297ca9d1d7
Since LinearHistogram and its descendants inherit ranges_ from Histogram, and we wanted to replace the copying into a std::vec for Histogram, the simplest approach seemed to just be to precompute ranges for all histograms, exponential or otherwise. This should have the added benefit of reducing the memory footprint for those histograms, since they will benefit from the deduplication work that the precomputing script already does. MozReview-Commit-ID: JTV5Dej5ZIb
209 lines
7.3 KiB
Python
209 lines
7.3 KiB
Python
# This Source Code Form is subject to the terms of the Mozilla Public
|
|
# License, v. 2.0. If a copy of the MPL was not distributed with this
|
|
# file, You can obtain one at http://mozilla.org/MPL/2.0/.
|
|
|
|
# Write out histogram information for C++. The histograms are defined
|
|
# in a file provided as a command-line argument.
|
|
|
|
from __future__ import print_function
|
|
from shared_telemetry_utils import StringTable, static_assert, ParserError
|
|
|
|
import sys
|
|
import histogram_tools
|
|
|
|
banner = """/* This file is auto-generated, see gen-histogram-data.py. */
|
|
"""
|
|
|
|
|
|
def print_array_entry(output, histogram, name_index, exp_index, label_index,
|
|
label_count, key_index, key_count):
|
|
cpp_guard = histogram.cpp_guard()
|
|
if cpp_guard:
|
|
print("#if defined(%s)" % cpp_guard, file=output)
|
|
print(" { %s, %s, %s, %s, %d, %d, %s, %d, %d, %d, %d, %s, %s },"
|
|
% (histogram.low(),
|
|
histogram.high(),
|
|
histogram.n_buckets(),
|
|
histogram.nsITelemetry_kind(),
|
|
name_index,
|
|
exp_index,
|
|
histogram.dataset(),
|
|
label_index,
|
|
label_count,
|
|
key_index,
|
|
key_count,
|
|
" | ".join(histogram.record_in_processes_enum()),
|
|
"true" if histogram.keyed() else "false"), file=output)
|
|
if cpp_guard:
|
|
print("#endif", file=output)
|
|
|
|
|
|
def write_histogram_table(output, histograms):
|
|
string_table = StringTable()
|
|
label_table = []
|
|
label_count = 0
|
|
keys_table = []
|
|
keys_count = 0
|
|
|
|
print("constexpr HistogramInfo gHistogramInfos[] = {", file=output)
|
|
for histogram in histograms:
|
|
name_index = string_table.stringIndex(histogram.name())
|
|
exp_index = string_table.stringIndex(histogram.expiration())
|
|
|
|
labels = histogram.labels()
|
|
label_index = 0
|
|
if len(labels) > 0:
|
|
label_index = label_count
|
|
label_table.append((histogram.name(), string_table.stringIndexes(labels)))
|
|
label_count += len(labels)
|
|
|
|
keys = histogram.keys()
|
|
key_index = 0
|
|
if len(keys) > 0:
|
|
key_index = keys_count
|
|
keys_table.append((histogram.name(), string_table.stringIndexes(keys)))
|
|
keys_count += len(keys)
|
|
|
|
print_array_entry(output, histogram, name_index, exp_index,
|
|
label_index, len(labels), key_index, len(keys))
|
|
print("};\n", file=output)
|
|
|
|
strtab_name = "gHistogramStringTable"
|
|
string_table.writeDefinition(output, strtab_name)
|
|
static_assert(output, "sizeof(%s) <= UINT32_MAX" % strtab_name,
|
|
"index overflow")
|
|
|
|
print("\nconst uint32_t gHistogramLabelTable[] = {", file=output)
|
|
for name, indexes in label_table:
|
|
print("/* %s */ %s," % (name, ", ".join(map(str, indexes))), file=output)
|
|
print("};", file=output)
|
|
|
|
print("\nconst uint32_t gHistogramKeyTable[] = {", file=output)
|
|
for name, indexes in keys_table:
|
|
print("/* %s */ %s," % (name, ", ".join(map(str, indexes))), file=output)
|
|
print("};", file=output)
|
|
|
|
|
|
# Write out static asserts for histogram data. We'd prefer to perform
|
|
# these checks in this script itself, but since several histograms
|
|
# (generally enumerated histograms) use compile-time constants for
|
|
# their upper bounds, we have to let the compiler do the checking.
|
|
|
|
def static_asserts_for_boolean(output, histogram):
|
|
pass
|
|
|
|
|
|
def static_asserts_for_flag(output, histogram):
|
|
pass
|
|
|
|
|
|
def static_asserts_for_count(output, histogram):
|
|
pass
|
|
|
|
|
|
def static_asserts_for_enumerated(output, histogram):
|
|
n_values = histogram.high()
|
|
static_assert(output, "%s > 2" % n_values,
|
|
"Not enough values for %s" % histogram.name())
|
|
|
|
|
|
def shared_static_asserts(output, histogram):
|
|
name = histogram.name()
|
|
low = histogram.low()
|
|
high = histogram.high()
|
|
n_buckets = histogram.n_buckets()
|
|
static_assert(output, "%s < %s" % (low, high), "low >= high for %s" % name)
|
|
static_assert(output, "%s > 2" % n_buckets, "Not enough values for %s" % name)
|
|
static_assert(output, "%s >= 1" % low, "Incorrect low value for %s" % name)
|
|
static_assert(output, "%s > %s" % (high, n_buckets), "high must be > number of buckets for %s;"
|
|
" you may want an enumerated histogram" % name)
|
|
|
|
|
|
def static_asserts_for_linear(output, histogram):
|
|
shared_static_asserts(output, histogram)
|
|
|
|
|
|
def static_asserts_for_exponential(output, histogram):
|
|
shared_static_asserts(output, histogram)
|
|
|
|
|
|
def write_histogram_static_asserts(output, histograms):
|
|
print("""
|
|
// Perform the checks at the beginning of HistogramGet at
|
|
// compile time, so that incorrect histogram definitions
|
|
// give compile-time errors, not runtime errors.""", file=output)
|
|
|
|
table = {
|
|
'boolean': static_asserts_for_boolean,
|
|
'flag': static_asserts_for_flag,
|
|
'count': static_asserts_for_count,
|
|
'enumerated': static_asserts_for_enumerated,
|
|
'categorical': static_asserts_for_enumerated,
|
|
'linear': static_asserts_for_linear,
|
|
'exponential': static_asserts_for_exponential,
|
|
}
|
|
|
|
for histogram in histograms:
|
|
kind = histogram.kind()
|
|
if kind not in table:
|
|
raise Exception('Unknown kind "%s" for histogram "%s".' % (kind, histogram.name()))
|
|
fn = table[kind]
|
|
fn(output, histogram)
|
|
|
|
|
|
def write_histogram_ranges(output, histograms):
|
|
# This generates static data to avoid costly initialization of histograms
|
|
# (especially exponential ones which require log and exp calls) at runtime.
|
|
# The format must exactly match that required in histogram.cc, which is
|
|
# 0, buckets..., INT_MAX. Additionally, the list ends in a 0 to aid asserts
|
|
# that validate that the length of the ranges list is correct.U cache miss.
|
|
print("const int gHistogramBucketLowerBounds[] = {", file=output)
|
|
|
|
# Print the dummy buckets for expired histograms, and set the offset to match.
|
|
print("0,1,2,INT_MAX,", file=output)
|
|
offset = 4
|
|
ranges_offsets = {}
|
|
|
|
for histogram in histograms:
|
|
ranges = tuple(histogram.ranges())
|
|
if ranges not in ranges_offsets:
|
|
ranges_offsets[ranges] = offset
|
|
# Suffix each ranges listing with INT_MAX, to match histogram.cc's
|
|
# expected format.
|
|
offset += len(ranges) + 1
|
|
print(','.join(map(str, ranges)), ',INT_MAX,', file=output)
|
|
print("0};", file=output)
|
|
|
|
if offset > 32767:
|
|
raise Exception('Histogram offsets exceeded maximum value for an int16_t.')
|
|
|
|
print("const int16_t gHistogramBucketLowerBoundIndex[] = {", file=output)
|
|
for histogram in histograms:
|
|
cpp_guard = histogram.cpp_guard()
|
|
if cpp_guard:
|
|
print("#if defined(%s)" % cpp_guard, file=output)
|
|
|
|
our_offset = ranges_offsets[tuple(histogram.ranges())]
|
|
print("%d," % our_offset, file=output)
|
|
|
|
if cpp_guard:
|
|
print("#endif", file=output)
|
|
print("};", file=output)
|
|
|
|
|
|
def main(output, *filenames):
|
|
try:
|
|
histograms = list(histogram_tools.from_files(filenames))
|
|
except ParserError as ex:
|
|
print("\nError processing histograms:\n" + str(ex) + "\n")
|
|
sys.exit(1)
|
|
|
|
print(banner, file=output)
|
|
write_histogram_table(output, histograms)
|
|
write_histogram_ranges(output, histograms)
|
|
write_histogram_static_asserts(output, histograms)
|
|
|
|
|
|
if __name__ == '__main__':
|
|
main(sys.stdout, *sys.argv[1:])
|