John RegehrMar 2

won't say I'm totally proud of myself here, but once I saw that the Claude C compiler was super buggy according to YARPGen and Csmith, I had a hard time preventing myself from doing something about it

https://john.regehr.org/writing/claude_c_compiler.html

claude_c_compiler

Show threadzwarichMar 2
@regehr My first reaction when seeing the Claude C Compiler was “John Regehr is going to be able to milk this so hard”.
Show threadAssertionError("Joe Groff")Mar 3
@zwarich @regehr i just keep trying to find a backup of the classic "state of an 80s C compiler" tweet
Show threadzwarichMar 3
@joe @regehr Someone should prank Claude Code by giving it an ancient C compiler where struct field labels needed to be globally unique.
Show threadAssertionError("Joe Groff")Mar 3
@zwarich @regehr or the one where `x <<= n` takes O(`n`) time
Show threadJohn RegehrMar 3
@joe @zwarich ha--I know that one. let me dig it up.
Show threadzwarichMar 3
@regehr @joe You’ll also need an emulation environment for running the result. Might be easier to just hack up TCC to have the archaic behavior.
Show threadJohn RegehrMar 3

@zwarich @joe it's table 1 from this paper, right?

https://www.cs.swarthmore.edu/~bylvisa1/cs97/f13/Papers/DifferentialTestingForSoftware.pdf

Show threadPer Vognsen
@regehr @zwarich @joe Heh, speaking of 0.0f/0.0f crashing the tested compiler (one of the table 1 examples), you can still crash tcc today (last I checked) with INT_MIN/-1 on x86. I like to test any program that has a built-in expression evaluator with that and it's scary (but many not surprising) how many you can crash with it.
Mar 3 at 1:24amPhanpy
Show threadJohn RegehrMar 3
@pervognsen @zwarich @joe it's like shooting fish in a barrel with a nuclear bomb