zig/src/test.zig

922 lines
36 KiB
Zig

const std = @import("std");
const link = @import("link.zig");
const Compilation = @import("Compilation.zig");
const Allocator = std.mem.Allocator;
const zir = @import("zir.zig");
const Package = @import("Package.zig");
const introspect = @import("introspect.zig");
const build_options = @import("build_options");
const enable_qemu: bool = build_options.enable_qemu;
const enable_wine: bool = build_options.enable_wine;
const enable_wasmtime: bool = build_options.enable_wasmtime;
const glibc_multi_install_dir: ?[]const u8 = build_options.glibc_multi_install_dir;
const cheader = @embedFile("link/cbe.h");
test "self-hosted" {
var ctx = TestContext.init();
defer ctx.deinit();
try @import("stage2_tests").addCases(&ctx);
try ctx.run();
}
const ErrorMsg = union(enum) {
src: struct {
msg: []const u8,
line: u32,
column: u32,
},
plain: struct {
msg: []const u8,
},
fn init(other: Compilation.AllErrors.Message) ErrorMsg {
switch (other) {
.src => |src| return .{
.src = .{
.msg = src.msg,
.line = @intCast(u32, src.line),
.column = @intCast(u32, src.column),
},
},
.plain => |plain| return .{
.plain = .{
.msg = plain.msg,
},
},
}
}
pub fn format(
self: ErrorMsg,
comptime fmt: []const u8,
options: std.fmt.FormatOptions,
writer: anytype,
) !void {
switch (self) {
.src => |src| {
return writer.print(":{d}:{d}: error: {s}", .{
src.line + 1,
src.column + 1,
src.msg,
});
},
.plain => |plain| {
return writer.print("error: {s}", .{plain.msg});
},
}
}
};
pub const TestContext = struct {
/// TODO: find a way to treat cases as individual tests (shouldn't show "1 test passed" if there are 200 cases)
cases: std.ArrayList(Case),
pub const Update = struct {
/// The input to the current update. We simulate an incremental update
/// with the file's contents changed to this value each update.
///
/// This value can change entirely between updates, which would be akin
/// to deleting the source file and creating a new one from scratch; or
/// you can keep it mostly consistent, with small changes, testing the
/// effects of the incremental compilation.
src: [:0]const u8,
case: union(enum) {
/// A transformation update transforms the input and tests against
/// the expected output ZIR.
Transformation: [:0]const u8,
/// An error update attempts to compile bad code, and ensures that it
/// fails to compile, and for the expected reasons.
/// A slice containing the expected errors *in sequential order*.
Error: []const ErrorMsg,
/// An execution update compiles and runs the input, testing the
/// stdout against the expected results
/// This is a slice containing the expected message.
Execution: []const u8,
},
};
pub const File = struct {
/// Contents of the importable file. Doesn't yet support incremental updates.
src: [:0]const u8,
path: []const u8,
};
pub const TestType = enum {
Zig,
ZIR,
};
/// A Case consists of a set of *updates*. The same Compilation is used for each
/// update, so each update's source is treated as a single file being
/// updated by the test harness and incrementally compiled.
pub const Case = struct {
/// The name of the test case. This is shown if a test fails, and
/// otherwise ignored.
name: []const u8,
/// The platform the test targets. For non-native platforms, an emulator
/// such as QEMU is required for tests to complete.
target: std.zig.CrossTarget,
/// In order to be able to run e.g. Execution updates, this must be set
/// to Executable.
output_mode: std.builtin.OutputMode,
updates: std.ArrayList(Update),
extension: TestType,
cbe: bool = false,
files: std.ArrayList(File),
/// Adds a subcase in which the module is updated with `src`, and the
/// resulting ZIR is validated against `result`.
pub fn addTransform(self: *Case, src: [:0]const u8, result: [:0]const u8) void {
self.updates.append(.{
.src = src,
.case = .{ .Transformation = result },
}) catch unreachable;
}
/// Adds a subcase in which the module is updated with `src`, compiled,
/// run, and the output is tested against `result`.
pub fn addCompareOutput(self: *Case, src: [:0]const u8, result: []const u8) void {
self.updates.append(.{
.src = src,
.case = .{ .Execution = result },
}) catch unreachable;
}
/// Adds a subcase in which the module is updated with `src`, which
/// should contain invalid input, and ensures that compilation fails
/// for the expected reasons, given in sequential order in `errors` in
/// the form `:line:column: error: message`.
pub fn addError(self: *Case, src: [:0]const u8, errors: []const []const u8) void {
var array = self.updates.allocator.alloc(ErrorMsg, errors.len) catch unreachable;
for (errors) |e, i| {
if (e[0] != ':') {
array[i] = .{ .plain = .{ .msg = e } };
continue;
}
var cur = e[1..];
var line_index = std.mem.indexOf(u8, cur, ":");
if (line_index == null) {
@panic("Invalid test: error must be specified as follows:\n:line:column: error: message\n=========\n");
}
const line = std.fmt.parseInt(u32, cur[0..line_index.?], 10) catch @panic("Unable to parse line number");
cur = cur[line_index.? + 1 ..];
const column_index = std.mem.indexOf(u8, cur, ":");
if (column_index == null) {
@panic("Invalid test: error must be specified as follows:\n:line:column: error: message\n=========\n");
}
const column = std.fmt.parseInt(u32, cur[0..column_index.?], 10) catch @panic("Unable to parse column number");
cur = cur[column_index.? + 2 ..];
if (!std.mem.eql(u8, cur[0..7], "error: ")) {
@panic("Invalid test: error must be specified as follows:\n:line:column: error: message\n=========\n");
}
const msg = cur[7..];
if (line == 0 or column == 0) {
@panic("Invalid test: error line and column must be specified starting at one!");
}
array[i] = .{
.src = .{
.msg = msg,
.line = line - 1,
.column = column - 1,
},
};
}
self.updates.append(.{ .src = src, .case = .{ .Error = array } }) catch unreachable;
}
/// Adds a subcase in which the module is updated with `src`, and
/// asserts that it compiles without issue
pub fn compiles(self: *Case, src: [:0]const u8) void {
self.addError(src, &[_][]const u8{});
}
};
pub fn addExe(
ctx: *TestContext,
name: []const u8,
target: std.zig.CrossTarget,
T: TestType,
) *Case {
ctx.cases.append(Case{
.name = name,
.target = target,
.updates = std.ArrayList(Update).init(ctx.cases.allocator),
.output_mode = .Exe,
.extension = T,
.files = std.ArrayList(File).init(ctx.cases.allocator),
}) catch unreachable;
return &ctx.cases.items[ctx.cases.items.len - 1];
}
/// Adds a test case for Zig input, producing an executable
pub fn exe(ctx: *TestContext, name: []const u8, target: std.zig.CrossTarget) *Case {
return ctx.addExe(name, target, .Zig);
}
/// Adds a test case for ZIR input, producing an executable
pub fn exeZIR(ctx: *TestContext, name: []const u8, target: std.zig.CrossTarget) *Case {
return ctx.addExe(name, target, .ZIR);
}
pub fn addObj(
ctx: *TestContext,
name: []const u8,
target: std.zig.CrossTarget,
T: TestType,
) *Case {
ctx.cases.append(Case{
.name = name,
.target = target,
.updates = std.ArrayList(Update).init(ctx.cases.allocator),
.output_mode = .Obj,
.extension = T,
.files = std.ArrayList(File).init(ctx.cases.allocator),
}) catch unreachable;
return &ctx.cases.items[ctx.cases.items.len - 1];
}
/// Adds a test case for Zig input, producing an object file
pub fn obj(ctx: *TestContext, name: []const u8, target: std.zig.CrossTarget) *Case {
return ctx.addObj(name, target, .Zig);
}
/// Adds a test case for ZIR input, producing an object file
pub fn objZIR(ctx: *TestContext, name: []const u8, target: std.zig.CrossTarget) *Case {
return ctx.addObj(name, target, .ZIR);
}
pub fn addC(ctx: *TestContext, name: []const u8, target: std.zig.CrossTarget, T: TestType) *Case {
ctx.cases.append(Case{
.name = name,
.target = target,
.updates = std.ArrayList(Update).init(ctx.cases.allocator),
.output_mode = .Obj,
.extension = T,
.cbe = true,
.files = std.ArrayList(File).init(ctx.cases.allocator),
}) catch unreachable;
return &ctx.cases.items[ctx.cases.items.len - 1];
}
pub fn c(ctx: *TestContext, name: []const u8, target: std.zig.CrossTarget, src: [:0]const u8, comptime out: [:0]const u8) void {
ctx.addC(name, target, .Zig).addTransform(src, cheader ++ out);
}
pub fn addCompareOutput(
ctx: *TestContext,
name: []const u8,
T: TestType,
src: [:0]const u8,
expected_stdout: []const u8,
) void {
ctx.addExe(name, .{}, T).addCompareOutput(src, expected_stdout);
}
/// Adds a test case that compiles the Zig source given in `src`, executes
/// it, runs it, and tests the output against `expected_stdout`
pub fn compareOutput(
ctx: *TestContext,
name: []const u8,
src: [:0]const u8,
expected_stdout: []const u8,
) void {
return ctx.addCompareOutput(name, .Zig, src, expected_stdout);
}
/// Adds a test case that compiles the ZIR source given in `src`, executes
/// it, runs it, and tests the output against `expected_stdout`
pub fn compareOutputZIR(
ctx: *TestContext,
name: []const u8,
src: [:0]const u8,
expected_stdout: []const u8,
) void {
ctx.addCompareOutput(name, .ZIR, src, expected_stdout);
}
pub fn addTransform(
ctx: *TestContext,
name: []const u8,
target: std.zig.CrossTarget,
T: TestType,
src: [:0]const u8,
result: [:0]const u8,
) void {
ctx.addObj(name, target, T).addTransform(src, result);
}
/// Adds a test case that compiles the Zig given in `src` to ZIR and tests
/// the ZIR against `result`
pub fn transform(
ctx: *TestContext,
name: []const u8,
target: std.zig.CrossTarget,
src: [:0]const u8,
result: [:0]const u8,
) void {
ctx.addTransform(name, target, .Zig, src, result);
}
/// Adds a test case that cleans up the ZIR source given in `src`, and
/// tests the resulting ZIR against `result`
pub fn transformZIR(
ctx: *TestContext,
name: []const u8,
target: std.zig.CrossTarget,
src: [:0]const u8,
result: [:0]const u8,
) void {
ctx.addTransform(name, target, .ZIR, src, result);
}
pub fn addError(
ctx: *TestContext,
name: []const u8,
target: std.zig.CrossTarget,
T: TestType,
src: [:0]const u8,
expected_errors: []const []const u8,
) void {
ctx.addObj(name, target, T).addError(src, expected_errors);
}
/// Adds a test case that ensures that the Zig given in `src` fails to
/// compile for the expected reasons, given in sequential order in
/// `expected_errors` in the form `:line:column: error: message`.
pub fn compileError(
ctx: *TestContext,
name: []const u8,
target: std.zig.CrossTarget,
src: [:0]const u8,
expected_errors: []const []const u8,
) void {
ctx.addError(name, target, .Zig, src, expected_errors);
}
/// Adds a test case that ensures that the ZIR given in `src` fails to
/// compile for the expected reasons, given in sequential order in
/// `expected_errors` in the form `:line:column: error: message`.
pub fn compileErrorZIR(
ctx: *TestContext,
name: []const u8,
target: std.zig.CrossTarget,
src: [:0]const u8,
expected_errors: []const []const u8,
) void {
ctx.addError(name, target, .ZIR, src, expected_errors);
}
pub fn addCompiles(
ctx: *TestContext,
name: []const u8,
target: std.zig.CrossTarget,
T: TestType,
src: [:0]const u8,
) void {
ctx.addObj(name, target, T).compiles(src);
}
/// Adds a test case that asserts that the Zig given in `src` compiles
/// without any errors.
pub fn compiles(
ctx: *TestContext,
name: []const u8,
target: std.zig.CrossTarget,
src: [:0]const u8,
) void {
ctx.addCompiles(name, target, .Zig, src);
}
/// Adds a test case that asserts that the ZIR given in `src` compiles
/// without any errors.
pub fn compilesZIR(
ctx: *TestContext,
name: []const u8,
target: std.zig.CrossTarget,
src: [:0]const u8,
) void {
ctx.addCompiles(name, target, .ZIR, src);
}
/// Adds a test case that first ensures that the Zig given in `src` fails
/// to compile for the reasons given in sequential order in
/// `expected_errors` in the form `:line:column: error: message`, then
/// asserts that fixing the source (updating with `fixed_src`) isn't broken
/// by incremental compilation.
pub fn incrementalFailure(
ctx: *TestContext,
name: []const u8,
target: std.zig.CrossTarget,
src: [:0]const u8,
expected_errors: []const []const u8,
fixed_src: [:0]const u8,
) void {
var case = ctx.addObj(name, target, .Zig);
case.addError(src, expected_errors);
case.compiles(fixed_src);
}
/// Adds a test case that first ensures that the ZIR given in `src` fails
/// to compile for the reasons given in sequential order in
/// `expected_errors` in the form `:line:column: error: message`, then
/// asserts that fixing the source (updating with `fixed_src`) isn't broken
/// by incremental compilation.
pub fn incrementalFailureZIR(
ctx: *TestContext,
name: []const u8,
target: std.zig.CrossTarget,
src: [:0]const u8,
expected_errors: []const []const u8,
fixed_src: [:0]const u8,
) void {
var case = ctx.addObj(name, target, .ZIR);
case.addError(src, expected_errors);
case.compiles(fixed_src);
}
fn init() TestContext {
const allocator = std.heap.page_allocator;
return .{ .cases = std.ArrayList(Case).init(allocator) };
}
fn deinit(self: *TestContext) void {
for (self.cases.items) |case| {
for (case.updates.items) |u| {
if (u.case == .Error) {
case.updates.allocator.free(u.case.Error);
}
}
case.updates.deinit();
}
self.cases.deinit();
self.* = undefined;
}
fn run(self: *TestContext) !void {
var progress = std.Progress{};
const root_node = try progress.start("tests", self.cases.items.len);
defer root_node.end();
var zig_lib_directory = try introspect.findZigLibDir(std.testing.allocator);
defer zig_lib_directory.handle.close();
defer std.testing.allocator.free(zig_lib_directory.path.?);
const random_seed = blk: {
var random_seed: u64 = undefined;
try std.crypto.randomBytes(std.mem.asBytes(&random_seed));
break :blk random_seed;
};
var default_prng = std.rand.DefaultPrng.init(random_seed);
for (self.cases.items) |case| {
if (build_options.skip_non_native and case.target.getCpuArch() != std.Target.current.cpu.arch)
continue;
var prg_node = root_node.start(case.name, case.updates.items.len);
prg_node.activate();
defer prg_node.end();
// So that we can see which test case failed when the leak checker goes off,
// or there's an internal error
progress.initial_delay_ns = 0;
progress.refresh_rate_ns = 0;
try self.runOneCase(std.testing.allocator, &prg_node, case, zig_lib_directory, &default_prng.random);
}
}
fn runOneCase(
self: *TestContext,
allocator: *Allocator,
root_node: *std.Progress.Node,
case: Case,
zig_lib_directory: Compilation.Directory,
rand: *std.rand.Random,
) !void {
const target_info = try std.zig.system.NativeTargetInfo.detect(allocator, case.target);
const target = target_info.target;
var arena_allocator = std.heap.ArenaAllocator.init(allocator);
defer arena_allocator.deinit();
const arena = &arena_allocator.allocator;
var tmp = std.testing.tmpDir(.{});
defer tmp.cleanup();
var cache_dir = try tmp.dir.makeOpenPath("zig-cache", .{});
defer cache_dir.close();
const tmp_path = try std.fs.path.join(arena, &[_][]const u8{ ".", "zig-cache", "tmp", &tmp.sub_path });
const zig_cache_directory: Compilation.Directory = .{
.handle = cache_dir,
.path = try std.fs.path.join(arena, &[_][]const u8{ tmp_path, "zig-cache" }),
};
const tmp_src_path = switch (case.extension) {
.Zig => "test_case.zig",
.ZIR => "test_case.zir",
};
var root_pkg: Package = .{
.root_src_directory = .{ .path = tmp_path, .handle = tmp.dir },
.root_src_path = tmp_src_path,
};
const ofmt: ?std.builtin.ObjectFormat = if (case.cbe) .c else null;
const bin_name = try std.zig.binNameAlloc(arena, .{
.root_name = "test_case",
.target = target,
.output_mode = case.output_mode,
.object_format = ofmt,
});
const emit_directory: Compilation.Directory = .{
.path = tmp_path,
.handle = tmp.dir,
};
const emit_bin: Compilation.EmitLoc = .{
.directory = emit_directory,
.basename = bin_name,
};
const comp = try Compilation.create(allocator, .{
.local_cache_directory = zig_cache_directory,
.global_cache_directory = zig_cache_directory,
.zig_lib_directory = zig_lib_directory,
.rand = rand,
.root_name = "test_case",
.target = target,
// TODO: support tests for object file building, and library builds
// and linking. This will require a rework to support multi-file
// tests.
.output_mode = case.output_mode,
// TODO: support testing optimizations
.optimize_mode = .Debug,
.emit_bin = emit_bin,
.root_pkg = &root_pkg,
.keep_source_files_loaded = true,
.object_format = ofmt,
.is_native_os = case.target.isNativeOs(),
});
defer comp.destroy();
for (case.files.items) |file| {
try tmp.dir.writeFile(file.path, file.src);
}
for (case.updates.items) |update, update_index| {
var update_node = root_node.start("update", 3);
update_node.activate();
defer update_node.end();
var sync_node = update_node.start("write", null);
sync_node.activate();
try tmp.dir.writeFile(tmp_src_path, update.src);
sync_node.end();
var module_node = update_node.start("parse/analysis/codegen", null);
module_node.activate();
try comp.makeBinFileWritable();
try comp.update();
module_node.end();
if (update.case != .Error) {
var all_errors = try comp.getAllErrorsAlloc();
defer all_errors.deinit(allocator);
if (all_errors.list.len != 0) {
std.debug.print("\nErrors occurred updating the compilation:\n================\n", .{});
for (all_errors.list) |err_msg| {
switch (err_msg) {
.src => |src| {
std.debug.print(":{d}:{d}: error: {s}\n================\n", .{
src.line + 1, src.column + 1, src.msg,
});
},
.plain => |plain| {
std.debug.print("error: {s}\n================\n", .{plain.msg});
},
}
}
if (case.cbe) {
const C = comp.bin_file.cast(link.File.C).?;
std.debug.print("Generated C: \n===============\n{}\n\n===========\n\n", .{C.main.items});
}
std.debug.print("Test failed.\n", .{});
std.process.exit(1);
}
}
switch (update.case) {
.Transformation => |expected_output| {
if (case.cbe) {
// The C file is always closed after an update, because we don't support
// incremental updates
var file = try tmp.dir.openFile(bin_name, .{ .read = true });
defer file.close();
var out = file.reader().readAllAlloc(arena, 1024 * 1024) catch @panic("Unable to read C output!");
if (expected_output.len != out.len) {
std.debug.print("\nTransformed C length differs:\n================\nExpected:\n================\n{}\n================\nFound:\n================\n{}\n================\nTest failed.\n", .{ expected_output, out });
std.process.exit(1);
}
for (expected_output) |e, i| {
if (out[i] != e) {
std.debug.print("\nTransformed C differs:\n================\nExpected:\n================\n{}\n================\nFound:\n================\n{}\n================\nTest failed.\n", .{ expected_output, out });
std.process.exit(1);
}
}
} else {
update_node.estimated_total_items = 5;
var emit_node = update_node.start("emit", null);
emit_node.activate();
var new_zir_module = try zir.emit(allocator, comp.bin_file.options.module.?);
defer new_zir_module.deinit(allocator);
emit_node.end();
var write_node = update_node.start("write", null);
write_node.activate();
var out_zir = std.ArrayList(u8).init(allocator);
defer out_zir.deinit();
try new_zir_module.writeToStream(allocator, out_zir.outStream());
write_node.end();
var test_node = update_node.start("assert", null);
test_node.activate();
defer test_node.end();
if (expected_output.len != out_zir.items.len) {
std.debug.print("{}\nTransformed ZIR length differs:\n================\nExpected:\n================\n{}\n================\nFound:\n================\n{}\n================\nTest failed.\n", .{ case.name, expected_output, out_zir.items });
std.process.exit(1);
}
for (expected_output) |e, i| {
if (out_zir.items[i] != e) {
std.debug.print("{}\nTransformed ZIR differs:\n================\nExpected:\n================\n{}\n================\nFound:\n================\n{}\n================\nTest failed.\n", .{ case.name, expected_output, out_zir.items });
std.process.exit(1);
}
}
}
},
.Error => |e| {
var test_node = update_node.start("assert", null);
test_node.activate();
defer test_node.end();
var handled_errors = try arena.alloc(bool, e.len);
for (handled_errors) |*h| {
h.* = false;
}
var all_errors = try comp.getAllErrorsAlloc();
defer all_errors.deinit(allocator);
for (all_errors.list) |a| {
for (e) |ex, i| {
const a_tag: @TagType(@TypeOf(a)) = a;
const ex_tag: @TagType(@TypeOf(ex)) = ex;
switch (a) {
.src => |src| {
if (ex_tag != .src) continue;
if (src.line == ex.src.line and
src.column == ex.src.column and
std.mem.eql(u8, ex.src.msg, src.msg))
{
handled_errors[i] = true;
break;
}
},
.plain => |plain| {
if (ex_tag != .plain) continue;
if (std.mem.eql(u8, ex.plain.msg, plain.msg)) {
handled_errors[i] = true;
break;
}
},
}
} else {
std.debug.print(
"{s}\nUnexpected error:\n================\n{}\n================\nTest failed.\n",
.{ case.name, ErrorMsg.init(a) },
);
std.process.exit(1);
}
}
for (handled_errors) |h, i| {
if (!h) {
const er = e[i];
std.debug.print(
"{s}\nDid not receive error:\n================\n{}\n================\nTest failed.\n",
.{ case.name, er },
);
std.process.exit(1);
}
}
},
.Execution => |expected_stdout| {
std.debug.assert(!case.cbe);
update_node.estimated_total_items = 4;
var exec_result = x: {
var exec_node = update_node.start("execute", null);
exec_node.activate();
defer exec_node.end();
var argv = std.ArrayList([]const u8).init(allocator);
defer argv.deinit();
const exe_path = try std.fmt.allocPrint(arena, "." ++ std.fs.path.sep_str ++ "{}", .{bin_name});
switch (case.target.getExternalExecutor()) {
.native => try argv.append(exe_path),
.unavailable => {
try self.runInterpreterIfAvailable(allocator, &exec_node, case, tmp.dir, bin_name);
return; // Pass test.
},
.qemu => |qemu_bin_name| if (enable_qemu) {
// TODO Ability for test cases to specify whether to link libc.
const need_cross_glibc = false; // target.isGnuLibC() and self.is_linking_libc;
const glibc_dir_arg = if (need_cross_glibc)
glibc_multi_install_dir orelse return // glibc dir not available; pass test
else
null;
try argv.append(qemu_bin_name);
if (glibc_dir_arg) |dir| {
const linux_triple = try target.linuxTriple(arena);
const full_dir = try std.fs.path.join(arena, &[_][]const u8{
dir,
linux_triple,
});
try argv.append("-L");
try argv.append(full_dir);
}
try argv.append(exe_path);
} else {
return; // QEMU not available; pass test.
},
.wine => |wine_bin_name| if (enable_wine) {
try argv.append(wine_bin_name);
try argv.append(exe_path);
} else {
return; // Wine not available; pass test.
},
.wasmtime => |wasmtime_bin_name| if (enable_wasmtime) {
try argv.append(wasmtime_bin_name);
try argv.append("--dir=.");
try argv.append(exe_path);
} else {
return; // wasmtime not available; pass test.
},
}
try comp.makeBinFileExecutable();
break :x try std.ChildProcess.exec(.{
.allocator = allocator,
.argv = argv.items,
.cwd_dir = tmp.dir,
});
};
var test_node = update_node.start("test", null);
test_node.activate();
defer test_node.end();
defer allocator.free(exec_result.stdout);
defer allocator.free(exec_result.stderr);
switch (exec_result.term) {
.Exited => |code| {
if (code != 0) {
std.debug.print("elf file exited with code {}\n", .{code});
return error.BinaryBadExitCode;
}
},
else => return error.BinaryCrashed,
}
if (!std.mem.eql(u8, expected_stdout, exec_result.stdout)) {
std.debug.panic(
"update index {}, mismatched stdout\n====Expected (len={}):====\n{}\n====Actual (len={}):====\n{}\n========\n",
.{ update_index, expected_stdout.len, expected_stdout, exec_result.stdout.len, exec_result.stdout },
);
}
},
}
}
}
fn runInterpreterIfAvailable(
self: *TestContext,
gpa: *Allocator,
node: *std.Progress.Node,
case: Case,
tmp_dir: std.fs.Dir,
bin_name: []const u8,
) !void {
const arch = case.target.cpu_arch orelse return;
switch (arch) {
.spu_2 => return self.runSpu2Interpreter(gpa, node, case, tmp_dir, bin_name),
else => return,
}
}
fn runSpu2Interpreter(
self: *TestContext,
gpa: *Allocator,
update_node: *std.Progress.Node,
case: Case,
tmp_dir: std.fs.Dir,
bin_name: []const u8,
) !void {
const spu = @import("codegen/spu-mk2.zig");
if (case.target.os_tag) |os| {
if (os != .freestanding) {
std.debug.panic("Only freestanding makes sense for SPU-II tests!", .{});
}
} else {
std.debug.panic("SPU_2 has no native OS, check the test!", .{});
}
var interpreter = spu.Interpreter(struct {
RAM: [0x10000]u8 = undefined,
pub fn read8(bus: @This(), addr: u16) u8 {
return bus.RAM[addr];
}
pub fn read16(bus: @This(), addr: u16) u16 {
return std.mem.readIntLittle(u16, bus.RAM[addr..][0..2]);
}
pub fn write8(bus: *@This(), addr: u16, val: u8) void {
bus.RAM[addr] = val;
}
pub fn write16(bus: *@This(), addr: u16, val: u16) void {
std.mem.writeIntLittle(u16, bus.RAM[addr..][0..2], val);
}
}){
.bus = .{},
};
{
var load_node = update_node.start("load", null);
load_node.activate();
defer load_node.end();
var file = try tmp_dir.openFile(bin_name, .{ .read = true });
defer file.close();
const header = try std.elf.readHeader(file);
var iterator = header.program_header_iterator(file);
var none_loaded = true;
while (try iterator.next()) |phdr| {
if (phdr.p_type != std.elf.PT_LOAD) {
std.debug.print("Encountered unexpected ELF program header: type {}\n", .{phdr.p_type});
std.process.exit(1);
}
if (phdr.p_paddr != phdr.p_vaddr) {
std.debug.print("Physical address does not match virtual address in ELF header!\n", .{});
std.process.exit(1);
}
if (phdr.p_filesz != phdr.p_memsz) {
std.debug.print("Physical size does not match virtual size in ELF header!\n", .{});
std.process.exit(1);
}
if ((try file.pread(interpreter.bus.RAM[phdr.p_paddr .. phdr.p_paddr + phdr.p_filesz], phdr.p_offset)) != phdr.p_filesz) {
std.debug.print("Read less than expected from ELF file!", .{});
std.process.exit(1);
}
std.log.scoped(.spu2_test).debug("Loaded 0x{x} bytes to 0x{x:0<4}\n", .{ phdr.p_filesz, phdr.p_paddr });
none_loaded = false;
}
if (none_loaded) {
std.debug.print("No data found in ELF file!\n", .{});
std.process.exit(1);
}
}
var exec_node = update_node.start("execute", null);
exec_node.activate();
defer exec_node.end();
var blocks: u16 = 1000;
const block_size = 1000;
while (!interpreter.undefined0) {
const pre_ip = interpreter.ip;
if (blocks > 0) {
blocks -= 1;
try interpreter.ExecuteBlock(block_size);
if (pre_ip == interpreter.ip) {
std.debug.print("Infinite loop detected in SPU II test!\n", .{});
std.process.exit(1);
}
}
}
}
};