Chromium Code Reviews
chromiumcodereview-hr@appspot.gserviceaccount.com (chromiumcodereview-hr) | Please choose your nickname with Settings | Help | Chromium Project | Gerrit Changes | Sign out
(157)

Issues Search
    My Issues | Starred     Open | Closed | All

Unified Diff: base/process/process_metrics_linux.cc

Issue 2858213002: Clean up base/process/process_metrics_linux.cc. (Closed)
Patch Set: constexpr is too much Created 3 years, 7 months ago
Use n/p to move between diff chunks; N/P to move between comments. Draft comments are only viewable by you.
Jump to:
View side-by-side diff with in-line comments
Download patch
« no previous file with comments | « base/process/process_metrics.h ('k') | no next file » | no next file with comments »
Expand Comments ('e') | Collapse Comments ('c') | Show Comments Hide Comments ('s')
Index: base/process/process_metrics_linux.cc
diff --git a/base/process/process_metrics_linux.cc b/base/process/process_metrics_linux.cc
index eb7caf75fd9752f198c35c9b3c2c7a128ac6e33b..1dde0d957fdb7902b24f92a090cbffd1ea7e7744 100644
--- a/base/process/process_metrics_linux.cc
+++ b/base/process/process_metrics_linux.cc
@@ -32,60 +32,70 @@ namespace base {
namespace {
void TrimKeyValuePairs(StringPairs* pairs) {
- DCHECK(pairs);
- StringPairs& p_ref = *pairs;
- for (size_t i = 0; i < p_ref.size(); ++i) {
- TrimWhitespaceASCII(p_ref[i].first, TRIM_ALL, &p_ref[i].first);
- TrimWhitespaceASCII(p_ref[i].second, TRIM_ALL, &p_ref[i].second);
+ for (auto& pair : *pairs) {
+ TrimWhitespaceASCII(pair.first, TRIM_ALL, &pair.first);
+ TrimWhitespaceASCII(pair.second, TRIM_ALL, &pair.second);
}
}
#if defined(OS_CHROMEOS)
// Read a file with a single number string and return the number as a uint64_t.
-static uint64_t ReadFileToUint64(const FilePath file) {
- std::string file_as_string;
- if (!ReadFileToString(file, &file_as_string))
+uint64_t ReadFileToUint64(const FilePath& file) {
+ std::string file_contents;
+ if (!ReadFileToString(file, &file_contents))
return 0;
- TrimWhitespaceASCII(file_as_string, TRIM_ALL, &file_as_string);
- uint64_t file_as_uint64 = 0;
- if (!StringToUint64(file_as_string, &file_as_uint64))
+ TrimWhitespaceASCII(file_contents, TRIM_ALL, &file_contents);
+ uint64_t file_contents_uint64 = 0;
+ if (!StringToUint64(file_contents, &file_contents_uint64))
return 0;
- return file_as_uint64;
+ return file_contents_uint64;
}
#endif
-// Read /proc/<pid>/status and return the value for |field|, or 0 on failure.
-// Only works for fields in the form of "Field: value kB".
-size_t ReadProcStatusAndGetFieldAsSizeT(pid_t pid, const std::string& field) {
+// Read |filename| in /proc/<pid>/, split the entries into key/value pairs, and
+// trim the key and value. On success, return true and write the trimmed
+// key/value pairs into |key_value_pairs|.
+bool ReadProcFileToTrimmedStringPairs(pid_t pid,
+ StringPiece filename,
+ StringPairs* key_value_pairs) {
std::string status_data;
{
// Synchronously reading files in /proc does not hit the disk.
ThreadRestrictions::ScopedAllowIO allow_io;
- FilePath status_file = internal::GetProcPidDir(pid).Append("status");
+ FilePath status_file = internal::GetProcPidDir(pid).Append(filename);
if (!ReadFileToString(status_file, &status_data))
- return 0;
+ return false;
}
+ SplitStringIntoKeyValuePairs(status_data, ':', '\n', key_value_pairs);
+ TrimKeyValuePairs(key_value_pairs);
+ return true;
+}
+// Read /proc/<pid>/status and return the value for |field|, or 0 on failure.
+// Only works for fields in the form of "Field: value kB".
+size_t ReadProcStatusAndGetFieldAsSizeT(pid_t pid, StringPiece field) {
StringPairs pairs;
- SplitStringIntoKeyValuePairs(status_data, ':', '\n', &pairs);
- TrimKeyValuePairs(&pairs);
- for (size_t i = 0; i < pairs.size(); ++i) {
- const std::string& key = pairs[i].first;
- const std::string& value_str = pairs[i].second;
- if (key == field) {
- std::vector<StringPiece> split_value_str =
- SplitStringPiece(value_str, " ", TRIM_WHITESPACE, SPLIT_WANT_ALL);
- if (split_value_str.size() != 2 || split_value_str[1] != "kB") {
- NOTREACHED();
- return 0;
- }
- size_t value;
- if (!StringToSizeT(split_value_str[0], &value)) {
- NOTREACHED();
- return 0;
- }
- return value;
+ if (!ReadProcFileToTrimmedStringPairs(pid, "status", &pairs))
+ return 0;
+
+ for (const auto& pair : pairs) {
+ const std::string& key = pair.first;
+ const std::string& value_str = pair.second;
+ if (key != field)
+ continue;
+
+ std::vector<StringPiece> split_value_str =
+ SplitStringPiece(value_str, " ", TRIM_WHITESPACE, SPLIT_WANT_ALL);
+ if (split_value_str.size() != 2 || split_value_str[1] != "kB") {
+ NOTREACHED();
+ return 0;
}
+ size_t value;
+ if (!StringToSizeT(split_value_str[0], &value)) {
+ NOTREACHED();
+ return 0;
+ }
+ return value;
}
// This can be reached if the process dies when proc is read -- in that case,
// the kernel can return missing fields.
@@ -97,30 +107,23 @@ size_t ReadProcStatusAndGetFieldAsSizeT(pid_t pid, const std::string& field) {
// write the value for |field| into |result|.
// Only works for fields in the form of "field : uint_value"
bool ReadProcStatusAndGetFieldAsUint64(pid_t pid,
- const std::string& field,
+ StringPiece field,
uint64_t* result) {
- std::string status_data;
- {
- // Synchronously reading files in /proc does not hit the disk.
- ThreadRestrictions::ScopedAllowIO allow_io;
- FilePath status_file = internal::GetProcPidDir(pid).Append("status");
- if (!ReadFileToString(status_file, &status_data))
- return false;
- }
-
StringPairs pairs;
- SplitStringIntoKeyValuePairs(status_data, ':', '\n', &pairs);
- TrimKeyValuePairs(&pairs);
- for (size_t i = 0; i < pairs.size(); ++i) {
- const std::string& key = pairs[i].first;
- const std::string& value_str = pairs[i].second;
- if (key == field) {
- uint64_t value;
- if (!StringToUint64(value_str, &value))
- return false;
- *result = value;
- return true;
- }
+ if (!ReadProcFileToTrimmedStringPairs(pid, "status", &pairs))
+ return false;
+
+ for (const auto& pair : pairs) {
+ const std::string& key = pair.first;
+ const std::string& value_str = pair.second;
+ if (key != field)
+ continue;
+
+ uint64_t value;
+ if (!StringToUint64(value_str, &value))
+ return false;
+ *result = value;
+ return true;
}
return false;
}
@@ -169,24 +172,24 @@ std::unique_ptr<ProcessMetrics> ProcessMetrics::CreateProcessMetrics(
return WrapUnique(new ProcessMetrics(process));
}
-// On linux, we return vsize.
+// On Linux, return vsize.
size_t ProcessMetrics::GetPagefileUsage() const {
return internal::ReadProcStatsAndGetFieldAsSizeT(process_,
internal::VM_VSIZE);
}
-// On linux, we return the high water mark of vsize.
+// On Linux, return the high water mark of vsize.
size_t ProcessMetrics::GetPeakPagefileUsage() const {
return ReadProcStatusAndGetFieldAsSizeT(process_, "VmPeak") * 1024;
}
-// On linux, we return RSS.
+// On Linux, return RSS.
size_t ProcessMetrics::GetWorkingSetSize() const {
return internal::ReadProcStatsAndGetFieldAsSizeT(process_, internal::VM_RSS) *
getpagesize();
}
-// On linux, we return the high water mark of RSS.
+// On Linux, return the high water mark of RSS.
size_t ProcessMetrics::GetPeakWorkingSetSize() const {
return ReadProcStatusAndGetFieldAsSizeT(process_, "VmHWM") * 1024;
}
@@ -232,8 +235,8 @@ double ProcessMetrics::GetCPUUsage() {
int cpu = GetProcessCPU(process_);
- // We have the number of jiffies in the time period. Convert to percentage.
- // Note this means we will go *over* 100 in the case where multiple threads
+ // The number of jiffies in the time period. Convert to percentage.
+ // Note: this means this will go *over* 100 in the case where multiple threads
// are together adding to more than one CPU's worth.
TimeDelta cpu_time = internal::ClockTicksToTimeDelta(cpu);
TimeDelta last_cpu_time = internal::ClockTicksToTimeDelta(last_cpu_);
@@ -241,7 +244,7 @@ double ProcessMetrics::GetCPUUsage() {
// If the number of threads running in the process has decreased since the
// last time this function was called, |last_cpu_time| will be greater than
// |cpu_time| which will result in a negative value in the below percentage
- // calculation. We prevent this by clamping to 0. crbug.com/546565.
+ // calculation. Prevent this by clamping to 0. https://crbug.com/546565.
// This computation is known to be shaky when threads are destroyed between
// "last" and "now", but for our current purposes, it's all right.
double percentage = 0.0;
@@ -256,27 +259,20 @@ double ProcessMetrics::GetCPUUsage() {
return percentage;
}
-// To have /proc/self/io file you must enable CONFIG_TASK_IO_ACCOUNTING
-// in your kernel configuration.
+// For the /proc/self/io file to exist, the Linux kernel must have
+// CONFIG_TASK_IO_ACCOUNTING enabled.
bool ProcessMetrics::GetIOCounters(IoCounters* io_counters) const {
- // Synchronously reading files in /proc does not hit the disk.
- ThreadRestrictions::ScopedAllowIO allow_io;
-
- std::string proc_io_contents;
- FilePath io_file = internal::GetProcPidDir(process_).Append("io");
- if (!ReadFileToString(io_file, &proc_io_contents))
+ StringPairs pairs;
+ if (!ReadProcFileToTrimmedStringPairs(process_, "io", &pairs))
return false;
io_counters->OtherOperationCount = 0;
io_counters->OtherTransferCount = 0;
- StringPairs pairs;
- SplitStringIntoKeyValuePairs(proc_io_contents, ':', '\n', &pairs);
- TrimKeyValuePairs(&pairs);
- for (size_t i = 0; i < pairs.size(); ++i) {
- const std::string& key = pairs[i].first;
- const std::string& value_str = pairs[i].second;
- uint64_t* target_counter = NULL;
+ for (const auto& pair : pairs) {
+ const std::string& key = pair.first;
+ const std::string& value_str = pair.second;
+ uint64_t* target_counter = nullptr;
if (key == "syscr")
target_counter = &io_counters->ReadOperationCount;
else if (key == "syscw")
@@ -320,23 +316,22 @@ int ProcessMetrics::GetOpenFdSoftLimit() const {
if (!ReadFileToString(fd_path, &limits_contents))
return -1;
- for (const auto& line :
- base::SplitStringPiece(limits_contents, "\n", base::KEEP_WHITESPACE,
- base::SPLIT_WANT_NONEMPTY)) {
- if (line.starts_with("Max open files")) {
- auto tokens = base::SplitStringPiece(line, " ", base::TRIM_WHITESPACE,
- base::SPLIT_WANT_NONEMPTY);
- if (tokens.size() > 3) {
- int limit = -1;
- if (StringToInt(tokens[3], &limit))
- return limit;
+ for (const auto& line : SplitStringPiece(
+ limits_contents, "\n", KEEP_WHITESPACE, SPLIT_WANT_NONEMPTY)) {
+ if (!line.starts_with("Max open files"))
+ continue;
+
+ auto tokens =
+ SplitStringPiece(line, " ", TRIM_WHITESPACE, SPLIT_WANT_NONEMPTY);
+ if (tokens.size() > 3) {
+ int limit = -1;
+ if (!StringToInt(tokens[3], &limit))
return -1;
- }
+ return limit;
}
}
return -1;
}
-
#endif // defined(OS_LINUX) || defined(OS_AIX)
ProcessMetrics::ProcessMetrics(ProcessHandle process)
@@ -399,7 +394,7 @@ bool ProcessMetrics::GetWorkingSetKBytesTotmaps(WorkingSetKBytes *ws_usage)
ret &= StringToInt(totmaps_fields[kPrivate_DirtyIndex], &private_dirty);
ret &= StringToInt(totmaps_fields[kSwapIndex], &swap);
- // On ChromeOS swap is to zram. We count this as private / shared, as
+ // On ChromeOS, swap goes to zram. Count this as private / shared, as
// increased swap decreases available RAM to user processes, which would
// otherwise create surprising results.
ws_usage->priv = private_clean + private_dirty + swap;
@@ -417,7 +412,7 @@ bool ProcessMetrics::GetWorkingSetKBytesStatm(WorkingSetKBytes* ws_usage)
// a) Large and slow to parse.
// b) Unavailable in the SUID sandbox.
- // First we need to get the page size, since everything is measured in pages.
+ // First get the page size, since everything is measured in pages.
// For details, see: man 5 proc.
const int page_size_kb = getpagesize() / 1024;
if (page_size_kb <= 0)
@@ -436,9 +431,10 @@ bool ProcessMetrics::GetWorkingSetKBytesStatm(WorkingSetKBytes* ws_usage)
std::vector<StringPiece> statm_vec =
SplitStringPiece(statm, " ", TRIM_WHITESPACE, SPLIT_WANT_ALL);
if (statm_vec.size() != 7)
- return false; // Not the format we expect.
+ return false; // Not the expected format.
- int statm_rss, statm_shared;
+ int statm_rss;
+ int statm_shared;
bool ret = true;
ret &= StringToInt(statm_vec[1], &statm_rss);
ret &= StringToInt(statm_vec[2], &statm_shared);
@@ -464,7 +460,7 @@ size_t GetSystemCommitCharge() {
return meminfo.total - meminfo.free - meminfo.buffers - meminfo.cached;
}
-int ParseProcStatCPU(const std::string& input) {
+int ParseProcStatCPU(StringPiece input) {
// |input| may be empty if the process disappeared somehow.
// e.g. http://crbug.com/145811.
if (input.empty())
@@ -496,13 +492,13 @@ int ParseProcStatCPU(const std::string& input) {
return -1;
}
-const char kProcSelfExe[] = "/proc/self/exe";
-
int GetNumberOfThreads(ProcessHandle process) {
return internal::ReadProcStatsAndGetFieldAsInt64(process,
internal::VM_NUMTHREADS);
}
+const char kProcSelfExe[] = "/proc/self/exe";
+
namespace {
// The format of /proc/diskstats is:
@@ -559,8 +555,7 @@ const size_t kDiskWeightedIOTime = 13;
} // namespace
std::unique_ptr<Value> SystemMemoryInfoKB::ToValue() const {
- std::unique_ptr<DictionaryValue> res(new DictionaryValue());
-
+ auto res = base::MakeUnique<DictionaryValue>();
res->SetInteger("total", total);
res->SetInteger("free", free);
res->SetInteger("available", available);
@@ -588,9 +583,7 @@ std::unique_ptr<Value> SystemMemoryInfoKB::ToValue() const {
return std::move(res);
}
-// exposed for testing
-bool ParseProcMeminfo(const std::string& meminfo_data,
- SystemMemoryInfoKB* meminfo) {
+bool ParseProcMeminfo(StringPiece meminfo_data, SystemMemoryInfoKB* meminfo) {
// The format of /proc/meminfo is:
//
// MemTotal: 8235324 kB
@@ -601,23 +594,23 @@ bool ParseProcMeminfo(const std::string& meminfo_data,
// There is no guarantee on the ordering or position
// though it doesn't appear to change very often
- // As a basic sanity check, let's make sure we at least get non-zero
- // MemTotal value
+ // As a basic sanity check at the end, make sure the MemTotal value will be at
+ // least non-zero. So start off with a zero total.
meminfo->total = 0;
for (const StringPiece& line : SplitStringPiece(
meminfo_data, "\n", KEEP_WHITESPACE, SPLIT_WANT_NONEMPTY)) {
std::vector<StringPiece> tokens = SplitStringPiece(
line, kWhitespaceASCII, TRIM_WHITESPACE, SPLIT_WANT_NONEMPTY);
- // HugePages_* only has a number and no suffix so we can't rely on
- // there being exactly 3 tokens.
+ // HugePages_* only has a number and no suffix so there may not be exactly 3
+ // tokens.
if (tokens.size() <= 1) {
DLOG(WARNING) << "meminfo: tokens: " << tokens.size()
<< " malformed line: " << line.as_string();
continue;
}
- int* target = NULL;
+ int* target = nullptr;
if (tokens[0] == "MemTotal:")
target = &meminfo->total;
else if (tokens[0] == "MemFree:")
@@ -645,7 +638,7 @@ bool ParseProcMeminfo(const std::string& meminfo_data,
else if (tokens[0] == "SReclaimable:")
target = &meminfo->reclaimable;
#if defined(OS_CHROMEOS)
- // Chrome OS has a tweaked kernel that allows us to query Shmem, which is
+ // Chrome OS has a tweaked kernel that allows querying Shmem, which is
// usually video memory otherwise invisible to the OS.
else if (tokens[0] == "Shmem:")
target = &meminfo->shmem;
@@ -656,13 +649,11 @@ bool ParseProcMeminfo(const std::string& meminfo_data,
StringToInt(tokens[1], target);
}
- // Make sure we got a valid MemTotal.
+ // Make sure the MemTotal is valid.
return meminfo->total > 0;
}
-// exposed for testing
-bool ParseProcVmstat(const std::string& vmstat_data,
- SystemMemoryInfoKB* meminfo) {
+bool ParseProcVmstat(StringPiece vmstat_data, SystemMemoryInfoKB* meminfo) {
// The format of /proc/vmstat is:
//
// nr_free_pages 299878
@@ -671,9 +662,8 @@ bool ParseProcVmstat(const std::string& vmstat_data,
// nr_inactive_file 2015629
// ...
//
- // We iterate through the whole file because the position of the
+ // Iterate through the whole file because the position of the
// fields are dependent on the kernel version and configuration.
-
for (const StringPiece& line : SplitStringPiece(
vmstat_data, "\n", KEEP_WHITESPACE, SPLIT_WANT_NONEMPTY)) {
std::vector<StringPiece> tokens = SplitStringPiece(
@@ -782,7 +772,7 @@ SystemDiskInfo::SystemDiskInfo() {
SystemDiskInfo::SystemDiskInfo(const SystemDiskInfo& other) = default;
std::unique_ptr<Value> SystemDiskInfo::ToValue() const {
- std::unique_ptr<DictionaryValue> res(new DictionaryValue());
+ auto res = base::MakeUnique<DictionaryValue>();
// Write out uint64_t variables as doubles.
// Note: this may discard some precision, but for JS there's no other option.
@@ -801,9 +791,10 @@ std::unique_ptr<Value> SystemDiskInfo::ToValue() const {
return std::move(res);
}
-bool IsValidDiskName(const std::string& candidate) {
+bool IsValidDiskName(StringPiece candidate) {
if (candidate.length() < 3)
return false;
+
if (candidate[1] == 'd' &&
(candidate[0] == 'h' || candidate[0] == 's' || candidate[0] == 'v')) {
// [hsv]d[a-z]+ case
@@ -815,14 +806,11 @@ bool IsValidDiskName(const std::string& candidate) {
}
const char kMMCName[] = "mmcblk";
- const size_t kMMCNameLen = strlen(kMMCName);
- if (candidate.length() < kMMCNameLen + 1)
- return false;
- if (candidate.compare(0, kMMCNameLen, kMMCName) != 0)
+ if (!candidate.starts_with(kMMCName))
return false;
// mmcblk[0-9]+ case
- for (size_t i = kMMCNameLen; i < candidate.length(); ++i) {
+ for (size_t i = strlen(kMMCName); i < candidate.length(); ++i) {
if (!isdigit(candidate[i]))
return false;
}
@@ -842,7 +830,7 @@ bool GetSystemDiskInfo(SystemDiskInfo* diskinfo) {
std::vector<StringPiece> diskinfo_lines = SplitStringPiece(
diskinfo_data, "\n", KEEP_WHITESPACE, SPLIT_WANT_NONEMPTY);
- if (diskinfo_lines.size() == 0) {
+ if (diskinfo_lines.empty()) {
DLOG(WARNING) << "No lines found";
return false;
}
@@ -876,31 +864,32 @@ bool GetSystemDiskInfo(SystemDiskInfo* diskinfo) {
line, kWhitespaceASCII, TRIM_WHITESPACE, SPLIT_WANT_NONEMPTY);
// Fields may have overflowed and reset to zero.
- if (IsValidDiskName(disk_fields[kDiskDriveName].as_string())) {
- StringToUint64(disk_fields[kDiskReads], &reads);
- StringToUint64(disk_fields[kDiskReadsMerged], &reads_merged);
- StringToUint64(disk_fields[kDiskSectorsRead], &sectors_read);
- StringToUint64(disk_fields[kDiskReadTime], &read_time);
- StringToUint64(disk_fields[kDiskWrites], &writes);
- StringToUint64(disk_fields[kDiskWritesMerged], &writes_merged);
- StringToUint64(disk_fields[kDiskSectorsWritten], &sectors_written);
- StringToUint64(disk_fields[kDiskWriteTime], &write_time);
- StringToUint64(disk_fields[kDiskIO], &io);
- StringToUint64(disk_fields[kDiskIOTime], &io_time);
- StringToUint64(disk_fields[kDiskWeightedIOTime], &weighted_io_time);
-
- diskinfo->reads += reads;
- diskinfo->reads_merged += reads_merged;
- diskinfo->sectors_read += sectors_read;
- diskinfo->read_time += read_time;
- diskinfo->writes += writes;
- diskinfo->writes_merged += writes_merged;
- diskinfo->sectors_written += sectors_written;
- diskinfo->write_time += write_time;
- diskinfo->io += io;
- diskinfo->io_time += io_time;
- diskinfo->weighted_io_time += weighted_io_time;
- }
+ if (!IsValidDiskName(disk_fields[kDiskDriveName].as_string()))
+ continue;
+
+ StringToUint64(disk_fields[kDiskReads], &reads);
+ StringToUint64(disk_fields[kDiskReadsMerged], &reads_merged);
+ StringToUint64(disk_fields[kDiskSectorsRead], &sectors_read);
+ StringToUint64(disk_fields[kDiskReadTime], &read_time);
+ StringToUint64(disk_fields[kDiskWrites], &writes);
+ StringToUint64(disk_fields[kDiskWritesMerged], &writes_merged);
+ StringToUint64(disk_fields[kDiskSectorsWritten], &sectors_written);
+ StringToUint64(disk_fields[kDiskWriteTime], &write_time);
+ StringToUint64(disk_fields[kDiskIO], &io);
+ StringToUint64(disk_fields[kDiskIOTime], &io_time);
+ StringToUint64(disk_fields[kDiskWeightedIOTime], &weighted_io_time);
+
+ diskinfo->reads += reads;
+ diskinfo->reads_merged += reads_merged;
+ diskinfo->sectors_read += sectors_read;
+ diskinfo->read_time += read_time;
+ diskinfo->writes += writes;
+ diskinfo->writes_merged += writes_merged;
+ diskinfo->sectors_written += sectors_written;
+ diskinfo->write_time += write_time;
+ diskinfo->io += io;
+ diskinfo->io_time += io_time;
+ diskinfo->weighted_io_time += weighted_io_time;
}
return true;
@@ -912,7 +901,7 @@ TimeDelta GetUserCpuTimeSinceBoot() {
#if defined(OS_CHROMEOS)
std::unique_ptr<Value> SwapInfo::ToValue() const {
- std::unique_ptr<DictionaryValue> res(new DictionaryValue());
+ auto res = base::MakeUnique<DictionaryValue>();
// Write out uint64_t variables as doubles.
// Note: this may discard some precision, but for JS there's no other option.
@@ -921,11 +910,10 @@ std::unique_ptr<Value> SwapInfo::ToValue() const {
res->SetDouble("orig_data_size", static_cast<double>(orig_data_size));
res->SetDouble("compr_data_size", static_cast<double>(compr_data_size));
res->SetDouble("mem_used_total", static_cast<double>(mem_used_total));
- if (compr_data_size > 0)
- res->SetDouble("compression_ratio", static_cast<double>(orig_data_size) /
- static_cast<double>(compr_data_size));
- else
- res->SetDouble("compression_ratio", 0);
+ double ratio = compr_data_size ? static_cast<double>(orig_data_size) /
+ static_cast<double>(compr_data_size)
+ : 0;
+ res->SetDouble("compression_ratio", ratio);
return std::move(res);
}
@@ -939,7 +927,7 @@ void GetSwapInfo(SwapInfo* swap_info) {
ReadFileToUint64(zram_path.Append("orig_data_size"));
if (orig_data_size <= 4096) {
// A single page is compressed at startup, and has a high compression
- // ratio. We ignore this as it doesn't indicate any real swapping.
+ // ratio. Ignore this as it doesn't indicate any real swapping.
swap_info->orig_data_size = 0;
swap_info->num_reads = 0;
swap_info->num_writes = 0;
« no previous file with comments | « base/process/process_metrics.h ('k') | no next file » | no next file with comments »

Powered by Google App Engine
This is Rietveld 408576698