Re: [avr-gcc-list] Operation problem

[Gerard Marull Paretas]

> Paulo Marques:
> I think you just answered your own question: if all the variables involved are uint16_t, then that is the precision used for all he _expression_ and it will overflow on the multiplication.

The same happens whether if currentsong.length is uint32_t or uint16_t.


As you asked Eric, I attach the C source of the function, the two disassembly files from both codes, and a diff file between the two asm files.

The items of structure are defined as follows:

struct{
    uint16_t bitrate;
    uint32_t length;
    uint16_t frames;
    uint16_t samples_per_sec;
    uint16_t samples_per_frame;
    uint8_t mpeg;
    uint8_t layer;
    uint8_t channel;
}currentsong;

I hope this information is enough now. (Sorry to not provide something more before :) )

Thanks for your help! Gerard.

csource.c
Description: Text Data

asm_works.s
Description: Binary data

asm_notworks.s
Description: Binary data

asm_diff.s
Description: Binary data