Scribd
Upload
482 views

Axelspace: Coding Assignment For: Embedded Systems Software Engineer

The document provides coding assignments for an embedded systems software engineer position at Axelspace. It includes instructions to write code for packet detection and parsing functions, along with unit tests. It also describes a communication protocol involving fixed-size packets with header and payload data. The interface description outlines how the parser class would integrate with other system components to process incoming telemetry packets.

Uploaded by

kenneth
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
482 views

Axelspace: Coding Assignment For: Embedded Systems Software Engineer

The document provides coding assignments for an embedded systems software engineer position at Axelspace. It includes instructions to write code for packet detection and parsing functions, along with unit tests. It also describes a communication protocol involving fixed-size packets with header and payload data. The interface description outlines how the parser class would integrate with other system components to process incoming telemetry packets.

Uploaded by

kenneth
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

Axelspace

Coding assignment for:


Embedded Systems Software Engineer
Notice: Please do not share this assignment with anyone except Axelspace employees.

Part 1
Imagine that as a part of the team you were assigned to develop and test two functions:
1. For packet detection
2. For packet parsing
Functions have to be compatible with the provided interface to make integration possible.

Code submission
Code:
• Please write the code as if it was your real assignment.
• Please use C++11 or higher, and standard libraries of C/C++
• Please focus on the portability aspect. If possible, code should be portable between CPU
architectures, compilers and OSs.
• We will give additional points for low footprint and determinism of the solution.
Tests:
• Provide unit tests proving correctness of your solution
• Tests may be written for any OS, compiler and in any language as well as using framework
of your choice. Please send a short HOWTO explaining how to build and run tests.
Communication protocol description
Basic conditions:
• Packet size is fixed to 12 bytes
• Every packet starts with a “magic” header
• Byte order of data in the packet is from most significant to least significant
• You cannot assume that all incoming bytes are correct, consistent packets.
There may be other types of data between packets, for example:
[packet] [packet] [control sequence] [packet] [spurious bytes] [control sequence] [packet]
Packet layout:
Magic header Secondary Header Reserved Payload Checksum
[4B] [2B] [1B] [4B] [1B]
0xABBACFFC 0xFF

Secondary header layout:


Protocol version Subsystem ID Component ID Telemetry type
[3 bit] [5 bit] [5 bit] [3 bit]

Magic header:
• 0xABBACFFC marks the start of a packet
Secondary header:
• Protocol versions:
◦ Supported protocol versions: 1, 2
Both versions have the same packet layout.
• Valid subsystem IDs:
◦ 1 - AOCS
◦ 3 - CDH
◦ 5 - COM
• Valid component IDs for each subsystem:
◦ AOCS: 20, 21, 22, 23, 30, 31
◦ CDH: 0
◦ COM: 1, 2, 10, 20
• Valid telemetry type:
◦ 1 - temperature (implies data of type IEEE 754 single precision floating-point)
Reserved field:
• One byte reserved for future protocol expansion.
◦ Must be set to 0xFF
Payload:
• If telemetry types is ‘1’ (temperature), the value in the 4 bytes of payload is IEEE 754 single
precision floating-point.
Checksum:
• Simple exclusive-or (xor) of all bytes from the start of secondary header to the end of
payload (including reserved field).
Interface description
For the exemplary code block fragment, please design and implement Parser class.

tlm_parser.h:
class Parser {
// Implement me
};

tlm.h:
typedef struct {
uint8_t subsys_id;
uint8_t compo_id;
Float temperature;
} CompoTlm_t;

tlm_proc.cpp:
...
...
while ( 1 ) {
while (iface.existRcvData()) {
uint8_t byte = iface.getByte();
// Parse incoming bytes and detect packet
if (parser.detectPkt(byte)) {
// One, correct packet has been detected
CompoTlm_t tlm = {};
int err = parser.extractData(tlm);
if (err == 0 ) {
// Pass telemetry info to the upper layer
system.tlm.update(tlm);
} else {
// Error handling, etc.
...
...
}
}
}
system.time.sleep_ms( 100 );
}
...
...
• iface is an object of a class that is used to communicate with the peripheral devices
◦ existRcvData() method returns true if there are bytes in the Rx FIFO
◦ getByte() method fetches one byte from the Rx FIFO
• parser is an object of a class that you are asked to implement
• system is an object of a class that communicates with the hypothetical OS
◦ tlm is an object used to process system’s telemetry.
◦ update() method can be used to send latest telemetry to the system control
Part 2
1. What is your favorite IPC mechanism and why ?

2. Imagine a program that realizes a very simple task a very large number of times;
for example, count up to 1 billion, one by one:
volatile int counter = 0;

int main(int argc, char** argv)


{
size_t nb_iter = 1000000000;

for (int i=0; i<nb_iter; i++)


{
counter ++;
}

printf("counter: %ld / %ld\n", counter, nb_iter);


return 0;
}

On a recent multi-core CPU, how could this program be enhanced to run faster ?
Please use C/C++ 11 or higher, and standard libraries of C/C++.

You might also like