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
|
import os
import glob
import subprocess
import csv
import io
import enum
import platform
import puremagic
import pathlib
from typing import Optional
from SCons.Node.FS import File, Dir
from typing_extensions import TypedDict
from .util import get_build_metric_dict
from .protocol import BuildMetricsCollector
class ArtifactType(str, enum.Enum):
UNKNOWN = "unknown"
PROGRAM = "Program" # .exe
LIBRARY = "Library" # .so, .a
ARCHIVE = "archive" # .zip, .tgz, not .a
OBJECT = "Object" # .o
TEXT = "text" # .h, .hpp, .cpp
# Types to run bloaty against
ARTIFACT_BIN_TYPES = [ArtifactType.PROGRAM, ArtifactType.LIBRARY, ArtifactType.OBJECT]
class BinSize(TypedDict):
vmsize: int
filesize: int
class BinMetrics(TypedDict, total=False):
text: BinSize
data: BinSize
rodata: BinSize
bss: BinSize
debug: BinSize
symtab: BinSize
dyntab: BinSize
def _run_bloaty(bloaty, target) -> Optional[BinMetrics]:
out = BinMetrics()
try:
# -n 0 -> do not collapse small sections into a section named [Other]
# --csv -> generate csv output to stdout
# -d sections -> only list sections, not symbols
proc = subprocess.run([bloaty, "-n", "0", "--csv", "-d", "sections",
str(target)], capture_output=True, universal_newlines=True)
if proc.returncode != 0:
# if we run bloaty against a thin archive, it will fail. Detect
# this and allow thin archives to pass, otherwise raise an
# exception.
# Note that our thin_archive tool sets the thin_archive
# attribute to True
if proc.stderr.startswith("bloaty: unknown file type for file") and getattr(
target.attributes, "thin_archive", False):
# this is a thin archive, pass it
return None
raise RuntimeError(f"Failed to call bloaty on '{str(target)}': {proc.stderr}")
for row in csv.DictReader(proc.stdout.splitlines()):
# sections,vmsize,filesize
section = row['sections']
vmsize = int(row['vmsize'])
filesize = int(row['filesize'])
binsize = BinSize(vmsize=vmsize, filesize=filesize)
if section == ".text":
out["text"] = binsize
elif section == ".data":
out["data"] = binsize
elif section == ".rodata":
out["rodata"] = binsize
elif section == ".bss":
out["bss"] = binsize
elif section.startswith(".debug"):
# there are multiple sections that start with .debug, and we
# need to sum them up.
if "debug" not in out:
out["debug"] = BinSize(vmsize=0, filesize=0)
out["debug"]["vmsize"] += vmsize
out["debug"]["filesize"] += filesize
elif section == ".symtab":
out["symtab"] = binsize
elif section == ".dyntab":
out["dyntab"] = binsize
return out
except FileNotFoundError:
if not _run_bloaty.printed_missing_bloaty_warning:
print(
"WARNING: could not find the bloaty binary. Binary section metrics will not be collected."
)
_run_bloaty.printed_missing_bloaty_warning = True
return None
_run_bloaty.printed_missing_bloaty_warning = False
class Artifact(TypedDict, total=False):
array_index: int
name: str
type: str
size: int
bin_metrics: BinMetrics
# First key: platform.system()
# Tuple key 1: ArtifactType
# Tuple Key 2: string to search for
_PLATFORM_LIBMAGIC_BINARY_IDENTITIES = {
"Windows": [(ArtifactType.LIBRARY, "executable (DLL)"), (ArtifactType.PROGRAM, "executable")],
"Linux": [(ArtifactType.PROGRAM, "interpreter"), (ArtifactType.LIBRARY, "shared object")],
"Darwin": [(ArtifactType.PROGRAM, "Mach-O universal binary"),
(ArtifactType.LIBRARY, "linked shared library")],
}
_ARTIFACT_TYPE_FROM_BUILDER = {
"SharedObject": ArtifactType.OBJECT, # .dyn.o
"StaticObject": ArtifactType.OBJECT, # .o
"StaticLibrary": ArtifactType.LIBRARY, # .a
"Idlc": ArtifactType.TEXT, # _gen.{h,cpp}
"Program": ArtifactType.PROGRAM, # .exe/*nix binaries
"Substfile": ArtifactType.TEXT, # build/opt/mongo/config.h and others
"InstallBuilder": ArtifactType.TEXT, # build/opt/third_party/wiredtiger/wiredtiger_ext.h
"Textfile": ArtifactType.TEXT, # build/opt/third_party/third_party_shim.cpp
}
_TEXT_IDENTIFIERS = ["ASCII text", "Unicode text"]
_EXTENSION_FALLBACK = {
".cpp": ArtifactType.TEXT,
".h": ArtifactType.TEXT,
".hpp": ArtifactType.TEXT,
".js": ArtifactType.TEXT,
".idl": ArtifactType.TEXT,
".so": ArtifactType.LIBRARY,
".o": ArtifactType.OBJECT,
# Windows
".obj": ArtifactType.OBJECT,
".lib": ArtifactType.LIBRARY,
# ilk, exp, pdb and res files on Windows have no appropriate tag, so we
# allow them to fallthrough to UNKNOWN
}
class CollectArtifacts(BuildMetricsCollector):
def __init__(self, env):
self._env = env
self._env = env
self._build_dir = env.get("BUILD_METRICS_ARTIFACTS_DIR", env.Dir('#').abspath)
self._artifacts = []
self._bloaty_bin = env.get("BUILD_METRICS_BLOATY", env.WhereIs('bloaty'))
self._metrics = {"total_artifact_size": 0, "num_artifacts": 0, "artifacts": []}
def get_name(self):
return "CollectArtifacts"
def walk(self, dirname):
for root, dirs, files in os.walk(dirname):
self._artifacts += list(map(lambda x: os.path.join(root, x), files))
def finalize(self):
self.walk(self._env.Dir(self._env.subst(self._build_dir)).abspath)
for artifact in self._artifacts:
artifact_dict = self._identify_artifact(artifact)
artifact_dict["array_index"] = len(self._metrics["artifacts"])
self._metrics["artifacts"].append(artifact_dict)
self._metrics["total_artifact_size"] += artifact_dict["size"]
self._metrics["num_artifacts"] = len(self._metrics["artifacts"])
return "artifact_metrics", self._metrics
def _identify_artifact(self, file_) -> Artifact:
def _type_from_builder(builder) -> ArtifactType:
name = builder.get_name(self._env)
return _ARTIFACT_TYPE_FROM_BUILDER.get(name, ArtifactType.UNKNOWN)
type_ = ArtifactType.UNKNOWN
file_str = str(file_)
node = self._env.File(file_)
builder = node.get_builder()
if builder is not None:
type_ = _type_from_builder(builder)
if type_ == ArtifactType.UNKNOWN:
try:
magic_out = puremagic.from_file(file_str)
system = platform.system()
for search_type in _PLATFORM_LIBMAGIC_BINARY_IDENTITIES.get(system):
if search_type[1] in magic_out:
type_ = search_type[0]
break
if type_ == ArtifactType.UNKNOWN and any(s in magic_out for s in _TEXT_IDENTIFIERS):
type_ = ArtifactType.TEXT
except (puremagic.main.PureError, ValueError):
# exception means that puremagic failed to id the filetype. We'll
# fallback to file extension in this case.
pass
if type_ == ArtifactType.UNKNOWN:
type_ = _EXTENSION_FALLBACK.get(pathlib.Path(file_str).suffix, ArtifactType.UNKNOWN)
out = Artifact({"name": file_, "type": type_, "size": node.get_size()})
if type_ in ARTIFACT_BIN_TYPES:
bin_metrics = _run_bloaty(self._bloaty_bin, node)
if bin_metrics is not None:
out["bin_metrics"] = bin_metrics
return out
|