introduce std.event.fs for async file system functions

only works on linux so far

CMakeLists.txt

@@ -460,6 +460,7 @@ set(ZIG_STD_FILES
     "empty.zig"
     "event.zig"
     "event/channel.zig"
+    "event/fs.zig"
     "event/future.zig"
     "event/group.zig"
     "event/lock.zig"

std/build.zig

@@ -603,10 +603,10 @@ pub const Builder = struct {
     }
 
     fn copyFile(self: *Builder, source_path: []const u8, dest_path: []const u8) !void {
-        return self.copyFileMode(source_path, dest_path, os.default_file_mode);
+        return self.copyFileMode(source_path, dest_path, os.File.default_mode);
     }
 
-    fn copyFileMode(self: *Builder, source_path: []const u8, dest_path: []const u8, mode: os.FileMode) !void {
+    fn copyFileMode(self: *Builder, source_path: []const u8, dest_path: []const u8, mode: os.File.Mode) !void {
         if (self.verbose) {
             warn("cp {} {}\n", source_path, dest_path);
         }

std/debug/index.zig

@@ -672,14 +672,10 @@ fn parseFormValueRef(allocator: *mem.Allocator, in_stream: var, comptime T: type
 
 const ParseFormValueError = error{
     EndOfStream,
-    Io,
-    BadFd,
-    Unexpected,
     InvalidDebugInfo,
     EndOfFile,
-    IsDir,
     OutOfMemory,
-};
+} || std.os.File.ReadError;
 
 fn parseFormValue(allocator: *mem.Allocator, in_stream: var, form_id: u64, is_64: bool) ParseFormValueError!FormValue {
     return switch (form_id) {

std/event.zig

@@ -1,17 +1,19 @@
+pub const Channel = @import("event/channel.zig").Channel;
+pub const Future = @import("event/future.zig").Future;
+pub const Group = @import("event/group.zig").Group;
+pub const Lock = @import("event/lock.zig").Lock;
 pub const Locked = @import("event/locked.zig").Locked;
 pub const Loop = @import("event/loop.zig").Loop;
-pub const Lock = @import("event/lock.zig").Lock;
+pub const fs = @import("event/fs.zig");
 pub const tcp = @import("event/tcp.zig");
-pub const Channel = @import("event/channel.zig").Channel;
-pub const Group = @import("event/group.zig").Group;
-pub const Future = @import("event/future.zig").Future;
 
 test "import event tests" {
+    _ = @import("event/channel.zig");
+    _ = @import("event/fs.zig");
+    _ = @import("event/future.zig");
+    _ = @import("event/group.zig");
+    _ = @import("event/lock.zig");
     _ = @import("event/locked.zig");
     _ = @import("event/loop.zig");
-    _ = @import("event/lock.zig");
     _ = @import("event/tcp.zig");
-    _ = @import("event/channel.zig");
-    _ = @import("event/group.zig");
-    _ = @import("event/future.zig");
 }

std/event/fs.zig

@@ -0,0 +1,343 @@
+const std = @import("../index.zig");
+const event = std.event;
+const assert = std.debug.assert;
+const os = std.os;
+const mem = std.mem;
+
+pub const RequestNode = std.atomic.Queue(Request).Node;
+
+pub const Request = struct {
+    msg: Msg,
+    finish: Finish,
+
+    pub const Finish = union(enum) {
+        TickNode: event.Loop.NextTickNode,
+        DeallocCloseOperation: *CloseOperation,
+        NoAction,
+    };
+
+    pub const Msg = union(enum) {
+        PWriteV: PWriteV,
+        PReadV: PReadV,
+        OpenRead: OpenRead,
+        Close: Close,
+        WriteFile: WriteFile,
+        End, // special - means the fs thread should exit
+
+        pub const PWriteV = struct {
+            fd: os.FileHandle,
+            data: []const []const u8,
+            offset: usize,
+            result: Error!void,
+
+            pub const Error = error{};
+        };
+
+        pub const PReadV = struct {
+            fd: os.FileHandle,
+            iov: []os.linux.iovec,
+            offset: usize,
+            result: Error!usize,
+
+            pub const Error = os.File.ReadError;
+        };
+
+        pub const OpenRead = struct {
+            /// must be null terminated. TODO https://github.com/ziglang/zig/issues/265
+            path: []const u8,
+            result: Error!os.FileHandle,
+
+            pub const Error = os.File.OpenError;
+        };
+
+        pub const WriteFile = struct {
+            /// must be null terminated. TODO https://github.com/ziglang/zig/issues/265
+            path: []const u8,
+            contents: []const u8,
+            mode: os.File.Mode,
+            result: Error!void,
+
+            pub const Error = os.File.OpenError || os.File.WriteError;
+        };
+
+        pub const Close = struct {
+            fd: os.FileHandle,
+        };
+    };
+};
+
+/// data - both the outer and inner references - must live until pwritev promise completes.
+pub async fn pwritev(loop: *event.Loop, fd: os.FileHandle, offset: usize, data: []const []const u8) !void {
+    //const data_dupe = try mem.dupe(loop.allocator, []const u8, data);
+    //defer loop.allocator.free(data_dupe);
+
+    // workaround for https://github.com/ziglang/zig/issues/1194
+    var my_handle: promise = undefined;
+    suspend |p| {
+        my_handle = p;
+        resume p;
+    }
+
+    var req_node = RequestNode{
+        .next = undefined,
+        .data = Request{
+            .msg = Request.Msg{
+                .PWriteV = Request.Msg.PWriteV{
+                    .fd = fd,
+                    .data = data,
+                    .offset = offset,
+                    .result = undefined,
+                },
+            },
+            .finish = Request.Finish{
+                .TickNode = event.Loop.NextTickNode{
+                    .next = undefined,
+                    .data = my_handle,
+                },
+            },
+        },
+    };
+
+    suspend |_| {
+        loop.linuxFsRequest(&req_node);
+    }
+
+    return req_node.data.msg.PWriteV.result;
+}
+
+/// data - just the inner references - must live until pwritev promise completes.
+pub async fn preadv(loop: *event.Loop, fd: os.FileHandle, offset: usize, data: []const []u8) !usize {
+    //const data_dupe = try mem.dupe(loop.allocator, []const u8, data);
+    //defer loop.allocator.free(data_dupe);
+
+    // workaround for https://github.com/ziglang/zig/issues/1194
+    var my_handle: promise = undefined;
+    suspend |p| {
+        my_handle = p;
+        resume p;
+    }
+
+    const iovecs = try loop.allocator.alloc(os.linux.iovec, data.len);
+    defer loop.allocator.free(iovecs);
+
+    for (data) |buf, i| {
+        iovecs[i] = os.linux.iovec{
+            .iov_base = buf.ptr,
+            .iov_len = buf.len,
+        };
+    }
+
+    var req_node = RequestNode{
+        .next = undefined,
+        .data = Request{
+            .msg = Request.Msg{
+                .PReadV = Request.Msg.PReadV{
+                    .fd = fd,
+                    .iov = iovecs,
+                    .offset = offset,
+                    .result = undefined,
+                },
+            },
+            .finish = Request.Finish{
+                .TickNode = event.Loop.NextTickNode{
+                    .next = undefined,
+                    .data = my_handle,
+                },
+            },
+        },
+    };
+
+    suspend |_| {
+        loop.linuxFsRequest(&req_node);
+    }
+
+    return req_node.data.msg.PReadV.result;
+}
+
+pub async fn openRead(loop: *event.Loop, path: []const u8) os.File.OpenError!os.FileHandle {
+    // workaround for https://github.com/ziglang/zig/issues/1194
+    var my_handle: promise = undefined;
+    suspend |p| {
+        my_handle = p;
+        resume p;
+    }
+
+    var req_node = RequestNode{
+        .next = undefined,
+        .data = Request{
+            .msg = Request.Msg{
+                .OpenRead = Request.Msg.OpenRead{
+                    .path = path,
+                    .result = undefined,
+                },
+            },
+            .finish = Request.Finish{
+                .TickNode = event.Loop.NextTickNode{
+                    .next = undefined,
+                    .data = my_handle,
+                },
+            },
+        },
+    };
+
+    suspend |_| {
+        loop.linuxFsRequest(&req_node);
+    }
+
+    return req_node.data.msg.OpenRead.result;
+}
+
+/// This abstraction helps to close file handles in defer expressions
+/// without suspending. Start a CloseOperation before opening a file.
+pub const CloseOperation = struct {
+    loop: *event.Loop,
+    have_fd: bool,
+    close_req_node: RequestNode,
+
+    pub fn create(loop: *event.Loop) (error{OutOfMemory}!*CloseOperation) {
+        const self = try loop.allocator.createOne(CloseOperation);
+        self.* = CloseOperation{
+            .loop = loop,
+            .have_fd = false,
+            .close_req_node = RequestNode{
+                .next = undefined,
+                .data = Request{
+                    .msg = Request.Msg{
+                        .Close = Request.Msg.Close{ .fd = undefined },
+                    },
+                    .finish = Request.Finish{ .DeallocCloseOperation = self },
+                },
+            },
+        };
+        return self;
+    }
+
+    /// Defer this after creating.
+    pub fn deinit(self: *CloseOperation) void {
+        if (self.have_fd) {
+            self.loop.linuxFsRequest(&self.close_req_node);
+        } else {
+            self.loop.allocator.destroy(self);
+        }
+    }
+
+    pub fn setHandle(self: *CloseOperation, handle: os.FileHandle) void {
+        self.close_req_node.data.msg.Close.fd = handle;
+        self.have_fd = true;
+    }
+};
+
+/// contents must remain alive until writeFile completes.
+pub async fn writeFile(loop: *event.Loop, path: []const u8, contents: []const u8) !void {
+    return await (async writeFileMode(loop, path, contents, os.File.default_mode) catch unreachable);
+}
+
+/// contents must remain alive until writeFile completes.
+pub async fn writeFileMode(loop: *event.Loop, path: []const u8, contents: []const u8, mode: os.File.Mode) !void {
+    // workaround for https://github.com/ziglang/zig/issues/1194
+    var my_handle: promise = undefined;
+    suspend |p| {
+        my_handle = p;
+        resume p;
+    }
+
+    const path_with_null = try std.cstr.addNullByte(loop.allocator, path);
+    defer loop.allocator.free(path_with_null);
+
+    var req_node = RequestNode{
+        .next = undefined,
+        .data = Request{
+            .msg = Request.Msg{
+                .WriteFile = Request.Msg.WriteFile{
+                    .path = path_with_null[0..path.len],
+                    .contents = contents,
+                    .mode = mode,
+                    .result = undefined,
+                },
+            },
+            .finish = Request.Finish{
+                .TickNode = event.Loop.NextTickNode{
+                    .next = undefined,
+                    .data = my_handle,
+                },
+            },
+        },
+    };
+
+    suspend |_| {
+        loop.linuxFsRequest(&req_node);
+    }
+
+    return req_node.data.msg.WriteFile.result;
+}
+
+/// The promise resumes when the last data has been confirmed written, but before the file handle
+/// is closed.
+pub async fn readFile(loop: *event.Loop, file_path: []const u8, max_size: usize) ![]u8 {
+    var close_op = try CloseOperation.create(loop);
+    defer close_op.deinit();
+
+    const path_with_null = try std.cstr.addNullByte(loop.allocator, file_path);
+    defer loop.allocator.free(path_with_null);
+
+    const fd = try await (async openRead(loop, path_with_null[0..file_path.len]) catch unreachable);
+    close_op.setHandle(fd);
+
+    var list = std.ArrayList(u8).init(loop.allocator);
+    defer list.deinit();
+
+    while (true) {
+        try list.ensureCapacity(list.len + os.page_size);
+        const buf = list.items[list.len..];
+        const buf_array = [][]u8{buf};
+        const amt = try await (async preadv(loop, fd, list.len, buf_array) catch unreachable);
+        list.len += amt;
+        if (amt < buf.len) {
+            return list.toOwnedSlice();
+        }
+    }
+}
+
+const test_tmp_dir = "std_event_fs_test";
+
+test "write a file, watch it, write it again" {
+    var da = std.heap.DirectAllocator.init();
+    defer da.deinit();
+
+    const allocator = &da.allocator;
+
+    // TODO move this into event loop too
+    try os.makePath(allocator, test_tmp_dir);
+    defer os.deleteTree(allocator, test_tmp_dir) catch {};
+
+    var loop: event.Loop = undefined;
+    try loop.initMultiThreaded(allocator);
+    defer loop.deinit();
+
+    var result: error!void = undefined;
+    const handle = try async<allocator> testFsWatchCantFail(&loop, &result);
+    defer cancel handle;
+
+    loop.run();
+    return result;
+}
+
+async fn testFsWatchCantFail(loop: *event.Loop, result: *(error!void)) void {
+    result.* = await async testFsWatch(loop) catch unreachable;
+}
+
+async fn testFsWatch(loop: *event.Loop) !void {
+    const file_path = try os.path.join(loop.allocator, test_tmp_dir, "file.txt");
+    defer loop.allocator.free(file_path);
+
+    const contents =
+        \\line 1
+        \\line 2
+    ;
+
+    // first just write then read the file
+    try await try async writeFile(loop, file_path, contents);
+
+    const read_contents = try await try async readFile(loop, file_path, 1024 * 1024);
+    assert(mem.eql(u8, read_contents, contents));
+}

std/event/loop.zig

@@ -2,10 +2,12 @@ const std = @import("../index.zig");
 const builtin = @import("builtin");
 const assert = std.debug.assert;
 const mem = std.mem;
-const posix = std.os.posix;
-const windows = std.os.windows;
 const AtomicRmwOp = builtin.AtomicRmwOp;
 const AtomicOrder = builtin.AtomicOrder;
+const fs = std.event.fs;
+const os = std.os;
+const posix = os.posix;
+const windows = os.windows;
 
 pub const Loop = struct {
     allocator: *mem.Allocator,
@@ -13,7 +15,7 @@ pub const Loop = struct {
     os_data: OsData,
     final_resume_node: ResumeNode,
     pending_event_count: usize,
-    extra_threads: []*std.os.Thread,
+    extra_threads: []*os.Thread,
 
     // pre-allocated eventfds. all permanently active.
     // this is how we send promises to be resumed on other threads.
@@ -65,7 +67,7 @@ pub const Loop = struct {
     /// TODO copy elision / named return values so that the threads referencing *Loop
     /// have the correct pointer value.
     pub fn initMultiThreaded(self: *Loop, allocator: *mem.Allocator) !void {
-        const core_count = try std.os.cpuCount(allocator);
+        const core_count = try os.cpuCount(allocator);
         return self.initInternal(allocator, core_count);
     }
 
@@ -92,7 +94,7 @@ pub const Loop = struct {
         );
         errdefer self.allocator.free(self.eventfd_resume_nodes);
 
-        self.extra_threads = try self.allocator.alloc(*std.os.Thread, extra_thread_count);
+        self.extra_threads = try self.allocator.alloc(*os.Thread, extra_thread_count);
         errdefer self.allocator.free(self.extra_threads);
 
         try self.initOsData(extra_thread_count);
@@ -104,17 +106,34 @@ pub const Loop = struct {
         self.allocator.free(self.extra_threads);
     }
 
-    const InitOsDataError = std.os.LinuxEpollCreateError || mem.Allocator.Error || std.os.LinuxEventFdError ||
-        std.os.SpawnThreadError || std.os.LinuxEpollCtlError || std.os.BsdKEventError ||
-        std.os.WindowsCreateIoCompletionPortError;
+    const InitOsDataError = os.LinuxEpollCreateError || mem.Allocator.Error || os.LinuxEventFdError ||
+        os.SpawnThreadError || os.LinuxEpollCtlError || os.BsdKEventError ||
+        os.WindowsCreateIoCompletionPortError;
 
     const wakeup_bytes = []u8{0x1} ** 8;
 
     fn initOsData(self: *Loop, extra_thread_count: usize) InitOsDataError!void {
         switch (builtin.os) {
             builtin.Os.linux => {
+                self.os_data.fs_queue = std.atomic.Queue(fs.Request).init();
+                self.os_data.fs_queue_len = 0;
+                // we need another thread for the file system because Linux does not have an async
+                // file system I/O API.
+                self.os_data.fs_end_request = fs.RequestNode{
+                    .next = undefined,
+                    .data = fs.Request{
+                        .msg = fs.Request.Msg.End,
+                        .finish = fs.Request.Finish.NoAction,
+                    },
+                };
+                self.os_data.fs_thread = try os.spawnThread(self, linuxFsRun);
                 errdefer {
-                    while (self.available_eventfd_resume_nodes.pop()) |node| std.os.close(node.data.eventfd);
+                    self.linuxFsRequest(&self.os_data.fs_end_request);
+                    self.os_data.fs_thread.wait();
+                }
+
+                errdefer {
+                    while (self.available_eventfd_resume_nodes.pop()) |node| os.close(node.data.eventfd);
                 }
                 for (self.eventfd_resume_nodes) |*eventfd_node| {
                     eventfd_node.* = std.atomic.Stack(ResumeNode.EventFd).Node{
@@ -123,7 +142,7 @@ pub const Loop = struct {
                                 .id = ResumeNode.Id.EventFd,
                                 .handle = undefined,
                             },
-                            .eventfd = try std.os.linuxEventFd(1, posix.EFD_CLOEXEC | posix.EFD_NONBLOCK),
+                            .eventfd = try os.linuxEventFd(1, posix.EFD_CLOEXEC | posix.EFD_NONBLOCK),
                             .epoll_op = posix.EPOLL_CTL_ADD,
                         },
                         .next = undefined,
@@ -131,17 +150,17 @@ pub const Loop = struct {
                     self.available_eventfd_resume_nodes.push(eventfd_node);
                 }
 
-                self.os_data.epollfd = try std.os.linuxEpollCreate(posix.EPOLL_CLOEXEC);
-                errdefer std.os.close(self.os_data.epollfd);
+                self.os_data.epollfd = try os.linuxEpollCreate(posix.EPOLL_CLOEXEC);
+                errdefer os.close(self.os_data.epollfd);
 
-                self.os_data.final_eventfd = try std.os.linuxEventFd(0, posix.EFD_CLOEXEC | posix.EFD_NONBLOCK);
-                errdefer std.os.close(self.os_data.final_eventfd);
+                self.os_data.final_eventfd = try os.linuxEventFd(0, posix.EFD_CLOEXEC | posix.EFD_NONBLOCK);
+                errdefer os.close(self.os_data.final_eventfd);
 
                 self.os_data.final_eventfd_event = posix.epoll_event{
                     .events = posix.EPOLLIN,
                     .data = posix.epoll_data{ .ptr = @ptrToInt(&self.final_resume_node) },
                 };
-                try std.os.linuxEpollCtl(
+                try os.linuxEpollCtl(
                     self.os_data.epollfd,
                     posix.EPOLL_CTL_ADD,
                     self.os_data.final_eventfd,
@@ -151,19 +170,19 @@ pub const Loop = struct {
                 var extra_thread_index: usize = 0;
                 errdefer {
                     // writing 8 bytes to an eventfd cannot fail
-                    std.os.posixWrite(self.os_data.final_eventfd, wakeup_bytes) catch unreachable;
+                    os.posixWrite(self.os_data.final_eventfd, wakeup_bytes) catch unreachable;
                     while (extra_thread_index != 0) {
                         extra_thread_index -= 1;
                         self.extra_threads[extra_thread_index].wait();
                     }
                 }
                 while (extra_thread_index < extra_thread_count) : (extra_thread_index += 1) {
-                    self.extra_threads[extra_thread_index] = try std.os.spawnThread(self, workerRun);
+                    self.extra_threads[extra_thread_index] = try os.spawnThread(self, workerRun);
                 }
             },
             builtin.Os.macosx => {
-                self.os_data.kqfd = try std.os.bsdKQueue();
-                errdefer std.os.close(self.os_data.kqfd);
+                self.os_data.kqfd = try os.bsdKQueue();
+                errdefer os.close(self.os_data.kqfd);
 
                 self.os_data.kevents = try self.allocator.alloc(posix.Kevent, extra_thread_count);
                 errdefer self.allocator.free(self.os_data.kevents);
@@ -191,7 +210,7 @@ pub const Loop = struct {
                     };
                     self.available_eventfd_resume_nodes.push(eventfd_node);
                     const kevent_array = (*[1]posix.Kevent)(&eventfd_node.data.kevent);
-                    _ = try std.os.bsdKEvent(self.os_data.kqfd, kevent_array, eventlist, null);
+                    _ = try os.bsdKEvent(self.os_data.kqfd, kevent_array, eventlist, null);
                     eventfd_node.data.kevent.flags = posix.EV_CLEAR | posix.EV_ENABLE;
                     eventfd_node.data.kevent.fflags = posix.NOTE_TRIGGER;
                     // this one is for waiting for events
@@ -216,30 +235,30 @@ pub const Loop = struct {
                     .udata = @ptrToInt(&self.final_resume_node),
                 };
                 const kevent_array = (*[1]posix.Kevent)(&self.os_data.final_kevent);
-                _ = try std.os.bsdKEvent(self.os_data.kqfd, kevent_array, eventlist, null);
+                _ = try os.bsdKEvent(self.os_data.kqfd, kevent_array, eventlist, null);
                 self.os_data.final_kevent.flags = posix.EV_ENABLE;
                 self.os_data.final_kevent.fflags = posix.NOTE_TRIGGER;
 
                 var extra_thread_index: usize = 0;
                 errdefer {
-                    _ = std.os.bsdKEvent(self.os_data.kqfd, kevent_array, eventlist, null) catch unreachable;
+                    _ = os.bsdKEvent(self.os_data.kqfd, kevent_array, eventlist, null) catch unreachable;
                     while (extra_thread_index != 0) {
                         extra_thread_index -= 1;
                         self.extra_threads[extra_thread_index].wait();
                     }
                 }
                 while (extra_thread_index < extra_thread_count) : (extra_thread_index += 1) {
-                    self.extra_threads[extra_thread_index] = try std.os.spawnThread(self, workerRun);
+                    self.extra_threads[extra_thread_index] = try os.spawnThread(self, workerRun);
                 }
             },
             builtin.Os.windows => {
-                self.os_data.io_port = try std.os.windowsCreateIoCompletionPort(
+                self.os_data.io_port = try os.windowsCreateIoCompletionPort(
                     windows.INVALID_HANDLE_VALUE,
                     null,
                     undefined,
                     undefined,
                 );
-                errdefer std.os.close(self.os_data.io_port);
+                errdefer os.close(self.os_data.io_port);
 
                 for (self.eventfd_resume_nodes) |*eventfd_node, i| {
                     eventfd_node.* = std.atomic.Stack(ResumeNode.EventFd).Node{
@@ -262,7 +281,7 @@ pub const Loop = struct {
                     while (i < extra_thread_index) : (i += 1) {
                         while (true) {
                             const overlapped = @intToPtr(?*windows.OVERLAPPED, 0x1);
-                            std.os.windowsPostQueuedCompletionStatus(self.os_data.io_port, undefined, @ptrToInt(&self.final_resume_node), overlapped) catch continue;
+                            os.windowsPostQueuedCompletionStatus(self.os_data.io_port, undefined, @ptrToInt(&self.final_resume_node), overlapped) catch continue;
                             break;
                         }
                     }
@@ -272,7 +291,7 @@ pub const Loop = struct {
                     }
                 }
                 while (extra_thread_index < extra_thread_count) : (extra_thread_index += 1) {
-                    self.extra_threads[extra_thread_index] = try std.os.spawnThread(self, workerRun);
+                    self.extra_threads[extra_thread_index] = try os.spawnThread(self, workerRun);
                 }
             },
             else => {},
@@ -282,17 +301,17 @@ pub const Loop = struct {
     fn deinitOsData(self: *Loop) void {
         switch (builtin.os) {
             builtin.Os.linux => {
-                std.os.close(self.os_data.final_eventfd);
-                while (self.available_eventfd_resume_nodes.pop()) |node| std.os.close(node.data.eventfd);
-                std.os.close(self.os_data.epollfd);
+                os.close(self.os_data.final_eventfd);
+                while (self.available_eventfd_resume_nodes.pop()) |node| os.close(node.data.eventfd);
+                os.close(self.os_data.epollfd);
                 self.allocator.free(self.eventfd_resume_nodes);
             },
             builtin.Os.macosx => {
                 self.allocator.free(self.os_data.kevents);
-                std.os.close(self.os_data.kqfd);
+                os.close(self.os_data.kqfd);
             },
             builtin.Os.windows => {
-                std.os.close(self.os_data.io_port);
+                os.close(self.os_data.io_port);
             },
             else => {},
         }
@@ -307,17 +326,17 @@ pub const Loop = struct {
         try self.modFd(
             fd,
             posix.EPOLL_CTL_ADD,
-            std.os.linux.EPOLLIN | std.os.linux.EPOLLOUT | std.os.linux.EPOLLET,
+            os.linux.EPOLLIN | os.linux.EPOLLOUT | os.linux.EPOLLET,
             resume_node,
         );
     }
 
     pub fn modFd(self: *Loop, fd: i32, op: u32, events: u32, resume_node: *ResumeNode) !void {
-        var ev = std.os.linux.epoll_event{
+        var ev = os.linux.epoll_event{
             .events = events,
-            .data = std.os.linux.epoll_data{ .ptr = @ptrToInt(resume_node) },
+            .data = os.linux.epoll_data{ .ptr = @ptrToInt(resume_node) },
         };
-        try std.os.linuxEpollCtl(self.os_data.epollfd, op, fd, &ev);
+        try os.linuxEpollCtl(self.os_data.epollfd, op, fd, &ev);
     }
 
     pub fn removeFd(self: *Loop, fd: i32) void {
@@ -326,7 +345,7 @@ pub const Loop = struct {
     }
 
     fn removeFdNoCounter(self: *Loop, fd: i32) void {
-        std.os.linuxEpollCtl(self.os_data.epollfd, std.os.linux.EPOLL_CTL_DEL, fd, undefined) catch {};
+        os.linuxEpollCtl(self.os_data.epollfd, os.linux.EPOLL_CTL_DEL, fd, undefined) catch {};
     }
 
     pub async fn waitFd(self: *Loop, fd: i32) !void {
@@ -353,7 +372,7 @@ pub const Loop = struct {
                 builtin.Os.macosx => {
                     const kevent_array = (*[1]posix.Kevent)(&eventfd_node.kevent);
                     const eventlist = ([*]posix.Kevent)(undefined)[0..0];
-                    _ = std.os.bsdKEvent(self.os_data.kqfd, kevent_array, eventlist, null) catch {
+                    _ = os.bsdKEvent(self.os_data.kqfd, kevent_array, eventlist, null) catch {
                         self.next_tick_queue.unget(next_tick_node);
                         self.available_eventfd_resume_nodes.push(resume_stack_node);
                         return;
@@ -361,8 +380,8 @@ pub const Loop = struct {
                 },
                 builtin.Os.linux => {
                     // the pending count is already accounted for
-                    const epoll_events = posix.EPOLLONESHOT | std.os.linux.EPOLLIN | std.os.linux.EPOLLOUT |
-                        std.os.linux.EPOLLET;
+                    const epoll_events = posix.EPOLLONESHOT | os.linux.EPOLLIN | os.linux.EPOLLOUT |
+                        os.linux.EPOLLET;
                     self.modFd(
                         eventfd_node.eventfd,
                         eventfd_node.epoll_op,
@@ -379,7 +398,7 @@ pub const Loop = struct {
                     // the consumer code can decide whether to read the completion key.
                     // it has to do this for normal I/O, so we match that behavior here.
                     const overlapped = @intToPtr(?*windows.OVERLAPPED, 0x1);
-                    std.os.windowsPostQueuedCompletionStatus(
+                    os.windowsPostQueuedCompletionStatus(
                         self.os_data.io_port,
                         undefined,
                         eventfd_node.completion_key,
@@ -406,6 +425,9 @@ pub const Loop = struct {
         self.finishOneEvent(); // the reference we start with
 
         self.workerRun();
+
+        self.os_data.fs_thread.wait();
+
         for (self.extra_threads) |extra_thread| {
             extra_thread.wait();
         }
@@ -453,15 +475,16 @@ pub const Loop = struct {
             // cause all the threads to stop
             switch (builtin.os) {
                 builtin.Os.linux => {
+                    self.linuxFsRequest(&self.os_data.fs_end_request);
                     // writing 8 bytes to an eventfd cannot fail
-                    std.os.posixWrite(self.os_data.final_eventfd, wakeup_bytes) catch unreachable;
+                    os.posixWrite(self.os_data.final_eventfd, wakeup_bytes) catch unreachable;
                     return;
                 },
                 builtin.Os.macosx => {
                     const final_kevent = (*[1]posix.Kevent)(&self.os_data.final_kevent);
                     const eventlist = ([*]posix.Kevent)(undefined)[0..0];
                     // cannot fail because we already added it and this just enables it
-                    _ = std.os.bsdKEvent(self.os_data.kqfd, final_kevent, eventlist, null) catch unreachable;
+                    _ = os.bsdKEvent(self.os_data.kqfd, final_kevent, eventlist, null) catch unreachable;
                     return;
                 },
                 builtin.Os.windows => {
@@ -469,7 +492,7 @@ pub const Loop = struct {
                     while (i < self.extra_threads.len + 1) : (i += 1) {
                         while (true) {
                             const overlapped = @intToPtr(?*windows.OVERLAPPED, 0x1);
-                            std.os.windowsPostQueuedCompletionStatus(self.os_data.io_port, undefined, @ptrToInt(&self.final_resume_node), overlapped) catch continue;
+                            os.windowsPostQueuedCompletionStatus(self.os_data.io_port, undefined, @ptrToInt(&self.final_resume_node), overlapped) catch continue;
                             break;
                         }
                     }
@@ -492,8 +515,8 @@ pub const Loop = struct {
             switch (builtin.os) {
                 builtin.Os.linux => {
                     // only process 1 event so we don't steal from other threads
-                    var events: [1]std.os.linux.epoll_event = undefined;
-                    const count = std.os.linuxEpollWait(self.os_data.epollfd, events[0..], -1);
+                    var events: [1]os.linux.epoll_event = undefined;
+                    const count = os.linuxEpollWait(self.os_data.epollfd, events[0..], -1);
                     for (events[0..count]) |ev| {
                         const resume_node = @intToPtr(*ResumeNode, ev.data.ptr);
                         const handle = resume_node.handle;
@@ -516,7 +539,7 @@ pub const Loop = struct {
                 },
                 builtin.Os.macosx => {
                     var eventlist: [1]posix.Kevent = undefined;
-                    const count = std.os.bsdKEvent(self.os_data.kqfd, self.os_data.kevents, eventlist[0..], null) catch unreachable;
+                    const count = os.bsdKEvent(self.os_data.kqfd, self.os_data.kevents, eventlist[0..], null) catch unreachable;
                     for (eventlist[0..count]) |ev| {
                         const resume_node = @intToPtr(*ResumeNode, ev.udata);
                         const handle = resume_node.handle;
@@ -541,9 +564,9 @@ pub const Loop = struct {
                     while (true) {
                         var nbytes: windows.DWORD = undefined;
                         var overlapped: ?*windows.OVERLAPPED = undefined;
-                        switch (std.os.windowsGetQueuedCompletionStatus(self.os_data.io_port, &nbytes, &completion_key, &overlapped, windows.INFINITE)) {
-                            std.os.WindowsWaitResult.Aborted => return,
-                            std.os.WindowsWaitResult.Normal => {},
+                        switch (os.windowsGetQueuedCompletionStatus(self.os_data.io_port, &nbytes, &completion_key, &overlapped, windows.INFINITE)) {
+                            os.WindowsWaitResult.Aborted => return,
+                            os.WindowsWaitResult.Normal => {},
                         }
                         if (overlapped != null) break;
                     }
@@ -569,11 +592,73 @@ pub const Loop = struct {
         }
     }
 
+    fn linuxFsRequest(self: *Loop, request_node: *fs.RequestNode) void {
+        _ = @atomicRmw(usize, &self.pending_event_count, AtomicRmwOp.Add, 1, AtomicOrder.SeqCst);
+        self.os_data.fs_queue.put(request_node);
+        _ = @atomicRmw(i32, &self.os_data.fs_queue_len, AtomicRmwOp.Add, 1, AtomicOrder.SeqCst); // let this wrap
+        const rc = os.linux.futex_wake(@ptrToInt(&self.os_data.fs_queue_len), os.linux.FUTEX_WAKE, 1);
+        switch (os.linux.getErrno(rc)) {
+            0 => {},
+            posix.EINVAL => unreachable,
+            else => unreachable,
+        }
+    }
+
+    fn linuxFsRun(self: *Loop) void {
+        var processed_count: i32 = 0; // we let this wrap
+        while (true) {
+            while (self.os_data.fs_queue.get()) |node| {
+                processed_count +%= 1;
+                switch (node.data.msg) {
+                    @TagType(fs.Request.Msg).PWriteV => @panic("TODO"),
+                    @TagType(fs.Request.Msg).PReadV => |*msg| {
+                        msg.result = os.posix_preadv(msg.fd, msg.iov.ptr, msg.iov.len, msg.offset);
+                    },
+                    @TagType(fs.Request.Msg).OpenRead => |*msg| {
+                        const flags = posix.O_LARGEFILE | posix.O_RDONLY;
+                        msg.result = os.posixOpenC(msg.path.ptr, flags, 0);
+                    },
+                    @TagType(fs.Request.Msg).Close => |*msg| os.close(msg.fd),
+                    @TagType(fs.Request.Msg).WriteFile => |*msg| blk: {
+                        const flags = posix.O_LARGEFILE | posix.O_WRONLY | posix.O_CREAT |
+                            posix.O_CLOEXEC | posix.O_TRUNC;
+                        const fd = os.posixOpenC(msg.path.ptr, flags, msg.mode) catch |err| {
+                            msg.result = err;
+                            break :blk;
+                        };
+                        defer os.close(fd);
+                        msg.result = os.posixWrite(fd, msg.contents);
+                    },
+                    @TagType(fs.Request.Msg).End => return,
+                }
+                switch (node.data.finish) {
+                    @TagType(fs.Request.Finish).TickNode => |*tick_node| self.onNextTick(tick_node),
+                    @TagType(fs.Request.Finish).DeallocCloseOperation => |close_op| {
+                        self.allocator.destroy(close_op);
+                    },
+                    @TagType(fs.Request.Finish).NoAction => {},
+                }
+                self.finishOneEvent();
+            }
+            const rc = os.linux.futex_wait(@ptrToInt(&self.os_data.fs_queue_len), os.linux.FUTEX_WAIT, processed_count, null);
+            switch (os.linux.getErrno(rc)) {
+                0 => continue,
+                posix.EINTR => continue,
+                posix.EAGAIN => continue,
+                else => unreachable,
+            }
+        }
+    }
+
     const OsData = switch (builtin.os) {
         builtin.Os.linux => struct {
             epollfd: i32,
             final_eventfd: i32,
-            final_eventfd_event: std.os.linux.epoll_event,
+            final_eventfd_event: os.linux.epoll_event,
+            fs_thread: *os.Thread,
+            fs_queue_len: i32, // we let this wrap
+            fs_queue: std.atomic.Queue(fs.Request),
+            fs_end_request: fs.RequestNode,
         },
         builtin.Os.macosx => MacOsData,
         builtin.Os.windows => struct {

std/io.zig

@@ -415,7 +415,6 @@ pub fn PeekStream(comptime buffer_size: usize, comptime InStreamError: type) typ
             self.at_end = (read < left);
             return pos + read;
         }
-
     };
 }
 
@@ -481,8 +480,7 @@ pub const SliceOutStream = struct {
 
         assert(self.pos <= self.slice.len);
 
-        const n =
-            if (self.pos + bytes.len <= self.slice.len)
+        const n = if (self.pos + bytes.len <= self.slice.len)
             bytes.len
         else
             self.slice.len - self.pos;
@@ -586,7 +584,7 @@ pub const BufferedAtomicFile = struct {
         });
         errdefer allocator.destroy(self);
 
-        self.atomic_file = try os.AtomicFile.init(allocator, dest_path, os.default_file_mode);
+        self.atomic_file = try os.AtomicFile.init(allocator, dest_path, os.File.default_mode);
         errdefer self.atomic_file.deinit();
 
         self.file_stream = FileOutStream.init(&self.atomic_file.file);

std/os/file.zig

@@ -15,10 +15,21 @@ pub const File = struct {
     /// The OS-specific file descriptor or file handle.
     handle: os.FileHandle,
 
+    pub const Mode = switch (builtin.os) {
+        Os.windows => void,
+        else => u32,
+    };
+
+    pub const default_mode = switch (builtin.os) {
+        Os.windows => {},
+        else => 0o666,
+    };
+
     pub const OpenError = os.WindowsOpenError || os.PosixOpenError;
 
     /// `path` needs to be copied in memory to add a null terminating byte, hence the allocator.
     /// Call close to clean up.
+    /// TODO deprecated, just use open
     pub fn openRead(allocator: *mem.Allocator, path: []const u8) OpenError!File {
         if (is_posix) {
             const flags = posix.O_LARGEFILE | posix.O_RDONLY;
@@ -39,16 +50,18 @@ pub const File = struct {
         }
     }
 
-    /// Calls `openWriteMode` with os.default_file_mode for the mode.
+    /// Calls `openWriteMode` with os.File.default_mode for the mode.
+    /// TODO deprecated, just use open
     pub fn openWrite(allocator: *mem.Allocator, path: []const u8) OpenError!File {
-        return openWriteMode(allocator, path, os.default_file_mode);
+        return openWriteMode(allocator, path, os.File.default_mode);
     }
 
     /// If the path does not exist it will be created.
     /// If a file already exists in the destination it will be truncated.
     /// `path` needs to be copied in memory to add a null terminating byte, hence the allocator.
     /// Call close to clean up.
-    pub fn openWriteMode(allocator: *mem.Allocator, path: []const u8, file_mode: os.FileMode) OpenError!File {
+    /// TODO deprecated, just use open
+    pub fn openWriteMode(allocator: *mem.Allocator, path: []const u8, file_mode: Mode) OpenError!File {
         if (is_posix) {
             const flags = posix.O_LARGEFILE | posix.O_WRONLY | posix.O_CREAT | posix.O_CLOEXEC | posix.O_TRUNC;
             const fd = try os.posixOpen(allocator, path, flags, file_mode);
@@ -72,7 +85,8 @@ pub const File = struct {
     /// If a file already exists in the destination this returns OpenError.PathAlreadyExists
     /// `path` needs to be copied in memory to add a null terminating byte, hence the allocator.
     /// Call close to clean up.
-    pub fn openWriteNoClobber(allocator: *mem.Allocator, path: []const u8, file_mode: os.FileMode) OpenError!File {
+    /// TODO deprecated, just use open
+    pub fn openWriteNoClobber(allocator: *mem.Allocator, path: []const u8, file_mode: Mode) OpenError!File {
         if (is_posix) {
             const flags = posix.O_LARGEFILE | posix.O_WRONLY | posix.O_CREAT | posix.O_CLOEXEC | posix.O_EXCL;
             const fd = try os.posixOpen(allocator, path, flags, file_mode);
@@ -282,7 +296,7 @@ pub const File = struct {
         Unexpected,
     };
 
-    pub fn mode(self: *File) ModeError!os.FileMode {
+    pub fn mode(self: *File) ModeError!Mode {
         if (is_posix) {
             var stat: posix.Stat = undefined;
             const err = posix.getErrno(posix.fstat(self.handle, &stat));
@@ -296,7 +310,7 @@ pub const File = struct {
 
             // TODO: we should be able to cast u16 to ModeError!u32, making this
             // explicit cast not necessary
-            return os.FileMode(stat.mode);
+            return Mode(stat.mode);
         } else if (is_windows) {
             return {};
         } else {
@@ -305,9 +319,11 @@ pub const File = struct {
     }
 
     pub const ReadError = error{
-        BadFd,
-        Io,
+        FileClosed,
+        InputOutput,
         IsDir,
+        WouldBlock,
+        SystemResources,
 
         Unexpected,
     };
@@ -323,9 +339,12 @@ pub const File = struct {
                         posix.EINTR => continue,
                         posix.EINVAL => unreachable,
                         posix.EFAULT => unreachable,
-                        posix.EBADF => return error.BadFd,
-                        posix.EIO => return error.Io,
+                        posix.EAGAIN => return error.WouldBlock,
+                        posix.EBADF => return error.FileClosed,
+                        posix.EIO => return error.InputOutput,
                         posix.EISDIR => return error.IsDir,
+                        posix.ENOBUFS => return error.SystemResources,
+                        posix.ENOMEM => return error.SystemResources,
                         else => return os.unexpectedErrorPosix(read_err),
                     }
                 }

std/os/index.zig

@@ -38,16 +38,6 @@ pub const path = @import("path.zig");
 pub const File = @import("file.zig").File;
 pub const time = @import("time.zig");
 
-pub const FileMode = switch (builtin.os) {
-    Os.windows => void,
-    else => u32,
-};
-
-pub const default_file_mode = switch (builtin.os) {
-    Os.windows => {},
-    else => 0o666,
-};
-
 pub const page_size = 4 * 1024;
 
 pub const UserInfo = @import("get_user_id.zig").UserInfo;
@@ -256,6 +246,26 @@ pub fn posixRead(fd: i32, buf: []u8) !void {
     }
 }
 
+pub fn posix_preadv(fd: i32, iov: [*]const posix.iovec, count: usize, offset: u64) !usize {
+    while (true) {
+        const rc = posix.preadv(fd, iov, count, offset);
+        const err = posix.getErrno(rc);
+        switch (err) {
+            0 => return rc,
+            posix.EINTR => continue,
+            posix.EINVAL => unreachable,
+            posix.EFAULT => unreachable,
+            posix.EAGAIN => return error.WouldBlock,
+            posix.EBADF => return error.FileClosed,
+            posix.EIO => return error.InputOutput,
+            posix.EISDIR => return error.IsDir,
+            posix.ENOBUFS => return error.SystemResources,
+            posix.ENOMEM => return error.SystemResources,
+            else => return unexpectedErrorPosix(err),
+        }
+    }
+}
+
 pub const PosixWriteError = error{
     WouldBlock,
     FileClosed,
@@ -853,7 +863,7 @@ pub fn copyFile(allocator: *Allocator, source_path: []const u8, dest_path: []con
 /// Guaranteed to be atomic. However until https://patchwork.kernel.org/patch/9636735/ is
 /// merged and readily available,
 /// there is a possibility of power loss or application termination leaving temporary files present
-pub fn copyFileMode(allocator: *Allocator, source_path: []const u8, dest_path: []const u8, mode: FileMode) !void {
+pub fn copyFileMode(allocator: *Allocator, source_path: []const u8, dest_path: []const u8, mode: File.Mode) !void {
     var in_file = try os.File.openRead(allocator, source_path);
     defer in_file.close();
 
@@ -879,7 +889,7 @@ pub const AtomicFile = struct {
 
     /// dest_path must remain valid for the lifetime of AtomicFile
     /// call finish to atomically replace dest_path with contents
-    pub fn init(allocator: *Allocator, dest_path: []const u8, mode: FileMode) !AtomicFile {
+    pub fn init(allocator: *Allocator, dest_path: []const u8, mode: File.Mode) !AtomicFile {
         const dirname = os.path.dirname(dest_path);
 
         var rand_buf: [12]u8 = undefined;

std/os/linux/index.zig

@@ -692,6 +692,10 @@ pub fn futex_wait(uaddr: usize, futex_op: u32, val: i32, timeout: ?*timespec) us
     return syscall4(SYS_futex, uaddr, futex_op, @bitCast(u32, val), @ptrToInt(timeout));
 }
 
+pub fn futex_wake(uaddr: usize, futex_op: u32, val: i32) usize {
+    return syscall3(SYS_futex, uaddr, futex_op, @bitCast(u32, val));
+}
+
 pub fn getcwd(buf: [*]u8, size: usize) usize {
     return syscall2(SYS_getcwd, @ptrToInt(buf), size);
 }
@@ -742,6 +746,10 @@ pub fn read(fd: i32, buf: [*]u8, count: usize) usize {
     return syscall3(SYS_read, @intCast(usize, fd), @ptrToInt(buf), count);
 }
 
+pub fn preadv(fd: i32, iov: [*]const iovec, count: usize, offset: u64) usize {
+    return syscall4(SYS_preadv, @intCast(usize, fd), @ptrToInt(iov), count, offset);
+}
+
 // TODO https://github.com/ziglang/zig/issues/265
 pub fn rmdir(path: [*]const u8) usize {
     return syscall1(SYS_rmdir, @ptrToInt(path));