RGBCube/serenity
1
Fork 0
mirror of https://github.com/RGBCube/serenity synced 2025-07-25 05:17:35 +00:00
Code Issues Activity Actions

Kernel: The inode fault handler should grab the VMObject lock earlier

Browse source 
It doesn't look healthy to create raw references into an array before
a temporary unlock. In fact, that temporary unlock looks generally
unhealthy, but it's a different problem.
This commit is contained in:
Andreas Kling 2020-02-08 12:54:06 +01:00
parent a9d7902bb7
commit 00d8ec3ead
1 changed files with 3 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

5
Kernel/VM/Region.cpp
View file

@ -433,13 +433,14 @@ PageFaultResponse Region::handle_inode_fault(size_t page_index_in_region)
{ {
ASSERT_INTERRUPTS_DISABLED(); ASSERT_INTERRUPTS_DISABLED();
ASSERT(vmobject().is_inode()); ASSERT(vmobject().is_inode());
auto& inode_vmobject = static_cast<InodeVMObject&>(vmobject());
auto& vmobject_physical_page_entry = inode_vmobject.physical_pages()[first_page_index() + page_index_in_region];
sti(); sti();
LOCKER(vmobject().m_paging_lock); LOCKER(vmobject().m_paging_lock);
cli(); cli();
auto& inode_vmobject = static_cast<InodeVMObject&>(vmobject());
auto& vmobject_physical_page_entry = inode_vmobject.physical_pages()[first_page_index() + page_index_in_region];
#ifdef PAGE_FAULT_DEBUG #ifdef PAGE_FAULT_DEBUG
dbg() << "Inode fault in " << name() << " page index: " << page_index_in_region; dbg() << "Inode fault in " << name() << " page index: " << page_index_in_region;
#endif #endif