This is a really great response. I appreciated the high-level overview in one place like this, and I feel like I learned something. Thanks for writing it up, Ian.
On Tue, May 16, 2017 at 10:05 AM, Ian Lance Taylor <i...@golang.org> wrote: > On Tue, May 16, 2017 at 2:01 AM, <canjian...@qq.com> wrote: > > > > Generational and Compact gc have already been thought best practice. But > > golang doesn't adopt it. Who can tell me the reason? > > This has been discussed in the past. > > Ignoring details, the basic advantages of a compacting GC are 1) avoid > fragmentation, and 2) permit the use of a simple and efficient bump > allocator. However, modern memory allocation algorithms, like the > tcmalloc-based approach used by the Go runtime, have essentially no > fragmentation issues. And while a bump allocator can be simple and > efficient for a single-threaded program, in a multi-threaded program > like Go it requires locks. In general it's likely to be more > efficient to allocate memory using a set of per-thread caches, and at > that point you've lost the advantages of a bump allocator. So I would > assert that, in general, with many caveats, there is no real advantage > to using a compacting memory allocator for a multi-threaded program > today. I don't mean that there is anything wrong with using a > compacting allocator, I'm just claiming that it doesn't bring any big > advantage over a non-compacting one. > > Now let's consider a generational GC. The point of a generational GC > relies on the generational hypothesis: that most values allocated in a > program are quickly unused, so there is an advantage for the GC to > spend more time looking at recently allocated objects. Here Go > differs from many garbage collected languages in that many objects are > allocated directly on the program stack. The Go compiler uses escape > analysis to find objects whose lifetime is known at compile time, and > allocates them on the stack rather than in garbage collected memory. > So in general, in Go, compared to other languages, a larger percentage > of the quickly-unused values that a generational GC looks for are > never allocated in GC memory in the first place. So a generational GC > would likely bring less advantage to Go than it does for other > languages. > > More subtly, the implicit point of most generational GC > implementations is to reduce the amount of time that a program pauses > for garbage collection. By looking at only the youngest generation > during a pause, the pause is kept short. However, Go uses a > concurrent garbage collector, and in Go the pause time is independent > of the size of the youngest generation, or of any generation. Go is > basically assuming that in a multi-threaded program it is better > overall to spend slightly more total CPU time on GC, by running GC in > parallel on a different core, rather than to minimize GC time but to > pause overall program execution for longer. > > All that said, generational GC could perhaps still bring significant > value to Go, by reducing the amount of work the GC has to do even in > parallel. It's a hypothesis that needs to be tested. Current GC work > in Go is actually looking closely at a related but different > hypothesis: that Go programs may tend to allocate memory on a > per-request basis. This is described at > https://docs.google.com/document/d/1gCsFxXamW8RRvOe5hECz98Ftk- > tcRRJcDFANj2VwCB0/view > . This is work in progress and it remains to be seen whether it will > be advantageous in reality. > > Ian > > -- > You received this message because you are subscribed to the Google Groups > "golang-nuts" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to golang-nuts+unsubscr...@googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > -- You received this message because you are subscribed to the Google Groups "golang-nuts" group. To unsubscribe from this group and stop receiving emails from it, send an email to golang-nuts+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
