TIL mmap() with MAP_POPULATE and huge tlb will return a valid pointer and then SIGBUS you if you have non left in your NUMA node. Aaaaaargh!
-
-
Replying to @mattgodbolt
Is this intentional (eew) or a kernel bug that will be fixed?
1 reply 0 retweets 0 likes -
Replying to @RichFelker
very unclear. linus's comment is characteristically vague, but leaves thinking it's "working as intended"
1 reply 0 retweets 0 likes -
Replying to @mattgodbolt
Also, do you have overcommit disabled? If so it's surely a bug, regardless of whether he calls it one.
1 reply 0 retweets 0 likes -
Replying to @pkhuong @mattgodbolt
Regardless, signal after "successful" alloc is "not a bug" with overcommit on, but is a very serious bug with oc off.
4:38 AM - 24 Feb 2017
0 replies
0 retweets
0 likes
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.