Nice, an operating system that kills desktop X apps when they exceed 500MB and no mention in the install docs.
#OpenBSD
-
-
Yes, I meant swap. malloc is returning NULL if swapped? I forgot about limited 32bit addr. space. That's a likely case.
-
No, I just mean having (nontrivial amounts of) swap is a very bad idea. Instead of malloc failing & being able to report failure...
-
...you can get the system into a state where it would take weeks to successfully login & kill whatever allocated so much.
-
Yes, I've been in such situations. They aren't cool. What's suggested on Linux to do on malloc failure? In games it's easy, you often have a
-
Library level - back out the operation and return an error (you did remember not to commit/free anything before finishing alloc, right?).
-
Application level - depends on the app. Those with no valuable data of any sort can just try to show an error and exit...
-
But if you have any data the user might be upset to lose, you need at least an allocation-free emergency-save/recovery-dump code path.
-
Or you need to ensure that data is constantly saved in a form suitable for recovery after abort (see browsers' tab-restore features).
End of conversation
New conversation -
-
-
And virtual *address space* if that's what you meant is highly finite on 32 bit platforms or ilp32 abis, & will run out long before ram/swap
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.