Timeline for Build a Compiler Bomb
Current License: CC BY-SA 3.0
8 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Feb 6, 2017 at 15:57 | comment | added | Yakk | Unless you are really careful, the limitation here will be template depth based. Which means that the byte count will grow with the template depth parameter more than anything; only exponential. Seems sad to use Ackermann to get only exponential growth. | |
| Jan 15, 2016 at 13:40 | comment | added | Toby Speight | But we want bigger code! Fibonacci gives almost the same size as pure linear code (but longer compile time that the linear). You could certainly have fun with a static array of size A+B in each class, now I think of it... | |
| Jan 14, 2016 at 21:45 | comment | added | Will Ness | won't Fibonacci give you a smaller code and better output size control? | |
| Jan 13, 2016 at 10:29 | history | edited | Toby Speight | CC BY-SA 3.0 | added 132 characters in body |
| Jan 13, 2016 at 9:42 | comment | added | Mark | I was expecting something absurd to come out of C++ templates. I wasn't expecting the Ackermann function. | |
| Jan 13, 2016 at 0:34 | comment | added | Dave | Update: As Ben Voigt points out on his answer, GCC on Linux does generate ELF files as .o output, and I've been able to confirm the <3,14> variant with it, so yup - this is valid. | |
| Jan 12, 2016 at 23:01 | comment | added | Dave | I really like this one, but I'm not sure I can accept a .o output, since I did say ELF/.exe/etc. (and compiling this fully optimises it all out!). Still, +1 (and confirmed) | |
| Jan 12, 2016 at 17:50 | history | answered | Toby Speight | CC BY-SA 3.0 |