Re-arrange the logic in the NFS direct I/O path so that nfs_read/write_data
structs are allocated just before they are scheduled, rather than
allocating them all at once before we start scheduling requests.
Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Neil Brown observed that the kmalloc() in nfs_get_user_pages() is more
likely to fail if the I/O is large enough to require the allocation of more
than a single page to keep track of all the pinned pages in the user's
buffer.
Instead of tracking one large page array per dreq/iocb, track pages per
nfs_read/write_data, just like the cached I/O path does. An array for
pages is already allocated for us by nfs_readdata_alloc() (and the write
and commit equivalents).
This is also required for adding support for vectored I/O to the NFS direct
I/O path.
The original reason to pin the user buffer and allocate all the NFS data
structures before trying to schedule I/O was to ensure all needed resources
are allocated on the client before starting to send requests. This reduces
the chance that resource exhaustion on the client will cause a short read
or write.
On the other hand, for an application making very large application I/O
requests, this means that it will be nearly impossible for the application
to make forward progress on a resource-limited client.
Thus, moving the buffer pinning functionality into the I/O scheduling
loops should be good for scalability. The next patch will do the same for
NFS data structure allocation.
Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Clean-up and fix a minor bug: the logic was dirtying page cache pages on
both read and write operations.
Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Make the user_addr, user_count, and pos parameters explicit to the
scheduler routines, and remove the fields from nfs_direct_req. The
iovec API will be passing in a series of these, not just one set.
Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
An NFSv3/v4 client must reschedule on-the-wire writes if the writes are
UNSTABLE, and the server reboots before the client can complete a
subsequent COMMIT request.
To support direct asynchronous scatter-gather writes, the write
rescheduler in fs/nfs/direct.c must not depend on the I/O parameters
in the controlling nfs_direct_req structure. iovecs can be somewhat
arbitrarily complex, so there could be an unbounded amount of information
to save for a rarely encountered requirement.
Refactor the direct write rescheduler so it uses information from each
nfs_write_data structure to reschedule writes, instead of caching that
information in the controlling nfs_direct_req structure.
Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Factor out the logic that increments and decrements the outstanding I/O
count. This will be a commonly used bit of code in upcoming patches.
Also make this an atomic_t again, since it will be very often manipulated
outside dreq->spin lock.
Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Patch from Andrew Victor
Remove the remaining legacy __mem_isa() definitions for the ARM
platforms.
Signed-off-by: Andrew Victor <andrew@sanpeople.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Patch from Lennert Buytenhek
The ep93xx ohci bits have been merged into the gregkh-2.6 tree, which
means that they will probably go upstream soon, so make the core ep93xx
code instantiate an appropriate ep93xx-ohci platform device.
Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Patch from Lennert Buytenhek
Add an initial implementation of the clk_* API for the cirrus ep93xx
to the tree. The initial implementation is somewhat minimal, with the
intention of extending it as we go along.
Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Patch from Lennert Buytenhek
Commit d6551e884c forgot to update the
description of what goes into r2 when calling iwmmxt_task_restore().
Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
i915 vblanks can be generated from either pipe a or b, however a disabled
pipe generates no interrupts. This change allows the X server to select
which pipe generates vblank interrupts.
From: Keith Packard <keith.packard@intel.com> via DRM CVS
Signed-off-by: Dave Airlie <airlied@linux.ie>
Add support for r200 vertex programs (R200_EMIT_VAP_PVS_CNTL, and new
packet type for making it possible to address whole tcl vector space
and have a larger count)
From: Roland Scheidegger (DRM CVS)
Signed-off-by: Dave Airlie <airlied@linux.ie>
Do a tcl state flush before accessing tcl vector space. This fixes some
more problems with flickering (bug #6637). drm may not be appropriate
place for this, since doing that flush there might both be overkill and
insufficient in some cases. However, it's hard to figure out when that
flush is needed, so this has to suffice. There does not seem to be a
performance penalty associated with it.
From: Roland Scheidegger (DRM CVS)
Signed-off-by: Dave Airlie <airlied@linux.ie>
ring_head is offset in card memory, not iomem pointer. Fixed, removed
fuckloads of amazingly bogus casts somebody had sprinkled all over the
place.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Dave Airlie <airlied@linux.ie>
