M7: synchronous IPC — endpoints, handle table, IPC_Call/ReplyWait

The microkernel message backbone the VFS server and drivers will ride on.

- src/kernel/ipc_sync.zig: Endpoint (sender FIFO threaded via Task.next +
  a recv WaitQueue for servers + a small notification ring). call() (client
  sends, wakes a server, blocks) and replyWait() (server replies to the held
  caller, then receives the next). Reply routing keys on Task.ipc_client —
  synchronous IPC owes one reply at a time. Payloads copy frame-to-frame
  through the physmap (copyAcross on arch.translate, added in M5); an unmapped
  page fails the copy instead of #PF-ing. MSG_MAX 256.
- Bootstrap naming: an integer name registry (danos.ServiceId, vfs=1) with
  create_endpoint / ipc_register / ipc_lookup — any process finds a server
  without threading a handle through spawn.
- notifyFromIsr(): ISR-safe async wake (badge with the high bit set), the
  hook M10's IRQ-as-message needs. Unused/untested until then.
- Task gains handles[16] (opaque *Endpoint, to avoid a sched<->ipc import
  cycle) + ipc_client/send/reply/status fields; sched gains
  blockCurrentLocked/readyLocked; arch gains setSyscallResult2 (rdx badge).
- Syscalls 6..10 wired in process.zig; exit() now drops the caller's endpoint
  refs. lib/ipc.zig: user-side createEndpoint/register/lookup/call
  (server-side replyWait lands with the first server in M9).
- New `ipc-call` test: two kernel tasks ping-pong 100 calls, every reply
  request+1. Suite 29/29.

lib/ipc.zig

@@ -1,10 +1,16 @@
-//! User-space IPC helpers. The kernel's synchronous IPC syscalls
-//! (create_endpoint, ipc_register/lookup, ipc_call, ipc_reply_wait) arrive in a
-//! later milestone; this reserves the module boundary now so drivers and servers
-//! can be written against `rt.ipc` without restructuring once they light up.
+//! User-space IPC helpers over the kernel's synchronous IPC syscalls. A client
+//! `call`s an endpoint (send + block for reply); the VFS server and drivers are
+//! reached this way. The server side (`replyWait`, which returns two values) is
+//! added with the first server binary.
 
-/// A fixed-size, register-friendly message payload. The wire format for the VFS
-/// and driver protocols is layered on top of this by the servers themselves.
+const danos = @import("danos");
+const sc = @import("syscall.zig");
+
+/// A small-int handle into the calling process's handle table.
+pub const Handle = usize;
+
+/// A fixed-size, register-friendly message payload. Server protocols (VFS, driver)
+/// layer their own wire format on top of the bytes a call carries.
 pub const Message = extern struct {
     tag: u64 = 0,
     a: u64 = 0,
@@ -12,4 +18,37 @@ pub const Message = extern struct {
     c: u64 = 0,
 };
 
-// call() / replyWait() / createEndpoint() land with the kernel IPC syscalls.
+/// Whether a syscall return value is a wrapped -errno (lands in the top page).
+inline fn failed(r: usize) bool {
+    return r > ~@as(usize, 0) - 4095;
+}
+
+/// Create a new endpoint owned by this process; returns its handle.
+pub fn createEndpoint() ?Handle {
+    const r = sc.syscall0(.create_endpoint);
+    return if (failed(r)) null else r;
+}
+
+/// Publish endpoint `h` under a well-known service id so other processes find it.
+pub fn register(id: danos.ServiceId, h: Handle) bool {
+    return !failed(sc.syscall2(.ipc_register, @intFromEnum(id), h));
+}
+
+/// Find the endpoint published under `id`, installing a handle to it in this
+/// process.
+pub fn lookup(id: danos.ServiceId) ?Handle {
+    const r = sc.syscall1(.ipc_lookup, @intFromEnum(id));
+    return if (failed(r)) null else r;
+}
+
+pub const CallError = error{Failed};
+
+/// Send `msg` to endpoint `h` and block until the server replies into `reply`.
+/// Returns the reply length.
+pub fn call(h: Handle, msg: []const u8, reply: []u8) CallError!usize {
+    const r = sc.syscall5(.ipc_call, h, @intFromPtr(msg.ptr), msg.len, @intFromPtr(reply.ptr), reply.len);
+    return if (failed(r)) error.Failed else r;
+}
+
+// replyWait() (server side — returns message length + sender badge) lands with
+// the first server binary (M9), where it can be exercised end to end.

src/kernel/arch/x86_64/cpu.zig

@@ -79,6 +79,14 @@ pub fn setSyscallResult(state: *CpuState, value: u64) void {
     state.rax = value;
 }
 
+/// Write a *second* syscall return value (rdx here — restored by both the
+/// syscall/sysret and int-0x80 entry paths; unlike rcx/r11 it is not consumed by
+/// sysretq). Used by IPC_ReplyWait to hand back the sender's badge alongside the
+/// message length in rax.
+pub fn setSyscallResult2(state: *CpuState, value: u64) void {
+    state.rdx = value;
+}
+
 /// Bring up the serial port (the kernel's machine-readable log). No dependencies,
 /// so it can be the very first thing called.
 pub fn serialInit() void {

src/kernel/ipc_sync.zig

@@ -0,0 +1,277 @@
+//! Synchronous IPC: the microkernel message backbone. An `Endpoint` is a
+//! rendezvous point; a client `call`s it (send a message, block for a reply) and
+//! a server `replyWait`s on it (reply to the last client, then block for the next
+//! request). This is the substrate the user-space VFS server and device drivers
+//! are reached through — `open`/`read`/`write` become user-space wrappers that
+//! marshal a request into a `call`.
+//!
+//! Design (see docs/syscall.md, the plan):
+//! - **Copy method, no bounce buffer.** Payloads are copied frame-to-frame
+//!   through the physmap (`copyAcross`), which is mapped in every address space's
+//!   shared kernel half — so the kernel reads/writes either process's user memory
+//!   without a CR3 switch, and an unmapped page fails the copy instead of #PF-ing.
+//! - **Reply routing on the server.** IPC is synchronous, so a server owes a reply
+//!   to exactly one client at a time; that caller is held in `Task.ipc_client`.
+//! - **Sender FIFO on the endpoint.** A blocked caller must be *received without
+//!   becoming runnable*, which a WaitQueue can't express, so callers queue on the
+//!   endpoint's own FIFO (threaded through the otherwise-idle `Task.next`); servers
+//!   waiting for work use a normal WaitQueue.
+//!
+//! Trust model (bring-up): copies honour only page presence and a user-half bound,
+//! not the leaf U/S or R/W bits and not SMAP — a #PF-tolerant, permission-checked
+//! copy is a later security-track item, matching the existing debug_write gap.
+
+const std = @import("std");
+const danos = @import("danos");
+const arch = @import("arch");
+const sched = @import("scheduler.zig");
+const sync = @import("sync.zig");
+const heap = @import("heap.zig");
+
+const page_size = danos.page_size;
+const Task = sched.Task;
+
+/// Largest message a single call/reply may carry. Bumping it is trivial; kept
+/// small because the copy runs under the big kernel lock.
+pub const MSG_MAX: usize = 256;
+
+pub const max_handles = sched.ipc_max_handles;
+pub const max_services = 8;
+
+/// Errno-style failures, returned as `-value` in the syscall result register.
+pub const EBADF: i64 = 1; // bad handle
+pub const E2BIG: i64 = 2; // message exceeds MSG_MAX
+pub const EFAULT: i64 = 3; // buffer unmapped / out of the user half
+pub const ENOENT: i64 = 4; // no such registered service
+pub const ENOSPC: i64 = 5; // handle table or registry full
+pub const ENOMEM: i64 = 6; // out of memory
+
+/// A badge with this bit set is an asynchronous notification (e.g. an IRQ), not a
+/// message from a client — there is no reply owed. The low bits carry the source
+/// (a GSI for IRQs). Used by `notifyFromIsr`/M10; the message path uses a plain
+/// task-id badge with this bit clear.
+pub const notify_badge_bit: u64 = 1 << 63;
+
+/// End of the user (low) canonical half — user buffers must lie below it.
+const user_half_end: u64 = 0x0000_8000_0000_0000;
+
+/// A rendezvous endpoint. Allocated from the kernel heap; referenced by handle
+/// (per process) and/or by a registry slot, counted by `refcount`.
+pub const Endpoint = struct {
+    refcount: u32 = 1,
+    // Callers blocked in `call`, awaiting receive, in FIFO order (threaded via
+    // Task.next; each such task is .blocked and in no scheduler queue).
+    sender_head: ?*Task = null,
+    sender_tail: ?*Task = null,
+    // Servers blocked in `replyWait` awaiting a request.
+    recv_wq: sched.WaitQueue = .{},
+    // Pending asynchronous notifications (badges), a small coalescing ring.
+    notify_buf: [8]u64 = undefined,
+    notify_head: u8 = 0,
+    notify_tail: u8 = 0,
+};
+
+pub fn createEndpoint() ?*Endpoint {
+    const ep = heap.allocator().create(Endpoint) catch return null;
+    ep.* = .{};
+    return ep;
+}
+
+/// Drop a reference; free the endpoint when the last one goes. (Frames are leaked
+/// today like other kernel objects — but the refcount bookkeeping lands now.)
+pub fn dropRef(ep: *Endpoint) void {
+    if (ep.refcount > 1) {
+        ep.refcount -= 1;
+    } else {
+        heap.allocator().destroy(ep);
+    }
+}
+
+// --- sender FIFO (endpoint-local, via Task.next) ----------------------------
+
+fn enqueueSender(ep: *Endpoint, t: *Task) void {
+    t.next = null;
+    if (ep.sender_tail) |tail| tail.next = t else ep.sender_head = t;
+    ep.sender_tail = t;
+}
+
+fn dequeueSender(ep: *Endpoint) ?*Task {
+    const t = ep.sender_head orelse return null;
+    ep.sender_head = t.next;
+    if (ep.sender_head == null) ep.sender_tail = null;
+    t.next = null;
+    return t;
+}
+
+// --- cross-address-space copy ----------------------------------------------
+
+/// Copy `len` bytes from `src_va` in address space `src_as` to `dst_va` in
+/// `dst_as`, walking each side's page tables through the physmap (no CR3 switch).
+/// `*_as == 0` means the kernel address space (for kernel-task endpoints). User
+/// buffers must lie in the low half. Returns false — never #PFs — if any page is
+/// unmapped or out of range. Handles page-straddling buffers.
+fn copyAcross(src_as: u64, src_va: u64, dst_as: u64, dst_va: u64, len: usize) bool {
+    const src_root = if (src_as != 0) src_as else arch.kernelPageTable();
+    const dst_root = if (dst_as != 0) dst_as else arch.kernelPageTable();
+    if (src_as != 0 and (src_va >= user_half_end or src_va + len > user_half_end)) return false;
+    if (dst_as != 0 and (dst_va >= user_half_end or dst_va + len > user_half_end)) return false;
+
+    var off: usize = 0;
+    while (off < len) {
+        const s = arch.translate(src_root, src_va + off) orelse return false;
+        const d = arch.translate(dst_root, dst_va + off) orelse return false;
+        const s_left = page_size - ((src_va + off) & (page_size - 1));
+        const d_left = page_size - ((dst_va + off) & (page_size - 1));
+        const n = @min(@min(s_left, d_left), len - off);
+        const src: [*]const u8 = @ptrFromInt(danos.physToVirt(s));
+        const dst: [*]u8 = @ptrFromInt(danos.physToVirt(d));
+        @memcpy(dst[0..n], src[0..n]);
+        off += n;
+    }
+    return true;
+}
+
+// --- the two IPC operations -------------------------------------------------
+
+/// Client side of IPC_Call: send `[msg_ptr, msg_len)` to `ep` and block until a
+/// server replies into `[reply_ptr, reply_cap)`. Returns the reply length, or a
+/// negative errno. Runs as the current task.
+pub fn call(ep: *Endpoint, msg_ptr: u64, msg_len: u64, reply_ptr: u64, reply_cap: u64) i64 {
+    if (msg_len > MSG_MAX or reply_cap > MSG_MAX) return -E2BIG;
+    const flags = sync.enter();
+    defer sync.leave(flags);
+
+    const me = sched.cur();
+    me.ipc_send_ptr = msg_ptr;
+    me.ipc_send_len = msg_len;
+    me.ipc_reply_ptr = reply_ptr;
+    me.ipc_reply_cap = reply_cap;
+    me.ipc_status = 0;
+
+    enqueueSender(ep, me); // join the FIFO, then...
+    sched.wakeLocked(&ep.recv_wq); // ...wake a waiting server (no-op if none)
+    sched.blockCurrentLocked(); // block until the reply readies us again
+
+    return me.ipc_status; // reply length or -errno, written by the replier
+}
+
+/// Server side of IPC_ReplyWait: deliver `[reply_ptr, reply_len)` to the client
+/// we currently owe (if any), then receive the next request into
+/// `[recv_ptr, recv_cap)`, blocking until one arrives. Writes the sender's badge
+/// to `out_badge` and returns the request length, or a negative errno. A pending
+/// notification is delivered ahead of client requests (length 0, badge with
+/// `notify_badge_bit` set, no reply owed).
+pub fn replyWait(ep: *Endpoint, reply_ptr: u64, reply_len: u64, recv_ptr: u64, recv_cap: u64, out_badge: *u64) i64 {
+    if (reply_len > MSG_MAX or recv_cap > MSG_MAX) return -E2BIG;
+    const flags = sync.enter();
+    defer sync.leave(flags);
+
+    const me = sched.cur();
+
+    // (1) Reply to the client we're still holding, if any.
+    if (me.ipc_client) |client| {
+        me.ipc_client = null;
+        const n = @min(reply_len, client.ipc_reply_cap);
+        if (copyAcross(me.aspace, reply_ptr, client.aspace, client.ipc_reply_ptr, n)) {
+            client.ipc_status = @intCast(n);
+        } else {
+            client.ipc_status = -EFAULT;
+        }
+        sched.readyLocked(client); // its `call` now returns
+    }
+
+    // (2) Receive the next request (or notification), blocking until one is ready.
+    while (true) {
+        if (popNotify(ep)) |badge| {
+            out_badge.* = badge | notify_badge_bit;
+            return 0; // notification: no payload, no reply owed
+        }
+        if (dequeueSender(ep)) |caller| {
+            const n = @min(caller.ipc_send_len, recv_cap);
+            if (!copyAcross(caller.aspace, caller.ipc_send_ptr, me.aspace, recv_ptr, n)) {
+                caller.ipc_status = -EFAULT; // bad sender buffer: fail it, keep serving
+                sched.readyLocked(caller);
+                continue;
+            }
+            me.ipc_client = caller; // remember who to reply to
+            out_badge.* = caller.id;
+            return @intCast(n);
+        }
+        sched.waitLocked(&ep.recv_wq); // nothing yet — sleep until woken, then retry
+    }
+}
+
+// --- asynchronous notification (for IRQ-as-message, M10) --------------------
+
+fn popNotify(ep: *Endpoint) ?u64 {
+    if (ep.notify_head == ep.notify_tail) return null;
+    const badge = ep.notify_buf[ep.notify_head % ep.notify_buf.len];
+    ep.notify_head +%= 1;
+    return badge;
+}
+
+/// Post an asynchronous notification carrying `badge` to `ep` and wake a waiting
+/// receiver. ISR-safe: takes the big kernel lock and releases it without touching
+/// the interrupt flag (the ISR's iretq restores it), exactly like the timer tick.
+/// A full ring drops the notification (the driver re-reads device state anyway).
+pub fn notifyFromIsr(ep: *Endpoint, badge: u64) void {
+    _ = sync.enter();
+    if (ep.notify_tail -% ep.notify_head < ep.notify_buf.len) {
+        ep.notify_buf[ep.notify_tail % ep.notify_buf.len] = badge;
+        ep.notify_tail +%= 1;
+    }
+    sched.wakeLocked(&ep.recv_wq);
+    sync.leaveIsr();
+}
+
+// --- per-process handle table + name registry -------------------------------
+
+/// Install `ep` in task `t`'s handle table; returns the small-int handle or
+/// -ENOSPC. The caller has already taken/holds the reference the slot represents.
+pub fn installHandle(t: *Task, ep: *Endpoint) i64 {
+    for (&t.handles, 0..) |*slot, i| {
+        if (slot.* == null) {
+            slot.* = @ptrCast(ep);
+            return @intCast(i);
+        }
+    }
+    return -ENOSPC;
+}
+
+/// Resolve a handle to its endpoint, or null if out of range / unused.
+pub fn resolveHandle(t: *Task, h: u64) ?*Endpoint {
+    if (h >= t.handles.len) return null;
+    const slot = t.handles[@intCast(h)] orelse return null;
+    return @ptrCast(@alignCast(slot));
+}
+
+/// Drop every endpoint reference an exiting task holds. Called from the scheduler
+/// exit path so a dead server's endpoints don't linger referenced.
+pub fn closeHandles(t: *Task) void {
+    for (&t.handles) |*slot| {
+        if (slot.*) |p| {
+            dropRef(@ptrCast(@alignCast(p)));
+            slot.* = null;
+        }
+    }
+}
+
+var registry: [max_services]?*Endpoint = .{null} ** max_services;
+
+/// Publish `ep` under well-known `id` (takes a reference). Returns 0 or -errno.
+pub fn register(id: u32, ep: *Endpoint) i64 {
+    if (id >= max_services) return -ENOENT;
+    if (registry[id]) |old| dropRef(old);
+    ep.refcount += 1;
+    registry[id] = ep;
+    return 0;
+}
+
+/// Find the endpoint published under `id`, taking a reference for the caller to
+/// install in its handle table. Null if nothing is registered there.
+pub fn lookup(id: u32) ?*Endpoint {
+    if (id >= max_services) return null;
+    const ep = registry[id] orelse return null;
+    ep.refcount += 1;
+    return ep;
+}

src/kernel/process.zig

@@ -26,6 +26,7 @@ const arch = @import("arch");
 const pmm = @import("pmm.zig");
 const sched = @import("scheduler.zig");
 const sync = @import("sync.zig");
+const ipc = @import("ipc_sync.zig");
 const log = @import("log.zig");
 
 const page_size = danos.page_size;
@@ -89,9 +90,13 @@ fn syscall(state: *arch.CpuState) void {
     switch (@as(Syscall, @enumFromInt(arch.syscallNumber(state)))) {
         .exit => {
             exit_code = arch.syscallArg(state, 0);
-            // A scheduled process frees its address space and reschedules; a
-            // borrowed test thread unwinds back to the kernel that entered it.
-            if (sched.currentIsUserProcess()) sched.exitUser() else arch.userExit();
+            // A scheduled process drops its endpoint references, frees its address
+            // space, and reschedules; a borrowed test thread unwinds back to the
+            // kernel that entered it.
+            if (sched.currentIsUserProcess()) {
+                ipc.closeHandles(sched.cur());
+                sched.exitUser();
+            } else arch.userExit();
         },
         .yield => {
             sched.yield();
@@ -104,10 +109,72 @@ fn syscall(state: *arch.CpuState) void {
         .debug_write => sysDebugWrite(state),
         .mmap => sysMmap(state),
         .munmap => sysMunmap(state),
+        .create_endpoint => sysCreateEndpoint(state),
+        .ipc_register => sysIpcRegister(state),
+        .ipc_lookup => sysIpcLookup(state),
+        .ipc_call => sysIpcCall(state),
+        .ipc_reply_wait => sysIpcReplyWait(state),
         _ => fail(state),
     }
 }
 
+/// Return `-errno` in the syscall result register.
+fn failErr(state: *arch.CpuState, errno: i64) void {
+    arch.setSyscallResult(state, @bitCast(-errno));
+}
+
+/// create_endpoint() -> handle: allocate an endpoint and install it in the
+/// caller's handle table.
+fn sysCreateEndpoint(state: *arch.CpuState) void {
+    const ep = ipc.createEndpoint() orelse return failErr(state, ipc.ENOMEM);
+    const h = ipc.installHandle(sched.cur(), ep);
+    if (h < 0) {
+        ipc.dropRef(ep);
+        return failErr(state, ipc.ENOSPC);
+    }
+    arch.setSyscallResult(state, @intCast(h));
+}
+
+/// ipc_register(service_id, handle): publish the caller's endpoint under a
+/// well-known id so other processes can find it.
+fn sysIpcRegister(state: *arch.CpuState) void {
+    const id: u32 = @truncate(arch.syscallArg(state, 0));
+    const ep = ipc.resolveHandle(sched.cur(), arch.syscallArg(state, 1)) orelse return failErr(state, ipc.EBADF);
+    arch.setSyscallResult(state, @bitCast(ipc.register(id, ep)));
+}
+
+/// ipc_lookup(service_id) -> handle: find a published endpoint and install a
+/// handle to it in the caller.
+fn sysIpcLookup(state: *arch.CpuState) void {
+    const id: u32 = @truncate(arch.syscallArg(state, 0));
+    const ep = ipc.lookup(id) orelse return failErr(state, ipc.ENOENT);
+    const h = ipc.installHandle(sched.cur(), ep);
+    if (h < 0) {
+        ipc.dropRef(ep);
+        return failErr(state, ipc.ENOSPC);
+    }
+    arch.setSyscallResult(state, @intCast(h));
+}
+
+/// ipc_call(handle, msg_ptr, msg_len, reply_ptr, reply_cap) -> reply_len.
+/// Blocks until the server replies; the trap frame lives on this task's kernel
+/// stack, so it survives the block and receives the result on resume.
+fn sysIpcCall(state: *arch.CpuState) void {
+    const ep = ipc.resolveHandle(sched.cur(), arch.syscallArg(state, 0)) orelse return failErr(state, ipc.EBADF);
+    const r = ipc.call(ep, arch.syscallArg(state, 1), arch.syscallArg(state, 2), arch.syscallArg(state, 3), arch.syscallArg(state, 4));
+    arch.setSyscallResult(state, @bitCast(r));
+}
+
+/// ipc_reply_wait(handle, reply_ptr, reply_len, recv_ptr, recv_cap) -> recv_len,
+/// with the sender's badge in the secondary result register (rdx).
+fn sysIpcReplyWait(state: *arch.CpuState) void {
+    const ep = ipc.resolveHandle(sched.cur(), arch.syscallArg(state, 0)) orelse return failErr(state, ipc.EBADF);
+    var badge: u64 = 0;
+    const r = ipc.replyWait(ep, arch.syscallArg(state, 1), arch.syscallArg(state, 2), arch.syscallArg(state, 3), arch.syscallArg(state, 4), &badge);
+    arch.setSyscallResult(state, @bitCast(r));
+    arch.setSyscallResult2(state, badge);
+}
+
 /// debug_write(ptr, len): copy bytes from user memory into the kernel log.
 /// A bring-up diagnostic — real output goes through the VFS/console later.
 ///

src/kernel/scheduler.zig

@@ -50,9 +50,26 @@ pub const Task = struct {
     // process.zig lazily seeds it to the arena base on the first mmap). Bumped up
     // as the user heap grows; user task only.
     heap_next: u64 = 0,
-    next: ?*Task = null, // ready-queue link
+    // --- synchronous IPC (ipc_sync.zig) ---
+    // Per-process handle table: small-int handle -> *ipc_sync.Endpoint, kept
+    // opaque here so the scheduler and IPC modules don't import each other.
+    handles: [ipc_max_handles]?*anyopaque = .{null} ** ipc_max_handles,
+    // A server holds the caller it currently owes a reply to (set by ReplyWait's
+    // receive, cleared when it replies). A client, while blocked in Call, records
+    // its message + reply buffers here and its result lands in `ipc_status`.
+    ipc_client: ?*Task = null,
+    ipc_send_ptr: u64 = 0, // client: outgoing message (vaddr in this task's AS)
+    ipc_send_len: u64 = 0,
+    ipc_reply_ptr: u64 = 0, // client: reply buffer (vaddr)
+    ipc_reply_cap: u64 = 0,
+    ipc_status: i64 = 0, // client: reply length / -errno, written by the replier
+    next: ?*Task = null, // ready-queue link (also the endpoint sender-FIFO link)
 };
 
+/// Size of each task's IPC handle table. Kept here (not in ipc_sync.zig) because
+/// it dimensions a field of `Task`; ipc_sync.zig re-exports it.
+pub const ipc_max_handles = 16;
+
 var tasks = [_]Task{.{}} ** max_tasks;
 var next_id: u32 = 1;
 
@@ -399,6 +416,23 @@ pub fn wakeLocked(wq: *WaitQueue) void {
     enqueue(t);
 }
 
+/// Block the current task and switch away, without putting it on any wait queue —
+/// the caller has already linked it wherever it belongs (e.g. an endpoint's sender
+/// FIFO). Precondition: the big kernel lock is held; still held on return (when the
+/// task is made ready again). The IPC layer's counterpart to `waitLocked`.
+pub fn blockCurrentLocked() void {
+    cur().state = .blocked;
+    schedule();
+}
+
+/// Make a specific (currently blocked) task ready to run again. Precondition: the
+/// big kernel lock is held. Used by the IPC layer to wake a specific caller/server
+/// rather than "some waiter on a queue".
+pub fn readyLocked(t: *Task) void {
+    t.state = .ready;
+    enqueue(t);
+}
+
 /// Block on `wq` (a self-contained critical section).
 pub fn wait(wq: *WaitQueue) void {
     const flags = sync.enter();

src/kernel/tests.zig

@@ -17,6 +17,7 @@ const pmm = @import("pmm.zig");
 const heap = @import("heap.zig");
 const sched = @import("scheduler.zig");
 const ipc = @import("ipc.zig");
+const ipcsync = @import("ipc_sync.zig");
 const process = @import("process.zig");
 
 /// Formatted write straight to serial, independent of the framebuffer console.
@@ -73,6 +74,8 @@ pub fn run(case: []const u8, boot_info: *const BootInfo) void {
         eventTest();
     } else if (eql(case, "ipc")) {
         ipcTest();
+    } else if (eql(case, "ipc-call")) {
+        ipcCallTest();
     } else if (eql(case, "smp")) {
         smpTest();
     } else if (eql(case, "affinity")) {
@@ -774,6 +777,65 @@ fn userMemTest() void {
     result();
 }
 
+// --- synchronous IPC --------------------------------------------------------
+
+var ipc_ep: *ipcsync.Endpoint = undefined;
+var ipc_replies_ok: bool = false;
+var ipc_done: bool = false;
+
+/// Echo-increment server: reply to each request with request+1, forever.
+fn ipcServer() void {
+    var reply_buf: [8]u8 = undefined;
+    var reply_len: u64 = 0;
+    var badge: u64 = 0;
+    while (true) {
+        var recv: [8]u8 = undefined;
+        const n = ipcsync.replyWait(ipc_ep, @intFromPtr(&reply_buf), reply_len, @intFromPtr(&recv), recv.len, &badge);
+        if (n < 0) sched.exit();
+        const v = std.mem.readInt(u64, recv[0..8], .little);
+        std.mem.writeInt(u64, reply_buf[0..8], v + 1, .little);
+        reply_len = 8;
+    }
+}
+
+/// Client: make 100 synchronous calls, checking every reply is request+1.
+fn ipcClient() void {
+    var ok = true;
+    var i: u64 = 0;
+    while (i < 100) : (i += 1) {
+        var msg: [8]u8 = undefined;
+        std.mem.writeInt(u64, msg[0..8], i, .little);
+        var reply: [8]u8 = undefined;
+        const n = ipcsync.call(ipc_ep, @intFromPtr(&msg), 8, @intFromPtr(&reply), reply.len);
+        if (n != 8 or std.mem.readInt(u64, reply[0..8], .little) != i + 1) ok = false;
+    }
+    ipc_replies_ok = ok;
+    ipc_done = true;
+    sched.exit();
+}
+
+/// Synchronous IPC: a client and a server (two kernel tasks) ping-pong 100 calls
+/// through one Endpoint. Each round exercises the full rendezvous — the client
+/// blocks in `call`, the server blocks in `replyWait`, the reply is routed back to
+/// the exact caller, and `copyAcross` moves the bytes — so 100 correct replies
+/// prove the block/wake and reply-routing paths. (Kernel tasks, so no user ELF.)
+fn ipcCallTest() void {
+    log("DANOS-TEST-BEGIN: ipc-call\n", .{});
+    ipc_ep = ipcsync.createEndpoint().?;
+    ipc_replies_ok = false;
+    ipc_done = false;
+    sched.spawn(ipcServer, 5); // above this task, so the workers run
+    sched.spawn(ipcClient, 5);
+
+    const done: *volatile bool = &ipc_done;
+    var spins: u64 = 0;
+    while (!done.* and spins < 100_000_000) : (spins += 1) sched.yield();
+
+    check("client completed 100 synchronous calls", ipc_done);
+    check("every reply was request+1 (rendezvous + reply routing intact)", ipc_replies_ok);
+    result();
+}
+
 var proc_worker_run: bool = true;
 var proc_worker_ran: bool = false;
 

src/root.zig

@@ -70,6 +70,19 @@ pub const Syscall = enum(u64) {
     sleep = 3, // sleep(ms): block the caller for ms milliseconds
     mmap = 4, // mmap(len, prot) -> base: grant zeroed, page-aligned user pages
     munmap = 5, // munmap(base, len): release pages from a prior mmap
+    create_endpoint = 6, // create_endpoint() -> handle: a new IPC endpoint
+    ipc_register = 7, // ipc_register(service_id, handle): publish an endpoint by well-known id
+    ipc_lookup = 8, // ipc_lookup(service_id) -> handle: find a published endpoint
+    ipc_call = 9, // ipc_call(h, msg, len, reply, cap) -> reply_len: send + block for reply
+    ipc_reply_wait = 10, // ipc_reply_wait(h, reply, len, recv, cap) -> recv_len (+badge in rdx)
+    _,
+};
+
+/// Well-known IPC service ids for the bootstrap name registry (create_endpoint +
+/// ipc_register/ipc_lookup). Small integers, so no string interning is needed
+/// during bring-up. The VFS server registers under `vfs`; clients look it up.
+pub const ServiceId = enum(u32) {
+    vfs = 1,
     _,
 };
 

test/qemu_test.py

@@ -116,6 +116,11 @@ CASES = [
     {"name": "ipc",
      "expect": r"DANOS-TEST-RESULT: PASS",
      "fail":   r"DANOS-TEST-RESULT: FAIL"},
+    # Synchronous IPC: a client and server ping-pong 100 calls through an
+    # endpoint (rendezvous, reply routing, cross-address-space copy).
+    {"name": "ipc-call",
+     "expect": r"DANOS-TEST-RESULT: PASS",
+     "fail":   r"DANOS-TEST-RESULT: FAIL"},
     # Parallelism: needs more than one core, so this case boots with -smp 4.
     {"name": "smp",
      "smp":    4,