standalone proc/self/maps parsing code to try instead of less reliable current_exe.

Author: pnkfelix

src/symbolize/gimli.rs

@@ -30,6 +30,8 @@ cfg_if::cfg_if! {
     if #[cfg(windows)] {
         #[path = "gimli/mmap_windows.rs"]
         mod mmap;
+        #[path = "gimli/parse_running_mmaps_noop.rs"]
+        mod parse_running_mmaps;
     } else if #[cfg(any(
         target_os = "android",
         target_os = "freebsd",
@@ -44,9 +46,13 @@ cfg_if::cfg_if! {
     ))] {
         #[path = "gimli/mmap_unix.rs"]
         mod mmap;
+        #[path = "gimli/parse_running_mmaps_unix.rs"]
+        mod parse_running_mmaps;
     } else {
         #[path = "gimli/mmap_fake.rs"]
         mod mmap;
+        #[path = "gimli/parse_running_mmaps_noop.rs"]
+        mod parse_running_mmaps;
     }
 }
 

src/symbolize/gimli/libs_dl_iterate_phdr.rs

@@ -17,6 +17,23 @@ pub(super) fn native_libraries() -> Vec<Library> {
     return ret;
 }
 
+fn infer_current_exe(base_addr: usize) -> OsString {
+    if let Ok(entries) = super::parse_running_mmaps::parse_maps() {
+        let opt_path = entries.iter()
+            .find(|e| { // first entry with matching address and useful path
+                e.ip_matches(base_addr) &&
+                    e.pathname().map(|p|p.len() > 0).unwrap_or(false)}
+            )
+            .map(|e|e.pathname())
+            .flatten()
+            .cloned();
+        if let Some(path) = opt_path {
+            return path;
+        }
+    }
+    env::current_exe().map(|e| e.into()).unwrap_or_default()
+}
+
 // `info` should be a valid pointers.
 // `vec` should be a valid pointer to a `std::Vec`.
 unsafe extern "C" fn callback(
@@ -28,8 +45,12 @@ unsafe extern "C" fn callback(
     let libs = &mut *(vec as *mut Vec<Library>);
     let is_main_prog = info.dlpi_name.is_null() || *info.dlpi_name == 0;
     let name = if is_main_prog {
+        // The man page for dl_iterate_phdr says that the first object visited by
+        // callback is the main program; so the first time we encounter a
+        // nameless entry, we can assume its the main program and try to infer its path.
+        // After that, we cannot continue that assumption, and we use an empty string.
         if libs.is_empty() {
-            env::current_exe().map(|e| e.into()).unwrap_or_default()
+            infer_current_exe(info.dlpi_addr as usize)
         } else {
             OsString::new()
         }

src/symbolize/gimli/parse_running_mmaps_noop.rs

@@ -0,0 +1,8 @@
+use super::{OsString, Vec};
+
+pub(super) struct MapsEntry { }
+pub(super) fn parse_maps() -> Result<Vec<MapsEntry>, &'static str> { Err("this is a noop") }
+impl MapsEntry {
+    pub(super) fn pathname(&self) -> Option<&OsString> { None }
+    pub(super) fn ip_matches(&self, _ip: usize) -> bool { false }
+}

src/symbolize/gimli/parse_running_mmaps_unix.rs

@@ -0,0 +1,146 @@
+use super::mystd::io::BufRead;
+use super::{OsString, Vec};
+
+#[derive(PartialEq, Eq, Debug)]
+pub(super) struct MapsEntry {
+    /// start (inclusive) and limit (exclusive) of address range.
+    address: (usize, usize),
+    /// The perms field are the permissions for the entry
+    ///
+    /// r = read
+    /// w = write
+    /// x = execute
+    /// s = shared
+    /// p = private (copy on write)
+    perms: [char; 4],
+    /// Offset into the file (or "whatever").
+    offset: usize,
+    /// device (major, minor)
+    dev: (usize, usize),
+    /// inode on the device. 0 indicates that no inode is associated with the memory region (e.g. uninitalized data aka BSS).
+    inode: usize,
+    /// Usually the file backing the mapping.
+    ///
+    /// Note: The man page for proc includes a note about "coordination" by
+    /// using readelf to see the Offset field in ELF program headers. pnkfelix
+    /// is not yet sure if that is intended to be a comment on pathname, or what
+    /// form/purpose such coordination is meant to have.
+    ///
+    /// There are also some pseudo-paths:
+    /// "[stack]": The initial process's (aka main thread's) stack.
+    /// "[stack:<tid>]": a specific thread's stack. (This was only present for a limited range of Linux verisons; it was determined to be too expensive to provide.)
+    /// "[vdso]": Virtual dynamically linked shared object
+    /// "[heap]": The process's heap
+    ///
+    /// The pathname can be blank, which means it is an anonymous mapping
+    /// obtained via mmap.
+    ///
+    /// Newlines in pathname are replaced with an octal escape sequence.
+    ///
+    /// The pathname may have "(deleted)" appended onto it if the file-backed
+    /// path has been deleted.
+    ///
+    /// Note that modifications like the latter two indicated above imply that
+    /// in general the pathname may be ambiguous. (I.e. you cannot tell if the
+    /// denoted filename actually ended with the text "(deleted)", or if that
+    /// was added by the maps rendering.
+    pathname: OsString,
+}
+
+pub(super) fn parse_maps() -> Result<Vec<MapsEntry>, &'static str> {
+    let mut v = Vec::new();
+    let proc_self_maps = std::fs::File::open("/proc/self/maps").map_err(|_| "couldnt open /proc/self/maps")?;
+    let proc_self_maps = std::io::BufReader::new(proc_self_maps);
+    for line in proc_self_maps.lines() {
+        let line = line.map_err(|_io_error| "couldnt read line from /proc/self/maps")?;
+        v.push(line.parse()?);
+    }
+
+    Ok(v)
+}
+
+impl MapsEntry {
+    pub(super) fn pathname(&self) -> Option<&OsString> {
+        Some(&self.pathname)
+    }
+
+    pub(super) fn ip_matches(&self, ip: usize) -> bool {
+        self.address.0 <= ip && ip < self.address.1
+    }
+}
+
+impl std::str::FromStr for MapsEntry {
+    type Err = &'static str;
+
+    // Format: address perms offset dev inode pathname
+    // e.g.: "ffffffffff600000-ffffffffff601000 --xp 00000000 00:00 0                  [vsyscall]"
+    // e.g.: "7f5985f46000-7f5985f48000 rw-p 00039000 103:06 76021795                  /usr/lib/x86_64-linux-gnu/ld-linux-x86-64.so.2"
+    // e.g.: "35b1a21000-35b1a22000 rw-p 00000000 00:00 0"
+    fn from_str(s: &str) -> Result<Self, Self::Err> {
+        let mut parts = s
+            .split(' ') // space-separated fields
+            .filter(|s| s.len() > 0); // multiple spaces implies empty strings that need to be skipped.
+        let range_str = parts.next().ok_or("Couldn't find address")?;
+        let perms_str = parts.next().ok_or("Couldn't find permissions")?;
+        let offset_str = parts.next().ok_or("Couldn't find offset")?;
+        let dev_str = parts.next().ok_or("Couldn't find dev")?;
+        let inode_str = parts.next().ok_or("Couldn't find inode")?;
+        let pathname_str = parts.next().unwrap_or(""); // pathname may be omitted.
+
+        let hex = |s| usize::from_str_radix(s, 16).map_err(|_| "couldnt parse hex number");
+        let address = {
+            let (start, limit) = range_str.split_once('-').ok_or("Couldn't parse address range")?;
+            (hex(start)?, hex(limit)?)
+        };
+        let perms: [char; 4] = {
+            let mut chars = perms_str.chars();
+            let mut c = || chars.next().ok_or("insufficient perms");
+            let perms = [c()?, c()?, c()?, c()?];
+            if chars.next().is_some() { return Err("too many perms"); }
+            perms
+        };
+        let offset = hex(offset_str)?;
+        let dev = {
+            let (major, minor) = dev_str.split_once(':').ok_or("Couldn't parse dev")?;
+            (hex(major)?, hex(minor)?)
+        };
+        let inode = hex(inode_str)?;
+        let pathname = pathname_str.into();
+
+        Ok(MapsEntry { address, perms, offset, dev, inode, pathname })
+    }
+}
+
+#[test]
+fn check_maps_entry_parsing() {
+    assert_eq!("ffffffffff600000-ffffffffff601000 --xp 00000000 00:00 0                  \
+                [vsyscall]".parse::<MapsEntry>().unwrap(),
+               MapsEntry {
+                   address: (0xffffffffff600000, 0xffffffffff601000),
+                   perms: ['-','-','x','p'],
+                   offset: 0x00000000,
+                   dev: (0x00, 0x00),
+                   inode: 0x0,
+                   pathname: "[vsyscall]".into(),
+               });
+
+    assert_eq!("7f5985f46000-7f5985f48000 rw-p 00039000 103:06 76021795                  \
+                /usr/lib/x86_64-linux-gnu/ld-linux-x86-64.so.2".parse::<MapsEntry>().unwrap(),
+                 MapsEntry {
+                     address: (0x7f5985f46000, 0x7f5985f48000),
+                     perms: ['r','w','-','p'],
+                     offset: 0x00039000,
+                     dev: (0x103, 0x06),
+                     inode: 0x76021795,
+                     pathname: "/usr/lib/x86_64-linux-gnu/ld-linux-x86-64.so.2".into(),
+                 });
+    assert_eq!("35b1a21000-35b1a22000 rw-p 00000000 00:00 0".parse::<MapsEntry>().unwrap(),
+                 MapsEntry {
+                     address: (0x35b1a21000, 0x35b1a22000),
+                     perms: ['r','w','-','p'],
+                     offset: 0x00000000,
+                     dev: (0x00,0x00),
+                     inode: 0x0,
+                     pathname: Default::default(),
+                 });
+}