This is somewhat modelled after the powerpc version, and differs from
the legacy fallback in use fls64 instead of pointlessly splitting up the
address into low and high dwords and in that it takes (__)phys_to_dma
into account.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Reviewed-by: Robin Murphy <robin.murphy@arm.com>
This save some duplication for ia64, and makes the interface more
general. In the long run we want each dma_map_ops instance to fill this
out, but this will take a little more prep work.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
This commit introduced per-cpu cgroup local storage.
Per-cpu cgroup local storage is very similar to simple cgroup storage
(let's call it shared), except all the data is per-cpu.
The main goal of per-cpu variant is to implement super fast
counters (e.g. packet counters), which don't require neither
lookups, neither atomic operations.
>From userspace's point of view, accessing a per-cpu cgroup storage
is similar to other per-cpu map types (e.g. per-cpu hashmaps and
arrays).
Writing to a per-cpu cgroup storage is not atomic, but is performed
by copying longs, so some minimal atomicity is here, exactly
as with other per-cpu maps.
Signed-off-by: Roman Gushchin <guro@fb.com>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: Alexei Starovoitov <ast@kernel.org>
Acked-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
To simplify the following introduction of per-cpu cgroup storage,
let's rework a bit a mechanism of passing a pointer to a cgroup
storage into the bpf_get_local_storage(). Let's save a pointer
to the corresponding bpf_cgroup_storage structure, instead of
a pointer to the actual buffer.
It will help us to handle per-cpu storage later, which has
a different way of accessing to the actual data.
Signed-off-by: Roman Gushchin <guro@fb.com>
Acked-by: Song Liu <songliubraving@fb.com>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
In order to introduce per-cpu cgroup storage, let's generalize
bpf cgroup core to support multiple cgroup storage types.
Potentially, per-node cgroup storage can be added later.
This commit is mostly a formal change that replaces
cgroup_storage pointer with a array of cgroup_storage pointers.
It doesn't actually introduce a new storage type,
it will be done later.
Each bpf program is now able to have one cgroup storage of each type.
Signed-off-by: Roman Gushchin <guro@fb.com>
Acked-by: Song Liu <songliubraving@fb.com>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Add support for handling 52bit guest physical address to the
VGIC layer. So far we have limited the guest physical address
to 48bits, by explicitly masking the upper bits. This patch
removes the restriction. We do not have to check if the host
supports 52bit as the gpa is always validated during an access.
(e.g, kvm_{read/write}_guest, kvm_is_visible_gfn()).
Also, the ITS table save-restore is also not affected with
the enhancement. The DTE entries already store the bits[51:8]
of the ITT_addr (with a 256byte alignment).
Cc: Marc Zyngier <marc.zyngier@arm.com>
Cc: Christoffer Dall <cdall@kernel.org>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Kristina Martsenko <kristina.martsenko@arm.com>
[ Macro clean ups, fix PROPBASER and PENDBASER accesses ]
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
With the flush queue infrastructure already abstracted into IOVA
domains, hooking it up in iommu-dma is pretty simple. Since there is a
degree of dependency on the IOMMU driver knowing what to do to play
along, we key the whole thing off a domain attribute which will be set
on default DMA ops domains to request non-strict invalidation. That way,
drivers can indicate the appropriate support by acknowledging the
attribute, and we can easily fall back to strict invalidation otherwise.
The flush queue callback needs a handle on the iommu_domain which owns
our cookie, so we have to add a pointer back to that, but neatly, that's
also sufficient to indicate whether we're using a flush queue or not,
and thus which way to release IOVAs. The only slight subtlety is
switching __iommu_dma_unmap() from calling iommu_unmap() to explicit
iommu_unmap_fast()/iommu_tlb_sync() so that we can elide the sync
entirely in non-strict mode.
Signed-off-by: Zhen Lei <thunder.leizhen@huawei.com>
[rm: convert to domain attribute, tweak comments and commit message]
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
This is the 4.19-rc6 release
I needed to merge this in because of extensive conflicts in
the MSM and Intel pin control drivers. I know how to resolve
them, so let's do it like this.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
The sigaltstack(2) system call fails with -ENOMEM if the new alternative
signal stack is found to be smaller than SIGMINSTKSZ. On architectures
such as arm64, where the native value for SIGMINSTKSZ is larger than
the compat value, this can result in an unexpected error being reported
to a compat task. See, for example:
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=904385
This patch fixes the problem by extending do_sigaltstack to take the
minimum signal stack size as an additional parameter, allowing the
native and compat system call entry code to pass in their respective
values. COMPAT_SIGMINSTKSZ is just defined as SIGMINSTKSZ if it has not
been defined by the architecture.
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Dominik Brodowski <linux@dominikbrodowski.net>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Oleg Nesterov <oleg@redhat.com>
Reported-by: Steve McIntyre <steve.mcintyre@arm.com>
Tested-by: Steve McIntyre <93sam@debian.org>
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Pull operating performance points (OPP) material for 4.20 from Viresh Kumar.
"This contains patches that fix several bugs in the OPP core and
makes it more stable."
* 'opp/linux-next' of git://git.kernel.org/pub/scm/linux/kernel/git/vireshk/pm:
OPP: Pass OPP table to _of_add_opp_table_v{1|2}()
OPP: Prevent creating multiple OPP tables for devices sharing OPP nodes
OPP: Use a single mechanism to free the OPP table
OPP: Don't remove dynamic OPPs from _dev_pm_opp_remove_table()
cpufreq: mvebu: Remove OPPs using dev_pm_opp_remove()
OPP: Create separate kref for static OPPs list
OPP: Don't take OPP table's kref for static OPPs
OPP: Parse OPP table's DT properties from _of_init_opp_table()
OPP: Pass index to _of_init_opp_table()
OPP: Protect dev_list with opp_table lock
OPP: Don't try to remove all OPP tables on failure
OPP: Free OPP table properly on performance state irregularities
Internal helper function gpiod_set_array_value_complex() was changed to
return an error value, but not all gpiolib callers were updated to
propagate the new error up.
Fixes: 3027743f83 ("gpio: Remove VLA from gpiolib")
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Replace FUSE_MAX_PAGES_PER_REQ with the configurable parameter max_pages to
improve performance.
Old RFC with detailed description of the problem and many fixes by Mitsuo
Hayasaka (mitsuo.hayasaka.hu@hitachi.com):
- https://lkml.org/lkml/2012/7/5/136
We've encountered performance degradation and fixed it on a big and complex
virtual environment.
Environment to reproduce degradation and improvement:
1. Add lag to user mode FUSE
Add nanosleep(&(struct timespec){ 0, 1000 }, NULL); to xmp_write_buf in
passthrough_fh.c
2. patch UM fuse with configurable max_pages parameter. The patch will be
provided latter.
3. run test script and perform test on tmpfs
fuse_test()
{
cd /tmp
mkdir -p fusemnt
passthrough_fh -o max_pages=$1 /tmp/fusemnt
grep fuse /proc/self/mounts
dd conv=fdatasync oflag=dsync if=/dev/zero of=fusemnt/tmp/tmp \
count=1K bs=1M 2>&1 | grep -v records
rm fusemnt/tmp/tmp
killall passthrough_fh
}
Test results:
passthrough_fh /tmp/fusemnt fuse.passthrough_fh \
rw,nosuid,nodev,relatime,user_id=0,group_id=0 0 0
1073741824 bytes (1.1 GB) copied, 1.73867 s, 618 MB/s
passthrough_fh /tmp/fusemnt fuse.passthrough_fh \
rw,nosuid,nodev,relatime,user_id=0,group_id=0,max_pages=256 0 0
1073741824 bytes (1.1 GB) copied, 1.15643 s, 928 MB/s
Obviously with bigger lag the difference between 'before' and 'after'
will be more significant.
Mitsuo Hayasaka, in 2012 (https://lkml.org/lkml/2012/7/5/136),
observed improvement from 400-550 to 520-740.
Signed-off-by: Constantine Shulyupin <const@MakeLinux.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Current implementation allows each qp to have only one send engine. As
such, each qp has only one list to queue prebuilt packets when send engine
resources are not available. To improve performance, it is desired to
support multiple send engines for each qp.
This patch creates the framework to support two send engines
(two legs) for each qp for the TID RDMA protocol, which can be easily
extended to support more send engines. It achieves the goal by creating a
leg specific struct, iowait_work in the iowait struct, to hold the
work_struct and the tx_list as well as a pointer to the parent iowait
struct.
The hfi1_pkt_state now has an additional field to record the current legs
work structure and that is now passed to all egress waiters to determine
the leg that needs to wait via a new iowait helper. The APIs are adjusted
to use the new leg specific struct as required.
Many new and modified helpers are added to support this change.
Reviewed-by: Mitko Haralanov <mitko.haralanov@intel.com>
Signed-off-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Signed-off-by: Kaike Wan <kaike.wan@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
The driver-provided function check_send_wqe allows the hardware driver to
check and set up the incoming send wqe before it is inserted into the swqe
ring. This patch will rename it as setup_wqe to better reflect its
usage. In addition, this function is only called when all setup is
complete in rdmavt.
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Signed-off-by: Kaike Wan <kaike.wan@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Clang warns when one enumerated type is explicitly converted to another.
drivers/infiniband/hw/qedr/qedr_roce_cm.c:198:28: warning: implicit
conversion from enumeration type 'enum qed_roce_ll2_tx_dest' to
different enumeration type 'enum qed_ll2_tx_dest' [-Wenum-conversion]
ll2_tx_pkt.tx_dest = pkt->tx_dest;
~ ~~~~~^~~~~~~
1 warning generated.
Turns out that QED_ROCE_LL2_TX_DEST_NW and QED_ROCE_LL2_TX_DEST_LB are
only used once in the whole tree and QED_ROCE_LL2_TX_DEST_MAX is used
nowhere. Remove them and use the equivalent values from qed_ll2_tx_dest
in their place.
Reported-by: Nick Desaulniers <ndesaulniers@google.com>
Signed-off-by: Nathan Chancellor <natechancellor@gmail.com>
Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
Acked-by: Michal Kalderon <michal.kalderon@cavium.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Now that each struct nfs_pgio_header corresponds to one RPC call, we
only have one writer to the struct nfs_pgio_header.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Save a few bytes by allowing the read/write specific fields of the
structures to share storage.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
When the server fails to return post-op attributes, the client's
attempt to place read data directly in the page cache fails, and
so we have to do an extra copy in order to realign the data with
page borders.
This patch attempts to detect servers that don't return post-op
attributes on read (e.g. for pNFS) and adjusts the placement
calculation accordingly.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Reduce contention on the inode->i_lock by ensuring that we use RCU
when looking up the NFS open context.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Speed up lookups of an existing lock context by avoiding the inode->i_lock,
and using RCU instead.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Add a bvec array to struct xdr_buf, and have the client allocate it
when we need to receive data into pages.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
If the RPC call relies on the receive call allocating pages as buffers,
then let's label it so that we
a) Don't leak memory by allocating pages for requests that do not expect
this behaviour
b) Can optimise for the common case where calls do not require allocation.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Fix up the priority queue to not batch by owner, but by queue, so that
we allow '1 << priority' elements to be dequeued before switching to
the next priority queue.
The owner field is still used to wake up requests in round robin order
by owner to avoid single processes hogging the RPC layer by loading the
queues.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
If the server is slow, we can find ourselves with quite a lot of entries
on the receive queue. Converting the search from an O(n) to O(log(n))
can make a significant difference, particularly since we have to hold
a number of locks while searching.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Treat socket write space handling in the same way we now treat transport
congestion: by denying the XPRT_LOCK until the transport signals that it
has free buffer space.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
The theory was that we would need to grab the socket lock anyway, so we
might as well use it to gate the allocation of RPC slots for a TCP
socket.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Both RDMA and UDP transports require the request to get a "congestion control"
credit before they can be transmitted. Right now, this is done when
the request locks the socket. We'd like it to happen when a request attempts
to be transmitted for the first time.
In order to support retransmission of requests that already hold such
credits, we also want to ensure that they get queued first, so that we
don't deadlock with requests that have yet to obtain a credit.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
One of the intentions with the priority queues was to ensure that no
single process can hog the transport. The field task->tk_owner therefore
identifies the RPC call's origin, and is intended to allow the RPC layer
to organise queues for fairness.
This commit therefore modifies the transmit queue to group requests
by task->tk_owner, and ensures that we round robin among those groups.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
When we shift to using the transmit queue, then the task that holds the
write lock will not necessarily be the same as the one being transmitted.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Fix up the back channel code to recognise that it has already been
transmitted, so does not need to be called again.
Also ensure that we set req->rq_task.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
When storing a struct rpc_rqst on the slot allocation list, we currently
use the same field 'rq_list' as we use to store the request on the
receive queue. Since the structure is never on both lists at the same
time, this is OK.
However, for clarity, let's make that a union with different names for
the different lists so that we can more easily distinguish between
the two states.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Allow the caller in clnt.c to call into the code to wait for a reply
after calling xprt_transmit(). Again, the reason is that the backchannel
code does not need this functionality.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Separate out the action of adding a request to the reply queue so that the
backchannel code can simply skip calling it altogether.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Add a helper that will wake up a task that is sleeping on a specific
queue, and will set the value of task->tk_status. This is mainly
intended for use by the transport layer to notify the task of an
error condition.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Since we will want to introduce similar TCP state variables for the
transmission of requests, let's rename the existing ones to label
that they are for the receive side.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
If a message has been encoded using RPCSEC_GSS, the server is
maintaining a window of sequence numbers that it considers valid.
The client should normally be tracking that window, and needs to
verify that the sequence number used by the message being transmitted
still lies inside the window of validity.
So far, we've been able to assume this condition would be realised
automatically, since the client has been encoding the message only
after taking the socket lock. Once we change that condition, we
will need the explicit check.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
