Commit Graph

18 Commits

Author SHA1 Message Date
Andrew Tridgell 3f5534eed5 AP_HAL: use calloc in preferance to malloc 2018-01-17 11:22:05 +11:00
Lucas De Marchi 05a7eef1a2 RingBuffer: use explict "if" and return early
Remove confusing use of ternary operator together with assign + load.
No change in behavior.
2016-10-27 10:35:01 -03:00
Murilo Belluzzo 8526b25654 RingBuffer: Add a faster method to read a single byte 2016-10-27 14:23:42 +11:00
Murilo Belluzzo b856d26087 RingBuffer: ::set_size now returns true or false 2016-08-03 01:44:02 -03:00
Murilo Belluzzo f7f203efa9 RingBuffer: Add ::clear() method to discard the buffer content
The same could be achieved with ::set_size(::get_size()), but was not
obvious and hurts code readability.
2016-08-03 01:44:02 -03:00
Murilo Belluzzo 75a1b102fb RingBuffer: Handle zero sized better
Sometimes (like in DataFlash) the size of the ring buffer will be
determined in run time and the object can have size zero until proper
initialization. When this was the case, an underflow in ::get_size would
mess with the initializing algorithm.

Another issue was that the 'new' operator could fail what was not being
handled. Now, we only set the size member after we are sure 'new'
successfully allocated memory.
2016-08-03 01:44:02 -03:00
Murilo Belluzzo 3e1acdcbbf AP_HAL: Use atomic instead volatile on RingBuffer head/tail
Volatile will provide protection to sequence re-ordering and guarantee
the variable is fetched from memory, but it won't provide the memory
barrier needed to ensure that no re-ordering (by either the compiler or
the CPU) will happen among other threads of execution
accessing the same variables.

For more info about this effect can be found on articles about
std::memory_order.
2016-07-08 16:12:53 -03:00
Murilo Belluzzo b7dd4dad64 AP_HAL: Fix ByteBuffer::reserve() breaking buffer state
When using reserved(), the reserved memory cannot be read before it's
written, therefore we cannot update 'tail' until the caller of
reserved() is done writing.

To solve that, a method called 'commit()' was added so the caller can
inform that is done with the memory usage and is safe to update 'tail'.
The caller also has to inform the length that was actually written.

This solution was developed to work considering the usage context of
this class: 1 reader and 1 writer **only**.
2016-07-08 13:10:16 -03:00
Murilo Belluzzo 43d4012884 AP_HAL: Fix TOCTOU in ByteBuffer::peekiovec() 2016-07-08 13:10:16 -03:00
Murilo Belluzzo 9951b94d40 AP_HAL: Change the return type of 'peekiovec'
Possible values are '0', '1' and '2' so uint8_t is a better fit. Also,
invert 'if' condition so it's clear that is returning 0, 1 or 2.
2016-07-08 13:10:16 -03:00
Murilo Belluzzo 625f47def7 AP_HAL: Improve ByteBuffer::set_size
So it doesn't delete and re-create the buffer if the size
happens to be the same. Still resets the buffer content.
2016-07-08 13:10:16 -03:00
Murilo Belluzzo e9e31172c0 AP_HAL: Simplify ByteBuffer::readptr logic 2016-07-08 13:10:16 -03:00
Leandro Pereira fbefe32017 AP_HAL: Add method to reserve space in the ring buffer
Adds a method called `reserve()`, that will take a ByteBuffer::IoVec
array of at least two elements, and return the number of elements
filled out.  0 will be returned if `len` is over the total space of
the buffer; 1 will be returned if there's enough contiguous bytes in
the buffer; 2 will be returned if there are two non-contiguous blocks
of memory.

This method is suitable to be used with POSIX system calls such as
readv(), and is an optimization to not require temporary memory copies
while reading from a file descriptor.

Also modify the write() method to use reserve(), so that similar checks
are performed only in one place.
2016-07-08 13:10:16 -03:00
Leandro Pereira e3b676ba89 AP_HAL: Add method to peek non-contiguous parts of a ByteBuffer
Modify ByteBuffer class to have a `peekiovec()` method, that takes in a
`struct IoVec` array (similar to `struct iovec` from POSIX), and a
number of bytes, and returns the number of elements from this array
that have been filled out.  It is either 0 (buffer is empty), 1
(there's enough contiguous bytes to read that amount) or 2 (ring buffer
is wrapping around).

This enables using scatter-gather I/O (i.e. writev()), removing calls
to memcpy().  That's one call when no wrap-around is happening, and
two calls if it is.

Also, rewrite `ByteBuffer::peekbytes()` to use `peekiovec()`, so that
some of the checks performed by the former are not replicated in the
latter.
2016-07-08 13:10:16 -03:00
Andrew Tridgell 595d5c0002 AP_HAL: added set_size() to RingBuffer API 2016-05-23 23:41:02 +10:00
Andrew Tridgell 1df2512935 AP_HAL: added update() method for object ringbuffer
to support updating objects for GCS work Tom is doing
2016-02-23 16:34:06 +11:00
Andrew Tridgell 4ff396dfa8 AP_HAL: added force() and peek() method for object ringbuffers 2016-02-22 12:34:32 +11:00
Andrew Tridgell ec6a679482 AP_HAL: added ByteBuffer and ObjectBuffer in RingBuffer
much better API than old macros
2015-12-20 07:33:54 +11:00