RGBCube/serenity
1
Fork 0
mirror of https://github.com/RGBCube/serenity synced 2025-07-25 16:47:36 +00:00
Code Issues Activity Actions

Kernel: Implement lazy committed page allocation

Browse source 
By designating a committed page pool we can guarantee to have physical
pages available for lazy allocation in mappings. However, when forking
we will overcommit. The assumption is that worst-case it's better for
the fork to die due to insufficient physical memory on COW access than
the parent that created the region. If a fork wants to ensure that all
memory is available (trigger a commit) then it can use madvise.

This also means that fork now can gracefully fail if we don't have
enough physical pages available.
This commit is contained in:
Tom 2020-09-04 21:12:25 -06:00 • committed by Andreas Kling
parent e21cc4cff6
commit b2a52f6208
20 changed files with 329 additions and 67 deletions
Download patch file Download diff file Expand all files Collapse all files

6
Kernel/VM/AnonymousVMObject.h
View file

@ -38,10 +38,12 @@ public:
static NonnullRefPtr<AnonymousVMObject> create_with_size(size_t);
static RefPtr<AnonymousVMObject> create_for_physical_range(PhysicalAddress, size_t);
static NonnullRefPtr<AnonymousVMObject> create_with_physical_page(PhysicalPage&);
virtual NonnullRefPtr<VMObject> clone() override;
virtual RefPtr<VMObject> clone() override;
virtual RefPtr<PhysicalPage> allocate_committed_page(size_t);
protected:
explicit AnonymousVMObject(size_t);
explicit AnonymousVMObject(size_t, bool initialize_pages = true);
explicit AnonymousVMObject(const AnonymousVMObject&);
virtual const char* class_name() const override { return "AnonymousVMObject"; }