Kernel+SystemMonitor: Add fault counters

This patch adds three separate per-process fault counters:

- Inode faults

    An inode fault happens when we've memory-mapped a file from disk
    and we end up having to load 1 page (4KB) of the file into memory.

- Zero faults

    Memory returned by mmap() is lazily zeroed out. Every time we have
    to zero out 1 page, we count a zero fault.

- CoW faults

    VM objects can be shared by multiple mappings that make their own
    unique copy iff they want to modify it. The typical reason here is
    memory shared between a parent and child process.

Kernel/FileSystem/ProcFS.cpp

@@ -685,6 +685,9 @@ Optional<KBuffer> procfs$all(InodeIdentifier)
         process_object.add("ticks", process.main_thread().ticks());
         process_object.add("priority", to_string(process.priority()));
         process_object.add("syscall_count", process.syscall_count());
+        process_object.add("inode_faults", process.inode_faults());
+        process_object.add("zero_faults", process.zero_faults());
+        process_object.add("cow_faults", process.cow_faults());
         process_object.add("icon_id", process.icon_id());
     };
     build_process(*Scheduler::colonel());

Kernel/Process.h

@@ -291,6 +291,12 @@ public:
 
     unsigned syscall_count() const { return m_syscall_count; }
     void did_syscall() { ++m_syscall_count; }
+    unsigned inode_faults() const { return m_inode_faults; }
+    void did_inode_fault() { ++m_inode_faults; }
+    unsigned zero_faults() const { return m_zero_faults; }
+    void did_zero_fault() { ++m_zero_faults; }
+    unsigned cow_faults() const { return m_cow_faults; }
+    void did_cow_fault() { ++m_cow_faults; }
 
     const ELFLoader* elf_loader() const { return m_elf_loader.ptr(); }
 
@@ -369,6 +375,9 @@ private:
     int m_next_tid { 0 };
 
     unsigned m_syscall_count { 0 };
+    unsigned m_inode_faults { 0 };
+    unsigned m_zero_faults { 0 };
+    unsigned m_cow_faults { 0 };
 
     RefPtr<ProcessTracer> m_tracer;
     OwnPtr<ELFLoader> m_elf_loader;

Kernel/VM/MemoryManager.cpp

@@ -296,6 +296,7 @@ bool MemoryManager::zero_page(Region& region, unsigned page_index_in_region)
     sti();
     LOCKER(vmo.m_paging_lock);
     cli();
+
     if (!vmo_page.is_null()) {
 #ifdef PAGE_FAULT_DEBUG
         dbgprintf("MM: zero_page() but page already present. Fine with me!\n");
@@ -303,6 +304,10 @@ bool MemoryManager::zero_page(Region& region, unsigned page_index_in_region)
         remap_region_page(region, page_index_in_region);
         return true;
     }
+
+    if (current)
+        current->process().did_zero_fault();
+
     auto physical_page = allocate_user_physical_page(ShouldZeroFill::Yes);
 #ifdef PAGE_FAULT_DEBUG
     dbgprintf("      >> ZERO P%p\n", physical_page->paddr().get());
@@ -325,6 +330,9 @@ bool MemoryManager::copy_on_write(Region& region, unsigned page_index_in_region)
         return true;
     }
 
+    if (current)
+        current->process().did_cow_fault();
+
 #ifdef PAGE_FAULT_DEBUG
     dbgprintf("    >> It's a COW page and it's time to COW!\n");
 #endif
@@ -367,6 +375,9 @@ bool MemoryManager::page_in_from_inode(Region& region, unsigned page_index_in_re
         return true;
     }
 
+    if (current)
+        current->process().did_inode_fault();
+
 #ifdef MM_DEBUG
     dbgprintf("MM: page_in_from_inode ready to read from inode\n");
 #endif