diff options
| author | Andreas Kling <awesomekling@gmail.com> | 2019-11-03 23:54:35 +0100 |
|---|---|---|
| committer | Andreas Kling <awesomekling@gmail.com> | 2019-11-03 23:54:35 +0100 |
| commit | 98b328754ec8bd29bd495c6567343e2422a33fa5 (patch) | |
| tree | cf2daf1e4e4b2fb7934557145ab0e746497fb106 | |
| parent | 5b7f8634e3f62643fdfdb879f9fb3dc9163d40fe (diff) | |
| download | serenity-98b328754ec8bd29bd495c6567343e2422a33fa5.zip | |
Kernel: Fix bad setup of CoW faults for offset regions
Regions with an offset into their VMObject were incorrectly adding the
page offset when indexing into the CoW bitmap.
| -rw-r--r-- | Kernel/VM/MemoryManager.cpp | 3 |
1 files changed, 1 insertions, 2 deletions
diff --git a/Kernel/VM/MemoryManager.cpp b/Kernel/VM/MemoryManager.cpp index 1d7f87e8c3..f52e0c7ba5 100644 --- a/Kernel/VM/MemoryManager.cpp +++ b/Kernel/VM/MemoryManager.cpp @@ -697,8 +697,7 @@ void MemoryManager::map_region_at_address(PageDirectory& page_directory, Region& if (physical_page) { pte.set_physical_page_base(physical_page->paddr().get()); pte.set_present(true); // FIXME: Maybe we should use the is_readable flag here? - // FIXME: It seems wrong that the *region* cow map is essentially using *VMO* relative indices. - if (region.should_cow(region.first_page_index() + i)) + if (region.should_cow(i)) pte.set_writable(false); else pte.set_writable(region.is_writable()); |