a

Cargo.lock

@@ -25,6 +25,12 @@ version = "1.0.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "baf1de4339761588bc0619e3cbc0120ee582ebb74b53b4efbf79117bd2da40fd"
 
+[[package]]
+name = "convert_case"
+version = "0.4.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "6245d59a3e82a7fc217c5828a6692dbc6dfb63a0c8c90495621f7b9d79704a0e"
+
 [[package]]
 name = "delegate"
 version = "0.9.0"
@@ -36,6 +42,19 @@ dependencies = [
  "syn 1.0.109",
 ]
 
+[[package]]
+name = "derive_more"
+version = "0.99.17"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "4fb810d30a7c1953f91334de7244731fc3f3c10d7fe163338a35b9f640960321"
+dependencies = [
+ "convert_case",
+ "proc-macro2",
+ "quote",
+ "rustc_version",
+ "syn 1.0.109",
+]
+
 [[package]]
 name = "fnv"
 version = "1.0.7"
@@ -70,6 +89,7 @@ name = "hbvm"
 version = "0.1.0"
 dependencies = [
  "delegate",
+ "derive_more",
  "hashbrown",
  "hbbytecode",
  "log",
@@ -163,6 +183,21 @@ version = "0.6.29"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "f162c6dd7b008981e4d40210aca20b4bd0f9b60ca9271061b07f78537722f2e1"
 
+[[package]]
+name = "rustc_version"
+version = "0.4.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "bfa0f585226d2e68097d4f95d113b15b83a82e819ab25717ec0590d9584ef366"
+dependencies = [
+ "semver",
+]
+
+[[package]]
+name = "semver"
+version = "1.0.17"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "bebd363326d05ec3e2f532ab7660680f3b02130d780c299bca73469d521bc0ed"
+
 [[package]]
 name = "static_assertions"
 version = "1.1.0"

hbvm/Cargo.toml

@@ -8,6 +8,7 @@ lto = true
 
 [dependencies]
 delegate          = "0.9"
+derive_more       = "0.99"
 hashbrown         = "0.13"
 hbbytecode.path   = "../hbbytecode"
 log               = "0.4"

hbvm/src/lib.rs

@@ -3,20 +3,3 @@ extern crate alloc;
 
 pub mod validate;
 pub mod vm;
-
-#[derive(Debug, PartialEq)]
-pub enum RuntimeErrors {
-    InvalidOpcodePair(u8, u8),
-    RegisterTooSmall,
-    HostError(u64),
-    PageNotMapped(u64),
-    InvalidJumpAddress(u64),
-    InvalidSystemCall(u8),
-}
-
-// If you solve the halting problem feel free to remove this
-#[derive(PartialEq, Debug)]
-pub enum HaltStatus {
-    Halted,
-    Running,
-}

hbvm/src/main.rs

@@ -1,3 +1,9 @@
+use hbvm::vm::{
+    mem::{Memory, MemoryAccessReason, PageSize},
+    trap::HandleTrap,
+    value::Value,
+};
+
 use {
     hbvm::{validate::validate, vm::Vm},
     std::io::{stdin, Read},
@@ -6,13 +12,13 @@ use {
 fn main() -> Result<(), Box<dyn std::error::Error>> {
     let mut prog = vec![];
     stdin().read_to_end(&mut prog)?;
-    
+
     if let Err(e) = validate(&prog) {
         eprintln!("Program validation error: {e:?}");
         return Ok(());
     } else {
         unsafe {
-            let mut vm = Vm::new_unchecked(&prog);
+            let mut vm = Vm::new_unchecked(&prog, TestTrapHandler);
             vm.memory.insert_test_page();
             println!("Program interrupt: {:?}", vm.run());
             println!("{:?}", vm.registers);
@@ -24,3 +30,30 @@ fn main() -> Result<(), Box<dyn std::error::Error>> {
 pub fn time() -> u32 {
     9
 }
+
+struct TestTrapHandler;
+impl HandleTrap for TestTrapHandler {
+    fn page_fault(
+        &mut self,
+        _: MemoryAccessReason,
+        _: &mut Memory,
+        _: u64,
+        _: PageSize,
+        _: *mut u8,
+    ) -> bool {
+        false
+    }
+
+    fn invalid_op(&mut self, _: &mut [Value; 256], _: &mut usize, _: &mut Memory, _: u8) -> bool
+    where
+        Self: Sized,
+    {
+        false
+    }
+
+    fn ecall(&mut self, _: &mut [Value; 256], _: &mut usize, _: &mut Memory)
+    where
+        Self: Sized,
+    {
+    }
+}

hbvm/src/validate.rs

@@ -1,21 +1,31 @@
+/// Program validation error kind
 #[derive(Clone, Copy, Debug, PartialEq, Eq)]
 pub enum ErrorKind {
+    /// Unknown opcode
     InvalidInstruction,
+    /// VM doesn't implement this valid opcode
     Unimplemented,
+    /// Attempted to copy over register boundary
     RegisterArrayOverflow,
 }
 
+/// Error
 #[derive(Clone, Copy, Debug, PartialEq, Eq)]
 pub struct Error {
+    /// Kind
     pub kind: ErrorKind,
+    /// Location in bytecode
     pub index: usize,
 }
 
+/// Perform bytecode validation. If it passes, the program should be
+/// sound to execute.
 pub fn validate(mut program: &[u8]) -> Result<(), Error> {
     use hbbytecode::opcode::*;
 
     let start = program;
     loop {
+        // Match on instruction types and perform necessary checks
         program = match program {
             [] => return Ok(()),
             [LD..=ST, reg, _, _, _, _, _, _, _, _, _, count, ..]

hbvm/src/vm/mem/mod.rs

@@ -1,10 +1,16 @@
 mod paging;
 
-use self::paging::{PageTable, Permission, PtEntry};
-use alloc::boxed::Box;
+use core::mem::MaybeUninit;
 
+use self::paging::{PageTable, Permission, PtEntry};
+use super::{trap::HandleTrap, VmRunError};
+use alloc::boxed::Box;
+use derive_more::Display;
+
+/// HoleyBytes virtual memory
 #[derive(Clone, Debug)]
 pub struct Memory {
+    /// Root page table
     root_pt: *mut PageTable,
 }
 
@@ -47,13 +53,23 @@ impl Memory {
                 entry = PtEntry::new(Box::into_raw(pt) as _, Permission::Node);
             }
 
-            self.root_pt_mut()[0] = entry;
+            (*self.root_pt)[0] = entry;
         }
     }
 
     /// Load value from an address
-    pub unsafe fn load(&self, addr: u64, target: *mut u8, count: usize) -> Result<(), ()> {
+    ///
+    /// # Safety
+    /// Applies same conditions as for [`core::ptr::copy_nonoverlapping`]
+    pub unsafe fn load(
+        &mut self,
+        addr: u64,
+        target: *mut u8,
+        count: usize,
+        traph: &mut impl HandleTrap,
+    ) -> Result<(), LoadError> {
         self.memory_access(
+            MemoryAccessReason::Load,
             addr,
             target,
             count,
@@ -64,96 +80,193 @@ impl Memory {
                 )
             },
             |src, dst, count| core::ptr::copy_nonoverlapping(src, dst, count),
+            traph,
         )
+        .map_err(|_| LoadError)
     }
 
     /// Store value to an address
-    pub unsafe fn store(&mut self, addr: u64, source: *const u8, count: usize) -> Result<(), ()> {
+    ///
+    /// # Safety
+    /// Applies same conditions as for [`core::ptr::copy_nonoverlapping`]
+    pub unsafe fn store(
+        &mut self,
+        addr: u64,
+        source: *const u8,
+        count: usize,
+        traph: &mut impl HandleTrap,
+    ) -> Result<(), StoreError> {
         self.memory_access(
+            MemoryAccessReason::Store,
             addr,
             source.cast_mut(),
             count,
             |perm| perm == Permission::Write,
             |dst, src, count| core::ptr::copy_nonoverlapping(src, dst, count),
+            traph,
         )
+        .map_err(|_| StoreError)
     }
 
     /// Copy a block of memory
-    pub unsafe fn block_copy(&mut self, src: u64, dst: u64, count: u64) -> Result<(), ()> {
-        let count = usize::try_from(count).expect("?conradluget a better CPU");
+    ///
+    /// # Safety
+    /// - Same as for [`Self::load`] and [`Self::store`]
+    /// - Your faith in the gods of UB
+    ///     - Addr-san claims it's fine but who knows is she isn't lying :ferrisSus:
+    pub unsafe fn block_copy(
+        &mut self,
+        src: u64,
+        dst: u64,
+        count: usize,
+        traph: &mut impl HandleTrap,
+    ) -> Result<(), MemoryAccessReason> {
+        // Yea, i know it is possible to do this more efficiently, but I am too lazy.
 
-        let mut srcs = PageSplitter::new(src, count, self.root_pt);
-        let mut dsts = PageSplitter::new(dst, count, self.root_pt);
-        let mut c_src = srcs.next().ok_or(())?;
-        let mut c_dst = dsts.next().ok_or(())?;
+        const STACK_BUFFER_SIZE: usize = 512;
 
-        loop {
-            let min_size = c_src.size.min(c_dst.size);
+        // Decide if to use stack-allocated buffer or to heap allocate
+        // Deallocation is again decided on size at the end of the function
+        let mut buf = MaybeUninit::<[u8; STACK_BUFFER_SIZE]>::uninit();
+        let buf = if count <= STACK_BUFFER_SIZE {
+            buf.as_mut_ptr().cast()
+        } else {
             unsafe {
-                core::ptr::copy(c_src.ptr, c_dst.ptr, min_size);
-            }
+                let layout = core::alloc::Layout::from_size_align_unchecked(count, 1);
+                let ptr = alloc::alloc::alloc(layout);
+                if ptr.is_null() {
+                    alloc::alloc::handle_alloc_error(layout);
+                }
 
-            match (
-                match c_src.size.saturating_sub(min_size) {
-                    0 => srcs.next(),
-                    size => Some(PageSplitResult { size, ..c_src }),
-                },
-                match c_dst.size.saturating_sub(min_size) {
-                    0 => dsts.next(),
-                    size => Some(PageSplitResult { size, ..c_dst }),
-                },
-            ) {
-                (None, None) => return Ok(()),
-                (Some(src), Some(dst)) => (c_src, c_dst) = (src, dst),
-                _ => return Err(()),
+                ptr
             }
+        };
+
+        // Perform memory block transfer
+        let status = (|| {
+            // Load to buffer
+            self.memory_access(
+                MemoryAccessReason::Load,
+                src,
+                buf,
+                count,
+                |perm| {
+                    matches!(
+                        perm,
+                        Permission::Readonly | Permission::Write | Permission::Exec
+                    )
+                },
+                |src, dst, count| core::ptr::copy(src, dst, count),
+                traph,
+            )
+            .map_err(|_| MemoryAccessReason::Load)?;
+
+            // Store from buffer
+            self.memory_access(
+                MemoryAccessReason::Store,
+                dst,
+                buf,
+                count,
+                |perm| perm == Permission::Write,
+                |dst, src, count| core::ptr::copy(src, dst, count),
+                traph,
+            )
+            .map_err(|_| MemoryAccessReason::Store)?;
+
+            Ok::<_, MemoryAccessReason>(())
+        })();
+
+        // Deallocate if used heap-allocated array
+        if count > STACK_BUFFER_SIZE {
+            alloc::alloc::dealloc(
+                buf,
+                core::alloc::Layout::from_size_align_unchecked(count, 1),
+            );
         }
+
+        status
     }
 
-    #[inline]
-    pub fn root_pt(&self) -> &PageTable {
-        unsafe { &*self.root_pt }
-    }
-
-    #[inline]
-    pub fn root_pt_mut(&mut self) -> &mut PageTable {
-        unsafe { &mut *self.root_pt }
-    }
-
+    /// Split address to pages, check their permissions and feed pointers with offset
+    /// to a specified function.
+    ///
+    /// If page is not found, execute page fault trap handler.
+    #[allow(clippy::too_many_arguments)] // Silence peasant
     fn memory_access(
-        &self,
+        &mut self,
+        reason: MemoryAccessReason,
         src: u64,
         mut dst: *mut u8,
         len: usize,
-        permission_check: impl Fn(Permission) -> bool,
-        action: impl Fn(*mut u8, *mut u8, usize),
+        permission_check: fn(Permission) -> bool,
+        action: fn(*mut u8, *mut u8, usize),
+        traph: &mut impl HandleTrap,
     ) -> Result<(), ()> {
-        for PageSplitResult { ptr, size, perm } in PageSplitter::new(src, len, self.root_pt) {
-            if !permission_check(perm) {
-                return Err(());
+        let mut pspl = AddrSplitter::new(src, len, self.root_pt);
+        loop {
+            match pspl.next() {
+                // Page found
+                Some(Ok(AddrSplitOk { ptr, size, perm })) => {
+                    if !permission_check(perm) {
+                        return Err(());
+                    }
+
+                    // Perform memory action and bump dst pointer
+                    action(ptr, dst, size);
+                    dst = unsafe { dst.add(size) };
+                }
+                Some(Err(AddrSplitError { addr, size })) => {
+                    // Execute page fault handler
+                    if traph.page_fault(reason, self, addr, size, dst) {
+                        // Shift the splitter address
+                        pspl.bump(size);
+
+                        // Bump dst pointer
+                        dst = unsafe { dst.add(size as _) };
+                    } else {
+                        return Err(()); // Unhandleable
+                    }
+                }
+                None => return Ok(()),
             }
-
-            action(ptr, dst, size);
-            dst = unsafe { dst.add(size) };
         }
-
-        Ok(())
     }
 }
 
-struct PageSplitResult {
+/// Result from address split
+struct AddrSplitOk {
+    /// Pointer to the start for perform operation
     ptr: *mut u8,
+
+    /// Size to the end of page / end of desired size
     size: usize,
+
+    /// Page permission
     perm: Permission,
 }
 
-struct PageSplitter {
+struct AddrSplitError {
+    /// Address of failure
     addr: u64,
+
+    /// Requested page size
+    size: PageSize,
+}
+
+/// Address splitter into pages
+struct AddrSplitter {
+    /// Current address
+    addr: u64,
+
+    /// Size left
     size: usize,
+
+    /// Page table
     pagetable: *const PageTable,
 }
 
-impl PageSplitter {
+impl AddrSplitter {
+    /// Create a new page splitter
     pub const fn new(addr: u64, size: usize, pagetable: *const PageTable) -> Self {
         Self {
             addr,
@@ -161,19 +274,29 @@ impl PageSplitter {
             pagetable,
         }
     }
+
+    /// Bump address by size X
+    fn bump(&mut self, page_size: PageSize) {
+        self.addr += page_size as u64;
+        self.size = self.size.saturating_sub(page_size as _);
+    }
 }
 
-impl Iterator for PageSplitter {
-    type Item = PageSplitResult;
+impl Iterator for AddrSplitter {
+    type Item = Result<AddrSplitOk, AddrSplitError>;
 
     fn next(&mut self) -> Option<Self::Item> {
+        // The end, everything is fine
         if self.size == 0 {
             return None;
         }
 
         let (base, perm, size, offset) = 'a: {
             let mut current_pt = self.pagetable;
+
+            // Walk the page table
             for lvl in (0..5).rev() {
+                // Get an entry
                 unsafe {
                     let entry = (*current_pt).get_unchecked(
                         usize::try_from((self.addr >> (lvl * 9 + 12)) & ((1 << 9) - 1))
@@ -182,34 +305,102 @@ impl Iterator for PageSplitter {
 
                     let ptr = entry.ptr();
                     match entry.permission() {
-                        Permission::Empty => return None,
+                        // No page → page fault
+                        Permission::Empty => {
+                            return Some(Err(AddrSplitError {
+                                addr: self.addr,
+                                size: PageSize::from_lvl(lvl)?,
+                            }))
+                        }
+
+                        // Node → proceed waking
                         Permission::Node => current_pt = ptr as _,
+
+                        // Leaft → return relevant data
                         perm => {
                             break 'a (
+                                // Pointer in host memory
                                 ptr as *mut u8,
                                 perm,
-                                match lvl {
-                                    0 => 4096,
-                                    1 => 1024_usize.pow(2) * 2,
-                                    2 => 1024_usize.pow(3),
-                                    _ => return None,
-                                },
+                                PageSize::from_lvl(lvl)?,
+                                // In-page offset
                                 self.addr as usize & ((1 << (lvl * 9 + 12)) - 1),
-                            )
+                            );
                         }
                     }
                 }
             }
-            return None;
+            return None; // Reached the end (should not happen)
         };
 
-        let avail = (size - offset).clamp(0, self.size);
-        self.addr += size as u64;
-        self.size = self.size.saturating_sub(size);
-        Some(PageSplitResult {
-            ptr: unsafe { base.add(offset) },
+        // Get available byte count in the selected page with offset
+        let avail = (size as usize - offset).clamp(0, self.size);
+        self.bump(size);
+
+        Some(Ok(AddrSplitOk {
+            ptr: unsafe { base.add(offset) }, // Return pointer to the start of region
             size: avail,
             perm,
-        })
+        }))
+    }
+}
+
+/// Page size
+#[derive(Clone, Copy, Debug, PartialEq, Eq)]
+pub enum PageSize {
+    /// 4 KiB page (on level 0)
+    Size4K = 4096,
+
+    /// 2 MiB page (on level 1)
+    Size2M = 1024 * 1024 * 2,
+
+    /// 1 GiB page (on level 2)
+    Size1G = 1024 * 1024 * 1024,
+}
+
+impl PageSize {
+    /// Convert page table level to size of page
+    fn from_lvl(lvl: u8) -> Option<Self> {
+        match lvl {
+            0 => Some(PageSize::Size4K),
+            1 => Some(PageSize::Size2M),
+            2 => Some(PageSize::Size1G),
+            _ => None,
+        }
+    }
+}
+
+/// Unhandled load access trap
+#[derive(Clone, Copy, Display, Debug, PartialEq, Eq)]
+pub struct LoadError;
+
+/// Unhandled store access trap
+#[derive(Clone, Copy, Display, Debug, PartialEq, Eq)]
+pub struct StoreError;
+
+#[derive(Clone, Copy, Display, Debug, PartialEq, Eq)]
+pub enum MemoryAccessReason {
+    Load,
+    Store,
+}
+
+impl From<MemoryAccessReason> for VmRunError {
+    fn from(value: MemoryAccessReason) -> Self {
+        match value {
+            MemoryAccessReason::Load => Self::LoadAccessEx,
+            MemoryAccessReason::Store => Self::StoreAccessEx,
+        }
+    }
+}
+
+impl From<LoadError> for VmRunError {
+    fn from(_: LoadError) -> Self {
+        Self::LoadAccessEx
+    }
+}
+
+impl From<StoreError> for VmRunError {
+    fn from(_: StoreError) -> Self {
+        Self::StoreAccessEx
     }
 }

hbvm/src/vm/mem/paging.rs

@@ -6,30 +6,40 @@ use core::{
 };
 use delegate::delegate;
 
+/// Page entry permission
 #[derive(Clone, Copy, Debug, Default, PartialEq, Eq)]
 #[repr(u8)]
 pub enum Permission {
+    /// No page present
     #[default]
     Empty,
+    /// Points to another pagetable
     Node,
+    /// Page is read only
     Readonly,
+    /// Page is readable and writable
     Write,
+    /// Page is readable and executable
     Exec,
 }
 
+/// Page table entry
 #[derive(Clone, Copy, Default, PartialEq, Eq)]
 pub struct PtEntry(u64);
 impl PtEntry {
+    /// Create new
     #[inline]
     pub unsafe fn new(ptr: *mut PtPointedData, permission: Permission) -> Self {
         Self(ptr as u64 | permission as u64)
     }
 
+    /// Get permission
     #[inline]
     pub fn permission(&self) -> Permission {
         unsafe { core::mem::transmute(self.0 as u8 & 0b111) }
     }
 
+    /// Get pointer to the data (leaf) or next page table (node)
     #[inline]
     pub fn ptr(&self) -> *mut PtPointedData {
         (self.0 & !((1 << 12) - 1)) as _
@@ -45,6 +55,8 @@ impl Debug for PtEntry {
     }
 }
 
+
+/// Page table
 #[derive(Clone, Copy, Debug, PartialEq, Eq)]
 #[repr(align(4096))]
 pub struct PageTable([PtEntry; 512]);
@@ -89,13 +101,17 @@ where
 
 impl Default for PageTable {
     fn default() -> Self {
+        // SAFETY: It's fine, zeroed page table entry is valid (= empty)
         Self(unsafe { MaybeUninit::zeroed().assume_init() })
     }
 }
 
+/// Data page table entry can possibly point to
 #[derive(Clone, Copy)]
 #[repr(C, align(4096))]
 pub union PtPointedData {
+    /// Node - next page table
     pub pt: PageTable,
+    /// Leaf
     pub page: u8,
 }

hbvm/src/vm/mod.rs

@@ -11,8 +11,11 @@
 //   program size. If you are (rightfully) worried about the UB, for now just
 //   append your program with 11 zeroes.
 
-mod mem;
-mod value;
+use self::trap::HandleTrap;
+
+pub mod mem;
+pub mod trap;
+pub mod value;
 
 use {
     crate::validate,
@@ -23,6 +26,7 @@ use {
     value::Value,
 };
 
+/// Extract a parameter from program
 macro_rules! param {
     ($self:expr, $ty:ty) => {{
         assert_impl_one!($ty: OpParam);
@@ -37,6 +41,7 @@ macro_rules! param {
     }};
 }
 
+/// Perform binary operation `#0 ← #1 OP #2`
 macro_rules! binary_op {
     ($self:expr, $ty:ident, $handler:expr) => {{
         let ParamBBB(tg, a0, a1) = param!($self, ParamBBB);
@@ -51,6 +56,7 @@ macro_rules! binary_op {
     }};
 }
 
+/// Perform binary operation with immediate `#0 ← #1 OP imm #2`
 macro_rules! binary_op_imm {
     ($self:expr, $ty:ident, $handler:expr) => {{
         let ParamBBD(tg, a0, imm) = param!($self, ParamBBD);
@@ -61,6 +67,7 @@ macro_rules! binary_op_imm {
     }};
 }
 
+/// Jump at `#3` if ordering on `#0 <=> #1` is equal to expected
 macro_rules! cond_jump {
     ($self:expr, $ty:ident, $expected:ident) => {{
         let ParamBBD(a0, a1, jt) = param!($self, ParamBBD);
@@ -72,36 +79,59 @@ macro_rules! cond_jump {
     }};
 }
 
-pub struct Vm<'a> {
+/// HoleyBytes Virtual Machine
+pub struct Vm<'a, T> {
+    /// Holds 256 registers
+    ///
+    /// Writing to register 0 is considered undefined behaviour
+    /// in terms of HoleyBytes program execution
     pub registers: [Value; 256],
+
+    /// Memory implementation
     pub memory: Memory,
+
+    /// Trap handler
+    pub traph: T,
+
+    // Program counter
     pc: usize,
+
+    /// Program
     program: &'a [u8],
 }
 
-impl<'a> Vm<'a> {
+impl<'a, T: HandleTrap> Vm<'a, T> {
+    /// Create a new VM with program and trap handler
+    ///
     /// # Safety
     /// Program code has to be validated
-    pub unsafe fn new_unchecked(program: &'a [u8]) -> Self {
+    pub unsafe fn new_unchecked(program: &'a [u8], traph: T) -> Self {
         Self {
             registers: [Value::from(0_u64); 256],
             memory: Default::default(),
+            traph,
             pc: 0,
             program,
         }
     }
 
-    pub fn new_validated(program: &'a [u8]) -> Result<Self, validate::Error> {
+    /// Create a new VM with program and trap handler only if it passes validation
+    pub fn new_validated(program: &'a [u8], traph: T) -> Result<Self, validate::Error> {
         validate::validate(program)?;
-        Ok(unsafe { Self::new_unchecked(program) })
+        Ok(unsafe { Self::new_unchecked(program, traph) })
     }
 
-    pub fn run(&mut self) -> HaltReason {
+    /// Execute program
+    ///
+    /// Program can return [`VmRunError`] if a trap handling failed
+    pub fn run(&mut self) -> Result<(), VmRunError> {
         use hbbytecode::opcode::*;
         loop {
+            // Fetch instruction
             let Some(&opcode) = self.program.get(self.pc)
-                else { return HaltReason::ProgramEnd };
+                else { return Ok(()) };
 
+            // Big match
             unsafe {
                 match opcode {
                     NOP => param!(self, ()),
@@ -196,45 +226,30 @@ impl<'a> Vm<'a> {
                             _ => 0,
                         };
 
-                        if self
-                            .memory
-                            .load(
-                                self.read_reg(base).as_u64() + off + n as u64,
-                                self.registers.as_mut_ptr().add(usize::from(dst) + n).cast(),
-                                usize::from(count).saturating_sub(n),
-                            )
-                            .is_err()
-                        {
-                            return HaltReason::LoadAccessEx;
-                        }
+                        self.memory.load(
+                            self.read_reg(base).as_u64() + off + n as u64,
+                            self.registers.as_mut_ptr().add(usize::from(dst) + n).cast(),
+                            usize::from(count).saturating_sub(n),
+                            &mut self.traph,
+                        )?;
                     }
                     ST => {
                         let ParamBBDH(dst, base, off, count) = param!(self, ParamBBDH);
-                        if self
-                            .memory
-                            .store(
-                                self.read_reg(base).as_u64() + off,
-                                self.registers.as_ptr().add(usize::from(dst)).cast(),
-                                count.into(),
-                            )
-                            .is_err()
-                        {
-                            return HaltReason::LoadAccessEx;
-                        }
+                        self.memory.store(
+                            self.read_reg(base).as_u64() + off,
+                            self.registers.as_ptr().add(usize::from(dst)).cast(),
+                            count.into(),
+                            &mut self.traph,
+                        )?;
                     }
                     BMC => {
                         let ParamBBD(src, dst, count) = param!(self, ParamBBD);
-                        if self
-                            .memory
-                            .block_copy(
-                                self.read_reg(src).as_u64(),
-                                self.read_reg(dst).as_u64(),
-                                count,
-                            )
-                            .is_err()
-                        {
-                            return HaltReason::LoadAccessEx;
-                        }
+                        self.memory.block_copy(
+                            self.read_reg(src).as_u64(),
+                            self.read_reg(dst).as_u64(),
+                            count as _,
+                            &mut self.traph,
+                        )?;
                     }
                     BRC => {
                         let ParamBBB(src, dst, count) = param!(self, ParamBBB);
@@ -261,7 +276,8 @@ impl<'a> Vm<'a> {
                     JGTU => cond_jump!(self, sint, Greater),
                     ECALL => {
                         param!(self, ());
-                        return HaltReason::Ecall;
+                        self.traph
+                            .ecall(&mut self.registers, &mut self.pc, &mut self.memory);
                     }
                     ADDF => binary_op!(self, as_f64, ops::Add::add),
                     MULF => binary_op!(self, as_f64, ops::Mul::mul),
@@ -274,21 +290,29 @@ impl<'a> Vm<'a> {
                     }
                     ADDFI => binary_op_imm!(self, as_f64, ops::Add::add),
                     MULFI => binary_op_imm!(self, as_f64, ops::Mul::mul),
-                    _ => return HaltReason::InvalidOpcode,
+                    op => {
+                        if !self.traph.invalid_op(
+                            &mut self.registers,
+                            &mut self.pc,
+                            &mut self.memory,
+                            op,
+                        ) {
+                            return Err(VmRunError::InvalidOpcodeEx);
+                        }
+                    }
                 }
             }
         }
     }
 
+    /// Read register
     #[inline]
     unsafe fn read_reg(&self, n: u8) -> Value {
-        if n == 0 {
-            0_u64.into()
-        } else {
-            *self.registers.get_unchecked(n as usize)
-        }
+        *self.registers.get_unchecked(n as usize)
     }
 
+    /// Write a register.
+    /// Writing to register 0 is no-op.
     #[inline]
     unsafe fn write_reg(&mut self, n: u8, value: Value) {
         if n != 0 {
@@ -297,12 +321,16 @@ impl<'a> Vm<'a> {
     }
 }
 
+/// Virtual machine halt error
 #[derive(Copy, Clone, Debug, PartialEq, Eq)]
 #[repr(u8)]
-pub enum HaltReason {
-    ProgramEnd,
-    Ecall,
-    InvalidOpcode,
+pub enum VmRunError {
+    /// Unhandled invalid opcode exceptions
+    InvalidOpcodeEx,
+
+    /// Unhandled load access exception
     LoadAccessEx,
+
+    /// Unhandled store access exception
     StoreAccessEx,
 }

hbvm/src/vm/trap.rs

@@ -0,0 +1,33 @@
+use super::{
+    mem::{Memory, MemoryAccessReason, PageSize},
+    value::Value,
+};
+
+/// Handle VM traps
+pub trait HandleTrap {
+    /// Handle page fault
+    fn page_fault(
+        &mut self,
+        reason: MemoryAccessReason,
+        memory: &mut Memory,
+        vaddr: u64,
+        size: PageSize,
+        dataptr: *mut u8,
+    ) -> bool;
+
+    /// Handle invalid opcode exception
+    fn invalid_op(
+        &mut self,
+        regs: &mut [Value; 256],
+        pc: &mut usize,
+        memory: &mut Memory,
+        op: u8,
+    ) -> bool
+    where
+        Self: Sized;
+
+    /// Handle environment calls
+    fn ecall(&mut self, regs: &mut [Value; 256], pc: &mut usize, memory: &mut Memory)
+    where
+        Self: Sized;
+}

hbvm/src/vm/value.rs

@@ -1,7 +1,13 @@
 use core::fmt::Debug;
 
+/// Define [`Value`] union
+///
+/// # Safety
+/// Union variants have to be sound to byte-reinterpretate
+/// between each other. Otherwise the behaviour is undefined.
 macro_rules! value_def {
     ($($ty:ident),* $(,)?) => {
+        /// HBVM register value
         #[derive(Copy, Clone)]
         #[repr(packed)]
         pub union Value {
@@ -10,6 +16,7 @@ macro_rules! value_def {
 
         paste::paste! {
             impl Value {$(
+                #[doc = "Byte-reinterpret [`Value`] as [`" $ty "`]"]
                 #[inline]
                 pub fn [<as_ $ty>](&self) -> $ty {
                     unsafe { self.$ty }
@@ -29,9 +36,11 @@ macro_rules! value_def {
 }
 
 value_def!(u64, i64, f64);
+static_assertions::const_assert_eq!(core::mem::size_of::<Value>(), 8);
 
 impl Debug for Value {
     fn fmt(&self, f: &mut core::fmt::Formatter<'_>) -> core::fmt::Result {
-        self.as_u64().fmt(f)
+        // Print formatted as hexadecimal, unsigned integer
+        write!(f, "{:x}", self.as_u64())
     }
 }