It'd be nice to hide details of how rsrc nodes are used by a request
from rsrc.c, specifically which request fields store them, and what bits
are signifying if there is a node in a request. It rather belong to
generic request handling, so move the helper to io_uring.c. While doing
so remove clearing of ->buf_node as it's controlled by REQ_F_BUF_NODE
and doesn't require zeroing.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/bb73fb42baf825edb39344365aff48cdfdd4c692.1746533789.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
dmabuf backed area will be taking an offset instead of addresses, and
io_buffer_validate() is not flexible enough to facilitate it. It also
takes an iovec, which may truncate the u64 length zcrx takes. Add a new
helper function for validation.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/0b3b735391a0a8f8971bf0121c19765131fddd3b.1746097431.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
There are two helpers here, one assigns and increments the node ref
count, and the other is simply a wrapper around that for the buffer node
handling.
The buffer node assignment benefits from checking and setting
REQ_F_BUF_NODE together, otherwise stalls have been observed on setting
that flag later in the process. Hence re-do it so that it's set when
checked, and cleared in case of (unlikely) failure. With that, the
buffer node helper can go, and then drop the generic
io_req_assign_rsrc_node() helper as well as there's only a single user
of it left at that point.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
All vectored reg buffer users should use io_import_reg_vec() for iovec
imports, since iovec placement is the function's responsibility and
callers shouldn't know much about it, drop the offset parameter from
io_prep_reg_vec() and calculate it inside.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/08ed87ca4bbc06724373b6ce06f36b703fe60c4e.1741457480.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Add io_import_reg_vec(), which will be responsible for importing
vectored registered buffers. The function might reallocate the vector,
but it'd try to do the conversion in place first, which is why it's
required of the user to pad the iovec to the right border of the cache.
Overlapping also depends on struct iovec being larger than bvec, which
is not the case on e.g. 32 bit architectures. Don't try to complicate
this case and make sure vectors never overlap, it'll be improved later.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/60bd246b1249476a6996407c1dbc38ef6febad14.1741362889.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
io_uring/rsrc.h uses several types from include/linux/io_uring_types.h.
Include io_uring_types.h explicitly in rsrc.h to avoid depending on
users of rsrc.h including io_uring_types.h first.
Signed-off-by: Caleb Sander Mateos <csander@purestorage.com>
Reviewed-by: Li Zetao <lizetao1@huawei.com>
Link: https://lore.kernel.org/r/20250301183612.937529-1-csander@purestorage.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Declare io_find_buf_node() in io_uring/rsrc.h so it can be called from
other files.
Signed-off-by: Caleb Sander Mateos <csander@purestorage.com>
Link: https://lore.kernel.org/r/20250301001610.678223-1-csander@purestorage.com
[axboe: keep the inline for local hot path usage]
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Frequent alloc/free cycles on these is pretty costly. Use an io cache to
more efficiently reuse these buffers.
Signed-off-by: Keith Busch <kbusch@kernel.org>
Link: https://lore.kernel.org/r/20250227223916.143006-7-kbusch@meta.com
[axboe: fix imu leak]
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Provide an interface for the kernel to leverage the existing
pre-registered buffers that io_uring provides. User space can reference
these later to achieve zero-copy IO.
User space must register an empty fixed buffer table with io_uring in
order for the kernel to make use of it.
Signed-off-by: Keith Busch <kbusch@kernel.org>
Link: https://lore.kernel.org/r/20250227223916.143006-5-kbusch@meta.com
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Merge mainline fixes into 6.15 branch, as upcoming patches depend on
fixes that went into the 6.14 mainline branch.
* io_uring-6.14:
io_uring/net: save msg_control for compat
io_uring/rw: clean up mshot forced sync mode
io_uring/rw: move ki_complete init into prep
io_uring/rw: don't directly use ki_complete
io_uring/rw: forbid multishot async reads
io_uring/rsrc: remove unused constants
io_uring: fix spelling error in uapi io_uring.h
io_uring: prevent opcode speculation
io-wq: backoff when retrying worker creation
Registered buffer are currently imported in two steps, first we lookup
a rsrc node and then use it to set up the iterator. The first part is
usually done at the prep stage, and import happens whenever it's needed.
As we want to defer binding to a node so that it works with linked
requests, combine both steps into a single helper.
Reviewed-by: Keith Busch <kbusch@kernel.org>
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Link: https://lore.kernel.org/r/20250224213116.3509093-6-kbusch@meta.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
IO_NODE_ALLOC_CACHE_MAX has been unused since commit fbbb8e991d
("io_uring/rsrc: get rid of io_rsrc_node allocation cache") removed the
rsrc_node_cache.
IO_RSRC_TAG_TABLE_SHIFT and IO_RSRC_TAG_TABLE_MASK have been unused
since commit 7029acd8a9 ("io_uring/rsrc: get rid of per-ring
io_rsrc_node list") removed the separate tag table for registered nodes.
Signed-off-by: Caleb Sander Mateos <csander@purestorage.com>
Reviewed-by: Li Zetao <lizetao1@huawei.com>
Link: https://lore.kernel.org/r/20250219033444.2020136-1-csander@purestorage.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Add io_zcrx_area that represents a region of userspace memory that is
used for zero copy. During ifq registration, userspace passes in the
uaddr and len of userspace memory, which is then pinned by the kernel.
Each net_iov is mapped to one of these pages.
The freelist is a spinlock protected list that keeps track of all the
net_iovs/pages that aren't used.
For now, there is only one area per ifq and area registration happens
implicitly as part of ifq registration. There is no API for
adding/removing areas yet. The struct for area registration is there for
future extensibility once we support multiple areas and TCP devmem.
Reviewed-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: David Wei <dw@davidwei.uk>
Acked-by: Jakub Kicinski <kuba@kernel.org>
Link: https://lore.kernel.org/r/20250215000947.789731-3-dw@davidwei.uk
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Most callers of io_put_rsrc_node() already check that node is non-NULL:
- io_rsrc_data_free()
- io_sqe_buffer_register()
- io_reset_rsrc_node()
- io_req_put_rsrc_nodes() (REQ_F_BUF_NODE indicates non-NULL buf_node)
Only io_splice_cleanup() can call io_put_rsrc_node() with a NULL node.
So move the NULL check there.
Signed-off-by: Caleb Sander Mateos <csander@purestorage.com>
Link: https://lore.kernel.org/r/20250216225900.1075446-1-csander@purestorage.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Checking for lockdep_assert_held(&ctx->uring_lock) in io_free_rsrc_node()
means that the assertion is only checked when the resource drops to zero
references.
Move the lockdep assertion up into the caller io_put_rsrc_node() so that it
instead happens on every reference count decrement.
Signed-off-by: Jann Horn <jannh@google.com>
Link: https://lore.kernel.org/r/20250120-uring-lockdep-assert-earlier-v1-1-68d8e071a4bb@google.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
io_uring_ctx parameter for io_rsrc_node_alloc() is unused for now.
This patch removes the parameter and fixes the callers accordingly.
Signed-off-by: Sidong Yang <sidong.yang@furiosa.ai>
Link: https://lore.kernel.org/r/20250115142033.658599-1-sidong.yang@furiosa.ai
Signed-off-by: Jens Axboe <axboe@kernel.dk>
io_try_coalesce_buffer() is a useful helper collecting useful info about
a set of pages, I want to reuse it for analysing ring/etc. mappings. I
don't need the entire thing and only interested if it can be coalesced
into a single page, but that's better than duplicating the parsing.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/353b447953cd5d34c454a7d909bb6024c391d6e2.1732886067.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The following pattern becomes more and more:
+ io_req_assign_rsrc_node(&req->buf_node, node);
+ req->flags |= REQ_F_BUF_NODE;
so make it a helper, which is less fragile to use than above code, for
example, the BUF_NODE flag is even missed in current io_uring_cmd_prep().
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Link: https://lore.kernel.org/r/20241107110149.890530-4-ming.lei@redhat.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
`io_rsrc_node` instance won't be shared among different io_uring ctxs,
and its allocation 'ctx' is always same with the user's 'ctx', so it is
safe to pass user 'ctx' reference to rsrc helpers. Even in io_clone_buffers(),
`io_rsrc_node` instance is allocated actually for destination io_uring_ctx.
Then io_rsrc_node_ctx() can be removed, and the 8 bytes `ctx` pointer will be
removed from `io_rsrc_node` in the following patch.
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Link: https://lore.kernel.org/r/20241107110149.890530-2-ming.lei@redhat.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Currently the io_rsrc_node assignment in io_kiocb is an array of two
pointers, as two nodes may be assigned to a request - one file node,
and one buffer node. However, the buffer node can co-exist with the
provided buffers, as currently it's not supported to use both provided
and registered buffers at the same time.
This crucially brings struct io_kiocb down to 4 cache lines again, as
before it spilled into the 5th cacheline.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Rather than keep the type field separate rom ctx, use the fact that we
can encode up to 4 types of nodes in the LSB of the ctx pointer. Doesn't
reclaim any space right now on 64-bit archs, but it leaves a full int
for future use.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The empty node was used as a placeholder for a sparse entry, but it
didn't really solve any issues. The caller still has to check for
whether it's the empty node or not, it may as well just check for a NULL
return instead.
The dummy_ubuf was used for a sparse buffer entry, but NULL will serve
the same purpose there of ensuring an -EFAULT on attempted import.
Just use NULL for a sparse node, regardless of whether or not it's a
file or buffer resource.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Puts and reset an existing node in a slot, if one exists. Returns true
if a node was there, false if not. This helps cleanup some of the code
that does a lookup just to clear an existing node.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
There are lots of spots open-coding this functionality, add a generic
helper that does the node lookup in a speculation safe way.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
For files, there's nr_user_files/file_table/file_data, and buffers have
nr_user_bufs/user_bufs/buf_data. There's no reason why file_table and
file_data can't be the same thing, and ditto for the buffer side. That
gets rid of more io_ring_ctx state that's in two spots rather than just
being in one spot, as it should be. Put all the registered file data in
one locations, and ditto on the buffer front.
This also avoids having both io_rsrc_data->nodes being an allocated
array, and ->user_bufs[] or ->file_table.nodes. There's no reason to
have this information duplicated. Keep it in one spot, io_rsrc_data,
along with how many resources are available.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Add the empty node initializing to the preinit part of the io_kiocb
allocation, and reset them if they have been used.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Rather than allocate an io_rsrc_node for an empty/sparse buffer entry,
add a const entry that can be used for that. This just needs checking
for writing the tag, and the put check needs to check for that sparse
node rather than NULL for validity.
This avoids allocating rsrc nodes for sparse buffer entries.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Work in progress, but get rid of the per-ring serialization of resource
nodes, like registered buffers and files. Main issue here is that one
node can otherwise hold up a bunch of other nodes from getting freed,
which is especially a problem for file resource nodes and networked
workloads where some descriptors may not see activity in a long time.
As an example, instantiate an io_uring ring fd and create a sparse
registered file table. Even 2 will do. Then create a socket and register
it as fixed file 0, F0. The number of open files in the app is now 5,
with 0/1/2 being the usual stdin/out/err, 3 being the ring fd, and 4
being the socket. Register this socket (eg "the listener") in slot 0 of
the registered file table. Now add an operation on the socket that uses
slot 0. Finally, loop N times, where each loop creates a new socket,
registers said socket as a file, then unregisters the socket, and
finally closes the socket. This is roughly similar to what a basic
accept loop would look like.
At the end of this loop, it's not unreasonable to expect that there
would still be 5 open files. Each socket created and registered in the
loop is also unregistered and closed. But since the listener socket
registered first still has references to its resource node due to still
being active, each subsequent socket unregistration is stuck behind it
for reclaim. Hence 5 + N files are still open at that point, where N is
awaiting the final put held up by the listener socket.
Rewrite the io_rsrc_node handling to NOT rely on serialization. Struct
io_kiocb now gets explicit resource nodes assigned, with each holding a
reference to the parent node. A parent node is either of type FILE or
BUFFER, which are the two types of nodes that exist. A request can have
two nodes assigned, if it's using both registered files and buffers.
Since request issue and task_work completion is both under the ring
private lock, no atomics are needed to handle these references. It's a
simple unlocked inc/dec. As before, the registered buffer or file table
each hold a reference as well to the registered nodes. Final put of the
node will remove the node and free the underlying resource, eg unmap the
buffer or put the file.
Outside of removing the stall in resource reclaim described above, it
has the following advantages:
1) It's a lot simpler than the previous scheme, and easier to follow.
No need to specific quiesce handling anymore.
2) There are no resource node allocations in the fast path, all of that
happens at resource registration time.
3) The structs related to resource handling can all get simplified
quite a bit, like io_rsrc_node and io_rsrc_data. io_rsrc_put can
go away completely.
4) Handling of resource tags is much simpler, and doesn't require
persistent storage as it can simply get assigned up front at
registration time. Just copy them in one-by-one at registration time
and assign to the resource node.
The only real downside is that a request is now explicitly limited to
pinning 2 resources, one file and one buffer, where before just
assigning a resource node to a request would pin all of them. The upside
is that it's easier to follow now, as an individual resource is
explicitly referenced and assigned to the request.
With this in place, the above mentioned example will be using exactly 5
files at the end of the loop, not N.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
It's only used from __io_req_set_rsrc_node(), and it takes both the ctx
and node itself, while never using the ctx. Just open-code the basic
refs++ in __io_req_set_rsrc_node() instead.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
All callers already hold the ring lock and hence are passing '0',
remove the argument and the conditional locking that it controlled.
Suggested-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
If we change it to tracking ubuf->start + ubuf->len, then we can reduce
the size of struct io_mapped_ubuf by another 4 bytes, effectively 8
bytes, as a hole is eliminated too.
This shrinks io_mapped_ubuf to 32 bytes.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We don't really need to cache this, let's reclaim 8 bytes from struct
io_mapped_ubuf and just calculate it when we need it. The only hot path
here is io_import_fixed().
Signed-off-by: Jens Axboe <axboe@kernel.dk>
A recent commit added support for copying registered buffers from one
ring to another. But that term is a bit confusing, as no copying of
buffer data is done here. What is being done is simply cloning the
buffer registrations from one ring to another.
Rename it while we still can, so that it's more descriptive. No
functional changes in this patch.
Fixes: 7cc2a6eadc ("io_uring: add IORING_REGISTER_COPY_BUFFERS method")
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Buffers can get registered with io_uring, which allows to skip the
repeated pin_pages, unpin/unref pages for each O_DIRECT operation. This
reduces the overhead of O_DIRECT IO.
However, registrering buffers can take some time. Normally this isn't an
issue as it's done at initialization time (and hence less critical), but
for cases where rings can be created and destroyed as part of an IO
thread pool, registering the same buffers for multiple rings become a
more time sensitive proposition. As an example, let's say an application
has an IO memory pool of 500G. Initial registration takes:
Got 500 huge pages (each 1024MB)
Registered 500 pages in 409 msec
or about 0.4 seconds. If we go higher to 900 1GB huge pages being
registered:
Registered 900 pages in 738 msec
which is, as expected, a fully linear scaling.
Rather than have each ring pin/map/register the same buffer pool,
provide an io_uring_register(2) opcode to simply duplicate the buffers
that are registered with another ring. Adding the same 900GB of
registered buffers to the target ring can then be accomplished in:
Copied 900 pages in 17 usec
While timing differs a bit, this provides around a 25,000-40,000x
speedup for this use case.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Currently there's a single ring owner of a mapped buffer, and hence the
reference count will always be 1 when it's torn down and freed. However,
in preparation for being able to link io_mapped_ubuf to different spots,
add a reference count to manage the lifetime of it.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Add support for checking and coalescing multi-hugepage-backed fixed
buffers. The coalescing optimizes both time and space consumption caused
by mapping and storing multi-hugepage fixed buffers.
A coalescable multi-hugepage buffer should fully cover its folios
(except potentially the first and last one), and these folios should
have the same size. These requirements are for easier processing later,
also we need same size'd chunks in io_import_fixed for fast iov_iter
adjust.
Signed-off-by: Chenliang Li <cliang01.li@samsung.com>
Reviewed-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/20240731090133.4106-3-cliang01.li@samsung.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Store the folio shift and folio mask into imu struct and use it in
iov_iter adjust, as we will have non PAGE_SIZE'd chunks if a
multi-hugepage buffer get coalesced.
Signed-off-by: Chenliang Li <cliang01.li@samsung.com>
Reviewed-by: Anuj Gupta <anuj20.g@samsung.com>
Reviewed-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/20240731090133.4106-2-cliang01.li@samsung.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
io_req_put_rsrc_locked() is a weird shim function around
io_req_put_rsrc(). All calls to io_req_put_rsrc() require holding
->uring_lock, so we can just use it directly.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/a195bc78ac3d2c6fbaea72976e982fe51e50ecdd.1712331455.git.asml.silence@gmail.com
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Currently lists are being used to manage this, but best practice is
usually to have these in an array instead as that it cheaper to manage.
Outside of that detail, games are also played with KASAN as the list
is inside the cached entry itself.
Finally, all users of this need a struct io_cache_entry embedded in
their struct, which is union'ized with something else in there that
isn't used across the free -> realloc cycle.
Get rid of all of that, and simply have it be an array. This will not
change the memory used, as we're just trading an 8-byte member entry
for the per-elem array size.
This reduces the overhead of the recycled allocations, and it reduces
the amount of code code needed to support recycling to about half of
what it currently is.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Changes to AF_UNIX trigger rebuild of io_uring, but io_uring does
not use AF_UNIX anymore.
Let's not include af_unix.h and instead include necessary headers.
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://lore.kernel.org/r/20240212234236.63714-1-kuniyu@amazon.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
For the normal read/write path, we have already locked the ring
submission side when assigning the file. This causes branch
mispredictions when we then check and try and lock again in
io_req_set_rsrc_node(). As this is a very hot path, this matters.
Add a basic helper that already assumes we already have it locked,
and use that in io_file_get_fixed().
Signed-off-by: Jens Axboe <axboe@kernel.dk>
