IDR IDs are signed, so struct ib_uobject.id should be signed.  This
avoids some sparse pointer signedness warnings.
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

Avoid confusion about what it might mean, since it's never initialized.
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

Add a device capability to show when it can handle checksum offload.
Also add a send flag for inserting checksums and a csum_ok field to
the completion record.
Signed-off-by: NEli Cohen <eli@mellanox.co.il>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

This is based on user feedback from Doug Ledford at RedHat:

Events that occur on an rdma_cm_id are reported to userspace through an
event channel.  Connection request events are reported on the event
channel associated with the listen.  When the connection is accepted, a
new rdma_cm_id is created and automatically uses the listen event
channel.  This is suboptimal where the user only wants listen events on
that channel.

Additionally, it may be desirable to have events related to connection
establishment use a different event channel than those related to
already established connections.

Allow the user to migrate an rdma_cm_id between event channels. All
pending events associated with the rdma_cm_id are moved to the new event
channel.
Signed-off-by: NSean Hefty <sean.hefty@intel.com>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

To allow ULPs to tune timeout values and capture retry statistics,
report the number of times that a mad send operation was retried.

For RMPP mads, report the total number of times that the any portion
(send window) of the send operation was retried.
Signed-off-by: NSean Hefty <sean.hefty@intel.com>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

Stop using kobject_register, as this way we can control the sending of
the uevent properly, after everything is properly initialized.

Cc: Roland Dreier <rolandd@cisco.com>
Cc: Sean Hefty <mshefty@ichips.intel.com>
Cc: Hal Rosenstock <hal.rosenstock@gmail.com>
Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Not architecture specific code should not #include <asm/scatterlist.h>.

This patch therefore either replaces them with
#include <linux/scatterlist.h> or simply removes them if they were
unused.
Signed-off-by: NAdrian Bunk <bunk@kernel.org>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

The IB CM provides a message received acknowledged (MRA) message that
can be sent to indicate that a REQ or REP message has been received, but
will require more time to process than the timeout specified by those
messages.  In many cases, the application may not know how long it will
take to respond to a CM message, but the majority of the time, it will
usually respond before a retry has been sent.  Rather than sending an
MRA in response to all messages just to handle the case where a longer
timeout is needed, it is more efficient to queue the MRA for sending in
case a duplicate message is received.

This avoids sending an MRA when it is not needed, but limits the number
of times that a REQ or REP will be resent.  It also provides for a
simpler implementation than generating the MRA based on a timer event.
(That is, trying to send the MRA after receiving the first REQ or REP if
a response has not been generated, so that it is received at the remote
side before a duplicate REQ or REP has been received)
Signed-off-by: NSean Hefty <sean.hefty@intel.com>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

The declaration of struct ib_user_mad_reg_req.method_mask[] exported
to userspace was an array of __u32, but the kernel internally treated
it as a bitmap made up of longs.  This makes a difference for 64-bit
big-endian kernels, where numbering the bits in an array of__u32 gives:

    |31.....0|63....31|95....64|127...96|

while numbering the bits in an array of longs gives:

    |63..............0|127............64|

64-bit userspace can handle this by just treating method_mask[] as an
array of longs, but 32-bit userspace is really stuck: the meaning of
the bits in method_mask[] depends on whether the kernel is 32-bit or
64-bit, and there's no sane way for userspace to know that.

Fix this by updating <rdma/ib_user_mad.h> to make it clear that
method_mask[] is an array of longs, and using a compat_ioctl method to
convert to an array of 64-bit longs to handle the 32-on-64 problem.
This fixes the interface description to match existing behavior (so
working binaries continue to work) in almost all situations, and gives
consistent semantics in the case of 32-bit userspace that can run on
either a 32-bit or 64-bit kernel, so that the same binary can work for
both 32-on-32 and 32-on-64 systems.
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

Add support for setting the P_Key index of sent MADs and getting the
P_Key index of received MADs.  This requires a change to the layout of
the ABI structure struct ib_user_mad_hdr, so to avoid breaking
compatibility, we default to the old (unchanged) ABI and add a new
ioctl IB_USER_MAD_ENABLE_PKEY that allows applications that are aware
of the new ABI to opt into using it.

We plan on switching to the new ABI by default in a year or so, and
this patch adds a warning that is printed when an application uses the
old ABI, to push people towards converting to the new ABI.
Signed-off-by: NRoland Dreier <rolandd@cisco.com>
Reviewed-by: NSean Hefty <sean.hefty@intel.com>
Reviewed-by: NHal Rosenstock <hal@xsigo.com>

During ib_umem_get(), determine whether all pages from the memory
region are hugetlb pages and report this in the "hugetlb" member.
Low-level drivers can use this information if they need it.
Signed-off-by: NJoachim Fenkes <fenkes@de.ibm.com>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

Export the ability to set the type of service to user space.  Model
the interface after setsockopt.
Signed-off-by: NSean Hefty <sean.hefty@intel.com>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

Provide support to specify a type of service for a communication
identifier.  A new function call is used when dealing with IPv4
addresses.  For IPv6 addresses, the ToS is specified through the
traffic class field in the sockaddr_in6 structure.
Signed-off-by: NSean Hefty <sean.hefty@intel.com>

[ The comments Eitan Zahavi and myself have made over the v1 post at 
  <http://lists.openfabrics.org/pipermail/general/2007-August/039247.html>
  were fully addressed. ]
 
Reviewed-by: Or Gerlitz <ogerlitz@voltaire.com> 
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

The QoS annex defines new fields for path records.  Add them to the
ib_sa for consumers that want to use them.
Signed-off-by: NSean Hefty <sean.hefty@intel.com>
Reviewed-by: NOr Gerlitz <ogerlitz@voltaire.com>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

After moving the definition of struct ib_umem_chunk from ib_verbs.h to
ib_umem.h there isn't any reason for the macro IB_UMEM_MAX_PAGE_CHUNK
to stay in ib_verbs.h.  Move the macro to umem.c, the only place where
it is used.
Signed-off-by: NDotan Barak <dotanb@dev.mellanox.co.il>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

ib_verbs.h uses struct list_head and rw_semaphore, so while the files
<linux/list.h> and <linux/rwsem.h> seem to be pulled in indirectly by
the other header files it includes, the right thing is to include
those files directly.
Signed-off-by: NDotan Barak <dotanb@dev.mellanox.co.il>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

ib_mad.h uses struct list_head, so while linux/list.h seems to be
pulled in indirectly by one of the headers it includes, the right
thing is to include linux/list.h directly.
Signed-off-by: NDotan Barak <dotanb@dev.mellanox.co.il>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

The IB CM should include the HCA ACK delay when calculating the local
ACK timeout value to use for RC QPs.  If the HCA ACK delay is large
enough relative to the packet life time, then if it is not taken into
account, the calculated timeout value ends up being too small, which
can result in "retry exceeded" errors.
Signed-off-by: NSean Hefty <sean.hefty@intel.com>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

MADs sent to the SA should use the the default P_Key (0x7fff/0xffff).
There's no requirement that the default P_Key is stored at index 0 in
the local P_Key table, so add code to the sa_query module to look up
the index of the default P_Key when creating an address handle for the
SA (which is done any time the P_Key table might change), and use this
index for all SA queries.
Signed-off-by: NSean Hefty <sean.hefty@intel.com>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

First thing mm.h does is including sched.h solely for can_do_mlock() inline
function which has "current" dereference inside. By dealing with can_do_mlock()
mm.h can be detached from sched.h which is good. See below, why.

This patch
a) removes unconditional inclusion of sched.h from mm.h
b) makes can_do_mlock() normal function in mm/mlock.c
c) exports can_do_mlock() to not break compilation
d) adds sched.h inclusions back to files that were getting it indirectly.
e) adds less bloated headers to some files (asm/signal.h, jiffies.h) that were
   getting them indirectly

Net result is:
a) mm.h users would get less code to open, read, preprocess, parse, ... if
   they don't need sched.h
b) sched.h stops being dependency for significant number of files:
   on x86_64 allmodconfig touching sched.h results in recompile of 4083 files,
   after patch it's only 3744 (-8.3%).

Cross-compile tested on

	all arm defconfigs, all mips defconfigs, all powerpc defconfigs,
	alpha alpha-up
	arm
	i386 i386-up i386-defconfig i386-allnoconfig
	ia64 ia64-up
	m68k
	mips
	parisc parisc-up
	powerpc powerpc-up
	s390 s390-up
	sparc sparc-up
	sparc64 sparc64-up
	um-x86_64
	x86_64 x86_64-up x86_64-defconfig x86_64-allnoconfig

as well as my two usual configs.
Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Add ib_find_gid() and ib_find_pkey() functions that use uncached device
queries.  The calls might block but the returns are always up-to-date.
Cache P_Key and GID table lengths in core to avoid extra port info queries.
Signed-off-by: NYosef Etigin <yosefe@voltaire.com>
Acked-by: NMichael S. Tsirkin <mst@dev.mellanox.co.il>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

When memory pinned with ib_umem_get() is released, ib_umem_release()
needs to subtract the amount of memory being unpinned from
mm->locked_vm.  However, ib_umem_release() may be called with
mm->mmap_sem already held for writing if the memory is being released
as part of an munmap() call, so it is sometimes necessary to defer
this accounting into a workqueue.

However, the work struct used to defer this accounting is dynamically
allocated before it is queued, so there is the possibility of failing
that allocation.  If the allocation fails, then ib_umem_release has no
choice except to bail out and leave the process with a permanently
elevated locked_vm.

Fix this by allocating the structure to defer accounting as part of
the original struct ib_umem, so there's no possibility of failing a
later allocation if creating the struct ib_umem and pinning memory
succeeds.
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

Export ib_umem_get()/ib_umem_release() and put low-level drivers in
control of when to call ib_umem_get() to pin and DMA map userspace,
rather than always calling it in ib_uverbs_reg_mr() before calling the
low-level driver's reg_user_mr method.

Also move these functions to be in the ib_core module instead of
ib_uverbs, so that driver modules using them do not depend on
ib_uverbs.

This has a number of advantages:
 - It is better design from the standpoint of making generic code a
   library that can be used or overridden by device-specific code as
   the details of specific devices dictate.
 - Drivers that do not need to pin userspace memory regions do not
   need to take the performance hit of calling ib_mem_get().  For
   example, although I have not tried to implement it in this patch,
   the ipath driver should be able to avoid pinning memory and just
   use copy_{to,from}_user() to access userspace memory regions.
 - Buffers that need special mapping treatment can be identified by
   the low-level driver.  For example, it may be possible to solve
   some Altix-specific memory ordering issues with mthca CQs in
   userspace by mapping CQ buffers with extra flags.
 - Drivers that need to pin and DMA map userspace memory for things
   other than memory regions can use ib_umem_get() directly, instead
   of hacks using extra parameters to their reg_phys_mr method.  For
   example, the mlx4 driver that is pending being merged needs to pin
   and DMA map QP and CQ buffers, but it does not need to create a
   memory key for these buffers.  So the cleanest solution is for mlx4
   to call ib_umem_get() in the create_qp and create_cq methods.
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

The semantics defined by the InfiniBand specification say that
completion events are only generated when a completions is added to a
completion queue (CQ) after completion notification is requested.  In
other words, this means that the following race is possible:

	while (CQ is not empty)
		ib_poll_cq(CQ);
	// new completion is added after while loop is exited
	ib_req_notify_cq(CQ);
	// no event is generated for the existing completion

To close this race, the IB spec recommends doing another poll of the
CQ after requesting notification.

However, it is not always possible to arrange code this way (for
example, we have found that NAPI for IPoIB cannot poll after
requesting notification).  Also, some hardware (eg Mellanox HCAs)
actually will generate an event for completions added before the call
to ib_req_notify_cq() -- which is allowed by the spec, since there's
no way for any upper-layer consumer to know exactly when a completion
was really added -- so the extra poll of the CQ is just a waste.

Motivated by this, we add a new flag "IB_CQ_REPORT_MISSED_EVENTS" for
ib_req_notify_cq() so that it can return a hint about whether the a
completion may have been added before the request for notification.
The return value of ib_req_notify_cq() is extended so:

	 < 0	means an error occurred while requesting notification
	== 0	means notification was requested successfully, and if
		IB_CQ_REPORT_MISSED_EVENTS was passed in, then no
		events were missed and it is safe to wait for another
		event.
	 > 0	is only returned if IB_CQ_REPORT_MISSED_EVENTS was
		passed in.  It means that the consumer must poll the
		CQ again to make sure it is empty to avoid the race
		described above.

We add a flag to enable this behavior rather than turning it on
unconditionally, because checking for missed events may incur
significant overhead for some low-level drivers, and consumers that
don't care about the results of this test shouldn't be forced to pay
for the test.
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

Add a num_comp_vectors member to struct ib_device and extend
ib_create_cq() to pass in a comp_vector parameter -- this parallels
the userspace libibverbs API.  Update all hardware drivers to set
num_comp_vectors to 1 and have all ULPs pass 0 for the comp_vector
value.  Pass the value of num_comp_vectors to userspace rather than
hard-coding a value of 1.

We want multiple CQ event vector support (via MSI-X or similar for
adapters that can generate multiple interrupts), but it's not clear
how many vectors we want, or how we want to deal with policy issues
such as how to decide which vector to use or how to set up interrupt
affinity.  This patch is useful for experimenting, since no core
changes will be necessary when updating a driver to support multiple
vectors, and we know that we want to make at least these changes
anyway.
Signed-off-by: NMichael S. Tsirkin <mst@dev.mellanox.co.il>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

I noticed that many source files include <linux/pci.h> while they do
not appear to need it. Here is an attempt to clean it all up.

In order to find all possibly affected files, I searched for all
files including <linux/pci.h> but without any other occurence of "pci"
or "PCI". I removed the include statement from all of these, then I
compiled an allmodconfig kernel on both i386 and x86_64 and fixed the
false positives manually.

My tests covered 66% of the affected files, so there could be false
positives remaining. Untested files are:

arch/alpha/kernel/err_common.c
arch/alpha/kernel/err_ev6.c
arch/alpha/kernel/err_ev7.c
arch/ia64/sn/kernel/huberror.c
arch/ia64/sn/kernel/xpnet.c
arch/m68knommu/kernel/dma.c
arch/mips/lib/iomap.c
arch/powerpc/platforms/pseries/ras.c
arch/ppc/8260_io/enet.c
arch/ppc/8260_io/fcc_enet.c
arch/ppc/8xx_io/enet.c
arch/ppc/syslib/ppc4xx_sgdma.c
arch/sh64/mach-cayman/iomap.c
arch/xtensa/kernel/xtensa_ksyms.c
arch/xtensa/platform-iss/setup.c
drivers/i2c/busses/i2c-at91.c
drivers/i2c/busses/i2c-mpc.c
drivers/media/video/saa711x.c
drivers/misc/hdpuftrs/hdpu_cpustate.c
drivers/misc/hdpuftrs/hdpu_nexus.c
drivers/net/au1000_eth.c
drivers/net/fec_8xx/fec_main.c
drivers/net/fec_8xx/fec_mii.c
drivers/net/fs_enet/fs_enet-main.c
drivers/net/fs_enet/mac-fcc.c
drivers/net/fs_enet/mac-fec.c
drivers/net/fs_enet/mac-scc.c
drivers/net/fs_enet/mii-bitbang.c
drivers/net/fs_enet/mii-fec.c
drivers/net/ibm_emac/ibm_emac_core.c
drivers/net/lasi_82596.c
drivers/parisc/hppb.c
drivers/sbus/sbus.c
drivers/video/g364fb.c
drivers/video/platinumfb.c
drivers/video/stifb.c
drivers/video/valkyriefb.c
include/asm-arm/arch-ixp4xx/dma.h
sound/oss/au1550_ac97.c

I would welcome test reports for these files. I am fine with removing
the untested files from the patch if the general opinion is that these
changes aren't safe. The tested part would still be nice to have.

Note that this patch depends on another header fixup patch I submitted
to LKML yesterday:
  [PATCH] scatterlist.h needs types.h
  http://lkml.org/lkml/2007/3/01/141Signed-off-by: NJean Delvare <khali@linux-fr.org>
Cc: Badari Pulavarty <pbadari@us.ibm.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Extend rdma_cm to support multicast communication.  Multicast support
is added to the existing RDMA_PS_UDP port space, as well as a new
RDMA_PS_IPOIB port space.  The latter port space allows joining the
multicast groups used by IPoIB, which enables offloading IPoIB traffic
to a separate QP.  The port space determines the signature used in the
MGID when joining the group.  The newly added RDMA_PS_IPOIB also
allows for unicast operations, similar to RDMA_PS_UDP.

Supporting the RDMA_PS_IPOIB requires changing how UD QPs are initialized,
since we can no longer assume that the qkey is constant.  This requires
saving the Q_Key to use when attaching to a device, so that it is
available when creating the QP.  The Q_Key information is exported to
the user through the existing rdma_init_qp_attr() interface.

Multicast support is also exported to userspace through the rdma_ucm.
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

The IB SA tracks multicast join/leave requests on a per port basis and
does not do any reference counting: if two users of the same port join
the same group, and one leaves that group, then the SA will remove the
port from the group even though there is one user who wants to stay a
member left.  Therefore, in order to support multiple users of the
same multicast group from the same port, we need to perform reference
counting locally.

To do this, add an multicast submodule to ib_sa to perform reference
counting of multicast join/leave operations.  Modify ib_ipoib (the
only in-kernel user of multicast) to use the new interface.
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

Make the untyped data region in ib_user_mad have type u64 so that it
gets aligned properly.  This avoids alignment faults in ib_umad when
casting the data field to an rmpp_mad and accessing the 64-bit tid
field on architectures like ia64.
Signed-off-by: NJason Gunthorpe <jgunthorpe@obsidianresearch.com>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

struct ib_wc currently only includes the local QP number: this matches
the IB spec, but seems mostly useless. The following patch replaces
this with the pointer to qp itself, and updates all low level drivers
and all users.

This has the following advantages:
- Ability to get a per-qp context through wc->qp->qp_context
- Existing drivers already have the qp pointer ready in poll cq, so
  this change actually saves a tiny bit (extra memory read) on data path
  (for ehca it would actually be expensive to find the QP pointer when
  polling a CQ, but ehca does not support SRQ so we can leave wc->qp as
  NULL for ehca)
- Users that need the QP number can still get it through wc->qp->qp_num

Use case:

In IPoIB connected mode code, I have a common CQ shared by multiple
QPs.  To track connection usage, I need a way to get at some per-QP
context upon the completion, and I would like to avoid allocating
context object per work request just to stick a QP pointer into it.
With this code, I can just use wc->qp->qp_context.
Signed-off-by: NMichael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

<rdma/ib_verbs.h> uses struct kref, so it should include <linux/kref.h>
explicitly to avoid hidden include dependencies.
Signed-off-by: NMichael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

The ib_dma_alloc_coherent() wrapper uses a u64* for the dma_handle
parameter, unlike dma_alloc_coherent, which uses dma_addr_t*.  This
means that we need a temporary variable to handle the case when
ib_dma_alloc_coherent() just falls through directly to
dma_alloc_coherent() on architectures where sizeof u64 != sizeof
dma_addr_t.
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

At least on PPC, the "op ? op : dma" construct causes a compile failure
because the dma_* is a do{}while(0) macro.

This turns all of them into proper if/else to avoid this problem.
Signed-off-by: NBen Collins <bcollins@ubuntu.com>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

The QLogic InfiniPath HCAs use programmed I/O instead of HW DMA.
This patch allows a verbs device driver to interpose on DMA mapping
function calls in order to avoid relying on bus_to_virt() and
phys_to_virt() to undo the mappings created by dma_map_single(),
dma_map_sg(), etc.
Signed-off-by: NRalph Campbell <ralph.campbell@qlogic.com>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

Export the rdma cm interfaces to userspace via a misc device.
Signed-off-by: NSean Hefty <sean.hefty@intel.com>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

Allow the use of UD QPs through the rdma_cm, in order to provide
address translation services for resolving IB addresses for datagram
messages using SIDR.
Signed-off-by: NSean Hefty <sean.hefty@intel.com>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

During connection establishment, the passive side of a connection can
receive messages from the active side before the connection event has
been delivered to the user.  Allow the passive side to send messages
in response to received data before the event is delivered.  To handle
the case where the connection messages are lost, a new rdma_notify()
function is added that users may invoke to force a connection into the
established state.
Signed-off-by: NSean Hefty <sean.hefty@intel.com>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

Connection information was never given to the recipient of a
connection request or reply message.  Only the event was delivered.
Report the connection data with the event to allows user to
reject the connection based on the requested parameters, or adjust
their resources to match the request.
Signed-off-by: NSean Hefty <sean.hefty@intel.com>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

The qp_type parameter into the rdma_cm is unneeded, and can be
misleading.  The QP type should be determined from the port space.
Signed-off-by: NSean Hefty <sean.hefty@intel.com>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

The ib_cm_establish() function is replaced with a more generic
ib_cm_notify().  This routine is used to notify the CM that failover
has occurred, so that future CM messages (LAP, DREQ) reach the remote
CM.  (Currently, we continue to use the original path)  This bumps the
userspace CM ABI.

New alternate path information is captured when a LAP message is sent
or received.  This allows QP attributes to be initialized for the user
when a new path is loaded after failover occurs.
Signed-off-by: NSean Hefty <sean.hefty@intel.com>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>