Timeline for Build a Compiler Bomb
Current License: CC BY-SA 3.0
22 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 4, 2019 at 21:54 | comment | added | Aleksi Torhamo | @Joshua: I'm sorry, but there's just no way that that can be correct, as there are 1TB SD cards available, and you could fit ~6 of those per one side of a RAM stick. (So clearly you can fit at least 12TB of storage in the space of a single RAM stick without hitting any fundamental physical limits, and most motherboards have more than one RAM slot) For one atom per bit, you could actually fit on the order of 100-1000TB per square centimeter (1 atom thick), or billions of TB per cubic centimeter. | |
| Jun 3, 2017 at 1:45 | comment | added | Joshua | @AlexBollbach: The compiler's been fixed to not require the RAM since then. But as for 16TB RAM that can't fit in current RAM physical sizes because that would make one bit smaller than one atom. It's not impossible that will be willing to make our computers big enough, but its unlikely. In addition, the actual code for this answer was for 32 bit when I first discovered it. You don't have 16 TB 32 bit RAM for a single process. | |
| Jun 2, 2017 at 23:36 | comment | added | Alex Bollbach | the (You Don't) makes this response one that will not age gracefully with time. I (and people of my current time period) still don't. but at some point every one will. | |
| May 22, 2017 at 21:06 | comment | added | MD XF | @x13 reproducing behavior here. | |
| Jan 20, 2016 at 10:41 | comment | added | x13 |
I tried to compile it to see how fast it would lock up, i included stdint.h because it didn't know the type uint32_t. After that it compiles within the second and generates a segmentation fault when i run it.
|
|
| Jan 15, 2016 at 18:16 | comment | added | Shelvacu | @Joshua That's why you also put a swapfile on the compressed FS. I'm pretty sure that would work but I'm not sure. Also possibly zram: en.wikipedia.org/wiki/Zram | |
| Jan 15, 2016 at 18:09 | comment | added | Joshua | @shelvacu: The more immediate problem is the code causes the compiler to allocate 16TB RAM at once. | |
| Jan 15, 2016 at 18:08 | comment | added | Shelvacu | I would assume that the 16TB of data is very very repetitive, therefor you should be able to put both a swapfile and the output file on a compressed FS and then be able to do this, but I'm not sure. | |
| Jan 15, 2016 at 7:59 | comment | added | n̴̖̋h̷͉̃a̷̭̿h̸̡̅ẗ̵̨́d̷̰̀ĥ̷̳ | @Joshua: Check the markdown diff. Mego only added the highlighting hint. | |
| Jan 14, 2016 at 20:36 | comment | added | Joshua | In which case, what edit did you make to my code Mego? The diff says you edited the whole thing but I can't spot any differences. | |
| Jan 14, 2016 at 20:27 | comment | added | user45941 | @BenVoigt Regardless, editing other people's code is never acceptable here. Leave a comment if there's an issue. Relevant meta post: meta.codegolf.stackexchange.com/questions/1615/… | |
| Jan 14, 2016 at 20:06 | comment | added | Ben Voigt | @Joshua: In the future when you review a suggested edit, be sure to read the explanatory comment. | |
| Jan 14, 2016 at 19:50 | comment | added | Joshua | @Mego: See the revision history. | |
| Jan 14, 2016 at 19:40 | history | edited | user45941 | CC BY-SA 3.0 |
added 29 characters in body
|
| Jan 14, 2016 at 16:45 | history | edited | Joshua | CC BY-SA 3.0 |
sloppy rekey
|
| Jan 14, 2016 at 14:56 | comment | added | Ben Voigt | @Joshua: I wasn't golfing the entry, right now it is broken, because the first three lines define a variable, not a type. Then the next structure fails because its member uses an incomplete type. | |
| Jan 13, 2016 at 2:23 | comment | added | Joshua | Please do NOT attempt to golf this entry; golfing defeats the intent of the code sample and there are no score benefits to doing so anyway. Code is already GPL'd as of 2005. | |
| Jan 13, 2016 at 0:23 | review | Suggested edits | |||
| Jan 13, 2016 at 2:19 | |||||
| Jan 13, 2016 at 0:22 | comment | added | Ben Voigt | In my C++ answer, I wanted to avoid doing this because you run afoul of implementation limits on size of a single object, so my templates instead generate large numbers of more reasonably-sized objects. | |
| Jan 12, 2016 at 23:57 | comment | added | Joshua | I discovered this one by accident and watched the compiler topple over when it ran out of address space. | |
| Jan 12, 2016 at 23:45 | comment | added | Dave | "Would produce a 16TB binary if you had 16TB RAM (you don't)." - neither do I have a 16TB hard drive! I can't really verify this one, but it's cool nonetheless. | |
| Jan 12, 2016 at 20:27 | history | answered | Joshua | CC BY-SA 3.0 |