Unsafe math operations #4261

angusluckmann · 2020-04-06T13:46:59Z

V version: 0.1.24
OS: Android 9, Termux (could have probably been any other)

What did you do?

fn add(num1 int, num2 int) int {
	return (num1   num2)
}

fn sub(num1 int, num2 int) int {
	return (num1 - num2)
}

fn mul(num1 int, num2 int) int {
	return (num1 * num2)
}

fn div(num1 int, num2 int) int {
	return (num1 / num2)
}

fn main() {
	// Define necessary constants
	max_possible_int32 := int(2147483647)
	min_possible_int32 := int(-2147483648)

	// Check direct addition
	println("========== DIRECT ADDITION TEST ==========")
	mut res := int(max_possible_int32   max_possible_int32)
	println(res)
	println("")

	// Check direct substraction
	println("========== DIRECT SUBSTRACTION TEST ==========")
	res = int(min_possible_int32 - max_possible_int32)
	println(res)
	println("")

	//Check direct multiplication
	println("========== DIRECT MULTIPLICATION TEST ==========")
	res = int(max_possible_int32 * max_possible_int32)
	println(res)
	println("")

	// Check indirect addition
	println("========== INDIRECT ADDITION TEST ==========")
	mut tmp := add(max_possible_int32, max_possible_int32)
	println(tmp)
	println("")

	// Check indirect substraction
	println("========== INDIRECT SUBSTRACTION TEST ==========")
	tmp = sub(min_possible_int32, max_possible_int32)
	println(tmp)
	println("")

	// Check indirect multiplication
	println("========== INDIRECT MULTIPLICATION TEST ==========")
	tmp = mul(max_possible_int32, max_possible_int32)
	println(tmp)
	println("")

	// Check direct division
	println("========== DIRECT DIVISION TEST ==========")
	mydiv := int(1/0)
	println(mydiv)
	println("")

	// Check indirect division
	println("========== INDIRECT DIVISION TEST ==========")
	mydiv := div(1,0)
	println(mydiv)
	println("")
}

What did you expect to see?
4 compiler errors/warnings and 4 runtime errors

What did you see instead?
1 compiler error (direct division test) and the program spitting out 7 wrong values to the screen.
(and a bunch of compiler warnings saying that "casting int to int is not needed")

The text was updated successfully, but these errors were encountered:

angusluckmann · 2020-04-06T13:54:55Z

Sorry if I created the issue the wrong way. I'm new to Github Issues :)

danieldaeschle · 2020-04-06T19:54:59Z

try it with the new compiler v0.1.26 please!

angusluckmann · 2020-04-07T10:53:58Z

V version: 0.1.26 4e01bf4.78440be
Discoveries:
Now I get a weird C error on the division function (even if I don't call it).
And all the tests (even the direct division test) seem to run through.

More info:

I did a build through ...

git clone https://github.com/vlang/v
cd v
make

In that state I got the C error and an error on the direct division test.

Then I ran ...
v up
And after the update, the C error was still there, but the direct division test passed with a weird number (-973237312)

The version number mentioned above is the one after the v up command.

angusluckmann · 2020-04-07T10:57:41Z

I obviously added the v command to path with ...
export PATH=$PWD:$PATH
... and checked the version correctness with ...
v --version

danieldaeschle · 2020-04-07T11:10:56Z

thank you. the new error should be reproduceable.

can you paste the new error which you get?

angusluckmann · 2020-04-07T11:29:16Z

Just for the record, I'm gonna repeat some of the information to some them up.
V version: V 0.1.26 4e01bf4.78440be
Source Code:

fn add(num1 int, num2 int) int {
	return (num1   num2)
}

fn sub(num1 int, num2 int) int {
	return (num1 - num2)
}

fn mul(num1 int, num2 int) int {
	return (num1 * num2)
}

fn div(num1 int, num2 int) int {
	return (num1 / num2)
}

fn main() {
	// Define necessary constants
	max_possible_int32 := int(2147483647)
	min_possible_int32 := int(-2147483648)

	// Check direct addition
	println("========== DITECT ADDITION TEST ==========")
	mut res := int(max_possible_int32   max_possible_int32)
	println(res)
	println("")

	// Check direct substraction
	println("========== DIRECT SUBSTRACTION TEST ==========")
	res = int(min_possible_int32 - max_possible_int32)
	println(res)
	println("")

	//Check direct multiplication
	println("========== DIRECT MULTIPLICATION TEST ==========")
	res = int(max_possible_int32 * max_possible_int32)
	println(res)
	println("")

	// Check indirect addition
	println("========== INDIRECT ADDITION TEST ==========")
	mut tmp := add(max_possible_int32, max_possible_int32)
	println(tmp)
	println("")

	// Check indirect substraction
	println("========== INDIRECT SUBSTRACTION TEST ==========")
	tmp = sub(min_possible_int32, max_possible_int32)
	println(tmp)
	println("")

	// Check indirect multiplication
	println("========== INDIRECT MULTIPLICATION TEST ==========")
	tmp = mul(max_possible_int32, max_possible_int32)
	println(tmp)
	println("")

	// Check direct division
	println("========== DIRECT DIVISION TEST ==========")
	mydiv := int(1/0)
	println(mydiv)
	println("")

	// Check indirect division
	println("========== INDIRECT DIVISION TEST ==========")
	mydiv = div(1,0)
	println(mydiv)
	println("")
}

Compiler output:

==================
                          ^
/data/data/com.termux/files/home/vtest2/tmp.c:4519:5: error: conflicting types for 'div'
int div(int num1, int num2) {
    ^
/data/data/com.termux/files/usr/include/stdlib.h:183:7: note: previous declaration is here
div_t div(int __numerator, int __denominator) __attribute_const__;
      ^                                         /data/data/com.termux/files/home/vtest2/tmp.c:4556:8: error: assigning to 'int' from incompatible type 'div_t'                                          mydiv = div(1, 0);
              ^ ~~~~~~~~~
/data/data/com.termux/files/home/vtest2/tmp.c:4552:23: warning: division by zero is undefined [-Wdivision-by-zero]
        int mydiv = ((int)(1 / 0));
...
==================
(Use `v -cg` to print the entire error message)

builder error: C error.

Please make sure that:
- You have all V dependencies installed.
- You did not declare a C function that was not included. (Try commenting your code that involves C interop)
- You are running the latest version of V. (Try running `v up` and rerunning your command)

If you're confident that all of the above is true, please try running V with the `-cg` option which enables more debugging capabilities.

angusluckmann · 2020-04-07T11:34:40Z

Second experiment

V version: V 0.1.26 4e01bf4.78440be
Source Code:

fn add(num1 int, num2 int) int {
	return (num1   num2)
}

fn sub(num1 int, num2 int) int {
	return (num1 - num2)
}

fn mul(num1 int, num2 int) int {
	return (num1 * num2)
}

fn div(num1 int, num2 int) int {
	return (num1 / num2)
}

fn main() {
	// Define necessary constants
	max_possible_int32 := int(2147483647)
	min_possible_int32 := int(-2147483648)

	// Check direct addition
	println("========== DITECT ADDITION TEST ==========")
	mut res := int(max_possible_int32   max_possible_int32)
	println(res)
	println("")

	// Check direct substraction
	println("========== DIRECT SUBSTRACTION TEST ==========")
	res = int(min_possible_int32 - max_possible_int32)
	println(res)
	println("")

	//Check direct multiplication
	println("========== DIRECT MULTIPLICATION TEST ==========")
	res = int(max_possible_int32 * max_possible_int32)
	println(res)
	println("")

	// Check indirect addition
	println("========== INDIRECT ADDITION TEST ==========")
	mut tmp := add(max_possible_int32, max_possible_int32)
	println(tmp)
	println("")

	// Check indirect substraction
	println("========== INDIRECT SUBSTRACTION TEST ==========")
	tmp = sub(min_possible_int32, max_possible_int32)
	println(tmp)
	println("")

	// Check indirect multiplication
	println("========== INDIRECT MULTIPLICATION TEST ==========")
	tmp = mul(max_possible_int32, max_possible_int32)
	println(tmp)
	println("")

	// Check direct division
	println("========== DIRECT DIVISION TEST ==========")
	mydiv := int(1/0)
	println(mydiv)
	println("")

	// Check indirect division
	// println("========== INDIRECT DIVISION TEST ==========")
	// mydiv = div(1,0)
	// println(mydiv)
	// println("")
}

Compiler output (probably the same as last one):

==================                                                        ^
/data/data/com.termux/files/home/vtest2/tmp02.c:4519:5: error: conflicting types for 'div'
int div(int num1, int num2) {
    ^
/data/data/com.termux/files/usr/include/stdlib.h:183:7: note: previous declaration is here
div_t div(int __numerator, int __denominator) __attribute_const__;
      ^                                         /data/data/com.termux/files/home/vtest2/tmp02.c:4552:23: warning: division by zero is undefined [-Wdivision-by-zero]                                    int mydiv = ((int)(1 / 0));
                             ^ ~
/data/data/com.termux/files/home/vtest2/tmp02.c:6440:407: warning: integer literal is too large to be represented in a signed integer type, interpreting as unsigned [-Wimplicitly-unsigned-literal]
((u64)(1)), ((u64)(10)), ((u64)(100)), ((u64)(1000)), ((u64)(10000)), ((u64)(100000)), ((u64)(1000000)), ((u64)(10000000)), ((u64)(100000000)), ((u64)(1000000000)), ((u64)(10000000000)), ((u64)(100000000000)), ((u64)(1000000000000)), ((u64)(10000000000000)), ((u64)(100000000000000)), ((u64)(1000000000000000)), ((u64)(10000000000000000)), ((u64)(100000000000000000)), ((u64)(1000000000000000000)), ((u64)(10000000000000000000)),
...
==================
(Use `v -cg` to print the entire error message)

builder error: C error.

Please make sure that:
- You have all V dependencies installed.
- You did not declare a C function that was not included. (Try commenting your code that involves C interop)
- You are running the latest version of V. (Try running `v up` and rerunning your command)

If you're confident that all of the above is true, please try running V with the `-cg` option which enables more debugging capabilities.

angusluckmann · 2020-04-07T11:39:26Z

Third experiment

V version: V 0.1.26 4e01bf4.78440be
Source Code:

fn add(num1 int, num2 int) int {
	return (num1   num2)
}

fn sub(num1 int, num2 int) int {
	return (num1 - num2)
}

fn mul(num1 int, num2 int) int {
	return (num1 * num2)
}

/*
fn div(num1 int, num2 int) int {
	return (num1 / num2)
}
*/

fn main() {
	// Define necessary constants
	max_possible_int32 := int(2147483647)
	min_possible_int32 := int(-2147483648)

	// Check direct addition
	println("========== DITECT ADDITION TEST ==========")
	mut res := int(max_possible_int32   max_possible_int32)
	println(res)
	println("")

	// Check direct substraction
	println("========== DIRECT SUBSTRACTION TEST ==========")
	res = int(min_possible_int32 - max_possible_int32)
	println(res)
	println("")

	//Check direct multiplication
	println("========== DIRECT MULTIPLICATION TEST ==========")
	res = int(max_possible_int32 * max_possible_int32)
	println(res)
	println("")

	// Check indirect addition
	println("========== INDIRECT ADDITION TEST ==========")
	mut tmp := add(max_possible_int32, max_possible_int32)
	println(tmp)
	println("")

	// Check indirect substraction
	println("========== INDIRECT SUBSTRACTION TEST ==========")
	tmp = sub(min_possible_int32, max_possible_int32)
	println(tmp)
	println("")

	// Check indirect multiplication
	println("========== INDIRECT MULTIPLICATION TEST ==========")
	tmp = mul(max_possible_int32, max_possible_int32)
	println(tmp)
	println("")

	// Check direct division
	println("========== DIRECT DIVISION TEST ==========")
	mydiv := int(1/0)
	println(mydiv)
	println("")

	// Check indirect division
	// println("========== INDIRECT DIVISION TEST ==========")
	// mydiv = div(1,0)
	// println(mydiv)
	// println("")
}

Compiler output: none (just successfully compiling)
Program output:

========== DITECT ADDITION TEST ==========
-2

========== DIRECT SUBSTRACTION TEST ==========
1

========== DIRECT MULTIPLICATION TEST ==========
1

========== INDIRECT ADDITION TEST ==========
-2

========== INDIRECT SUBSTRACTION TEST ==========
1                                               
========== INDIRECT MULTIPLICATION TEST ==========
1

========== DIRECT DIVISION TEST ==========
-65227424

angusluckmann · 2020-04-07T11:39:55Z

That should be it.

Ivo-Balbaert · 2020-05-25T14:36:10Z

The Direct Division Test now (v 0.1.27) gives a V error as it should: mydiv := int(1/0) // error: division by zero.
Is this issue now solved?

M4SSD35TRUCT10N assigned medvednikov Apr 6, 2020

M4SSD35TRUCT10N added Bug This tag is applied to issues which reports bugs. Security Bugs/feature requests, that are related to either vlib modules or the language as a whole. labels Apr 6, 2020

M4SSD35TRUCT10N added this to the Beta Release milestone Apr 6, 2020

M4SSD35TRUCT10N added Status: Confirmed This bug has been confirmed to be valid by a contributor. Status: Needs More Info The issue author is expected to provide more information. OS: Android Bugs/feature requests, that are specific to Android OS. labels Apr 6, 2020

ArtemkaKun removed this from the Beta Release milestone May 26, 2023

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Unsafe math operations #4261

Unsafe math operations #4261

angusluckmann commented Apr 6, 2020 •

edited

Loading

angusluckmann commented Apr 6, 2020

danieldaeschle commented Apr 6, 2020

angusluckmann commented Apr 7, 2020 •

edited

Loading

angusluckmann commented Apr 7, 2020

danieldaeschle commented Apr 7, 2020

angusluckmann commented Apr 7, 2020

angusluckmann commented Apr 7, 2020

angusluckmann commented Apr 7, 2020

angusluckmann commented Apr 7, 2020

Ivo-Balbaert commented May 25, 2020

Unsafe math operations #4261

Unsafe math operations #4261

Comments

angusluckmann commented Apr 6, 2020 • edited Loading

angusluckmann commented Apr 6, 2020

danieldaeschle commented Apr 6, 2020

angusluckmann commented Apr 7, 2020 • edited Loading

angusluckmann commented Apr 7, 2020

danieldaeschle commented Apr 7, 2020

angusluckmann commented Apr 7, 2020

angusluckmann commented Apr 7, 2020

Second experiment

angusluckmann commented Apr 7, 2020

Third experiment

angusluckmann commented Apr 7, 2020

Ivo-Balbaert commented May 25, 2020

angusluckmann commented Apr 6, 2020 •

edited

Loading

angusluckmann commented Apr 7, 2020 •

edited

Loading