SQE items with this OP will take the specified amount of time
(asynchronously) before completing. This allows to serve as an
asynchronous delay in between SQE items (e.g: A sensor measurement
requested, which requires 50-ms before having the result available).
Signed-off-by: Luis Ubieda <luisf@croxel.com>
Add an assert ensuring programming errors where the executor code
assumes there is a next node in the list for transaction/chained
submissions.
Signed-off-by: Tom Burdick <thomas.burdick@intel.com>
As to facilitate work items preemption, common for sensor read ops.
e.g:
- Sensor read (submitted through workqueue).
- SPI read (submitted through workqueue using default handler).
This requires at least 2 thread pools to allow SPI read to complete,
to then come back and complete sensor reading.
Signed-off-by: Luis Ubieda <luisf@croxel.com>
Otherwise the RTIO Workqueue is not functional for devices during init.
An example of this issue is devices using SPI transfers with default
spi_rtio.
Signed-off-by: Luis Ubieda <luisf@croxel.com>
This commit fixes the bug that the memory of the work request is
freed up in the work handler, before it is not needed anymore by the p4wq.
This is fixed now, by using the new done_handler in the p4wq
for freeing up that memory.
Signed-off-by: Florian Weber <Florian.Weber@live.de>
Transmit/write buffers are expected to be constant values given to the
operation to transmit but not mutate. Seperate the OP_TX and OP_RX
operation description unions.
Signed-off-by: Tom Burdick <thomas.burdick@intel.com>
The default behavior for thread pending of completions should use
semaphores rather than yield/wait looping when multithreading is
available.
Disable by default only when multithreading isn't available.
Signed-off-by: Tom Burdick <thomas.burdick@intel.com>
As the items come from the memory slab, their initialization state
can't be guaranteed. This is causing some devices triggering an assert
in p4wq where the item's thread is not null (not zero).
Signed-off-by: Luis Ubieda <luisf@croxel.com>
Adds ability to process synchronous requests in an asynchronous
fashion, relying on dedicated P4WQ's and pre-allocated work items.
Signed-off-by: Luis Ubieda <luisf@croxel.com>
Moves the rtio_ prefixed lockfree queues to sys alongside existing
mpsc/spsc pbuf, ringbuf, and similar queue-like data structures.
Signed-off-by: Tom Burdick <thomas.burdick@intel.com>
Callbacks were a bit neglected in terms of test coverage, especially
when used in chains. It was clear from the code that chained callbacks
may not actually work, and callback ordering then was hard to verify.
Test callbacks chained to transactions work as expected.
The test iodev had built up some cruft over time and in the process
showed a few bugs once callback chaining was fixed so the test iodev now
better matches typical iodev implementations at this point.
Cancellation testing now includes an added case for cancelling a the
second submission in the chain prior to calling submit noting that no
completions notifications should be given back for those.
Signed-off-by: Tom Burdick <thomas.burdick@intel.com>
Namespaced the generated headers with `zephyr` to prevent
potential conflict with other headers.
Introduce a temporary Kconfig `LEGACY_GENERATED_INCLUDE_PATH`
that is enabled by default. This allows the developers to
continue the use of the old include paths for the time being
until it is deprecated and eventually removed. The Kconfig will
generate a build-time warning message, similar to the
`CONFIG_TIMER_RANDOM_GENERATOR`.
Updated the includes path of in-tree sources accordingly.
Most of the changes here are scripted, check the PR for more
info.
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Move the syscall_handler.h header, used internally only to a dedicated
internal folder that should not be used outside of Zephyr.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
This adds a few line use zephyr_syscall_header() to include
headers containing syscall function prototypes.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Until now iterable sections APIs have been part of the toolchain
(common) headers. They are not strictly related to a toolchain, they
just rely on linker providing support for sections. Most files relied on
indirect includes to access the API, now, it is included as needed.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
- Add a new API `rtio_sqe_cancel` to attempt canceling a queued SQE
- Add a new syscall `rtio_sqe_copy_in_get_handles` which allows getting
back the SQE handles generated by the copy_in operation so that they
can be canceled.
Signed-off-by: Yuval Peress <peress@google.com>
Reworks the zephyr macros and pools to be objects in their own right. Each
pool can be statically defined with a Z_ private macro. The objects can
then be initialized with an rtio instance statically.
This cleans up a lot of code that was otherwise doing little bits of
management around allocation/freeing and reduces the scope those functions
has to the data it needs.
This should enable sharing the pools of sqe, cqe, and mem blocks among rtio
instances in a future improvement easily.
Signed-off-by: Tom Burdick <thomas.burdick@intel.com>
Rather than the rings, which weren't shared between userspace and kernel
space in Zephyr like they are in Linux with io_uring, use atomic mpsc
queues for submission and completion queues.
Most importantly this removes a potential head of line blocker in the
submission queue as the sqe would be held until a task is completed.
As additional bonuses this avoids some additional locks and restrictions
about what can be submitted and where. It also removes the need for
two executors as all chains/transactions are done concurrently.
Lastly this opens up the possibility for a common pool of sqe's to
allocate from potentially saving lots of memory.
Signed-off-by: Tom Burdick <thomas.burdick@intel.com>
It was previously assumed that the 'sys_mem_blocks' struct would maintain
information about contiguous blocks allocated so the release API only
took the starting address. This led to an issue where allocating 2+
blocks would end up with a memory leak because any block not being the
first would never be released.
Add the buffer length as an argument so the correct number of blocks can
be released. Also, ammend the tests to match and verify.
Signed-off-by: Yuval Peress <peress@google.com>
- Introduce a new Kconfig to enable mempool in RTIO
- Introduce a new RTIO_DEFINE_WITH_MEMPOOL to allocate an RTIO context
with an associated memory pool.
- Add a new sqe read function rtio_sqe_read_with_pool() for memory pool
enabled RTIO contexts
- Allow IODevs to allocate only the memory they need via rtio_sqe_rx_buf()
- Allow the consumer to get the allocated buffer via
rtio_cqe_get_mempool_buffer()
- Consumers need to release the buffer via rtio_release_buffer() when
processing is complete.
Signed-off-by: Yuval Peress <peress@google.com>
Update the policy such that every completed sqe has a parallel cqe.
This has the primary purpose of making any reads in the sqe visible
to the consumer (since they might have different buffers).
Signed-off-by: Yuval Peress <peress@google.com>
Adds the transceive op which is needed for full SPI support. Notably
in RTIO transceive is expected to be a balanced buffer pair of the same
length and have non-null tx/rx bufs.
Signed-off-by: Tom Burdick <thomas.burdick@intel.com>
Adds a callback op to RTIO enabling C logic to be interspersed with
I/O operations. This is not safe from userspace but allowable in kernel
space.
Signed-off-by: Tom Burdick <thomas.burdick@intel.com>
When sending small buffers out it makes sense to copy rather than
reference to avoid having to keep the small buffer around for the
lifetime of the write request.
Adjusts the op numbers to always be +1 from the previously defined op
id making it easier to re-arrange if needed in the future.
Signed-off-by: Tom Burdick <thomas.burdick@intel.com>
The pending_sqe logic to track where in the ring queue the concurrent
executor had left off was slightly flawed. It didn't account for starting
all sqes in the queue and ending back up at the beginning.
Instead track the last SQE in the queue, from which the next one in the
queue will the one to start next.
If we happen to sweep the last known SQE in the queue, reset it to NULL
so the next time prepare is called we start at the beginning of the queue
again.
Signed-off-by: Tom Burdick <thomas.burdick@intel.com>
Transactional submissions treat a sequence of sqes as a single atomic
submission given to a single iodev and expect as a reply a single
completion.
This is useful for scatter gather like APIs that exist in Zephyr already
for I2C and SPI.
Signed-off-by: Tom Burdick <thomas.burdick@intel.com>
By using an mpsc queue for each iodev, the iodev itself is shareable across
contexts. Since its lock free, submits may occur even from an ISR context.
Rather than a fixed size queue, and with it the possibility of running
out of pre-allocated spots, each iodev now holds a wait-free mpsc
queue head.
This changes the parameter of iodev submit to be a struct containing 4
pointers for the rtio context, the submission queue entry, and the mpsc
node for the iodevs submission queue.
This solves the problem involving busy iodevs working with real
devices. For example a busy SPI bus driver could enqueue, without locking,
a request to start once the current request is done.
The queue entries are expected to be owned and allocated by the
executor rather than the iodev. This helps simplify potential
tuning knobs to one place, the RTIO context and its executor an
application directly uses.
As the test case shows iodevs can operate effectively lock free
with the mpsc queue and a single atomic denoting the current task.
Signed-off-by: Tom Burdick <thomas.burdick@intel.com>
Add support for userspace with RTIO by making rtio and rtio_iodev
k_objects. As well as adding three syscalls for copying in submissions,
copying out completions, and starting tasks with submit.
For the small devices Zephyr typically runs on one of the most important
attributes tends to be low memory usage. To maintain the low footprint of
RTIO and its current executor implementations the rings are not shared with
userspace. Sharing the rings it turns out would require copying submissions
before working with them to avoid TOCTOU issues.
The API could still support shared rings in the future so that a
kernel thread could directly poll, copy, verify, and start the submitted
work. This would require a third executor implementation that maintains its
own copy of submissions similiar to how io_uring in Linux works.
Signed-off-by: Tom Burdick <thomas.burdick@intel.com>
When an application is waiting on a completion it may be expecting,
rightfully so, that a new submissions slot is available.
Frees the submission queue event prior to enqueuing
the completion queue event in the simple executor.
Signed-off-by: Tom Burdick <thomas.burdick@intel.com>
Add the "zephyr/" prefix to various #include statements that are
preventing the CI form running with LEGACY_INCLUDE_PATH=n.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
Any project with Kconfig option CONFIG_LEGACY_INCLUDE_PATH set to n
couldn't be built because some files were missing zephyr/ prefix in
includes
Re-run the migrate_includes.py script to fix all legacy include paths
Signed-off-by: Tomislav Milkovic <milkovic@byte-lab.com>
Schedules I/O chains in the same order as they arrive providing a fixed
amount of concurrency. The low memory cost comes at the cost of some
computational cost that is likely to be acceptable with small amounts
of concurrency.
The code cost is about 4x higher than the simple linear executor
which isn't entirely unexpected as the logic requirements are quite a bit
more than doing the next thing in the queue.
Signed-off-by: Tom Burdick <thomas.burdick@intel.com>
A DMA friendly Stream API for zephyr. Based on ideas from io_uring
and iio, a queue based API for I/O operations.
Provides a pair of fixed length ringbuffer backed queues for submitting
I/O requests and recieving I/O completions. The requests may be chained
together to ensure the next operation does not start until the current
one is complete.
Requests target an abstract rtio_iodev which is expected to wrap all
the hardware particulars of how to perform the operation. For example
with a SPI bus device, a description of what a read, and write mean
can be decided by the iodev wrapping a particular device
hanging off of a SPI controller.
The queue pair are submitted to an executor which may be a simple
inplace looping executor done in the callers execution context
(thread/stack) but other executors are expected. A threadpool executor
might for example allow for concurrent request chains to execute in
parallel. A DMA executor, in conjunction with DMA aware iodevs
would allow for hardware offloading of operations going so far as to
schedule with priority using hardware arbitration.
Both the iodev and executor are definable by a particular
SoC, meaning they can work in conjuction to perform IO operations
using a particular DMA controller or methodology if desired.
The application decides entirely how large the queues are, where
the buffers to read/write come from (some executors
may have particular demands!), and which executor to submit
requests to.
Signed-off-by: Tom Burdick <thomas.burdick@intel.com>
