module, and are
available also for transport dependent ULPs.

Signed-off-by: Tom Tucker <tom@opengridcomputing.com>
Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB: Whitespace fixes

Remove some trailing whitespace that has snuck in despite the best
efforts of whitespace=error-all.  Also fix a few other whitespace
bogosities.

Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/cm: Randomize starting comm ID

Randomize the starting local comm ID to avoid getting a rejected
connection due to a stale connection after a system reboot or
reloading of the ib_cm.

Signed-off-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/mad: Remove unused includes

The ib_mad module does not use a kthread function, but mad_priv.h
includes <linux/kthread.h>.  mad_rmpp.c does not do any DMA-related
stuff, but includes <linux/dma-mapping.h>.  Remove the unused includes.

Signed-off-by: James Lentini <jlentini@netapp.com>
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/mad: Add support for dual-sided RMPP transfers.

The implementation assumes that any RMPP request that requires a
response uses DS RMPP.  Based on the RMPP start-up scenarios defined
by the spec, this should be a valid assumption.  That is, there is no
start-up scenario defined where an RMPP request is followed by a
non-RMPP response.  By having this assumption we avoid any API
changes.

In order for a node that supports DS RMPP to communicate with one that
does not, RMPP responses assume a new window size of 1 if a DS ACK has
not been received.  (By DS ACK, I'm referring to the turn-around ACK
after the final ACK of the request.)  This is a slight spec deviation,
but is necessary to allow communication with nodes that do not
generate the DS ACK.  It also handles the case when a response is sent
after the request state has been discarded.

Signed-off-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/cm: Use correct reject code for invalid GID

Set the reject code properly when rejecting a request that contains an
invalid GID.  A suitable GID is returned by the IB CM in the
additional reject information (ARI).  This is a spec compliancy issue.

Signed-off-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/cm: Enable atomics along with RDMA reads

Enable atomic operations along with RDMA reads if a local RDMA
read/atomic depth is provided by the user.

Signed-off-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/mthca: Return correct number of bits for static rate in query_qp

Incorrect number of bits was taken for static_rate field.

Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/mthca: Return port number for unconnected QPs in query_qp

port_num was not being returned for unconnected QPs.

Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/mthca: Fix default static rate returned for Tavor in AV

When default static rate is returned for Tavor, need to translate it
to an ib rate value.

Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/ipath: control receive polarity inversion

Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/ipath: fix return value from ipath_poll

This stops the generic poll code from waiting for a timeout.

Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/ipath: allow SMA to be disabled

This is useful for testing purposes.

Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/ipath: handle sq_sig_all field correctly

Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/ipath: put a limit on the number of QPs that can be created

Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/ipath: validate path_mig_state properly

Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/ipath: be more strict about testing the modify QP verb

Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/ipath: add serial number to hardware freeze error message

Also added the word "Hardware" after "Fatal" to make it more obvious
that it's hardware, not software.

Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/ipath: support new QLogic product naming scheme

This patch only renames files, fixes product names, and updates
comments.

Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/ipath: account for attached QPs correctly

Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/ipath: do not allow use of CQ entries with invalid counts

Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/ipath: add new minor device to allow sending of diag packets

Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/ipath: trivial cleanups

Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/ipath: remove stale references to userspace SMA

When we first submitted a userspace subnet management agent, it was
rejected, so we left it out of the final driver submission.  This patch
removes a number of vestigial references to it.

Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/ipath: simplify debugging code after ipath_core and ib_ipath merger

Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/ipath: simplify layering code

A lot of ipath layer code was only called in one place. Now that the
ipath_core and ib_ipath drivers are merged, it's more sensible to simply
inline the simple stuff that the layer code was doing.

Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/ipath: merge ipath_core and ib_ipath drivers

There is little point in keeping the two drivers separate, so we are
merging them.

Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/ipath: drop requirement that PIO buffers be mmaped write-only

Some userlands try to mmap these pages read-write, so accommodate them.

Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/ipath: fix handling of kpiobufs

Change comment: no longer imply that user can set ipath_kpiobufs to zero.
Actually set ipath_kpiobufs from parameter. Previously only altered
per-device ipath_lastport_piobuf, which was over-written in chip init.

Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/ipath: fix for crash on module unload, if cfgports < portcnt

Allocate enough pointers for all possible ports, to avoid problems in
cleanup/unload.

Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/ipath: lock resource limit counters correctly

Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/ipath: More changes to support InfiniPath on PowerPC 970 systems

Ordering of writethrough store buffers needs to be forced, and we need
to use ifdef to get writethrough behavior to InfiniPath buffers, because
there is no generic way to specify that at this time (similar to code
in char/drm/drm_vm.c and block/z2ram.c).

Signed-off-by: John Gregor <john.gregor@qlogic.com>
Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/ipath: Performance improvements via mmap of queues

Improve performance of userspace post receive, post SRQ receive, and
poll CQ operations for ipath by allowing userspace to directly mmap()
receive queues and completion queues.  This eliminates the copying
between userspace and the kernel in the data path.

Signed-off-by: Ralph Campbell <ralph.campbell@qlogic.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/uverbs: Pass userspace data to modify_srq and modify_qp methods

Pass a struct ib_udata to the low-level driver's ->modify_srq() and
->modify_qp() methods, so that it can get to the device-specific data
passed in by the userspace driver.

Signed-off-by: Ralph Campbell <ralph.campbell@qlogic.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/uverbs: Allow resize CQ operation to return driver-specific data

Add a ib_uverbs_resize_cq_resp.driver_data field so that low-level
drivers can return data from a resize CQ operation to userspace.  Have
ib_uverbs_resize_cq() only copy the cqe field, to avoid having to bump
the userspace ABI.

Signed-off-by: Ralph Campbell <ralph.campbell@qlogic.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/ehca: Add driver for IBM eHCA InfiniBand adapters

Add a driver for IBM GX bus InfiniBand adapters, which are usable with
some pSeries/System p systems.

Signed-off-by: Heiko J Schick <schickhj.ibm.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/srp: Add port/device attributes

Add local_ib_device and local_ib_port attributes to srp scsi_host.
These are needed when we want to connect to the same target through
multiple distinct ports.

Signed-off-by: Ishai Rabinovitz <ishai@mellanox.co.il>
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/mthca: Include the header we really want

Signed-off-by: James Lentini <jlentini@netapp.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/uverbs: Fix lockdep warning when QP is created with 2 CQs

Lockdep warns when userspace creates a QP that uses different CQs for
send completions and receive completions, because both CQs are locked
and their mutexes belong to the same lock class.  However, we know
that the mutexes are distinct and the nesting is safe (there is no
possibility of AB-BA deadlock because the mutexes are locked with
down_read()), so annotate the situation with SINGLE_DEPTH_NESTING to
get rid of the lockdep warning.

Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/uverbs: Use idr_read_cq() where appropriate

There were two functions that open-coded idr_read_cq() in terms of
idr_read_uobj() rather than using the helper.

Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/ipoib: Fix flush/start xmit race (from code review)

Prevent flush task from freeing the ipoib_neigh pointer, while
ipoib_start_xmit() is accessing the ipoib_neigh through the pointer it
has loaded from the skb's hardware address.

Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/mthca: Don't use privileged UAR for kernel access

Make kernel use UAR2 instead of UAR1 for hardware access: this adds
sanity checking from the hardware side, without any performance cost.

Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

IB/mthca: Fix lid used for sending traps

The SM LID used to send traps to is incorrectly set to port LID.  This
is a regression from 2.6.17 -- after a PortInfo MAD is received, no
traps are sent to the SM LID.  The traps go to the loopback interface
instead, and are dropped there.  The SM LID should be taken from the
sm_lid of the PortInfo response.

The bug was introduced by commit 12bbb2b7be7f5564952ebe0196623e97464b8ac5:
IB/mthca: Add client reregister event generation

Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>