News for hackers. Stuff that matters.

A comment on Why you can have millions of goroutines but only thousands of Java threads. Parent comment

Re:Why you can have millions of goroutines but only thousand... by timClicks on Monday October 08, 2018 @07:02PM (#18172070)
But wouldn't the consequence of spawning millions of native threads be to spill those virtual addresses into many, many pages and incur many more TLB misses and page faults generally?
[ Parent | on HN ]
Re:Why you can have millions of goroutines but only thousand... by nostrademons on Monday October 08, 2018 @07:45PM (#18172258)
Only the portions of the stack that are actually touched trigger a TLB miss and page fault. If you've got a million threads but each only touches the first 4K of stack space, you end up only touching 4G of RAM even though you've mapped 1T of it.
[ Parent | on HN ]
Re:Why you can have millions of goroutines but only thousand... by cryptonector on Tuesday October 09, 2018 @01:14AM (#18173493)
When you have terabytes of RAM you want large pages anyways.
[ Parent | on HN ]
Re:Why you can have millions of goroutines but only thousand... by gpderetta on Tuesday October 09, 2018 @03:40AM (#18174006)
As each stack will need to use [1] at least one page of physical memory, large pages are not great for stacks as they will normally waste memory.

[1] for classic contiguous stacks of course. That doesn't apply for separately allocated frames or segmented stacks.

[ Parent | on HN ]