bug: fix uptime calculation for Linux (#1410)

* bug: fix uptime for linux

Use another calculation to determine the uptime of a process on
Linux.

* appease clippy

* changelog

* edit add

CHANGELOG.md

@@ -21,6 +21,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 - [#1314](https://github.com/ClementTsang/bottom/pull/1314): Fix fat32 mounts not showing up in macOS.
 - [#1355](https://github.com/ClementTsang/bottom/pull/1355): Reduce chances of non-D0 devices waking up due to temperature checks on Linux.
+- [#1410](https://github.com/ClementTsang/bottom/pull/1410): Fix uptime calculation for Linux.
 
 ## [0.9.6] - 2023-08-26
 

src/data_collection/processes.rs

@@ -74,7 +74,7 @@ pub struct ProcessHarvest {
     /// The current state of the process (e.g. zombie, asleep).
     pub process_state: (String, char),
 
-    /// Cumulative total CPU time used.
+    /// Cumulative process uptime.
     pub time: Duration,
 
     /// This is the *effective* user ID of the process. This is only used on Unix platforms.
@@ -109,7 +109,7 @@ impl ProcessHarvest {
         self.write_bytes_per_sec += rhs.write_bytes_per_sec;
         self.total_read_bytes += rhs.total_read_bytes;
         self.total_write_bytes += rhs.total_write_bytes;
-        self.time += rhs.time;
+        self.time = self.time.max(rhs.time);
         #[cfg(feature = "gpu")]
         {
             self.gpu_mem += rhs.gpu_mem;

src/data_collection/processes/linux.rs

@@ -132,9 +132,7 @@ fn get_linux_cpu_usage(
 }
 
 fn read_proc(
-    prev_proc: &PrevProcDetails, process: Process, cpu_usage: f64, cpu_fraction: f64,
-    use_current_cpu_total: bool, time_difference_in_secs: u64, total_memory: u64,
-    user_table: &mut UserTable,
+    prev_proc: &PrevProcDetails, process: Process, args: ReadProcArgs, user_table: &mut UserTable,
 ) -> error::Result<(ProcessHarvest, u64)> {
     let Process {
         pid: _,
@@ -144,6 +142,15 @@ fn read_proc(
         cmdline,
     } = process;
 
+    let ReadProcArgs {
+        use_current_cpu_total,
+        cpu_usage,
+        cpu_fraction,
+        total_memory,
+        time_difference_in_secs,
+        uptime,
+    } = args;
+
     let (command, name) = {
         let truncated_name = stat.comm.as_str();
         if let Ok(cmdline) = cmdline {
@@ -231,7 +238,7 @@ fn read_proc(
         if ticks_per_sec == 0 {
             Duration::ZERO
         } else {
-            Duration::from_secs(stat.utime + stat.stime) / ticks_per_sec
+            Duration::from_secs(uptime.saturating_sub(stat.start_time / ticks_per_sec as u64))
         }
     } else {
         Duration::ZERO
@@ -279,6 +286,17 @@ fn is_str_numeric(s: &str) -> bool {
     s.chars().all(|c| c.is_ascii_digit())
 }
 
+/// General args to keep around for reading proc data.
+#[derive(Copy, Clone)]
+pub(crate) struct ReadProcArgs {
+    pub(crate) use_current_cpu_total: bool,
+    pub(crate) cpu_usage: f64,
+    pub(crate) cpu_fraction: f64,
+    pub(crate) total_memory: u64,
+    pub(crate) time_difference_in_secs: u64,
+    pub(crate) uptime: u64,
+}
+
 pub(crate) fn linux_process_data(
     collector: &mut DataCollector, time_difference_in_secs: u64,
 ) -> error::Result<Vec<ProcessHarvest>> {
@@ -329,6 +347,15 @@ pub(crate) fn linux_process_data(
             }
         });
 
+        let args = ReadProcArgs {
+            use_current_cpu_total,
+            cpu_usage,
+            cpu_fraction,
+            total_memory,
+            time_difference_in_secs,
+            uptime: sysinfo::System::uptime(),
+        };
+
         let process_vector: Vec<ProcessHarvest> = pids
             .filter_map(|pid_path| {
                 if let Ok(process) = Process::from_path(pid_path) {
@@ -336,16 +363,9 @@ pub(crate) fn linux_process_data(
                     let prev_proc_details = pid_mapping.entry(pid).or_default();
 
                     #[allow(unused_mut)]
-                    if let Ok((mut process_harvest, new_process_times)) = read_proc(
-                        prev_proc_details,
-                        process,
-                        cpu_usage,
-                        cpu_fraction,
-                        use_current_cpu_total,
-                        time_difference_in_secs,
-                        total_memory,
-                        user_table,
-                    ) {
+                    if let Ok((mut process_harvest, new_process_times)) =
+                        read_proc(prev_proc_details, process, args, user_table)
+                    {
                         #[cfg(feature = "gpu")]
                         if let Some(gpus) = &collector.gpu_pids {
                             gpus.iter().for_each(|gpu| {

src/data_collection/processes/linux/process.rs

@@ -26,7 +26,8 @@ fn next_part<'a>(iter: &mut impl Iterator<Item = &'a str>) -> Result<&'a str, io
         .ok_or_else(|| io::Error::from(io::ErrorKind::InvalidData))
 }
 
-/// A wrapper around the data in `/proc/<PID>/stat`. For documentation, see [here](https://man7.org/linux/man-pages/man5/proc.5.html).
+/// A wrapper around the data in `/proc/<PID>/stat`. For documentation, see
+/// [here](https://man7.org/linux/man-pages/man5/proc.5.html).
 ///
 /// Note this does not necessarily get all fields, only the ones we use in bottom.
 pub(crate) struct Stat {
@@ -47,6 +48,9 @@ pub(crate) struct Stat {
 
     /// The resident set size, or the number of pages the process has in real memory.
     pub rss: u64,
+
+    /// The start time of the process, represented in clock ticks.
+    pub start_time: u64,
 }
 
 impl Stat {
@@ -76,18 +80,19 @@ impl Stat {
             .chars()
             .next()
             .ok_or_else(|| anyhow!("missing state"))?;
-
         let ppid: Pid = next_part(&mut rest)?.parse()?;
 
         // Skip 9 fields until utime (pgrp, session, tty_nr, tpgid, flags, minflt, cminflt, majflt, cmajflt).
         let mut rest = rest.skip(9);
-
         let utime: u64 = next_part(&mut rest)?.parse()?;
         let stime: u64 = next_part(&mut rest)?.parse()?;
 
-        // Skip 8 fields until rss (cutime, cstime, priority, nice, num_threads, itrealvalue, starttime, vsize).
-        let mut rest = rest.skip(8);
+        // Skip 6 fields until starttime (cutime, cstime, priority, nice, num_threads, itrealvalue).
+        let mut rest = rest.skip(6);
+        let start_time: u64 = next_part(&mut rest)?.parse()?;
 
+        // Skip one field until rss (vsize)
+        let mut rest = rest.skip(1);
         let rss: u64 = next_part(&mut rest)?.parse()?;
 
         Ok(Stat {
@@ -97,6 +102,7 @@ impl Stat {
             utime,
             stime,
             rss,
+            start_time,
         })
     }
 

src/widgets/process_table/proc_widget_data.rs

@@ -263,6 +263,7 @@ impl ProcWidgetData {
         self.wps += other.wps;
         self.total_read += other.total_read;
         self.total_write += other.total_write;
+        self.time = self.time.max(other.time);
         #[cfg(feature = "gpu")]
         {
             self.gpu_mem_usage = match (&self.gpu_mem_usage, &other.gpu_mem_usage) {