Kernel: Make Kernel::VMObject allocation functions return KResultOr

This makes for nicer handling of errors compared to checking whether a
RefPtr is null. Additionally, this will give way to return different
types of errors in the future.

Kernel/Bus/USB/UHCIController.cpp

@@ -119,13 +119,17 @@ KResult UHCIController::reset()
     }
 
     // Let's allocate the physical page for the Frame List (which is 4KiB aligned)
-    auto framelist_vmobj = Memory::AnonymousVMObject::try_create_physically_contiguous_with_size(PAGE_SIZE);
-    m_framelist = MM.allocate_kernel_region_with_vmobject(*framelist_vmobj, PAGE_SIZE, "UHCI Framelist", Memory::Region::Access::Write);
+    auto maybe_framelist_vmobj = Memory::AnonymousVMObject::try_create_physically_contiguous_with_size(PAGE_SIZE);
+    if (maybe_framelist_vmobj.is_error())
+        return maybe_framelist_vmobj.error();
+
+    m_framelist = MM.allocate_kernel_region_with_vmobject(maybe_framelist_vmobj.release_value(), PAGE_SIZE, "UHCI Framelist", Memory::Region::Access::Write);
     dbgln("UHCI: Allocated framelist at physical address {}", m_framelist->physical_page(0)->paddr());
     dbgln("UHCI: Framelist is at virtual address {}", m_framelist->vaddr());
     write_sofmod(64); // 1mS frame time
 
-    create_structures();
+    if (auto result = create_structures(); result.is_error())
+        return result;
     setup_schedule();
 
     write_flbaseadd(m_framelist->physical_page(0)->paddr().get()); // Frame list (physical) address
@@ -139,12 +143,15 @@ KResult UHCIController::reset()
     return KSuccess;
 }
 
-UNMAP_AFTER_INIT void UHCIController::create_structures()
+UNMAP_AFTER_INIT KResult UHCIController::create_structures()
 {
     // Let's allocate memory for both the QH and TD pools
     // First the QH pool and all of the Interrupt QH's
-    auto qh_pool_vmobject = Memory::AnonymousVMObject::try_create_physically_contiguous_with_size(2 * PAGE_SIZE);
-    m_qh_pool = MM.allocate_kernel_region_with_vmobject(*qh_pool_vmobject, 2 * PAGE_SIZE, "UHCI Queue Head Pool", Memory::Region::Access::Write);
+    auto maybe_qh_pool_vmobject = Memory::AnonymousVMObject::try_create_physically_contiguous_with_size(2 * PAGE_SIZE);
+    if (maybe_qh_pool_vmobject.is_error())
+        return maybe_qh_pool_vmobject.error();
+
+    m_qh_pool = MM.allocate_kernel_region_with_vmobject(maybe_qh_pool_vmobject.release_value(), 2 * PAGE_SIZE, "UHCI Queue Head Pool", Memory::Region::Access::Write);
     memset(m_qh_pool->vaddr().as_ptr(), 0, 2 * PAGE_SIZE); // Zero out both pages
 
     // Let's populate our free qh list (so we have some we can allocate later on)
@@ -163,8 +170,10 @@ UNMAP_AFTER_INIT void UHCIController::create_structures()
     m_dummy_qh = allocate_queue_head();
 
     // Now the Transfer Descriptor pool
-    auto td_pool_vmobject = Memory::AnonymousVMObject::try_create_physically_contiguous_with_size(2 * PAGE_SIZE);
-    m_td_pool = MM.allocate_kernel_region_with_vmobject(*td_pool_vmobject, 2 * PAGE_SIZE, "UHCI Transfer Descriptor Pool", Memory::Region::Access::Write);
+    auto maybe_td_pool_vmobject = Memory::AnonymousVMObject::try_create_physically_contiguous_with_size(2 * PAGE_SIZE);
+    if (maybe_td_pool_vmobject.is_error())
+        return maybe_td_pool_vmobject.error();
+    m_td_pool = MM.allocate_kernel_region_with_vmobject(maybe_td_pool_vmobject.release_value(), 2 * PAGE_SIZE, "UHCI Transfer Descriptor Pool", Memory::Region::Access::Write);
     memset(m_td_pool->vaddr().as_ptr(), 0, 2 * PAGE_SIZE);
 
     // Set up the Isochronous Transfer Descriptor list
@@ -209,6 +218,8 @@ UNMAP_AFTER_INIT void UHCIController::create_structures()
         dbgln("    qh_pool: {}, length: {}", PhysicalAddress(m_qh_pool->physical_page(0)->paddr()), m_qh_pool->range().size());
         dbgln("    td_pool: {}, length: {}", PhysicalAddress(m_td_pool->physical_page(0)->paddr()), m_td_pool->range().size());
     }
+
+    return KSuccess;
 }
 
 UNMAP_AFTER_INIT void UHCIController::setup_schedule()

Kernel/Bus/USB/UHCIController.h

@@ -69,7 +69,7 @@ private:
 
     virtual bool handle_irq(const RegisterState&) override;
 
-    void create_structures();
+    KResult create_structures();
     void setup_schedule();
     size_t poll_transfer_queue(QueueHead& transfer_queue);
 

Kernel/Devices/KCOVInstance.cpp

@@ -28,10 +28,11 @@ KResult KCOVInstance::buffer_allocate(size_t buffer_size_in_entries)
     // - we allocate one kernel region using that vmobject
     // - when an mmap call comes in, we allocate another userspace region,
     //   backed by the same vmobject
-    this->vmobject = Memory::AnonymousVMObject::try_create_with_size(
+    auto maybe_vmobject = Memory::AnonymousVMObject::try_create_with_size(
         this->m_buffer_size_in_bytes, AllocationStrategy::AllocateNow);
-    if (!this->vmobject)
-        return ENOMEM;
+    if (maybe_vmobject.is_error())
+        return maybe_vmobject.error();
+    this->vmobject = maybe_vmobject.release_value();
 
     this->m_kernel_region = MM.allocate_kernel_region_with_vmobject(
         *this->vmobject, this->m_buffer_size_in_bytes, String::formatted("kcov_{}", this->m_pid),

Kernel/Devices/MemoryDevice.cpp

@@ -47,13 +47,14 @@ KResultOr<Memory::Region*> MemoryDevice::mmap(Process& process, FileDescription&
         return EINVAL;
     }
 
-    auto vmobject = Memory::AnonymousVMObject::try_create_for_physical_range(viewed_address, range.size());
-    if (!vmobject)
-        return ENOMEM;
+    auto maybe_vmobject = Memory::AnonymousVMObject::try_create_for_physical_range(viewed_address, range.size());
+    if (maybe_vmobject.is_error())
+        return maybe_vmobject.error();
+
     dbgln("MemoryDevice: Mapped physical memory at {} for range of {} bytes", viewed_address, range.size());
     return process.address_space().allocate_region_with_vmobject(
         range,
-        vmobject.release_nonnull(),
+        maybe_vmobject.release_value(),
         0,
         "Mapped Physical Memory",
         prot,

Kernel/Devices/SB16.cpp

@@ -238,10 +238,11 @@ KResultOr<size_t> SB16::write(FileDescription&, u64, const UserOrKernelBuffer& d
         if (!page)
             return ENOMEM;
         auto nonnull_page = page.release_nonnull();
-        auto vmobject = Memory::AnonymousVMObject::try_create_with_physical_pages({ &nonnull_page, 1 });
-        if (!vmobject)
-            return ENOMEM;
-        m_dma_region = MM.allocate_kernel_region_with_vmobject(*vmobject, PAGE_SIZE, "SB16 DMA buffer", Memory::Region::Access::Write);
+        auto maybe_vmobject = Memory::AnonymousVMObject::try_create_with_physical_pages({ &nonnull_page, 1 });
+        if (maybe_vmobject.is_error())
+            return maybe_vmobject.error();
+
+        m_dma_region = MM.allocate_kernel_region_with_vmobject(maybe_vmobject.release_value(), PAGE_SIZE, "SB16 DMA buffer", Memory::Region::Access::Write);
         if (!m_dma_region)
             return ENOMEM;
     }

Kernel/Graphics/Bochs/GraphicsAdapter.cpp

@@ -88,7 +88,8 @@ UNMAP_AFTER_INIT void BochsGraphicsAdapter::initialize_framebuffer_devices()
 {
     // FIXME: Find a better way to determine default resolution...
     m_framebuffer_device = FramebufferDevice::create(*this, 0, PhysicalAddress(PCI::get_BAR0(pci_address()) & 0xfffffff0), 1024, 768, 1024 * sizeof(u32));
-    m_framebuffer_device->initialize();
+    // FIXME: Would be nice to be able to return a KResult here.
+    VERIFY(!m_framebuffer_device->initialize().is_error());
 }
 
 GraphicsDevice::Type BochsGraphicsAdapter::type() const

Kernel/Graphics/FramebufferDevice.cpp

@@ -36,18 +36,20 @@ KResultOr<Memory::Region*> FramebufferDevice::mmap(Process& process, FileDescrip
     if (range.size() != Memory::page_round_up(framebuffer_size_in_bytes()))
         return EOVERFLOW;
 
-    auto vmobject = Memory::AnonymousVMObject::try_create_for_physical_range(m_framebuffer_address, Memory::page_round_up(framebuffer_size_in_bytes()));
-    if (!vmobject)
-        return ENOMEM;
-    m_userspace_real_framebuffer_vmobject = vmobject;
+    auto maybe_vmobject = Memory::AnonymousVMObject::try_create_for_physical_range(m_framebuffer_address, Memory::page_round_up(framebuffer_size_in_bytes()));
+    if (maybe_vmobject.is_error())
+        return maybe_vmobject.error();
+    m_userspace_real_framebuffer_vmobject = maybe_vmobject.release_value();
 
-    m_real_framebuffer_vmobject = Memory::AnonymousVMObject::try_create_for_physical_range(m_framebuffer_address, Memory::page_round_up(framebuffer_size_in_bytes()));
-    if (!m_real_framebuffer_vmobject)
-        return ENOMEM;
+    auto maybe_real_framebuffer_vmobject = Memory::AnonymousVMObject::try_create_for_physical_range(m_framebuffer_address, Memory::page_round_up(framebuffer_size_in_bytes()));
+    if (maybe_real_framebuffer_vmobject.is_error())
+        return maybe_real_framebuffer_vmobject.error();
+    m_real_framebuffer_vmobject = maybe_real_framebuffer_vmobject.release_value();
 
-    m_swapped_framebuffer_vmobject = Memory::AnonymousVMObject::try_create_with_size(Memory::page_round_up(framebuffer_size_in_bytes()), AllocationStrategy::AllocateNow);
-    if (!m_swapped_framebuffer_vmobject)
-        return ENOMEM;
+    auto maybe_swapped_framebuffer_vmobject = Memory::AnonymousVMObject::try_create_with_size(Memory::page_round_up(framebuffer_size_in_bytes()), AllocationStrategy::AllocateNow);
+    if (maybe_swapped_framebuffer_vmobject.is_error())
+        return maybe_swapped_framebuffer_vmobject.error();
+    m_swapped_framebuffer_vmobject = maybe_swapped_framebuffer_vmobject.release_value();
 
     m_real_framebuffer_region = MM.allocate_kernel_region_with_vmobject(*m_real_framebuffer_vmobject, Memory::page_round_up(framebuffer_size_in_bytes()), "Framebuffer", Memory::Region::Access::ReadWrite);
     if (!m_real_framebuffer_region)
@@ -107,16 +109,29 @@ String FramebufferDevice::device_name() const
     return String::formatted("fb{}", minor());
 }
 
-UNMAP_AFTER_INIT void FramebufferDevice::initialize()
+UNMAP_AFTER_INIT KResult FramebufferDevice::initialize()
 {
-    m_real_framebuffer_vmobject = Memory::AnonymousVMObject::try_create_for_physical_range(m_framebuffer_address, Memory::page_round_up(framebuffer_size_in_bytes()));
-    VERIFY(m_real_framebuffer_vmobject);
+    // FIXME: Would be nice to be able to unify this with mmap above, but this
+    //        function is UNMAP_AFTER_INIT for the time being.
+    auto maybe_real_framebuffer_vmobject = Memory::AnonymousVMObject::try_create_for_physical_range(m_framebuffer_address, Memory::page_round_up(framebuffer_size_in_bytes()));
+    if (maybe_real_framebuffer_vmobject.is_error())
+        return maybe_real_framebuffer_vmobject.error();
+    m_real_framebuffer_vmobject = maybe_real_framebuffer_vmobject.release_value();
+
+    auto maybe_swapped_framebuffer_vmobject = Memory::AnonymousVMObject::try_create_with_size(Memory::page_round_up(framebuffer_size_in_bytes()), AllocationStrategy::AllocateNow);
+    if (maybe_swapped_framebuffer_vmobject.is_error())
+        return maybe_swapped_framebuffer_vmobject.error();
+    m_swapped_framebuffer_vmobject = maybe_swapped_framebuffer_vmobject.release_value();
+
     m_real_framebuffer_region = MM.allocate_kernel_region_with_vmobject(*m_real_framebuffer_vmobject, Memory::page_round_up(framebuffer_size_in_bytes()), "Framebuffer", Memory::Region::Access::ReadWrite);
-    VERIFY(m_real_framebuffer_region);
-    m_swapped_framebuffer_vmobject = Memory::AnonymousVMObject::try_create_with_size(Memory::page_round_up(framebuffer_size_in_bytes()), AllocationStrategy::AllocateNow);
-    VERIFY(m_swapped_framebuffer_vmobject);
+    if (!m_real_framebuffer_region)
+        return ENOMEM;
+
     m_swapped_framebuffer_region = MM.allocate_kernel_region_with_vmobject(*m_swapped_framebuffer_vmobject, Memory::page_round_up(framebuffer_size_in_bytes()), "Framebuffer Swap (Blank)", Memory::Region::Access::ReadWrite);
-    VERIFY(m_swapped_framebuffer_region);
+    if (!m_swapped_framebuffer_region)
+        return ENOMEM;
+
+    return KSuccess;
 }
 
 UNMAP_AFTER_INIT FramebufferDevice::FramebufferDevice(const GraphicsDevice& adapter, size_t output_port_index, PhysicalAddress addr, size_t width, size_t height, size_t pitch)

Kernel/Graphics/FramebufferDevice.h

@@ -34,7 +34,7 @@ public:
     size_t framebuffer_size_in_bytes() const;
 
     virtual ~FramebufferDevice() {};
-    void initialize();
+    KResult initialize();
 
 private:
     // ^File

Kernel/Graphics/Intel/NativeGraphicsAdapter.cpp

@@ -639,6 +639,7 @@ void IntelNativeGraphicsAdapter::initialize_framebuffer_devices()
     VERIFY(m_framebuffer_height != 0);
     VERIFY(m_framebuffer_width != 0);
     m_framebuffer_device = FramebufferDevice::create(*this, 0, address, m_framebuffer_width, m_framebuffer_height, m_framebuffer_pitch);
-    m_framebuffer_device->initialize();
+    // FIXME: Would be nice to be able to return a KResult here.
+    VERIFY(!m_framebuffer_device->initialize().is_error());
 }
 }

Kernel/Graphics/VGACompatibleAdapter.cpp

@@ -33,7 +33,8 @@ UNMAP_AFTER_INIT void VGACompatibleAdapter::initialize_framebuffer_devices()
     VERIFY(m_framebuffer_height != 0);
     VERIFY(m_framebuffer_pitch != 0);
     m_framebuffer_device = FramebufferDevice::create(*this, 0, m_framebuffer_address, m_framebuffer_width, m_framebuffer_height, m_framebuffer_pitch);
-    m_framebuffer_device->initialize();
+    // FIXME: Would be nice to be able to return KResult here.
+    VERIFY(!m_framebuffer_device->initialize().is_error());
 }
 
 UNMAP_AFTER_INIT VGACompatibleAdapter::VGACompatibleAdapter(PCI::Address address)

Kernel/Graphics/VirtIOGPU/FrameBufferDevice.cpp

@@ -15,15 +15,18 @@ FrameBufferDevice::FrameBufferDevice(GPU& virtio_gpu, ScanoutID scanout)
     , m_gpu(virtio_gpu)
     , m_scanout(scanout)
 {
-    if (display_info().enabled)
-        create_framebuffer();
+    if (display_info().enabled) {
+        // FIXME: This should be in a place where we can handle allocation failures.
+        auto result = create_framebuffer();
+        VERIFY(!result.is_error());
+    }
 }
 
 FrameBufferDevice::~FrameBufferDevice()
 {
 }
 
-void FrameBufferDevice::create_framebuffer()
+KResult FrameBufferDevice::create_framebuffer()
 {
     // First delete any existing framebuffers to free the memory first
     m_framebuffer = nullptr;
@@ -40,12 +43,17 @@ void FrameBufferDevice::create_framebuffer()
     for (auto i = 0u; i < num_needed_pages; ++i) {
         pages.append(write_sink_page);
     }
-    m_framebuffer_sink_vmobject = Memory::AnonymousVMObject::try_create_with_physical_pages(pages.span());
+    auto maybe_framebuffer_sink_vmobject = Memory::AnonymousVMObject::try_create_with_physical_pages(pages.span());
+    if (maybe_framebuffer_sink_vmobject.is_error())
+        return maybe_framebuffer_sink_vmobject.error();
+    m_framebuffer_sink_vmobject = maybe_framebuffer_sink_vmobject.release_value();
 
     MutexLocker locker(m_gpu.operation_lock());
     m_current_buffer = &buffer_from_index(m_last_set_buffer_index.load());
     create_buffer(m_main_buffer, 0, m_buffer_size);
     create_buffer(m_back_buffer, m_buffer_size, m_buffer_size);
+
+    return KSuccess;
 }
 
 void FrameBufferDevice::create_buffer(Buffer& buffer, size_t framebuffer_offset, size_t framebuffer_size)
@@ -124,7 +132,10 @@ bool FrameBufferDevice::try_to_set_resolution(size_t width, size_t height)
         .width = (u32)width,
         .height = (u32)height,
     };
-    create_framebuffer();
+
+    // FIXME: Would be nice to be able to return KResultOr here.
+    if (auto result = create_framebuffer(); result.is_error())
+        return false;
     return true;
 }
 
@@ -255,9 +266,18 @@ KResultOr<Memory::Region*> FrameBufferDevice::mmap(Process& process, FileDescrip
     if (m_userspace_mmap_region)
         return ENOMEM;
 
-    auto vmobject = m_are_writes_active ? m_framebuffer->vmobject().try_clone() : m_framebuffer_sink_vmobject;
+    RefPtr<Memory::VMObject> vmobject;
+    if (m_are_writes_active) {
+        auto maybe_vmobject = m_framebuffer->vmobject().try_clone();
+        if (maybe_vmobject.is_error())
+            return maybe_vmobject.error();
+
+        vmobject = maybe_vmobject.release_value();
+    } else {
+        vmobject = m_framebuffer_sink_vmobject;
         if (vmobject.is_null())
             return ENOMEM;
+    }
 
     auto result = process.address_space().allocate_region_with_vmobject(
         range,
@@ -277,9 +297,10 @@ void FrameBufferDevice::deactivate_writes()
     m_are_writes_active = false;
     if (m_userspace_mmap_region) {
         auto* region = m_userspace_mmap_region.unsafe_ptr();
-        auto vm_object = m_framebuffer_sink_vmobject->try_clone();
-        VERIFY(vm_object);
-        region->set_vmobject(vm_object.release_nonnull());
+        auto maybe_vm_object = m_framebuffer_sink_vmobject->try_clone();
+        // FIXME: Would be nice to be able to return a KResult here.
+        VERIFY(!maybe_vm_object.is_error());
+        region->set_vmobject(maybe_vm_object.release_value());
         region->remap();
     }
     set_buffer(0);

Kernel/Graphics/VirtIOGPU/FrameBufferDevice.h

@@ -56,7 +56,7 @@ private:
     Protocol::DisplayInfoResponse::Display const& display_info() const;
     Protocol::DisplayInfoResponse::Display& display_info();
 
-    void create_framebuffer();
+    KResult create_framebuffer();
     void create_buffer(Buffer&, size_t, size_t);
     void set_buffer(int);
 

Kernel/Interrupts/APIC.cpp

@@ -277,11 +277,13 @@ UNMAP_AFTER_INIT bool APIC::init_bsp()
 
 UNMAP_AFTER_INIT static NonnullOwnPtr<Memory::Region> create_identity_mapped_region(PhysicalAddress paddr, size_t size)
 {
-    auto vmobject = Memory::AnonymousVMObject::try_create_for_physical_range(paddr, size);
-    VERIFY(vmobject);
+    auto maybe_vmobject = Memory::AnonymousVMObject::try_create_for_physical_range(paddr, size);
+    // FIXME: Would be nice to be able to return a KResultOr from here.
+    VERIFY(!maybe_vmobject.is_error());
+
     auto region = MM.allocate_kernel_region_with_vmobject(
         Memory::VirtualRange { VirtualAddress { static_cast<FlatPtr>(paddr.get()) }, size },
-        vmobject.release_nonnull(),
+        maybe_vmobject.release_value(),
         {},
         Memory::Region::Access::ReadWriteExecute);
     VERIFY(region);

Kernel/Memory/AddressSpace.cpp

@@ -170,10 +170,10 @@ KResultOr<Region*> AddressSpace::try_allocate_split_region(Region const& source_
 KResultOr<Region*> AddressSpace::allocate_region(VirtualRange const& range, StringView name, int prot, AllocationStrategy strategy)
 {
     VERIFY(range.is_valid());
-    auto vmobject = AnonymousVMObject::try_create_with_size(range.size(), strategy);
-    if (!vmobject)
-        return ENOMEM;
-    auto region = Region::try_create_user_accessible(range, vmobject.release_nonnull(), 0, KString::try_create(name), prot_to_region_access_flags(prot), Region::Cacheable::Yes, false);
+    auto maybe_vmobject = AnonymousVMObject::try_create_with_size(range.size(), strategy);
+    if (maybe_vmobject.is_error())
+        return maybe_vmobject.error();
+    auto region = Region::try_create_user_accessible(range, maybe_vmobject.release_value(), 0, KString::try_create(name), prot_to_region_access_flags(prot), Region::Cacheable::Yes, false);
     if (!region)
         return ENOMEM;
     if (!region->map(page_directory()))

Kernel/Memory/AnonymousVMObject.cpp

@@ -13,7 +13,7 @@
 
 namespace Kernel::Memory {
 
-RefPtr<VMObject> AnonymousVMObject::try_clone()
+KResultOr<NonnullRefPtr<VMObject>> AnonymousVMObject::try_clone()
 {
     // We need to acquire our lock so we copy a sane state
     ScopedSpinLock lock(m_lock);
@@ -21,9 +21,11 @@ RefPtr<VMObject> AnonymousVMObject::try_clone()
     if (is_purgeable() && is_volatile()) {
         // If this object is purgeable+volatile, create a new zero-filled purgeable+volatile
         // object, effectively "pre-purging" it in the child process.
-        auto clone = try_create_purgeable_with_size(size(), AllocationStrategy::None);
-        if (!clone)
-            return {};
+        auto maybe_clone = try_create_purgeable_with_size(size(), AllocationStrategy::None);
+        if (maybe_clone.is_error())
+            return maybe_clone.error();
+
+        auto clone = maybe_clone.release_value();
         clone->m_volatile = true;
         return clone;
     }
@@ -38,7 +40,7 @@ RefPtr<VMObject> AnonymousVMObject::try_clone()
 
     auto committed_pages = MM.commit_user_physical_pages(new_cow_pages_needed);
     if (!committed_pages.has_value())
-        return {};
+        return ENOMEM;
 
     // Create or replace the committed cow pages. When cloning a previously
     // cloned vmobject, we want to essentially "fork", leaving us and the
@@ -49,11 +51,12 @@ RefPtr<VMObject> AnonymousVMObject::try_clone()
     auto new_shared_committed_cow_pages = try_create<SharedCommittedCowPages>(committed_pages.release_value());
 
     if (!new_shared_committed_cow_pages)
-        return {};
+        return ENOMEM;
 
-    auto clone = adopt_ref_if_nonnull(new (nothrow) AnonymousVMObject(*this, *new_shared_committed_cow_pages));
-    if (!clone)
-        return {};
+    auto maybe_clone = adopt_ref_if_nonnull(new (nothrow) AnonymousVMObject(*this, new_shared_committed_cow_pages.release_nonnull()));
+    if (!maybe_clone)
+        return ENOMEM;
+    auto clone = maybe_clone.release_nonnull();
 
     m_shared_committed_cow_pages = move(new_shared_committed_cow_pages);
 
@@ -76,52 +79,58 @@ RefPtr<VMObject> AnonymousVMObject::try_clone()
     return clone;
 }
 
-RefPtr<AnonymousVMObject> AnonymousVMObject::try_create_with_size(size_t size, AllocationStrategy strategy)
+KResultOr<NonnullRefPtr<AnonymousVMObject>> AnonymousVMObject::try_create_with_size(size_t size, AllocationStrategy strategy)
 {
     Optional<CommittedPhysicalPageSet> committed_pages;
     if (strategy == AllocationStrategy::Reserve || strategy == AllocationStrategy::AllocateNow) {
         committed_pages = MM.commit_user_physical_pages(ceil_div(size, static_cast<size_t>(PAGE_SIZE)));
         if (!committed_pages.has_value())
-            return {};
+            return ENOMEM;
     }
-    return adopt_ref_if_nonnull(new (nothrow) AnonymousVMObject(size, strategy, move(committed_pages)));
+
+    return adopt_nonnull_ref_or_enomem(new (nothrow) AnonymousVMObject(size, strategy, move(committed_pages)));
 }
 
-RefPtr<AnonymousVMObject> AnonymousVMObject::try_create_physically_contiguous_with_size(size_t size)
+KResultOr<NonnullRefPtr<AnonymousVMObject>> AnonymousVMObject::try_create_physically_contiguous_with_size(size_t size)
 {
     auto contiguous_physical_pages = MM.allocate_contiguous_supervisor_physical_pages(size);
     if (contiguous_physical_pages.is_empty())
-        return {};
-    return adopt_ref_if_nonnull(new (nothrow) AnonymousVMObject(contiguous_physical_pages.span()));
+        return ENOMEM;
+
+    return adopt_nonnull_ref_or_enomem(new (nothrow) AnonymousVMObject(contiguous_physical_pages.span()));
 }
 
-RefPtr<AnonymousVMObject> AnonymousVMObject::try_create_purgeable_with_size(size_t size, AllocationStrategy strategy)
+KResultOr<NonnullRefPtr<AnonymousVMObject>> AnonymousVMObject::try_create_purgeable_with_size(size_t size, AllocationStrategy strategy)
 {
     Optional<CommittedPhysicalPageSet> committed_pages;
     if (strategy == AllocationStrategy::Reserve || strategy == AllocationStrategy::AllocateNow) {
         committed_pages = MM.commit_user_physical_pages(ceil_div(size, static_cast<size_t>(PAGE_SIZE)));
         if (!committed_pages.has_value())
-            return {};
+            return ENOMEM;
     }
+
     auto vmobject = adopt_ref_if_nonnull(new (nothrow) AnonymousVMObject(size, strategy, move(committed_pages)));
     if (!vmobject)
-        return {};
+        return ENOMEM;
+
     vmobject->m_purgeable = true;
-    return vmobject;
+    return vmobject.release_nonnull();
 }
 
-RefPtr<AnonymousVMObject> AnonymousVMObject::try_create_with_physical_pages(Span<NonnullRefPtr<PhysicalPage>> physical_pages)
+KResultOr<NonnullRefPtr<AnonymousVMObject>> AnonymousVMObject::try_create_with_physical_pages(Span<NonnullRefPtr<PhysicalPage>> physical_pages)
 {
-    return adopt_ref_if_nonnull(new (nothrow) AnonymousVMObject(physical_pages));
+    return adopt_nonnull_ref_or_enomem(new (nothrow) AnonymousVMObject(physical_pages));
 }
 
-RefPtr<AnonymousVMObject> AnonymousVMObject::try_create_for_physical_range(PhysicalAddress paddr, size_t size)
+KResultOr<NonnullRefPtr<AnonymousVMObject>> AnonymousVMObject::try_create_for_physical_range(PhysicalAddress paddr, size_t size)
 {
     if (paddr.offset(size) < paddr) {
         dbgln("Shenanigans! try_create_for_physical_range({}, {}) would wrap around", paddr, size);
-        return nullptr;
+        // Since we can't wrap around yet, let's pretend to OOM.
+        return ENOMEM;
     }
-    return adopt_ref_if_nonnull(new (nothrow) AnonymousVMObject(paddr, size));
+
+    return adopt_nonnull_ref_or_enomem(new (nothrow) AnonymousVMObject(paddr, size));
 }
 
 AnonymousVMObject::AnonymousVMObject(size_t size, AllocationStrategy strategy, Optional<CommittedPhysicalPageSet> committed_pages)

Kernel/Memory/AnonymousVMObject.h

@@ -18,12 +18,12 @@ class AnonymousVMObject final : public VMObject {
 public:
     virtual ~AnonymousVMObject() override;
 
-    static RefPtr<AnonymousVMObject> try_create_with_size(size_t, AllocationStrategy);
-    static RefPtr<AnonymousVMObject> try_create_for_physical_range(PhysicalAddress paddr, size_t size);
-    static RefPtr<AnonymousVMObject> try_create_with_physical_pages(Span<NonnullRefPtr<PhysicalPage>>);
-    static RefPtr<AnonymousVMObject> try_create_purgeable_with_size(size_t, AllocationStrategy);
-    static RefPtr<AnonymousVMObject> try_create_physically_contiguous_with_size(size_t);
-    virtual RefPtr<VMObject> try_clone() override;
+    static KResultOr<NonnullRefPtr<AnonymousVMObject>> try_create_with_size(size_t, AllocationStrategy);
+    static KResultOr<NonnullRefPtr<AnonymousVMObject>> try_create_for_physical_range(PhysicalAddress paddr, size_t size);
+    static KResultOr<NonnullRefPtr<AnonymousVMObject>> try_create_with_physical_pages(Span<NonnullRefPtr<PhysicalPage>>);
+    static KResultOr<NonnullRefPtr<AnonymousVMObject>> try_create_purgeable_with_size(size_t, AllocationStrategy);
+    static KResultOr<NonnullRefPtr<AnonymousVMObject>> try_create_physically_contiguous_with_size(size_t);
+    virtual KResultOr<NonnullRefPtr<VMObject>> try_clone() override;
 
     [[nodiscard]] NonnullRefPtr<PhysicalPage> allocate_committed_page(Badge<Region>);
     PageFaultResponse handle_cow_fault(size_t, VirtualAddress);

Kernel/Memory/MemoryManager.cpp

@@ -706,38 +706,39 @@ OwnPtr<Region> MemoryManager::allocate_contiguous_kernel_region(size_t size, Str
     auto range = kernel_page_directory().range_allocator().allocate_anywhere(size);
     if (!range.has_value())
         return {};
-    auto vmobject = AnonymousVMObject::try_create_physically_contiguous_with_size(size);
-    if (!vmobject) {
+    auto maybe_vmobject = AnonymousVMObject::try_create_physically_contiguous_with_size(size);
+    if (maybe_vmobject.is_error()) {
         kernel_page_directory().range_allocator().deallocate(range.value());
+        // FIXME: Would be nice to be able to return a KResultOr from here.
         return {};
     }
-    return allocate_kernel_region_with_vmobject(range.value(), *vmobject, name, access, cacheable);
+    return allocate_kernel_region_with_vmobject(range.value(), maybe_vmobject.release_value(), name, access, cacheable);
 }
 
 OwnPtr<Region> MemoryManager::allocate_kernel_region(size_t size, StringView name, Region::Access access, AllocationStrategy strategy, Region::Cacheable cacheable)
 {
     VERIFY(!(size % PAGE_SIZE));
-    auto vm_object = AnonymousVMObject::try_create_with_size(size, strategy);
-    if (!vm_object)
+    auto maybe_vm_object = AnonymousVMObject::try_create_with_size(size, strategy);
+    if (maybe_vm_object.is_error())
         return {};
     ScopedSpinLock lock(kernel_page_directory().get_lock());
     auto range = kernel_page_directory().range_allocator().allocate_anywhere(size);
     if (!range.has_value())
         return {};
-    return allocate_kernel_region_with_vmobject(range.value(), vm_object.release_nonnull(), name, access, cacheable);
+    return allocate_kernel_region_with_vmobject(range.value(), maybe_vm_object.release_value(), name, access, cacheable);
 }
 
 OwnPtr<Region> MemoryManager::allocate_kernel_region(PhysicalAddress paddr, size_t size, StringView name, Region::Access access, Region::Cacheable cacheable)
 {
-    auto vm_object = AnonymousVMObject::try_create_for_physical_range(paddr, size);
-    if (!vm_object)
+    auto maybe_vm_object = AnonymousVMObject::try_create_for_physical_range(paddr, size);
+    if (maybe_vm_object.is_error())
         return {};
     VERIFY(!(size % PAGE_SIZE));
     ScopedSpinLock lock(kernel_page_directory().get_lock());
     auto range = kernel_page_directory().range_allocator().allocate_anywhere(size);
     if (!range.has_value())
         return {};
-    return allocate_kernel_region_with_vmobject(range.value(), *vm_object, name, access, cacheable);
+    return allocate_kernel_region_with_vmobject(range.value(), maybe_vm_object.release_value(), name, access, cacheable);
 }
 
 OwnPtr<Region> MemoryManager::allocate_kernel_region_with_vmobject(VirtualRange const& range, VMObject& vmobject, StringView name, Region::Access access, Region::Cacheable cacheable)

Kernel/Memory/PrivateInodeVMObject.cpp

@@ -14,9 +14,9 @@ RefPtr<PrivateInodeVMObject> PrivateInodeVMObject::try_create_with_inode(Inode&
     return adopt_ref_if_nonnull(new (nothrow) PrivateInodeVMObject(inode, inode.size()));
 }
 
-RefPtr<VMObject> PrivateInodeVMObject::try_clone()
+KResultOr<NonnullRefPtr<VMObject>> PrivateInodeVMObject::try_clone()
 {
-    return adopt_ref_if_nonnull(new (nothrow) PrivateInodeVMObject(*this));
+    return adopt_nonnull_ref_or_enomem<VMObject>(new (nothrow) PrivateInodeVMObject(*this));
 }
 
 PrivateInodeVMObject::PrivateInodeVMObject(Inode& inode, size_t size)

Kernel/Memory/PrivateInodeVMObject.h

@@ -19,7 +19,7 @@ public:
     virtual ~PrivateInodeVMObject() override;
 
     static RefPtr<PrivateInodeVMObject> try_create_with_inode(Inode&);
-    virtual RefPtr<VMObject> try_clone() override;
+    virtual KResultOr<NonnullRefPtr<VMObject>> try_clone() override;
 
 private:
     virtual bool is_private_inode() const override { return true; }

Kernel/Memory/Region.cpp

@@ -75,14 +75,14 @@ OwnPtr<Region> Region::clone()
     if (vmobject().is_inode())
         VERIFY(vmobject().is_private_inode());
 
-    auto vmobject_clone = vmobject().try_clone();
-    if (!vmobject_clone)
+    auto maybe_vmobject_clone = vmobject().try_clone();
+    if (maybe_vmobject_clone.is_error())
         return {};
 
     // Set up a COW region. The parent (this) region becomes COW as well!
     remap();
     auto clone_region = Region::try_create_user_accessible(
-        m_range, vmobject_clone.release_nonnull(), m_offset_in_vmobject, m_name ? m_name->try_clone() : OwnPtr<KString> {}, access(), m_cacheable ? Cacheable::Yes : Cacheable::No, m_shared);
+        m_range, maybe_vmobject_clone.release_value(), m_offset_in_vmobject, m_name ? m_name->try_clone() : OwnPtr<KString> {}, access(), m_cacheable ? Cacheable::Yes : Cacheable::No, m_shared);
     if (!clone_region) {
         dbgln("Region::clone: Unable to allocate new Region for COW");
         return nullptr;

Kernel/Memory/ScatterGatherList.cpp

@@ -10,10 +10,12 @@ namespace Kernel::Memory {
 
 RefPtr<ScatterGatherList> ScatterGatherList::try_create(AsyncBlockDeviceRequest& request, Span<NonnullRefPtr<PhysicalPage>> allocated_pages, size_t device_block_size)
 {
-    auto vm_object = AnonymousVMObject::try_create_with_physical_pages(allocated_pages);
-    if (!vm_object)
+    auto maybe_vm_object = AnonymousVMObject::try_create_with_physical_pages(allocated_pages);
+    if (maybe_vm_object.is_error()) {
+        // FIXME: Would be nice to be able to return a KResultOr here.
         return {};
-    return adopt_ref_if_nonnull(new (nothrow) ScatterGatherList(vm_object.release_nonnull(), request, device_block_size));
+    }
+    return adopt_ref_if_nonnull(new (nothrow) ScatterGatherList(maybe_vm_object.release_value(), request, device_block_size));
 }
 
 ScatterGatherList::ScatterGatherList(NonnullRefPtr<AnonymousVMObject> vm_object, AsyncBlockDeviceRequest& request, size_t device_block_size)

Kernel/Memory/SharedInodeVMObject.cpp

@@ -21,9 +21,9 @@ RefPtr<SharedInodeVMObject> SharedInodeVMObject::try_create_with_inode(Inode& in
     return vmobject;
 }
 
-RefPtr<VMObject> SharedInodeVMObject::try_clone()
+KResultOr<NonnullRefPtr<VMObject>> SharedInodeVMObject::try_clone()
 {
-    return adopt_ref_if_nonnull(new (nothrow) SharedInodeVMObject(*this));
+    return adopt_nonnull_ref_or_enomem<VMObject>(new (nothrow) SharedInodeVMObject(*this));
 }
 
 SharedInodeVMObject::SharedInodeVMObject(Inode& inode, size_t size)

Kernel/Memory/SharedInodeVMObject.h

@@ -17,7 +17,7 @@ class SharedInodeVMObject final : public InodeVMObject {
 
 public:
     static RefPtr<SharedInodeVMObject> try_create_with_inode(Inode&);
-    virtual RefPtr<VMObject> try_clone() override;
+    virtual KResultOr<NonnullRefPtr<VMObject>> try_clone() override;
 
 private:
     virtual bool is_shared_inode() const override { return true; }

Kernel/Memory/VMObject.h

@@ -33,7 +33,7 @@ class VMObject : public RefCounted<VMObject>
 public:
     virtual ~VMObject();
 
-    virtual RefPtr<VMObject> try_clone() = 0;
+    virtual KResultOr<NonnullRefPtr<VMObject>> try_clone() = 0;
 
     virtual bool is_anonymous() const { return false; }
     virtual bool is_inode() const { return false; }

Kernel/Syscalls/anon_create.cpp

@@ -29,11 +29,11 @@ KResultOr<FlatPtr> Process::sys$anon_create(size_t size, int options)
     if (new_fd_or_error.is_error())
         return new_fd_or_error.error();
     auto new_fd = new_fd_or_error.release_value();
-    auto vmobject = Memory::AnonymousVMObject::try_create_purgeable_with_size(size, AllocationStrategy::Reserve);
-    if (!vmobject)
-        return ENOMEM;
+    auto maybe_vmobject = Memory::AnonymousVMObject::try_create_purgeable_with_size(size, AllocationStrategy::Reserve);
+    if (maybe_vmobject.is_error())
+        return maybe_vmobject.error();
 
-    auto anon_file = AnonymousFile::create(vmobject.release_nonnull());
+    auto anon_file = AnonymousFile::create(maybe_vmobject.release_value());
     if (!anon_file)
         return ENOMEM;
     auto description_or_error = FileDescription::create(*anon_file);

Kernel/Syscalls/mmap.cpp

@@ -219,12 +219,17 @@ KResultOr<FlatPtr> Process::sys$mmap(Userspace<const Syscall::SC_mmap_params*> u
     if (map_anonymous) {
         auto strategy = map_noreserve ? AllocationStrategy::None : AllocationStrategy::Reserve;
         RefPtr<Memory::AnonymousVMObject> vmobject;
-        if (flags & MAP_PURGEABLE)
-            vmobject = Memory::AnonymousVMObject::try_create_purgeable_with_size(Memory::page_round_up(size), strategy);
-        else
-            vmobject = Memory::AnonymousVMObject::try_create_with_size(Memory::page_round_up(size), strategy);
-        if (!vmobject)
-            return ENOMEM;
+        if (flags & MAP_PURGEABLE) {
+            auto maybe_vmobject = Memory::AnonymousVMObject::try_create_purgeable_with_size(Memory::page_round_up(size), strategy);
+            if (maybe_vmobject.is_error())
+                return maybe_vmobject.error();
+            vmobject = maybe_vmobject.release_value();
+        } else {
+            auto maybe_vmobject = Memory::AnonymousVMObject::try_create_with_size(Memory::page_round_up(size), strategy);
+            if (maybe_vmobject.is_error())
+                return maybe_vmobject.error();
+            vmobject = maybe_vmobject.release_value();
+        }
         auto region_or_error = address_space().allocate_region_with_vmobject(range.value(), vmobject.release_nonnull(), 0, {}, prot, map_shared);
         if (region_or_error.is_error())
             return region_or_error.error().error();