Rollup merge of #125087 - tbu-:pr_file_stream_len, r=ChrisDenton

Optimize `Seek::stream_len` impl for `File`

It uses the file metadata on Unix with a fallback for files incorrectly reported as zero-sized. It uses `GetFileSizeEx` on Windows.

This reduces the number of syscalls needed for determining the file size of an open file from 3 to 1.

11 files changed, 90 insertions, 11 deletions

diff --git a/library/std/src/fs.rs b/library/std/src/fs.rs
index 711efc7d011..6cbf8301e01 100644
--- a/library/std/src/fs.rs
+++ b/library/std/src/fs.rs
@@ -1311,9 +1311,39 @@ impl Write for &File {
 }
 #[stable(feature = "rust1", since = "1.0.0")]
 impl Seek for &File {
+    /// Seek to an offset, in bytes in a file.
+    ///
+    /// See [`Seek::seek`] docs for more info.
+    ///
+    /// # Platform-specific behavior
+    ///
+    /// This function currently corresponds to the `lseek64` function on Unix
+    /// and the `SetFilePointerEx` function on Windows. Note that this [may
+    /// change in the future][changes].
+    ///
+    /// [changes]: io#platform-specific-behavior
     fn seek(&mut self, pos: SeekFrom) -> io::Result<u64> {
         self.inner.seek(pos)
     }
+
+    /// Returns the length of this file (in bytes).
+    ///
+    /// See [`Seek::stream_len`] docs for more info.
+    ///
+    /// # Platform-specific behavior
+    ///
+    /// This function currently corresponds to the `statx` function on Linux
+    /// (with fallbacks) and the `GetFileSizeEx` function on Windows. Note that
+    /// this [may change in the future][changes].
+    ///
+    /// [changes]: io#platform-specific-behavior
+    fn stream_len(&mut self) -> io::Result<u64> {
+        if let Some(result) = self.inner.size() {
+            return result;
+        }
+        io::stream_len_default(self)
+    }
+
     fn stream_position(&mut self) -> io::Result<u64> {
         self.inner.tell()
     }
@@ -1363,6 +1393,9 @@ impl Seek for File {
     fn seek(&mut self, pos: SeekFrom) -> io::Result<u64> {
         (&*self).seek(pos)
     }
+    fn stream_len(&mut self) -> io::Result<u64> {
+        (&*self).stream_len()
+    }
     fn stream_position(&mut self) -> io::Result<u64> {
         (&*self).stream_position()
     }
@@ -1412,6 +1445,9 @@ impl Seek for Arc<File> {
     fn seek(&mut self, pos: SeekFrom) -> io::Result<u64> {
         (&**self).seek(pos)
     }
+    fn stream_len(&mut self) -> io::Result<u64> {
+        (&**self).stream_len()
+    }
     fn stream_position(&mut self) -> io::Result<u64> {
         (&**self).stream_position()
     }
diff --git a/library/std/src/io/mod.rs b/library/std/src/io/mod.rs
index 03f5f838311..20c82b64bcc 100644
--- a/library/std/src/io/mod.rs
+++ b/library/std/src/io/mod.rs
@@ -2028,7 +2028,7 @@ pub trait Seek {
 
     /// Returns the length of this stream (in bytes).
     ///
-    /// This method is implemented using up to three seek operations. If this
+    /// The default implementation uses up to three seek operations. If this
     /// method returns successfully, the seek position is unchanged (i.e. the
     /// position before calling this method is the same as afterwards).
     /// However, if this method returns an error, the seek position is
@@ -2062,16 +2062,7 @@ pub trait Seek {
     /// ```
     #[unstable(feature = "seek_stream_len", issue = "59359")]
     fn stream_len(&mut self) -> Result<u64> {
-        let old_pos = self.stream_position()?;
-        let len = self.seek(SeekFrom::End(0))?;
-
-        // Avoid seeking a third time when we were already at the end of the
-        // stream. The branch is usually way cheaper than a seek operation.
-        if old_pos != len {
-            self.seek(SeekFrom::Start(old_pos))?;
-        }
-
-        Ok(len)
+        stream_len_default(self)
     }
 
     /// Returns the current seek position from the start of the stream.
@@ -2132,6 +2123,19 @@ pub trait Seek {
     }
 }
 
+pub(crate) fn stream_len_default<T: Seek + ?Sized>(self_: &mut T) -> Result<u64> {
+    let old_pos = self_.stream_position()?;
+    let len = self_.seek(SeekFrom::End(0))?;
+
+    // Avoid seeking a third time when we were already at the end of the
+    // stream. The branch is usually way cheaper than a seek operation.
+    if old_pos != len {
+        self_.seek(SeekFrom::Start(old_pos))?;
+    }
+
+    Ok(len)
+}
+
 /// Enumeration of possible methods to seek within an I/O object.
 ///
 /// It is used by the [`Seek`] trait.
diff --git a/library/std/src/sys/fs/hermit.rs b/library/std/src/sys/fs/hermit.rs
index a9774bef9e3..175d919c289 100644
--- a/library/std/src/sys/fs/hermit.rs
+++ b/library/std/src/sys/fs/hermit.rs
@@ -422,6 +422,10 @@ impl File {
         self.0.seek(pos)
     }
 
+    pub fn size(&self) -> Option<io::Result<u64>> {
+        None
+    }
+
     pub fn tell(&self) -> io::Result<u64> {
         self.0.tell()
     }
diff --git a/library/std/src/sys/fs/solid.rs b/library/std/src/sys/fs/solid.rs
index 3bfb39bac95..808a9582911 100644
--- a/library/std/src/sys/fs/solid.rs
+++ b/library/std/src/sys/fs/solid.rs
@@ -459,6 +459,10 @@ impl File {
         self.tell()
     }
 
+    pub fn size(&self) -> Option<io::Result<u64>> {
+        None
+    }
+
     pub fn tell(&self) -> io::Result<u64> {
         unsafe {
             let mut out_offset = MaybeUninit::uninit();
diff --git a/library/std/src/sys/fs/uefi.rs b/library/std/src/sys/fs/uefi.rs
index 416c90b98b6..5763d7862f5 100644
--- a/library/std/src/sys/fs/uefi.rs
+++ b/library/std/src/sys/fs/uefi.rs
@@ -280,6 +280,10 @@ impl File {
         self.0
     }
 
+    pub fn size(&self) -> Option<io::Result<u64>> {
+        self.0
+    }
+
     pub fn tell(&self) -> io::Result<u64> {
         self.0
     }
diff --git a/library/std/src/sys/fs/unix.rs b/library/std/src/sys/fs/unix.rs
index a3e520fdeef..dc278274f00 100644
--- a/library/std/src/sys/fs/unix.rs
+++ b/library/std/src/sys/fs/unix.rs
@@ -1464,6 +1464,15 @@ impl File {
         Ok(n as u64)
     }
 
+    pub fn size(&self) -> Option<io::Result<u64>> {
+        match self.file_attr().map(|attr| attr.size()) {
+            // Fall back to default implementation if the returned size is 0,
+            // we might be in a proc mount.
+            Ok(0) => None,
+            result => Some(result),
+        }
+    }
+
     pub fn tell(&self) -> io::Result<u64> {
         self.seek(SeekFrom::Current(0))
     }
diff --git a/library/std/src/sys/fs/unsupported.rs b/library/std/src/sys/fs/unsupported.rs
index 0ff9533c047..efaddb51b37 100644
--- a/library/std/src/sys/fs/unsupported.rs
+++ b/library/std/src/sys/fs/unsupported.rs
@@ -259,6 +259,10 @@ impl File {
         self.0
     }
 
+    pub fn size(&self) -> Option<io::Result<u64>> {
+        self.0
+    }
+
     pub fn tell(&self) -> io::Result<u64> {
         self.0
     }
diff --git a/library/std/src/sys/fs/wasi.rs b/library/std/src/sys/fs/wasi.rs
index ebfc7377a2e..b65d86de12a 100644
--- a/library/std/src/sys/fs/wasi.rs
+++ b/library/std/src/sys/fs/wasi.rs
@@ -516,6 +516,10 @@ impl File {
         self.fd.seek(pos)
     }
 
+    pub fn size(&self) -> Option<io::Result<u64>> {
+        None
+    }
+
     pub fn tell(&self) -> io::Result<u64> {
         self.fd.tell()
     }
diff --git a/library/std/src/sys/fs/windows.rs b/library/std/src/sys/fs/windows.rs
index d01a572ac73..a95709b4891 100644
--- a/library/std/src/sys/fs/windows.rs
+++ b/library/std/src/sys/fs/windows.rs
@@ -616,6 +616,14 @@ impl File {
         Ok(newpos as u64)
     }
 
+    pub fn size(&self) -> Option<io::Result<u64>> {
+        let mut result = 0;
+        Some(
+            cvt(unsafe { c::GetFileSizeEx(self.handle.as_raw_handle(), &mut result) })
+                .map(|_| result as u64),
+        )
+    }
+
     pub fn tell(&self) -> io::Result<u64> {
         self.seek(SeekFrom::Current(0))
     }
diff --git a/library/std/src/sys/pal/windows/c/bindings.txt b/library/std/src/sys/pal/windows/c/bindings.txt
index d5fbb453c6f..a99c474c763 100644
--- a/library/std/src/sys/pal/windows/c/bindings.txt
+++ b/library/std/src/sys/pal/windows/c/bindings.txt
@@ -2156,6 +2156,7 @@ GetExitCodeProcess
 GetFileAttributesW
 GetFileInformationByHandle
 GetFileInformationByHandleEx
+GetFileSizeEx
 GetFileType
 GETFINALPATHNAMEBYHANDLE_FLAGS
 GetFinalPathNameByHandleW
diff --git a/library/std/src/sys/pal/windows/c/windows_sys.rs b/library/std/src/sys/pal/windows/c/windows_sys.rs
index eb2914b8644..95bf8040229 100644
--- a/library/std/src/sys/pal/windows/c/windows_sys.rs
+++ b/library/std/src/sys/pal/windows/c/windows_sys.rs
@@ -44,6 +44,7 @@ windows_targets::link!("kernel32.dll" "system" fn GetExitCodeProcess(hprocess :
 windows_targets::link!("kernel32.dll" "system" fn GetFileAttributesW(lpfilename : PCWSTR) -> u32);
 windows_targets::link!("kernel32.dll" "system" fn GetFileInformationByHandle(hfile : HANDLE, lpfileinformation : *mut BY_HANDLE_FILE_INFORMATION) -> BOOL);
 windows_targets::link!("kernel32.dll" "system" fn GetFileInformationByHandleEx(hfile : HANDLE, fileinformationclass : FILE_INFO_BY_HANDLE_CLASS, lpfileinformation : *mut core::ffi::c_void, dwbuffersize : u32) -> BOOL);
+windows_targets::link!("kernel32.dll" "system" fn GetFileSizeEx(hfile : HANDLE, lpfilesize : *mut i64) -> BOOL);
 windows_targets::link!("kernel32.dll" "system" fn GetFileType(hfile : HANDLE) -> FILE_TYPE);
 windows_targets::link!("kernel32.dll" "system" fn GetFinalPathNameByHandleW(hfile : HANDLE, lpszfilepath : PWSTR, cchfilepath : u32, dwflags : GETFINALPATHNAMEBYHANDLE_FLAGS) -> u32);
 windows_targets::link!("kernel32.dll" "system" fn GetFullPathNameW(lpfilename : PCWSTR, nbufferlength : u32, lpbuffer : PWSTR, lpfilepart : *mut PWSTR) -> u32);