Possible miscompilation error? #164
Comments
|
Tried on ESP32-WROOM-32 and got |
|
I've updated the comment with my configuration |
|
Tried with both versions from https://wokwi.com/projects/345876589283639891 and also with pub fn calculate(data: &[u8]) -> u8 {
const CRC8_POLYNOMIAL: u8 = 0x31;
let mut crc: u8 = 0xff;
for byte in data {
crc ^= byte;
println!("crc tmp: {:02x}", crc);
for _ in 0..8 {
if (crc & 0x80) > 0 {
println!("top bit is high");
//crc = (crc.overflowing_shl(1)).0 ^ CRC8_POLYNOMIAL;
crc = (crc << 1) ^ CRC8_POLYNOMIAL;
} else {
println!("top bit is low");
//crc = crc.overflowing_shl(1).0;
crc <<= 1;
}
}
}
crc
}
pub fn calculate2(data: &[u8]) -> u8 {
const CRC8_POLYNOMIAL: u8 = 0x31;
let mut crc: u8 = 0xff;
for byte in data {
crc ^= byte;
for _ in 0..8 {
if (crc & 0x80) > 0 {
//crc = (crc.overflowing_shl(1)).0 ^ CRC8_POLYNOMIAL;
crc = (crc << 1) ^ CRC8_POLYNOMIAL;
} else {
//crc = crc.overflowing_shl(1).0;
crc <<= 1;
}
}
}
crc
} |
|
Hi @vojty, sorry you've been running into this issue. Fortunately, its a duplicate of #134 which we already have a fix for! It will be released in the next 2 weeks with the 1.65 Rust toolchain. I tested your code with the new LLVM version locally and it works correctly with your input data :). I'll close this for now, in favour of the tracking issue here: #134. |
|
@MabezDev Thank you! |
|
@MabezDev I've upgraded the toolchain to 1.65 using https://github.com/esp-rs/espup Unfortunately, the calculation is still wrong. For given The program executed on the ESP chip outputs Code: pub fn calculate(data: &[u8]) -> u8 {
const CRC8_POLYNOMIAL: u8 = 0x31;
let mut crc: u8 = 0xff;
for byte in data {
crc ^= byte;
for _ in 0..8 {
if (crc & 0x80) > 0 {
crc = (crc << 1) ^ CRC8_POLYNOMIAL;
} else {
crc <<= 1;
}
}
}
crc
}
fn main() -> Result<()> {
// Temporary. Will disappear once ESP-IDF 4.4 is released, but for now it is necessary to call this function once,
// or else some patches to the runtime implemented by esp-idf-sys might not link properly.
esp_idf_sys::link_patches();
// Bind the log crate to the ESP Logging facilities
esp_idf_svc::log::EspLogger::initialize_default();
println!("dec={}", calculate(&[171, 13]));
Ok(())
}Log Also the previous example EDIT: then the result is correct: |
|
Thanks for the update, it seems like it's not entirely fixed. For me, it works on all opt-levels apart from when |
|
I can confirm that |
|
With the 1.69 release, this should now be fixed. Please let me know if you still run into issues! |
Hello, I've found a strange error and I'm not sure what the problem might be. I've been trying to use ESP32 + SCD41 with no luck. I tracked down the problem to this piece of code:
For the given example data
[9, 94]the correct result should be35but I'm getting55. It works correctly on the blank Rust project usingcargo newbut it fails when I deploy the code to my ESP32.I've created an example here https://wokwi.com/projects/345876589283639891 with 2 identical
calculatefunctions, the only difference is that one of them hasprintln!macros inside. However, the result differs.Does anyone know what's going on? What am I missing?
Configuration:
Chip: Espressif ESP32-WROOM-32 (https://www.laskakit.cz/laskakit-esp-vindriktning-esp-32-i2c/)
The text was updated successfully, but these errors were encountered: