diff options
| author | Tom <tomut@yahoo.com> | 2020-09-04 21:12:25 -0600 |
|---|---|---|
| committer | Andreas Kling <kling@serenityos.org> | 2021-01-01 23:43:44 +0100 |
| commit | b2a52f62089e4799336cf0bf32eb307b077d2e72 (patch) | |
| tree | d0e3763a1a452357c76b5fe68fa3ab1b6bc83147 /Kernel/VM/VMObject.h | |
| parent | e21cc4cff63c36917d83730871dfff1a9f0eb927 (diff) | |
| download | serenity-b2a52f62089e4799336cf0bf32eb307b077d2e72.zip | |
Kernel: Implement lazy committed page allocation
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.
Diffstat (limited to 'Kernel/VM/VMObject.h')
| -rw-r--r-- | Kernel/VM/VMObject.h | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/Kernel/VM/VMObject.h b/Kernel/VM/VMObject.h index e126113f79..f7bedf0f4b 100644 --- a/Kernel/VM/VMObject.h +++ b/Kernel/VM/VMObject.h @@ -47,7 +47,7 @@ class VMObject : public RefCounted<VMObject> public: virtual ~VMObject(); - virtual NonnullRefPtr<VMObject> clone() = 0; + virtual RefPtr<VMObject> clone() = 0; virtual bool is_anonymous() const { return false; } virtual bool is_purgeable() const { return false; } |