David Howells [Wed, 23 Aug 2006 00:06:12 +0000 (20:06 -0400)]
NFS: Eliminate client_sys in favour of cl_rpcclient
Eliminate nfs_server::client_sys in favour of nfs_client::cl_rpcclient as we
only really need one per server that we're talking to since it doesn't have any
security on it.
The retransmission management variables are also moved to the common struct as
they're required to set up the cl_rpcclient connection.
The NFS2/3 client and client_acl connections are thenceforth derived by cloning
the cl_rpcclient connection and post-applying the authorisation flavour.
The code for setting up the initial common connection has been moved to
client.c as nfs_create_rpc_client(). All the NFS program definition tables are
also moved there as that's where they're now required rather than super.c.
Signed-Off-By: David Howells <dhowells@redhat.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
David Howells [Wed, 23 Aug 2006 00:06:10 +0000 (20:06 -0400)]
NFS: Use the dentry superblock directly in nfs_statfs()
Use the nominated dentry's superblock directly in the NFS statfs() op to get a
file handle, rather than using s_root (which will become a dummy dentry in a
future patch).
Signed-Off-By: David Howells <dhowells@redhat.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
David Howells [Wed, 23 Aug 2006 00:06:09 +0000 (20:06 -0400)]
NFS: Add a lookupfh NFS RPC op
Add a lookup filehandle NFS RPC op so that a file handle can be looked up
without requiring dentries and inodes and other VFS stuff when doing an NFS4
pathwalk during mounting.
Signed-Off-By: David Howells <dhowells@redhat.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
David Howells [Wed, 23 Aug 2006 00:06:08 +0000 (20:06 -0400)]
NFS: Fix NFS4 callback up/down prototypes
Make the nfs_callback_up()/down() prototypes just do nothing if NFS4 is not
enabled. Also make the down function void type since we can't really do
anything if it fails.
Signed-Off-By: David Howells <dhowells@redhat.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
David Howells [Wed, 23 Aug 2006 00:06:07 +0000 (20:06 -0400)]
NFS: Add dentry materialisation op
The attached patch adds a new directory cache management function that prepares
a disconnected anonymous function to be connected into the dentry tree. The
anonymous dentry is transferred the name and parentage from another dentry.
The following changes were made in [try #2]:
(*) d_materialise_dentry() now switches the parentage of the two nodes around
correctly when one or other of them is self-referential.
The following changes were made in [try #7]:
(*) d_instantiate_unique() has had the interior part split out as function
__d_instantiate_unique(). Callers of this latter function must be holding
the appropriate locks.
(*) _d_rehash() has been added as a wrapper around __d_rehash() to call it
with the most obvious hash list (the one from the name). d_rehash() now
calls _d_rehash().
(*) d_materialise_dentry() is now __d_materialise_dentry() and is static.
(*) d_materialise_unique() added to perform the combination of d_find_alias(),
d_materialise_dentry() and d_add_unique() that the NFS client was doing
twice, all within a single dcache_lock critical section. This reduces the
number of times two different spinlocks were being accessed.
The following further changes were made:
(*) Add the dentries onto their parents d_subdirs lists.
Signed-Off-By: David Howells <dhowells@redhat.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
A pinned inode may in theory end up filling memory with cached ACCESS
calls. This patch ensures that the VM may shrink away the cache in these
particular cases.
The shrinker works by iterating through the list of inodes on the global
nfs_access_lru_list, and removing the least recently used access
cache entry until it is done (or until the entire cache is empty).
NFS: Add a new ACCESS rpc call cache to the linux nfs client
The current access cache only allows one entry at a time to be cached for each
inode. Add a per-inode red-black tree in order to allow more than one to
be cached at a time.
Should significantly cut down the time spent in path traversal for shared
directories such as ${PATH}, /usr/share, etc.
* 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)
* 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.
Al Viro [Sat, 23 Sep 2006 00:26:02 +0000 (01:26 +0100)]
[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>
Al Viro [Sat, 23 Sep 2006 00:18:41 +0000 (01:18 +0100)]
[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>
Al Viro [Fri, 22 Sep 2006 23:10:18 +0000 (00:10 +0100)]
[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>
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.
Merge branch 'for-linus' of master.kernel.org:/pub/scm/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
...
Eli Cohen [Fri, 22 Sep 2006 22:22:56 +0000 (15:22 -0700)]
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>
Roland Dreier [Fri, 22 Sep 2006 22:22:56 +0000 (15:22 -0700)]
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.
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>
Roland Dreier [Fri, 22 Sep 2006 22:22:55 +0000 (15:22 -0700)]
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.
Jack Morgenstein [Tue, 15 Aug 2006 18:11:18 +0000 (21:11 +0300)]
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>
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>
Roland Dreier [Fri, 22 Sep 2006 22:22:52 +0000 (15:22 -0700)]
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.
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>
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>
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>
Sean Hefty [Fri, 1 Sep 2006 22:33:55 +0000 (15:33 -0700)]
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>
Tom Tucker [Fri, 22 Sep 2006 22:22:48 +0000 (15:22 -0700)]
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>
Tom Tucker [Thu, 3 Aug 2006 21:02:42 +0000 (16:02 -0500)]
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>
Tom Tucker [Thu, 3 Aug 2006 21:02:40 +0000 (16:02 -0500)]
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>
James Lentini [Mon, 28 Aug 2006 22:12:04 +0000 (15:12 -0700)]
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>
Sean Hefty [Mon, 28 Aug 2006 22:10:32 +0000 (15:10 -0700)]
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>
Sean Hefty [Mon, 28 Aug 2006 18:57:42 +0000 (11:57 -0700)]
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>
Bryan O'Sullivan [Fri, 25 Aug 2006 18:24:34 +0000 (11:24 -0700)]
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>
Bryan O'Sullivan [Fri, 25 Aug 2006 18:24:32 +0000 (11:24 -0700)]
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>
Bryan O'Sullivan [Fri, 25 Aug 2006 18:24:29 +0000 (11:24 -0700)]
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>
Bryan O'Sullivan [Fri, 25 Aug 2006 18:24:26 +0000 (11:24 -0700)]
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>
Ralph Campbell [Fri, 22 Sep 2006 22:22:26 +0000 (15:22 -0700)]
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>
Ralph Campbell [Fri, 11 Aug 2006 21:58:09 +0000 (14:58 -0700)]
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>
Ralph Campbell [Fri, 22 Sep 2006 22:22:24 +0000 (15:22 -0700)]
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>
Ishai Rabinovitz [Tue, 15 Aug 2006 14:34:52 +0000 (17:34 +0300)]
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>
Roland Dreier [Fri, 22 Sep 2006 22:17:20 +0000 (15:17 -0700)]
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.
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>
Jack Morgenstein [Tue, 19 Sep 2006 08:13:24 +0000 (11:13 +0300)]
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.
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>
projects / linux-2.6 / log