1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
|
const std = @import("../std.zig");
const Allocator = std.mem.Allocator;
const Hash = std.hash.Wyhash;
const Dwarf = std.debug.Dwarf;
const assert = std.debug.assert;
const Coverage = @This();
/// Provides a globally-scoped integer index for directories.
///
/// As opposed to, for example, a directory index that is compilation-unit
/// scoped inside a single ELF module.
///
/// String memory references the memory-mapped debug information.
///
/// Protected by `mutex`.
directories: std.ArrayHashMapUnmanaged(String, void, String.MapContext, false),
/// Provides a globally-scoped integer index for files.
///
/// String memory references the memory-mapped debug information.
///
/// Protected by `mutex`.
files: std.ArrayHashMapUnmanaged(File, void, File.MapContext, false),
string_bytes: std.ArrayList(u8),
/// Protects the other fields.
mutex: std.Thread.Mutex,
pub const init: Coverage = .{
.directories = .{},
.files = .{},
.mutex = .{},
.string_bytes = .{},
};
pub const String = enum(u32) {
_,
pub const MapContext = struct {
string_bytes: []const u8,
pub fn eql(self: @This(), a: String, b: String, b_index: usize) bool {
_ = b_index;
const a_slice = span(self.string_bytes[@intFromEnum(a)..]);
const b_slice = span(self.string_bytes[@intFromEnum(b)..]);
return std.mem.eql(u8, a_slice, b_slice);
}
pub fn hash(self: @This(), a: String) u32 {
return @truncate(Hash.hash(0, span(self.string_bytes[@intFromEnum(a)..])));
}
};
pub const SliceAdapter = struct {
string_bytes: []const u8,
pub fn eql(self: @This(), a_slice: []const u8, b: String, b_index: usize) bool {
_ = b_index;
const b_slice = span(self.string_bytes[@intFromEnum(b)..]);
return std.mem.eql(u8, a_slice, b_slice);
}
pub fn hash(self: @This(), a: []const u8) u32 {
_ = self;
return @truncate(Hash.hash(0, a));
}
};
};
pub const SourceLocation = extern struct {
file: File.Index,
line: u32,
column: u32,
pub const invalid: SourceLocation = .{
.file = .invalid,
.line = 0,
.column = 0,
};
};
pub const File = extern struct {
directory_index: u32,
basename: String,
pub const Index = enum(u32) {
invalid = std.math.maxInt(u32),
_,
};
pub const MapContext = struct {
string_bytes: []const u8,
pub fn hash(self: MapContext, a: File) u32 {
const a_basename = span(self.string_bytes[@intFromEnum(a.basename)..]);
return @truncate(Hash.hash(a.directory_index, a_basename));
}
pub fn eql(self: MapContext, a: File, b: File, b_index: usize) bool {
_ = b_index;
if (a.directory_index != b.directory_index) return false;
const a_basename = span(self.string_bytes[@intFromEnum(a.basename)..]);
const b_basename = span(self.string_bytes[@intFromEnum(b.basename)..]);
return std.mem.eql(u8, a_basename, b_basename);
}
};
pub const SliceAdapter = struct {
string_bytes: []const u8,
pub const Entry = struct {
directory_index: u32,
basename: []const u8,
};
pub fn hash(self: @This(), a: Entry) u32 {
_ = self;
return @truncate(Hash.hash(a.directory_index, a.basename));
}
pub fn eql(self: @This(), a: Entry, b: File, b_index: usize) bool {
_ = b_index;
if (a.directory_index != b.directory_index) return false;
const b_basename = span(self.string_bytes[@intFromEnum(b.basename)..]);
return std.mem.eql(u8, a.basename, b_basename);
}
};
};
pub fn deinit(cov: *Coverage, gpa: Allocator) void {
cov.directories.deinit(gpa);
cov.files.deinit(gpa);
cov.string_bytes.deinit(gpa);
cov.* = undefined;
}
pub fn fileAt(cov: *Coverage, index: File.Index) *File {
return &cov.files.keys()[@intFromEnum(index)];
}
pub fn stringAt(cov: *Coverage, index: String) [:0]const u8 {
return span(cov.string_bytes.items[@intFromEnum(index)..]);
}
pub const ResolveAddressesDwarfError = Dwarf.ScanError;
pub fn resolveAddressesDwarf(
cov: *Coverage,
gpa: Allocator,
endian: std.builtin.Endian,
/// Asserts the addresses are in ascending order.
sorted_pc_addrs: []const u64,
/// Asserts its length equals length of `sorted_pc_addrs`.
output: []SourceLocation,
d: *Dwarf,
) ResolveAddressesDwarfError!void {
assert(sorted_pc_addrs.len == output.len);
assert(d.ranges.items.len != 0); // call `populateRanges` first.
var range_i: usize = 0;
var range: *std.debug.Dwarf.Range = &d.ranges.items[0];
var line_table_i: usize = undefined;
var prev_pc: u64 = 0;
var prev_cu: ?*std.debug.Dwarf.CompileUnit = null;
// Protects directories and files tables from other threads.
cov.mutex.lock();
defer cov.mutex.unlock();
next_pc: for (sorted_pc_addrs, output) |pc, *out| {
assert(pc >= prev_pc);
prev_pc = pc;
while (pc >= range.end) {
range_i += 1;
if (range_i >= d.ranges.items.len) {
out.* = SourceLocation.invalid;
continue :next_pc;
}
range = &d.ranges.items[range_i];
}
if (pc < range.start) {
out.* = SourceLocation.invalid;
continue :next_pc;
}
const cu = &d.compile_unit_list.items[range.compile_unit_index];
if (cu != prev_cu) {
prev_cu = cu;
if (cu.src_loc_cache == null) {
cov.mutex.unlock();
defer cov.mutex.lock();
d.populateSrcLocCache(gpa, endian, cu) catch |err| switch (err) {
error.MissingDebugInfo, error.InvalidDebugInfo => {
out.* = SourceLocation.invalid;
continue :next_pc;
},
else => |e| return e,
};
}
const slc = &cu.src_loc_cache.?;
const table_addrs = slc.line_table.keys();
line_table_i = std.sort.upperBound(u64, table_addrs, pc, struct {
fn order(context: u64, item: u64) std.math.Order {
return std.math.order(context, item);
}
}.order);
}
const slc = &cu.src_loc_cache.?;
const table_addrs = slc.line_table.keys();
while (line_table_i < table_addrs.len and table_addrs[line_table_i] <= pc) line_table_i += 1;
const entry = slc.line_table.values()[line_table_i - 1];
const corrected_file_index = entry.file - @intFromBool(slc.version < 5);
const file_entry = slc.files[corrected_file_index];
const dir_path = slc.directories[file_entry.dir_index].path;
try cov.string_bytes.ensureUnusedCapacity(gpa, dir_path.len + file_entry.path.len + 2);
const dir_gop = try cov.directories.getOrPutContextAdapted(gpa, dir_path, String.SliceAdapter{
.string_bytes = cov.string_bytes.items,
}, String.MapContext{
.string_bytes = cov.string_bytes.items,
});
if (!dir_gop.found_existing)
dir_gop.key_ptr.* = addStringAssumeCapacity(cov, dir_path);
const file_gop = try cov.files.getOrPutContextAdapted(gpa, File.SliceAdapter.Entry{
.directory_index = @intCast(dir_gop.index),
.basename = file_entry.path,
}, File.SliceAdapter{
.string_bytes = cov.string_bytes.items,
}, File.MapContext{
.string_bytes = cov.string_bytes.items,
});
if (!file_gop.found_existing) file_gop.key_ptr.* = .{
.directory_index = @intCast(dir_gop.index),
.basename = addStringAssumeCapacity(cov, file_entry.path),
};
out.* = .{
.file = @enumFromInt(file_gop.index),
.line = entry.line,
.column = entry.column,
};
}
}
pub fn addStringAssumeCapacity(cov: *Coverage, s: []const u8) String {
const result: String = @enumFromInt(cov.string_bytes.items.len);
cov.string_bytes.appendSliceAssumeCapacity(s);
cov.string_bytes.appendAssumeCapacity(0);
return result;
}
fn span(s: []const u8) [:0]const u8 {
return std.mem.sliceTo(@as([:0]const u8, @ptrCast(s)), 0);
}
|