Clean-up: replace rpc_call() helper with direct call to rpc_call_sync.
This makes NFSv2 and NFSv3 synchronous calls more computationally
efficient, and reduces stack consumption in functions that used to
invoke rpc_call more than once.
Test plan:
Compile kernel with CONFIG_NFS enabled. Connectathon on NFS version 2,
version 3, and version 4 mount points.
Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Add fields to the rpc_procinfo struct that allow the display of a
human-readable name for each procedure in the rpc_iostats output.
Also fix it so that the NFSv4 stats are broken up correctly by
sub-procedure number. NFSv4 uses only two real RPC procedures:
NULL, and COMPOUND.
Test plan:
Mount with NFSv2, NFSv3, and NFSv4, and do "cat /proc/self/mountstats".
Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Add a simple mechanism for collecting stats in the RPC client. Stats are
tabulated during xprt_release. Note that per_cpu shenanigans are not
required here because the RPC client already serializes on the transport
write lock.
Test plan:
Compile kernel with CONFIG_NFS enabled. Basic performance regression
testing with high-speed networking and high performance server.
Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Account for various things that occur while an RPC task is executed.
Separate timers for RPC round trip and RPC execution time show how
long RPC requests wait in queue before being sent. Eventually these
will be accumulated at xprt_release time in one place where they can
be viewed from userland.
Test plan:
Compile kernel with CONFIG_NFS enabled.
Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Monitor generic transport events. Add a transport switch callout to
format transport counters for export to user-land.
Test plan:
Compile kernel with CONFIG_NFS enabled.
Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
RPC wait queue length will eventually be exported to userland via the RPC
iostats interface.
Test plan:
Compile kernel with CONFIG_NFS enabled.
Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Add a field in nfs_server to record a timestamp when a mount succeeds.
Report the number of seconds the file system has been mounted via
nfs_show_stats().
Test plan:
Mount an NFS file system, watch the mountstats reports and compare with
clock time.
Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Add a per-superblock performance counter facility to the NFS client. This
facility mimics the counters available for block devices and for
networking. Expose these new counters via the new /proc/self/mountstats
interface.
Thanks to Andrew Morton and Trond Myklebust for their review and comments.
Test plan:
fsx and iozone on UP and SMP systems, with and without pre-emption. Watch
for memory overwrite bugs, and performance loss (significantly more CPU
required per op).
Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Sometimes it's important to know the exact RPC retransmit settings the
kernel is using for an NFS mount point. Add this facility to the NFS
client's show_options method.
Test plan:
Set various retransmit settings via the mount command, and check that the
settings are reflected in /proc/mounts.
Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Create a new file under /proc/self, called mountstats, where mounted file
systems can export information (configuration options, performance counters,
and so on). Use a mechanism similar to /proc/mounts and s_ops->show_options.
This mechanism does not violate namespace security, and is safe to use while
other processes are unmounting file systems.
Thanks to Mike Waychison for his review and comments.
Test-plan:
Test concurrent mount/unmount operations while cat'ing /proc/self/mountstats.
Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
read_cache_mtime is no longer used in nfs_inode. This patch removes
references of read_cache_mtime in the code comments.
Signed-off-by: Goldwyn Rodrigues <rgoldwyn@gmail.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
The nfs_open_context may live longer than the file descriptor that spawned
it, so it needs to carry a reference to the vfsmount. If not, then
generic_shutdown_super() may end up being called before reads and writes
have been flushed out.
Make a couple of functions static while we're at it...
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Add support for sending and receiving large RMPP transfers. The old
code supports transfers only as large as a single contiguous kernel
memory allocation. This patch uses linked list of memory buffers when
sending and receiving data to avoid needing contiguous pages for
larger transfers.
Receive side: copy the arriving MADs in chunks instead of coalescing
to one large buffer in kernel space.
Send side: split a multipacket MAD buffer to a list of segments,
(multipacket_list) and send these using a gather list of size 2.
Also, save pointer to last sent segment, and retrieve requested
segments by walking list starting at last sent segment. Finally,
save pointer to last-acked segment. When retrying, retrieve
segments for resending relative to this pointer. When updating last
ack, start at this pointer.
Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Pass actual capacity of created SRQ back to userspace, so that
userspace can report accurate capacities. This requires an ABI bump,
to change struct ib_uverbs_create_srq_resp.
Signed-off-by: Dotan Barak <dotanb@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Have mthca's create_srq method return the actual capacity of the SRQ
that gets created. Also update comments in <rdma/ib_verbs.h> to
clarify that this is what is expected from ib_create_srq().
Signed-off-by: Dotan Barak <dotanb@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
The size of struct ib_uverbs_create_qp_resp is not even multiple of 8
bytes. This causes problems for low-level drivers that add private
data after the structure: 32-bit userspace will look in the wrong
place for a response from a 64-bit kernel. Fix this by adding a
reserved field. Also, bump the ABI version because this changes the
size of a structure.
Pointed out by Hoang-Nam Nguyen <HNGUYEN@de.ibm.com>.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Add support to uverbs to handle querying userspace SRQs (shared
receive queues), including adding an ABI for marshalling requests and
responses. The kernel midlayer already has the underlying
ib_query_srq() function.
Signed-off-by: Dotan Barak <dotanb@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Add support to uverbs to handle querying userspace QPs (queue pairs),
including adding an ABI for marshalling requests and responses. The
kernel midlayer already has the underlying ib_query_qp() function.
Signed-off-by: Dotan Barak <dotanb@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
The in-kernel mthca driver contains a table of which attributes are
valid for each queue pair state transition. It turns out that both
other IB drivers -- ipath and ehca -- which are being prepared for
merging have copied this table, errors and all.
To forestall this code duplication, move this table and the code to
check parameters against it into a midlayer library function,
ib_modify_qp_is_ok().
Signed-off-by: Roland Dreier <rolandd@cisco.com>
This patch allows the consumer to set the page size of "pages" mapped
by the pool FMRs, which is a feature already existing in the base
verbs API. On the cosmetic side it changes ib_fmr_attr.page_size field
to be named page_shift.
Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Expose a writable "node_desc" sysfs attribute for InfiniBand devices.
This allows userspace to update the node description with information
such as the node's hostname, so that IB network management software
can tie its view to the real world.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Add support to uverbs to handle resizing userspace CQs (completion
queues), including adding an ABI for marshalling requests and
responses. The kernel midlayer already has ib_resize_cq().
Signed-off-by: Roland Dreier <rolandd@cisco.com>
1) huge_pte_offset() did not check the page table hierarchy
elements as being empty correctly, resulting in an OOPS
2) Need platform specific hugetlb_get_unmapped_area() to handle
the top-down vs. bottom-up address space allocation strategies.
Signed-off-by: David S. Miller <davem@davemloft.net>
We only need to write an invalid tag every 16 bytes,
so taking advantage of this can save many instructions
compared to the simple memset() call we make now.
A prefetching implementation is implemented for sun4u
and a block-init store version if implemented for Niagara.
The next trick is to be able to perform an init and
a copy_tsb() in parallel when growing a TSB table.
Signed-off-by: David S. Miller <davem@davemloft.net>
Put it one page below the top of the 32-bit address space.
This gives us ~16MB more address space to work with.
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently allocations are very constrained for 32-bit processes.
It grows down-up from 0x70000000 to 0xf0000000 which gives about
2GB of stack + dynamic mmap() space.
So support the top-down method, and we need to override the
generic helper function in order to deal with D-cache coloring.
With these changes I was able to squeeze out a mmap() just over
3.6GB in size in a 32-bit process.
Signed-off-by: David S. Miller <davem@davemloft.net>
This is good for up to %50 performance improvement of some test cases.
The problem has been the race conditions, and hopefully I've plugged
them all up here.
1) There was a serious race in switch_mm() wrt. lazy TLB
switching to and from kernel threads.
We could erroneously skip a tsb_context_switch() and thus
use a stale TSB across a TSB grow event.
There is a big comment now in that function describing
exactly how it can happen.
2) All code paths that do something with the TSB need to be
guarded with the mm->context.lock spinlock. This makes
page table flushing paths properly synchronize with both
TSB growing and TLB context changes.
3) TSB growing events are moved to the end of successful fault
processing. Previously it was in update_mmu_cache() but
that is deadlock prone. At the end of do_sparc64_fault()
we hold no spinlocks that could deadlock the TSB grow
sequence. We also have dropped the address space semaphore.
While we're here, add prefetching to the copy_tsb() routine
and put it in assembler into the tsb.S file. This piece of
code is quite time critical.
There are some small negative side effects to this code which
can be improved upon. In particular we grab the mm->context.lock
even for the tsb insert done by update_mmu_cache() now and that's
a bit excessive. We can get rid of that locking, and the same
lock taking in flush_tsb_user(), by disabling PSTATE_IE around
the whole operation including the capturing of the tsb pointer
and tsb_nentries value. That would work because anyone growing
the TSB won't free up the old TSB until all cpus respond to the
TSB change cross call.
I'm not quite so confident in that optimization to put it in
right now, but eventually we might be able to and the description
is here for reference.
This code seems very solid now. It passes several parallel GCC
bootstrap builds, and our favorite "nut cruncher" stress test which is
a full "make -j8192" build of a "make allmodconfig" kernel. That puts
about 256 processes on each cpu's run queue, makes lots of process cpu
migrations occur, causes lots of page table and TLB flushing activity,
incurs many context version number changes, and it swaps the machine
real far out to disk even though there is 16GB of ram on this test
system. :-)
Signed-off-by: David S. Miller <davem@davemloft.net>
Report 'sun4v' when appropriate in /proc/cpuinfo
Remove all the verifications of the OBP version string. Just
make sure it's there, and report it raw in the bootup logs and
via /proc/cpuinfo.
Signed-off-by: David S. Miller <davem@davemloft.net>
The mapping is a simple "(cpuid >> 2) == core" for now.
Later we'll add more sophisticated code that will walk
the sun4v machine description and figure this out from
there.
We should also add core mappings for jaguar and panther
processors.
Signed-off-by: David S. Miller <davem@davemloft.net>
This has been pending for a long time, and the fact
that we waste a ton of ram on some configurations
kind of pushed things over the edge.
Signed-off-by: David S. Miller <davem@davemloft.net>
Don't piggy back the SMP receive signal code to do the
context version change handling.
Instead allocate another fixed PIL number for this
asynchronous cross-call. We can't use smp_call_function()
because this thing is invoked with interrupts disabled
and a few spinlocks held.
Also, fix smp_call_function_mask() to count "cpus" correctly.
There is no guarentee that the local cpu is in the mask
yet that is exactly what this code was assuming.
Signed-off-by: David S. Miller <davem@davemloft.net>
1) Always spin_lock_init() in init_context(). The caller essentially
clears it out, or copies the mm info from the parent. In both
cases we need to explicitly initialize the spinlock.
2) Always do explicit IRQ disabling while taking mm->context.lock
and ctx_alloc_lock.
Signed-off-by: David S. Miller <davem@davemloft.net>
The UltraSPARC T1 manual recommends this because the chip
could instruction prefetch into the VA hole, and this would
also make decoding certain kinds of memory access traps
more difficult (because the chip sign extends certain pieces
of trap state).
Signed-off-by: David S. Miller <davem@davemloft.net>
There were several bugs in the SUN4V cpu mondo dispatch code.
In fact, if we ever got a EWOULDBLOCK or other error from
the hypervisor call, we'd potentially send a cpu mondo multiple
times to the same cpu and even worse we could loop until the
timeout resending the same mondo over and over to such cpus.
So let's bulletproof this thing as follows:
1) Implement cpu_mondo_send() and cpu_state() hypervisor calls
in arch/sparc64/kernel/entry.S, add prototypes to asm/hypervisor.h
2) Don't build and update the cpulist using inline functions, this
was causing the cpu mask to not get updated in the caller.
3) Disable interrupts during the entire mondo send, otherwise our
cpu list and/or mondo block could get overwritten if we take
an interrupt and do a cpu mondo send on the current cpu.
4) Check for all possible error return types from the cpu_mondo_send()
hypervisor call. In particular:
HV_EOK) Our work is done, all cpus have received the mondo.
HV_CPUERROR) One or more of the cpus in the cpu list we passed
to the hypervisor are in error state. Use cpu_state()
calls over the entries in the cpu list to see which
ones. Record them in "error_mask" and report this
after we are done sending the mondo to cpus which are
not in error state.
HV_EWOULDBLOCK) We need to keep trying.
Any other error we consider fatal, we report the event and exit
immediately.
5) We only timeout if forward progress is not made. Forward progress
is defined as having at least one cpu get the mondo successfully
in a given cpu_mondo_send() call. Otherwise we bump a counter
and delay a little. If the counter hits a limit, we signal an
error and report the event.
Also, smp_call_function_mask() error handling reports the number
of cpus incorrectly.
Signed-off-by: David S. Miller <davem@davemloft.net>
The context allocation scheme we use depends upon there being a 1<-->1
mapping from cpu to physical TLB for correctness. Chips like Niagara
break this assumption.
So what we do is notify all cpus with a cross call when the context
version number changes, and if necessary this makes them allocate
a valid context for the address space they are running at the time.
Stress tested with make -j1024, make -j2048, and make -j4096 kernel
builds on a 32-strand, 8 core, T2000 with 16GB of ram.
Signed-off-by: David S. Miller <davem@davemloft.net>
Niagara helps us find a ancient bug in the sparc64 port :-)
The ASI_* values are plain constant defines, thus signed 32-bit
on sparc64. To put shift this into the regs->tstate value we were
doing or'ing "(ASI_PNF << 24)" into there.
ASI_PNF is 0x82 and shifted left by 24 makes that topmost bit the
sign bit in a 32-bit value. This would get sign extended to 64-bits
and thus corrupt the top-half of the reg->tstate value.
This never caused problems in pre-Niagara cpus because the only thing
up there were the condition code values. But Niagara has the global
register level field, and this all 1's value is illegal there so
Niagara gives an illegal instruction trap due to this bug.
I'm pretty sure this bug is about as old as the sparc64 port itself.
This also points out that we weren't setting ASI_PNF for 32-bit tasks.
We should, so fix that while we're here.
Signed-off-by: David S. Miller <davem@davemloft.net>
It can map all of the linear kernel mappings with zero TSB hash
conflicts for systems with 16GB or less ram. In such cases, on
SUN4V, once we load up this TSB the first time with all the
mappings, we never take a linear kernel mapping TLB miss ever
again, the hypervisor handles them all.
Signed-off-by: David S. Miller <davem@davemloft.net>
Set, but never used.
We used to use this for dynamic IRQ retargetting, but that
code died a long time ago.
Signed-off-by: David S. Miller <davem@davemloft.net>
The SUN4V convention with non-shared TSBs is that the context
bit of the TAG is clear. So we have to choose an "invalid"
bit and initialize new TSBs appropriately. Otherwise a zero
TAG looks "valid".
Make sure, for the window fixup cases, that we use the right
global registers and that we don't potentially trample on
the live global registers in etrap/rtrap handling (%g2 and
%g6) and that we put the missing virtual address properly
in %g5.
Signed-off-by: David S. Miller <davem@davemloft.net>
This gives more consistent bogomips and delay() semantics,
especially on sun4v. It gives weird looking values though...
Signed-off-by: David S. Miller <davem@davemloft.net>
Chuck Lever