RGBCube/serenity
1
Fork 0
mirror of https://github.com/RGBCube/serenity synced 2025-05-28 07:55:07 +00:00
Code Issues Activity Actions

Kernel: Start implementing purgeable memory support

Browse source 
It's now possible to get purgeable memory by using mmap(MAP_PURGEABLE).
Purgeable memory has a "volatile" flag that can be set using madvise():

- madvise(..., MADV_SET_VOLATILE)
- madvise(..., MADV_SET_NONVOLATILE)

When in the "volatile" state, the kernel may take away the underlying
physical memory pages at any time, without notifying the owner.
This gives you a guilt discount when caching very large things. :^)

Setting a purgeable region to non-volatile will return whether or not
the memory has been taken away by the kernel while being volatile.
Basically, if madvise(..., MADV_SET_NONVOLATILE) returns 1, that means
the memory was purged while volatile, and whatever was in that piece
of memory needs to be reconstructed before use.
This commit is contained in:
Andreas Kling 2019-12-09 19:12:38 +01:00
parent 7248c34e35
commit dbb644f20c
13 changed files with 196 additions and 9 deletions
Download patch file Download diff file Expand all files Collapse all files

7
Kernel/VM/Region.cpp
View file

@ -299,10 +299,11 @@ PageFaultResponse Region::handle_zero_fault(size_t page_index_in_region)
ASSERT_INTERRUPTS_DISABLED();
ASSERT(vmobject().is_anonymous());
auto& vmobject_physical_page_entry = vmobject().physical_pages()[first_page_index() + page_index_in_region];
sti();
LOCKER(vmobject().m_paging_lock);
cli();
// NOTE: We don't need to acquire the VMObject's lock.
// This function is already exclusive due to interrupts being blocked.
auto& vmobject_physical_page_entry = vmobject().physical_pages()[first_page_index() + page_index_in_region];
if (!vmobject_physical_page_entry.is_null()) {
#ifdef PAGE_FAULT_DEBUG