RE: embedded C++: add cheap bitwise conversion to a register structure

Jan Cumps — Thu, 17 Apr 2025 22:17:43 GMT

I reused this for a stepper motor design that I'm developing. But I used the constructor this time, to do the conversion.

[embed:dc8ab71f-3b98-42d9-b0f6-e21e02a0f8e2:86887152-f39d-4180-a51f-7f898ef4cc9d:type=c_cpp&text=%2F%2A%20%20Stepper%20motor%20command%20wrapper%0D%0A%20%20%20%20lightweight%20%0D%0A%2A%2F%0D%0Aclass%20command%20%7B%0D%0Apublic%3A%0D%0A%20%20%20%20inline%20command%28uint32_t%20steps%2C%20bool%20reverse%29%20%3A%20cmd_%28steps%20%3C%3C%201%20%7C%20%28reverse%20%3F%200%20%3A%201%29%29%20%7B%7D%0D%0A%20%20%20%20inline%20operator%20uint32_t%28%29%20const%20%7B%20return%20cmd_%3B%20%7D%0D%0Aprivate%3A%0D%0A%20%20%20%20uint32_t%20cmd_%3B%0D%0A%7D%3B]

My stepper motor driver uses 32 bit instructions.
Bit 0 is the direction (left/right)
Bit 31 - 1 are the number of steps.

During construction, the two parameters are immediately converted to a single uint32_t, ready for the stepper driver.

How much space does this object take?

data size (each object): same as an uint32_t
code size (one time) - and also the cost of a single conversion: same as a call to steps << 1 | (reverse ? 0 : 1);

example:

command cmd(steps, reverse);

At runtime (firmware), this uses exactly the same resources as when you'd use a uint32_t variable, and fill it by calling:

uint32_t cmd =  steps << 1 | (reverse ? 0 : 1)

The original post uses late conversion. Data is converted when the operator is called. In that design, The operator uint32_t() has the conversion logic.
It makes sense. because the developer may want to manipulate individual register parts, before writing.. And we're not expecting arrays of registers.

In the example that I use in this comment, we don't expect manipulation on the data once it's created. You collect steps and directions, then send it to the stepper motor.
In that case, it makes sense to convert to the smallest data size immediately, in the constructor, taking the conversion hit at declaration.
When the operator uint32_t() is called, it just passes the data.

RE: embedded C++: add cheap bitwise conversion to a register structure

Jan Cumps — Sat, 05 Apr 2025 20:45:07 GMT

The approach allowed me to restrict the code in my firmware to just this:

At the same time, I used way less clock ticks and other resources (init time, memory) than TI's example. Without adding any costs to their (good, I think, but aged) code.

RE: embedded C++: add cheap bitwise conversion to a register structure

michaelkellett — Mon, 31 Mar 2025 10:12:18 GMT

What happens if ENBL = 2u ?

and

what does the generated assembler code look like ?

and

why not use bit fields, although in C the assignment of an out of range value to a bit field is implementation dependent (don't know about C++).

RE: embedded C++: add cheap bitwise conversion to a register structure

shabaz — Mon, 31 Mar 2025 04:24:17 GMT

Nice work! Very clever how all this can be encapsulated when using C++, and structs, not just classes. Sometimes easy to forget. I understand why the const (I won't put it here in case others want to have a go).