Rust support for `struct iov_iter`
Subsystems (2)
Alice Ryhl
reviewerv2 cover Rust support for `struct iov_iter`
This series adds support for the struct iov_iter type. This type represents an IO buffer for reading or writing, and can be configured for either direction of communication.
In Rust, we define separate types for reading and writing. This will ensure that you cannot mix them up and e.g. call copy_from_iter in a read_iter syscall.
To use the new abstractions, miscdevices are given new methods read_iter and write_iter that can be used to implement the read/write syscalls on a miscdevice. The miscdevice sample is updated to provide read/write operations.
Depends on:
https://lore.kernel.org/all/[email protected]/
---
Changes in v2:
- Remove Send/Sync/Copy impls.
- Reword docs significantly.
- Rename Kiocb::private_data() to Kiocb::device().
- Rebase on v6.16-rc2.
- Link to v1: https://lore.kernel.org/r/[email protected]
---
Alice Ryhl (3):
rust: iov: add iov_iter abstractions for ITER_SOURCE
rust: iov: add iov_iter abstractions for ITER_DEST
rust: miscdevice: Provide additional abstractions for iov_iter and kiocb structures
Lee Jones (1):
samples: rust_misc_device: Expand the sample to support read()ing from userspace
rust/kernel/iov.rs | 289 +++++++++++++++++++++++++++++++++++++++ rust/kernel/lib.rs | 1 + rust/kernel/miscdevice.rs | 97 ++++++++++++- samples/rust/rust_misc_device.rs | 36 ++++- 4 files changed, 420 insertions(+), 3 deletions(-)
---
base-commit: e76fe965a0365e346cdaaf462847bac748354ab2
change-id: 20250311-iov-iter-c984aea07d18
prerequisite-change-id: 20250605-pointed-to-6170ae01520f:v3
prerequisite-patch-id: 11b11790787664c10cc5ea40ce7b65f35f442c02
prerequisite-patch-id: bd6b8bd6b811baa375b1d2eb51c1c7475bfdede5
Best regards,
Signature · 1 line
Alice Ryhl <[email protected]>
Alice Ryhl
reviewerPATCH v2 1/4 rust: iov: add iov_iter abstractions for ITER_SOURCE
This adds abstractions for the iov_iter type in the case where data_source is ITER_SOURCE. This will make Rust implementations of fops->write_iter possible.
This series only has support for using existing IO vectors created by C code. Additional abstractions will be needed to support the creation of IO vectors in Rust code.
These abstractions make the assumption that struct iov_iter does not have internal self-references, which implies that it is valid to move it between different local variables.
--- rust/kernel/iov.rs | 152 +++++++++++++++++++++++++++++++++++++++++++++++++++++ rust/kernel/lib.rs | 1 + 2 files changed, 153 insertions(+)
rust/kernel/iov.rs new file +152 -0
@@ -0,0 +1,152 @@ +// SPDX-License-Identifier: GPL-2.0 + +// Copyright (C) 2025 Google LLC. + +//! IO vectors. +//! +//! C headers: [`include/linux/iov_iter.h`](srctree/include/linux/iov_iter.h), +//! [`include/linux/uio.h`](srctree/include/linux/uio.h) + +use crate ::{ + alloc::{Allocator ,Flags }, + bindings, + prelude::*, + types::Opaque , +}; +use core::{marker::PhantomData , mem::MaybeUninit , slice}; + +const ITER_SOURCE :bool = bindings::ITER_SOURCE !=0 ; + +/// An IO vector that acts as a source of data. +/// +/// The data may come from many different sources. This includes both things in kernel-space and +/// reading from userspace. It's not necessarily the case that the data source is immutable, so +/// rewinding the IO vector to read the same data twice is not guaranteed to result in the same +/// bytes. It's also possible that the data source is mapped in a thread-local manner using e.g. +/// `kmap_local_page()`, so this type is not `Send` to ensure that the mapping is read from the +/// right context in that scenario. +/// +/// # Invariants +/// +/// Must hold a valid `struct iov_iter` with `data_source` set to `ITER_SOURCE`. For the duration +/// of `'data`, it must be safe to read the data in this IO vector. +#[repr(transparent)] +pub struct IovIterSource <'data> { + iov:Opaque <bindings::iov_iter >, +/// Represent to the type system that this value contains a pointer to readable data it does +/// not own. + _source:PhantomData <&'data [u8 ]>, +} + +impl <'data>IovIterSource <'data> { +/// Obtain an `IovIterSource` from a raw pointer. +/// +/// # Safety +/// +/// * For the duration of `'iov`, the `struct iov_iter` must remain valid and must not be +/// accessed except through the returned reference. +/// * For the duration of `'data`, the buffers backing this IO vector must be valid for +/// reading. +#[track_caller] +#[inline] +pub unsafe fn from_raw <'iov>(ptr: *mut bindings::iov_iter ) -> &'iovmut IovIterSource <'data> { +// SAFETY: The caller ensures that `ptr` is valid. +let data_source =unsafe { (*ptr).data_source }; +assert_eq! (data_source,ITER_SOURCE ); + +// SAFETY: The caller ensures the struct invariants for the right durations. +unsafe { &mut *ptr.cast ::<IovIterSource <'data>>() } + } + +/// Access this as a raw `struct iov_iter`. +#[inline] +pub fn as_raw (&mut self) -> *mut bindings::iov_iter { + self.iov.get () + } + +/// Returns the number of bytes available in this IO vector. +/// +/// Note that this may overestimate the number of bytes. For example, reading from userspace +/// memory could fail with `EFAULT`, which will be treated as the end of the IO vector. +#[inline] +pub fn len (&self) ->usize { +// SAFETY: It is safe to access the `count` field. +unsafe { + (*self.iov.get ()) + .__bindgen_anon_1 + .__bindgen_anon_1 + .as_ref () + .count + } + } + +/// Returns whether there are any bytes left in this IO vector. +/// +/// This may return `true` even if there are no more bytes available. For example, reading from +/// userspace memory could fail with `EFAULT`, which will be treated as the end of the IO vector. +#[inline] +pub fn is_empty (&self) ->bool { + self.len () ==0 + } + +/// Advance this IO vector by `bytes` bytes. +/// +/// If `bytes` is larger than the size of this IO vector, it is advanced to the end. +#[inline] +pub fn advance (&mut self, bytes:usize ) { +// SAFETY: `self.iov` is a valid IO vector. +unsafe { bindings::iov_iter_advance (self.as_raw (), bytes) }; + } + +/// Advance this IO vector backwards by `bytes` bytes. +/// +/// # Safety +/// +/// The IO vector must not be reverted to before its beginning. +#[inline] +pub unsafe fn revert (&mut self, bytes:usize ) { +// SAFETY: `self.iov` is a valid IO vector, and `bytes` is in bounds. +unsafe { bindings::iov_iter_revert (self.as_raw (), bytes) }; + } + +/// Read data from this IO vector. +/// +/// Returns the number of bytes that have been copied. +#[inline] +pub fn copy_from_iter (&mut self, out: &mut [u8 ]) ->usize { +// SAFETY: We will not write uninitialized bytes to `out`. +let out =unsafe { &mut *(outas *mut [u8 ]as *mut [MaybeUninit <u8 >]) }; + + self.copy_from_iter_raw (out).len () + } + +/// Read data from this IO vector and append it to a vector. +/// +/// Returns the number of bytes that have been copied. +#[inline] +pub fn copy_from_iter_vec <A :Allocator >( + &mut self, + out: &mut Vec <u8 ,A >, + flags:Flags , + ) ->Result <usize > { + out.reserve (self.len (), flags)?; +let len = self.copy_from_iter_raw (out.spare_capacity_mut ()).len (); +// SAFETY: The next `len` bytes of the vector have been initialized. +unsafe { out.inc_len (len) }; +Ok (len) + } + +/// Read data from this IO vector into potentially uninitialized memory. +/// +/// Returns the sub-slice of the output that has been initialized. If the returned slice is +/// shorter than the input buffer, then the entire IO vector has been read. +#[inline] +pub fn copy_from_iter_raw (&mut self, out: &mut [MaybeUninit <u8 >]) -> &mut [u8 ] { +// SAFETY: `out` is valid for `out.len()` bytes. +let len = +unsafe { bindings::_copy_from_iter (out.as_mut_ptr ().cast (), out.len (), self.as_raw ()) }; + +// SAFETY: We just initialized the first `len` bytes of `out`. +unsafe { slice::from_raw_parts_mut (out.as_mut_ptr ().cast (), len) } + } +}
rust/kernel/lib.rs +1 -0
@@ -81,6 +81,7 @@pub mod init;pub mod io;pub mod ioctl; +pub mod iov;pub mod jump_label;#[cfg( CONFIG_KUNIT )] pub mod kunit;
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
Andreas Hindborg
reviewerRe: [PATCH v2 1/4] rust: iov: add iov_iter abstractions for ITER_SOURCE
Quoted 46 lines · Alice Ryhl
This adds abstractions for the
iov_itertype in the case wheredata_sourceisITER_SOURCE. This will make Rust implementations offops->write_iterpossible.This series only has support for using existing IO vectors created by C code. Additional abstractions will be needed to support the creation of IO vectors in Rust code.
These abstractions make the assumption that
struct iov_iterdoes not have internal self-references, which implies that it is valid to move it between different local variables.Signed-off-by Alice Ryhl <[email protected]>--- rust/kernel/iov.rs | 152 +++++++++++++++++++++++++++++++++++++++++++++++++++++ rust/kernel/lib.rs | 1 + 2 files changed, 153 insertions(+)rust/kernel/iov.rs new file +22
-0@@ -0,0 +1,152 @@ +// SPDX-License-Identifier: GPL-2.0 + +// Copyright (C) 2025 Google LLC. + +//! IO vectors. +//! +//! C headers: [`include/linux/iov_iter.h`](srctree/include/linux/iov_iter.h), +//! [`include/linux/uio.h`](srctree/include/linux/uio.h) + +use crate ::{ + alloc::{Allocator ,Flags }, + bindings, + prelude::*, + types::Opaque , +}; +use core::{marker::PhantomData , mem::MaybeUninit , slice}; + +const ITER_SOURCE :bool = bindings::ITER_SOURCE !=0 ; + +/// An IO vector that acts as a source of data. +/// +/// The data may come from many different sources. This includes both things in kernel-space and
rust/kernel/iov.rs line 32
Quoted 10 lines
rust/kernel/iov.rs new file +10
-0+/// reading from userspace. It's not necessarily the case that the data source is immutable, so +/// rewinding the IO vector to read the same data twice is not guaranteed to result in the same +/// bytes. It's also possible that the data source is mapped in a thread-local manner using e.g. +/// `kmap_local_page()`, so this type is not `Send` to ensure that the mapping is read from the +/// right context in that scenario. +/// +/// # Invariants +/// +/// Must hold a valid `struct iov_iter` with `data_source` set to `ITER_SOURCE`. For the duration +/// of `'data`, it must be safe to read the data in this IO vector.
In my opinion, the phrasing you had in v1 was better:
The buffers referenced by the IO vector must be valid for reading for the duration of 'data.
That is, I would prefer "must be valid for reading" over "it must be safe to read ...".
Quoted 31 lines
rust/kernel/iov.rs new file +31
-0+#[repr(transparent)] +pub struct IovIterSource <'data> { + iov:Opaque <bindings::iov_iter >, +/// Represent to the type system that this value contains a pointer to readable data it does +/// not own. + _source:PhantomData <&'data [u8 ]>, +} + +impl <'data>IovIterSource <'data> { +/// Obtain an `IovIterSource` from a raw pointer. +/// +/// # Safety +/// +/// * For the duration of `'iov`, the `struct iov_iter` must remain valid and must not be +/// accessed except through the returned reference. +/// * For the duration of `'data`, the buffers backing this IO vector must be valid for +/// reading. +#[track_caller] +#[inline] +pub unsafe fn from_raw <'iov>(ptr: *mut bindings::iov_iter ) -> &'iovmut IovIterSource <'data> { +// SAFETY: The caller ensures that `ptr` is valid. +let data_source =unsafe { (*ptr).data_source }; +assert_eq! (data_source,ITER_SOURCE ); + +// SAFETY: The caller ensures the struct invariants for the right durations. +unsafe { &mut *ptr.cast ::<IovIterSource <'data>>() } + } + +/// Access this as a raw `struct iov_iter`. +#[inline] +pub fn as_raw (&mut self) -> *mut bindings::iov_iter {
rust/kernel/iov.rs line 73
Quoted 10 lines
rust/kernel/iov.rs new file +10
-0+ self.iov.get () + } + +/// Returns the number of bytes available in this IO vector. +/// +/// Note that this may overestimate the number of bytes. For example, reading from userspace +/// memory could fail with `EFAULT`, which will be treated as the end of the IO vector. +#[inline] +pub fn len (&self) ->usize { +// SAFETY: It is safe to access the `count` field.
Reiterating my comment from v1: Why?
Quoted 34 lines
rust/kernel/iov.rs new file +34
-0+unsafe { + (*self.iov.get ()) + .__bindgen_anon_1 + .__bindgen_anon_1 + .as_ref () + .count + } + } + +/// Returns whether there are any bytes left in this IO vector. +/// +/// This may return `true` even if there are no more bytes available. For example, reading from +/// userspace memory could fail with `EFAULT`, which will be treated as the end of the IO vector. +#[inline] +pub fn is_empty (&self) ->bool { + self.len () ==0 + } + +/// Advance this IO vector by `bytes` bytes. +/// +/// If `bytes` is larger than the size of this IO vector, it is advanced to the end. +#[inline] +pub fn advance (&mut self, bytes:usize ) { +// SAFETY: `self.iov` is a valid IO vector. +unsafe { bindings::iov_iter_advance (self.as_raw (), bytes) }; + } + +/// Advance this IO vector backwards by `bytes` bytes. +/// +/// # Safety +/// +/// The IO vector must not be reverted to before its beginning. +#[inline] +pub unsafe fn revert (&mut self, bytes:usize ) {
rust/kernel/iov.rs line 117
Quoted 10 lines
rust/kernel/iov.rs new file +10
-0+// SAFETY: `self.iov` is a valid IO vector, and `bytes` is in bounds. +unsafe { bindings::iov_iter_revert (self.as_raw (), bytes) }; + } + +/// Read data from this IO vector. +/// +/// Returns the number of bytes that have been copied. +#[inline] +pub fn copy_from_iter (&mut self, out: &mut [u8 ]) ->usize { +// SAFETY: We will not write uninitialized bytes to `out`.
Can you provide something to back this claim?
Best regards,
Andreas Hindborg
Alice Ryhl
reviewerRe: [PATCH v2 1/4] rust: iov: add iov_iter abstractions for ITER_SOURCE
Quoted 8 lines · Andreas Hindborg
rust/kernel/iov.rs line 32
Quoted 4 lines · Alice Ryhl
rust/kernel/iov.rs new file +4
-0+/// # Invariants +/// +/// Must hold a valid `struct iov_iter` with `data_source` set to `ITER_SOURCE`. For the duration +/// of `'data`, it must be safe to read the data in this IO vector. In my opinion, the phrasing you had in v1 was better:
Quoted 5 lines
The buffers referenced by the IO vector must be valid for reading for the duration of
'data.That is, I would prefer "must be valid for reading" over "it must be safe to read ...".
If it's backed by userspace data, then technically there aren't any buffers that are valid for reading in the usual sense. We need to call into special assembly to read it, and a normal pointer dereference would be illegal.
Quoted 9 lines
rust/kernel/iov.rs line 73
Quoted 7 lines
rust/kernel/iov.rs new file +7
-0+/// Returns the number of bytes available in this IO vector. +/// +/// Note that this may overestimate the number of bytes. For example, reading from userspace +/// memory could fail with `EFAULT`, which will be treated as the end of the IO vector. +#[inline] +pub fn len (&self) ->usize { +// SAFETY: It is safe to access the `count` field. Reiterating my comment from v1: Why?
It's the same reason as why this is safe:
struct HasLength {
length: usize,
}
impl HasLength {
fn len(&self) -> usize {
// why is this safe?
self.length
}
}I'm not sure how to say it concisely. I guess it's because all access to the iov_iter goes through the &IovIterSource.
Quoted 38 lines
Quoted 38 lines
rust/kernel/iov.rs new file +38
-0+unsafe { + (*self.iov.get ()) + .__bindgen_anon_1 + .__bindgen_anon_1 + .as_ref () + .count + } + } + +/// Returns whether there are any bytes left in this IO vector. +/// +/// This may return `true` even if there are no more bytes available. For example, reading from +/// userspace memory could fail with `EFAULT`, which will be treated as the end of the IO vector. +#[inline] +pub fn is_empty (&self) ->bool { + self.len () ==0 + } + +/// Advance this IO vector by `bytes` bytes. +/// +/// If `bytes` is larger than the size of this IO vector, it is advanced to the end. +#[inline] +pub fn advance (&mut self, bytes:usize ) { +// SAFETY: `self.iov` is a valid IO vector. +unsafe { bindings::iov_iter_advance (self.as_raw (), bytes) }; + } + +/// Advance this IO vector backwards by `bytes` bytes. +/// +/// # Safety +/// +/// The IO vector must not be reverted to before its beginning. +#[inline] +pub unsafe fn revert (&mut self, bytes:usize ) { +// SAFETY: `self.iov` is a valid IO vector, and `bytes` is in bounds. +unsafe { bindings::iov_iter_revert (self.as_raw (), bytes) }; + } +
Quoted 8 lines
rust/kernel/iov.rs line 117
Quoted 6 lines
rust/kernel/iov.rs new file +6
-0+/// Read data from this IO vector. +/// +/// Returns the number of bytes that have been copied. +#[inline] +pub fn copy_from_iter (&mut self, out: &mut [u8 ]) ->usize { +// SAFETY: We will not write uninitialized bytes to `out`. Can you provide something to back this claim?
I guess the logic could go along these lines:
* If the iov_iter reads from userspace, then it's because we always consider such reads to produce initialized data.
* If the iov_iter reads from a kernel buffer, then the creator of the iov_iter must provide an initialized buffer.
Ultimately, if we don't know that the bytes are initialized, then it's impossible to use the API correctly because you can never inspect the bytes in any way. I.e., any implementation of copy_from_iter that produces uninit data is necessarily buggy.
Alice
Andreas Hindborg
reviewerRe: [PATCH v2 1/4] rust: iov: add iov_iter abstractions for ITER_SOURCE
Quoted 12 lines · Alice Ryhl
Quoted 11 lines · Andreas Hindborg
rust/kernel/iov.rs line 32
Quoted 4 lines · Alice Ryhl
rust/kernel/iov.rs new file +4
-0+/// # Invariants +/// +/// Must hold a valid `struct iov_iter` with `data_source` set to `ITER_SOURCE`. For the duration +/// of `'data`, it must be safe to read the data in this IO vector. In my opinion, the phrasing you had in v1 was better:
The buffers referenced by the IO vector must be valid for reading for the duration of
'data.
Quoted 7 lines
Quoted 2 lines
That is, I would prefer "must be valid for reading" over "it must be safe to read ...".
If it's backed by userspace data, then technically there aren't any buffers that are valid for reading in the usual sense. We need to call into special assembly to read it, and a normal pointer dereference would be illegal.
If you go with "safe to read" for this reason, I think you should expand the statement along the lines you used here.
What is the special assembly that is used to read this data? From a quick scan it looks like that if CONFIG_UACCESS_MEMCPY is enabled, a regular memcpy call is used.
Quoted 23 lines
Quoted 10 lines
rust/kernel/iov.rs line 73
Quoted 8 lines
rust/kernel/iov.rs new file +7
-0+/// Returns the number of bytes available in this IO vector. +/// +/// Note that this may overestimate the number of bytes. For example, reading from userspace +/// memory could fail with `EFAULT`, which will be treated as the end of the IO vector. +#[inline] +pub fn len (&self) ->usize { +// SAFETY: It is safe to access the `count` field. Reiterating my comment from v1: Why?
It's the same reason as why this is safe:
struct HasLength {
length: usize,
}impl HasLength { fn len(&self) -> usize { // why is this safe? self.length } }
Quoted 2 lines
I'm not sure how to say it concisely. I guess it's because all access to the
iov_itergoes through the &IovIterSource.
So "By existence of a shared reference to self, count is valid for read."?
Quoted 48 lines
Quoted 39 lines
Quoted 39 lines
rust/kernel/iov.rs new file +38
-0+unsafe { + (*self.iov.get ()) + .__bindgen_anon_1 + .__bindgen_anon_1 + .as_ref () + .count + } + } + +/// Returns whether there are any bytes left in this IO vector. +/// +/// This may return `true` even if there are no more bytes available. For example, reading from +/// userspace memory could fail with `EFAULT`, which will be treated as the end of the IO vector. +#[inline] +pub fn is_empty (&self) ->bool { + self.len () ==0 + } + +/// Advance this IO vector by `bytes` bytes. +/// +/// If `bytes` is larger than the size of this IO vector, it is advanced to the end. +#[inline] +pub fn advance (&mut self, bytes:usize ) { +// SAFETY: `self.iov` is a valid IO vector. +unsafe { bindings::iov_iter_advance (self.as_raw (), bytes) }; + } + +/// Advance this IO vector backwards by `bytes` bytes. +/// +/// # Safety +/// +/// The IO vector must not be reverted to before its beginning. +#[inline] +pub unsafe fn revert (&mut self, bytes:usize ) { +// SAFETY: `self.iov` is a valid IO vector, and `bytes` is in bounds. +unsafe { bindings::iov_iter_revert (self.as_raw (), bytes) }; + } +Quoted 8 lines
rust/kernel/iov.rs line 117
Quoted 6 lines
rust/kernel/iov.rs new file +6
-0+/// Read data from this IO vector. +/// +/// Returns the number of bytes that have been copied. +#[inline] +pub fn copy_from_iter (&mut self, out: &mut [u8 ]) ->usize { +// SAFETY: We will not write uninitialized bytes to `out`. Can you provide something to back this claim?
Quoted 4 lines
I guess the logic could go along these lines:
* If the
iov_iterreads from userspace, then it's because we always consider such reads to produce initialized data.
I don't think it is enough to just state that we consider the reads to produce initialized data.
Quoted 7 lines
* If the
iov_iterreads from a kernel buffer, then the creator of theiov_itermust provide an initialized buffer.Ultimately, if we don't know that the bytes are initialized, then it's impossible to use the API correctly because you can never inspect the bytes in any way. I.e., any implementation of
copy_from_iterthat produces uninit data is necessarily buggy.
I would agree. How do we fix that? You are more knowledgeable than me in this field, so you probably have a better shot than me, at finding a solution.
As far as I can tell, we need to read from a place unknown to the rust abstract machine, and we need to be able to have the abstract machine consider the data initialized after the read.
Is this volatile memcpy [1], or would that only solve the data race problem, not uninitialized data problem?
Best regards,
Andreas Hindborg
Alice Ryhl
reviewerRe: [PATCH v2 1/4] rust: iov: add iov_iter abstractions for ITER_SOURCE
Quoted 22 lines · Andreas Hindborg
Quoted 12 lines · Alice Ryhl
Quoted 11 lines · Andreas Hindborg
rust/kernel/iov.rs line 32
Quoted 4 lines · Alice Ryhl
rust/kernel/iov.rs new file +4
-0+/// # Invariants +/// +/// Must hold a valid `struct iov_iter` with `data_source` set to `ITER_SOURCE`. For the duration +/// of `'data`, it must be safe to read the data in this IO vector. In my opinion, the phrasing you had in v1 was better:
The buffers referenced by the IO vector must be valid for reading for the duration of
'data.Quoted 7 lines
Quoted 2 lines
That is, I would prefer "must be valid for reading" over "it must be safe to read ...".
If it's backed by userspace data, then technically there aren't any buffers that are valid for reading in the usual sense. We need to call into special assembly to read it, and a normal pointer dereference would be illegal.
Quoted 6 lines
If you go with "safe to read" for this reason, I think you should expand the statement along the lines you used here.
What is the special assembly that is used to read this data? From a quick scan it looks like that if
CONFIG_UACCESS_MEMCPYis enabled, a regularmemcpycall is used.
When reading from userspace, you're given an arbitrary untrusted address that could point anywhere. The memory could be swapped out and need to be loaded back from disk. The memory could correspond to an mmap region for a file on a NFS mount and reading it could involve a network call. The address could be dangling, which must be properly handled and turned into an EFAULT error instead of UB. Every architecture has its own asm for handling all of this safely so that behavior is safe no matter what pointer we are given from userspace.
As for CONFIG_UACCESS_MEMCPY, I don't think it is used on any real system today. It would require you to be on a NOMMU system where the userspace and the kernel are in the same address space.
Quoted 75 lines
Quoted 22 lines
Quoted 9 lines
rust/kernel/iov.rs line 73
Quoted 7 lines
rust/kernel/iov.rs new file +7
-0+/// Returns the number of bytes available in this IO vector. +/// +/// Note that this may overestimate the number of bytes. For example, reading from userspace +/// memory could fail with `EFAULT`, which will be treated as the end of the IO vector. +#[inline] +pub fn len (&self) ->usize { +// SAFETY: It is safe to access the `count` field. Reiterating my comment from v1: Why?
It's the same reason as why this is safe:
struct HasLength {
length: usize,
}impl HasLength { fn len(&self) -> usize { // why is this safe? self.length } }Quoted 2 lines
I'm not sure how to say it concisely. I guess it's because all access to the
iov_itergoes through the &IovIterSource.So "By existence of a shared reference to
self,countis valid for read."?Quoted 48 lines
Quoted 47 lines
Quoted 35 lines
rust/kernel/iov.rs new file +34
-0+unsafe { + (*self.iov.get ()) + .__bindgen_anon_1 + .__bindgen_anon_1 + .as_ref () + .count + } + } + +/// Returns whether there are any bytes left in this IO vector. +/// +/// This may return `true` even if there are no more bytes available. For example, reading from +/// userspace memory could fail with `EFAULT`, which will be treated as the end of the IO vector. +#[inline] +pub fn is_empty (&self) ->bool { + self.len () ==0 + } + +/// Advance this IO vector by `bytes` bytes. +/// +/// If `bytes` is larger than the size of this IO vector, it is advanced to the end. +#[inline] +pub fn advance (&mut self, bytes:usize ) { +// SAFETY: `self.iov` is a valid IO vector. +unsafe { bindings::iov_iter_advance (self.as_raw (), bytes) }; + } + +/// Advance this IO vector backwards by `bytes` bytes. +/// +/// # Safety +/// +/// The IO vector must not be reverted to before its beginning. +#[inline] +pub unsafe fn revert (&mut self, bytes:usize ) {rust/kernel/iov.rs line 117
Quoted 10 lines
rust/kernel/iov.rs new file +10
-0+// SAFETY: `self.iov` is a valid IO vector, and `bytes` is in bounds. +unsafe { bindings::iov_iter_revert (self.as_raw (), bytes) }; + } + +/// Read data from this IO vector. +/// +/// Returns the number of bytes that have been copied. +#[inline] +pub fn copy_from_iter (&mut self, out: &mut [u8 ]) ->usize { +// SAFETY: We will not write uninitialized bytes to `out`. Can you provide something to back this claim?
Quoted 7 lines
Quoted 4 lines
I guess the logic could go along these lines:
* If the
iov_iterreads from userspace, then it's because we always consider such reads to produce initialized data.I don't think it is enough to just state that we consider the reads to produce initialized data.
See above re userspace.
Quoted 11 lines
Quoted 7 lines
* If the
iov_iterreads from a kernel buffer, then the creator of theiov_itermust provide an initialized buffer.Ultimately, if we don't know that the bytes are initialized, then it's impossible to use the API correctly because you can never inspect the bytes in any way. I.e., any implementation of
copy_from_iterthat produces uninit data is necessarily buggy.I would agree. How do we fix that? You are more knowledgeable than me in this field, so you probably have a better shot than me, at finding a solution.
I think there is nothing to fix. If there exists a callsite on the C side that creates an iov_iter that reads from an uninitialized kernel buffer, then we can fix that specific call-site. I don't think anything else needs to be done.
Quoted 8 lines
As far as I can tell, we need to read from a place unknown to the rust abstract machine, and we need to be able to have the abstract machine consider the data initialized after the read.
Is this volatile memcpy [1], or would that only solve the data race problem, not uninitialized data problem?
Volatile memcpy deals with data races.
In general, we can argue all we want about wording of these safety comments, but calling copy_from_iter is the right way to read from an iov_iter. If there is a problem, the problem is specific call-sites that construct an iov_iter with an uninit buffer. I don't know whether such call-sites exist.
Alice
Andreas Hindborg
reviewerRe: [PATCH v2 1/4] rust: iov: add iov_iter abstractions for ITER_SOURCE
Quoted 30 lines · Alice Ryhl
Quoted 22 lines · Andreas Hindborg
Quoted 12 lines · Alice Ryhl
Quoted 11 lines · Andreas Hindborg
rust/kernel/iov.rs line 32
Quoted 4 lines · Alice Ryhl
rust/kernel/iov.rs new file +4
-0+/// # Invariants +/// +/// Must hold a valid `struct iov_iter` with `data_source` set to `ITER_SOURCE`. For the duration +/// of `'data`, it must be safe to read the data in this IO vector. In my opinion, the phrasing you had in v1 was better:
The buffers referenced by the IO vector must be valid for reading for the duration of
'data.Quoted 7 lines
Quoted 2 lines
That is, I would prefer "must be valid for reading" over "it must be safe to read ...".
If it's backed by userspace data, then technically there aren't any buffers that are valid for reading in the usual sense. We need to call into special assembly to read it, and a normal pointer dereference would be illegal.
Quoted 6 lines
If you go with "safe to read" for this reason, I think you should expand the statement along the lines you used here.
What is the special assembly that is used to read this data? From a quick scan it looks like that if
CONFIG_UACCESS_MEMCPYis enabled, a regularmemcpycall is used.
Quoted 8 lines
When reading from userspace, you're given an arbitrary untrusted address that could point anywhere. The memory could be swapped out and need to be loaded back from disk. The memory could correspond to an mmap region for a file on a NFS mount and reading it could involve a network call. The address could be dangling, which must be properly handled and turned into an EFAULT error instead of UB. Every architecture has its own asm for handling all of this safely so that behavior is safe no matter what pointer we are given from userspace.
I don't think that is relevant. My point is, you can't reference "special assemby" without detailing what that means.
You have a safety requirement in from_raw:
/// * For the duration of `'data`, the buffers backing this IO vector must be valid for
/// reading.This should probably be promoted to invariant for the type, since from_raw is the only way to construct the type?
But are you saying that the referenced buffers need not be mapped and readable while this type exists? The mapping happens as part of bindings::_copy_to_iter?
Quoted 3 lines
As for
CONFIG_UACCESS_MEMCPY, I don't think it is used on any real system today. It would require you to be on a NOMMU system where the userspace and the kernel are in the same address space.
Ah. I was just browsing for the "special assembly", and that was all I could find.
Quoted 78 lines
Quoted 78 lines
Quoted 23 lines
Quoted 10 lines
rust/kernel/iov.rs line 73
Quoted 8 lines
rust/kernel/iov.rs new file +7
-0+/// Returns the number of bytes available in this IO vector. +/// +/// Note that this may overestimate the number of bytes. For example, reading from userspace +/// memory could fail with `EFAULT`, which will be treated as the end of the IO vector. +#[inline] +pub fn len (&self) ->usize { +// SAFETY: It is safe to access the `count` field. Reiterating my comment from v1: Why?
It's the same reason as why this is safe:
struct HasLength {
length: usize,
}impl HasLength { fn len(&self) -> usize { // why is this safe? self.length } }Quoted 2 lines
I'm not sure how to say it concisely. I guess it's because all access to the
iov_itergoes through the &IovIterSource.So "By existence of a shared reference to
self,countis valid for read."?Quoted 50 lines
Quoted 39 lines
Quoted 39 lines
rust/kernel/iov.rs new file +38
-0+unsafe { + (*self.iov.get ()) + .__bindgen_anon_1 + .__bindgen_anon_1 + .as_ref () + .count + } + } + +/// Returns whether there are any bytes left in this IO vector. +/// +/// This may return `true` even if there are no more bytes available. For example, reading from +/// userspace memory could fail with `EFAULT`, which will be treated as the end of the IO vector. +#[inline] +pub fn is_empty (&self) ->bool { + self.len () ==0 + } + +/// Advance this IO vector by `bytes` bytes. +/// +/// If `bytes` is larger than the size of this IO vector, it is advanced to the end. +#[inline] +pub fn advance (&mut self, bytes:usize ) { +// SAFETY: `self.iov` is a valid IO vector. +unsafe { bindings::iov_iter_advance (self.as_raw (), bytes) }; + } + +/// Advance this IO vector backwards by `bytes` bytes. +/// +/// # Safety +/// +/// The IO vector must not be reverted to before its beginning. +#[inline] +pub unsafe fn revert (&mut self, bytes:usize ) { +// SAFETY: `self.iov` is a valid IO vector, and `bytes` is in bounds. +unsafe { bindings::iov_iter_revert (self.as_raw (), bytes) }; + } +Quoted 8 lines
rust/kernel/iov.rs line 117
Quoted 6 lines
rust/kernel/iov.rs new file +6
-0+/// Read data from this IO vector. +/// +/// Returns the number of bytes that have been copied. +#[inline] +pub fn copy_from_iter (&mut self, out: &mut [u8 ]) ->usize { +// SAFETY: We will not write uninitialized bytes to `out`. Can you provide something to back this claim?
I guess the logic could go along these lines:
Quoted 7 lines
Quoted 5 lines
Quoted 2 lines
* If the
iov_iterreads from userspace, then it's because we always consider such reads to produce initialized data.I don't think it is enough to just state that we consider the reads to produce initialized data.
See above re userspace.
You actually have the safety requirement I was looking for in from_raw:
/// * For the duration of `'data`, the buffers backing this IO vector must be valid for
/// reading.But I am wondering whether this needs to align with the invariant, and not the other way around?
Quoted 9 lines
Quoted 9 lines
Quoted 8 lines
* If the
iov_iterreads from a kernel buffer, then the creator of theiov_itermust provide an initialized buffer.Ultimately, if we don't know that the bytes are initialized, then it's impossible to use the API correctly because you can never inspect the bytes in any way. I.e., any implementation of
copy_from_iterthat produces uninit data is necessarily buggy.
Quoted 8 lines
Quoted 3 lines
I would agree. How do we fix that? You are more knowledgeable than me in this field, so you probably have a better shot than me, at finding a solution.
I think there is nothing to fix. If there exists a callsite on the C side that creates an
iov_iterthat reads from an uninitialized kernel buffer, then we can fix that specific call-site. I don't think anything else needs to be done.
If soundness of this code hinges on specific call site behavior, this should be a safety requirement.
Quoted 10 lines
Quoted 9 lines
As far as I can tell, we need to read from a place unknown to the rust abstract machine, and we need to be able to have the abstract machine consider the data initialized after the read.
Is this volatile memcpy [1], or would that only solve the data race problem, not uninitialized data problem?
Quoted 7 lines
Volatile memcpy deals with data races.
In general, we can argue all we want about wording of these safety comments, but calling
copy_from_iteris the right way to read from aniov_iter. If there is a problem, the problem is specific call-sites that construct aniov_iterwith an uninit buffer. I don't know whether such call-sites exist.
I am not saying it is the wrong way. I am asking that we detail in the safety requirements _why_ it is the right way.
You have a type invariant
For the duration of 'data, it must be safe to read the data in this IO vector.
that says "safe to read" instead of "valid for read" because "special assembly" is used to read the data, and that somehow makes it OK. We should be more specific.
How about making the invariant:
For the duration of 'data, it must be safe to read the data in this IO vector with the C API _copy_from_iter.
And then your safety comment regarding uninit bytes can be:
We write out with copy_from_iter_raw, which transitively writes out using _copy_from_iter. By C API contract, _copy_from_iter does not write uninitialized bytes to out.
In this way we can defer to the implementation of _copy_from_user, which is what I think you want?
Best regards,
Andreas Hindborg
Alice Ryhl
reviewerRe: [PATCH v2 1/4] rust: iov: add iov_iter abstractions for ITER_SOURCE
Quoted 44 lines · Andreas Hindborg
Quoted 30 lines · Alice Ryhl
Quoted 22 lines · Andreas Hindborg
Quoted 12 lines · Alice Ryhl
Quoted 11 lines · Andreas Hindborg
rust/kernel/iov.rs line 32
Quoted 4 lines · Alice Ryhl
rust/kernel/iov.rs new file +4
-0+/// # Invariants +/// +/// Must hold a valid `struct iov_iter` with `data_source` set to `ITER_SOURCE`. For the duration +/// of `'data`, it must be safe to read the data in this IO vector. In my opinion, the phrasing you had in v1 was better:
The buffers referenced by the IO vector must be valid for reading for the duration of
'data.Quoted 7 lines
Quoted 2 lines
That is, I would prefer "must be valid for reading" over "it must be safe to read ...".
If it's backed by userspace data, then technically there aren't any buffers that are valid for reading in the usual sense. We need to call into special assembly to read it, and a normal pointer dereference would be illegal.
Quoted 6 lines
If you go with "safe to read" for this reason, I think you should expand the statement along the lines you used here.
What is the special assembly that is used to read this data? From a quick scan it looks like that if
CONFIG_UACCESS_MEMCPYis enabled, a regularmemcpycall is used.Quoted 8 lines
When reading from userspace, you're given an arbitrary untrusted address that could point anywhere. The memory could be swapped out and need to be loaded back from disk. The memory could correspond to an mmap region for a file on a NFS mount and reading it could involve a network call. The address could be dangling, which must be properly handled and turned into an EFAULT error instead of UB. Every architecture has its own asm for handling all of this safely so that behavior is safe no matter what pointer we are given from userspace.
I don't think that is relevant. My point is, you can't reference "special assemby" without detailing what that means.
Quoted 7 lines
You have a safety requirement in
from_raw:/// * For the duration of `'data`, the buffers backing this IO vector must be valid for /// reading.This should probably be promoted to invariant for the type, since
from_rawis the only way to construct the type?
Sure, let's get the wording consistent, but that was the purpose of this line in the invariants:
For the duration of 'data, it must be safe to read the data in this IO vector.
Quoted 3 lines
But are you saying that the referenced buffers need not be mapped and readable while this type exists? The mapping happens as part of
bindings::_copy_to_iter?
Ultimately, it's an implementation detail.
In our "# Invariants" section, we tend to "expand" the underlying C types and describe exactly what it means for that C type to be valid, even if those details are implementation details that nobody outside that C file should think about. Usually that's fine, but in this case, I don't think it is feasible.
The iov_iter type is like a giant enum with a bunch of different implementations. Some implementations just read from a simple kernel buffer that must, of course, be mapped. Some implementations traverse complex data structures and stitch the data together from multiple buffers. Other implementations map the data into memory on-demand inside the copy_from_iter call, without requiring it to be mapped at other times. And finally, some implementations perform IO by reading from userspace, in which case it's valid for the userspace pointer to be *literally any 64-bit integer*. If the address is dangling, that's caught inside the call to copy_from_iter and is not a safety issue.
I just want the type invariant to say that reading from it is valid, as long as the read happens before a certain lifetime expires, without elaborating on precisely what that means.
Quoted 12 lines
Quoted 12 lines
Quoted 11 lines
Quoted 7 lines
* If the
iov_iterreads from a kernel buffer, then the creator of theiov_itermust provide an initialized buffer.Ultimately, if we don't know that the bytes are initialized, then it's impossible to use the API correctly because you can never inspect the bytes in any way. I.e., any implementation of
copy_from_iterthat produces uninit data is necessarily buggy.I would agree. How do we fix that? You are more knowledgeable than me in this field, so you probably have a better shot than me, at finding a solution.
Quoted 7 lines
Quoted 4 lines
I think there is nothing to fix. If there exists a callsite on the C side that creates an
iov_iterthat reads from an uninitialized kernel buffer, then we can fix that specific call-site. I don't think anything else needs to be done.If soundness of this code hinges on specific call site behavior, this should be a safety requirement.
Yes, when we add Rust constructors for this type, they will need appropriate soundness checks or safety requirements to verify that the provided buffer is valid for the chosen iter_type.
For now, it is constructed in C and we usually don't have safety comments in C code.
Quoted 33 lines
Quoted 9 lines
Quoted 8 lines
As far as I can tell, we need to read from a place unknown to the rust abstract machine, and we need to be able to have the abstract machine consider the data initialized after the read.
Is this volatile memcpy [1], or would that only solve the data race problem, not uninitialized data problem?
Quoted 7 lines
Volatile memcpy deals with data races.
In general, we can argue all we want about wording of these safety comments, but calling
copy_from_iteris the right way to read from aniov_iter. If there is a problem, the problem is specific call-sites that construct aniov_iterwith an uninit buffer. I don't know whether such call-sites exist.I am not saying it is the wrong way. I am asking that we detail in the safety requirements _why_ it is the right way.
You have a type invariant
For the duration of
'data, it must be safe to read the data in this IO vector.that says "safe to read" instead of "valid for read" because "special assembly" is used to read the data, and that somehow makes it OK. We should be more specific.
How about making the invariant:
For the duration of
'data, it must be safe to read the data in this IO vector with the C API_copy_from_iter.
Quoted 8 lines
And then your safety comment regarding uninit bytes can be:
We write
outwithcopy_from_iter_raw, which transitively writesoutusing_copy_from_iter. By C API contract,_copy_from_iterdoes not write uninitialized bytes toout.In this way we can defer to the implementation of
_copy_from_user, which is what I think you want?
Yes, this is pretty much what I want except that _copy_from_user isn't the only C function you could call to read from an iov_iter.
Alice
Andreas Hindborg
reviewerRe: [PATCH v2 1/4] rust: iov: add iov_iter abstractions for ITER_SOURCE
Quoted 8 lines · Alice Ryhl
Quoted 7 lines · Andreas Hindborg
Quoted 5 lines · Alice Ryhl
Quoted 4 lines · Andreas Hindborg
Quoted 2 lines · Alice Ryhl
Quoted 1 line · Andreas Hindborg
"Alice Ryhl" <[email protected]> writes:
Quoted 10 lines
The
iov_itertype is like a giant enum with a bunch of different implementations. Some implementations just read from a simple kernel buffer that must, of course, be mapped. Some implementations traverse complex data structures and stitch the data together from multiple buffers. Other implementations map the data into memory on-demand inside thecopy_from_itercall, without requiring it to be mapped at other times. And finally, some implementations perform IO by reading from userspace, in which case it's valid for the userspace pointer to be *literally any 64-bit integer*. If the address is dangling, that's caught inside the call tocopy_from_iterand is not a safety issue.
At any rate, this is a very informative paragraph. It would be great if you could have this information in the documentation for this type.
Best regards,
Andreas Hindborg
Alice Ryhl
reviewerPATCH v2 2/4 rust: iov: add iov_iter abstractions for ITER_DEST
In reply to Alice RyhlThis adds abstractions for the iov_iter type in the case where data_source is ITER_DEST. This will make Rust implementations of fops->read_iter possible.
This series only has support for using existing IO vectors created by C code. Additional abstractions will be needed to support the creation of IO vectors in Rust code.
These abstractions make the assumption that struct iov_iter does not have internal self-references, which implies that it is valid to move it between different local variables.
--- rust/kernel/iov.rs | 137 +++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 137 insertions(+)
rust/kernel/iov.rs +137 -0
@@ -16,6 +16,15 @@use core::{marker::PhantomData , mem::MaybeUninit , slice};const ITER_SOURCE :bool = bindings::ITER_SOURCE !=0 ; +const ITER_DEST :bool = bindings::ITER_DEST !=0 ; + +// Compile-time assertion for the above constants. +const _: () = { +build_assert! ( +ITER_SOURCE !=ITER_DEST , +"ITER_DEST and ITER_SOURCE should be different." + ); +};/// An IO vector that acts as a source of data. ///
@@ -150,3 +159,131 @@ pub fn copy_from_iter_raw(&mut self, out: &mut [MaybeUninit<u8>]) -> &mut [u8] {unsafe { slice::from_raw_parts_mut (out.as_mut_ptr ().cast (), len) } } } + +/// An IO vector that acts as a destination for data. +/// +/// IO vectors support many different types of destinations. This includes both buffers in +/// kernel-space and writing to userspace. It's possible that the destination buffer is mapped in a +/// thread-local manner using e.g. `kmap_local_page()`, so this type is not `Send` to ensure that +/// the mapping is written to the right context in that scenario. +/// +/// # Invariants +/// +/// Must hold a valid `struct iov_iter` with `data_source` set to `ITER_DEST`. For the duration +/// of `'data`, it must be safe to write data to this IO vector. +#[repr(transparent)] +pub struct IovIterDest <'data> { + iov:Opaque <bindings::iov_iter >, +/// Represent to the type system that this value contains a pointer to writable data it does +/// not own. + _source:PhantomData <&'datamut [u8 ]>, +} + +impl <'data>IovIterDest <'data> { +/// Obtain an `IovIterDest` from a raw pointer. +/// +/// # Safety +/// +/// * For the duration of `'iov`, the `struct iov_iter` must remain valid and must not be +/// accessed except through the returned reference. +/// * For the duration of `'data`, the buffers backing this IO vector must be valid for +/// writing. +#[track_caller] +#[inline] +pub unsafe fn from_raw <'iov>(ptr: *mut bindings::iov_iter ) -> &'iovmut IovIterDest <'data> { +// SAFETY: The caller ensures that `ptr` is valid. +let data_source =unsafe { (*ptr).data_source }; +assert_eq! (data_source,ITER_DEST ); + +// SAFETY: The caller ensures the struct invariants for the right durations. +unsafe { &mut *ptr.cast ::<IovIterDest <'data>>() } + } + +/// Access this as a raw `struct iov_iter`. +#[inline] +pub fn as_raw (&mut self) -> *mut bindings::iov_iter { + self.iov.get () + } + +/// Returns the number of bytes available in this IO vector. +/// +/// Note that this may overestimate the number of bytes. For example, reading from userspace +/// memory could fail with EFAULT, which will be treated as the end of the IO vector. +#[inline] +pub fn len (&self) ->usize { +// SAFETY: It is safe to access the `count` field. +unsafe { + (*self.iov.get ()) + .__bindgen_anon_1 + .__bindgen_anon_1 + .as_ref () + .count + } + } + +/// Returns whether there are any bytes left in this IO vector. +/// +/// This may return `true` even if there are no more bytes available. For example, reading from +/// userspace memory could fail with EFAULT, which will be treated as the end of the IO vector. +#[inline] +pub fn is_empty (&self) ->bool { + self.len () ==0 + } + +/// Advance this IO vector by `bytes` bytes. +/// +/// If `bytes` is larger than the size of this IO vector, it is advanced to the end. +#[inline] +pub fn advance (&mut self, bytes:usize ) { +// SAFETY: `self.iov` is a valid IO vector. +unsafe { bindings::iov_iter_advance (self.as_raw (), bytes) }; + } + +/// Advance this IO vector backwards by `bytes` bytes. +/// +/// # Safety +/// +/// The IO vector must not be reverted to before its beginning. +#[inline] +pub unsafe fn revert (&mut self, bytes:usize ) { +// SAFETY: `self.iov` is a valid IO vector, and `bytes` is in bounds. +unsafe { bindings::iov_iter_revert (self.as_raw (), bytes) }; + } + +/// Write data to this IO vector. +/// +/// Returns the number of bytes that were written. If this is shorter than the provided slice, +/// then no more bytes can be written. +#[inline] +pub fn copy_to_iter (&mut self, input: &[u8 ]) ->usize { +// SAFETY: `input` is valid for `input.len()` bytes. +unsafe { bindings::_copy_to_iter (input.as_ptr ().cast (), input.len (), self.as_raw ()) } + } + +/// Utility for implementing `read_iter` given the full contents of the file. +/// +/// The full contents of the file being read from is represented by `contents`. This call will +/// write the appropriate sub-slice of `contents` and update the file position in `ppos` so +/// that the file will appear to contain `contents` even if takes multiple reads to read the +/// entire file. +#[inline] +pub fn simple_read_from_buffer (&mut self, ppos: &mut i64 , contents: &[u8 ]) ->Result <usize > { +if *ppos <0 { +return Err (EINVAL ); + } +let Ok (pos) = usize::try_from (*ppos)else { +return Ok (0 ); + }; +if pos >= contents.len () { +return Ok (0 ); + } + +// BOUNDS: We just checked that `pos < contents.len()` above. +let num_written = self.copy_to_iter (&contents[pos..]); + +// OVERFLOW: pos+num_written <= contents.len() <= isize::MAX <= i64::MAX + *ppos = (pos + num_written)as i64 ; + +Ok (num_written) + } +}
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
Andreas Hindborg
reviewerRe: [PATCH v2 2/4] rust: iov: add iov_iter abstractions for ITER_DEST
Quoted 8 lines · Alice Ryhl
This adds abstractions for the
iov_itertype in the case wheredata_sourceisITER_DEST. This will make Rust implementations offops->read_iterpossible.This series only has support for using existing IO vectors created by C code. Additional abstractions will be needed to support the creation of IO vectors in Rust code.
Quoted 5 lines
These abstractions make the assumption that
struct iov_iterdoes not have internal self-references, which implies that it is valid to move it between different local variables.Signed-off-by Alice Ryhl <[email protected]>
Please see the comments on v1 [1].
Best regards,
Andreas Hindborg
Alice Ryhl
reviewerRe: [PATCH v2 2/4] rust: iov: add iov_iter abstractions for ITER_DEST
Quoted 16 lines · Andreas Hindborg
Quoted 8 lines · Alice Ryhl
This adds abstractions for the
iov_itertype in the case wheredata_sourceisITER_DEST. This will make Rust implementations offops->read_iterpossible.This series only has support for using existing IO vectors created by C code. Additional abstractions will be needed to support the creation of IO vectors in Rust code.
Quoted 5 lines
These abstractions make the assumption that
struct iov_iterdoes not have internal self-references, which implies that it is valid to move it between different local variables.Signed-off-by Alice Ryhl <[email protected]>
Quoted 7 lines
Please see the comments on v1 [1].
Best regards,
Andreas Hindborg
Sorry I should probably have articulated this somewhere. After I looked at this code with fresh eyes, I realized that the only things the two types have in common are len, advance, and revert. I didn't think that attempting to deduplicate the logic was worth it.
Alice
Alice Ryhl
reviewerPATCH v2 3/4 rust: miscdevice: Provide additional abstractions for iov_iter and kiocb structures
In reply to Alice RyhlThese will be used for the read_iter() and write_iter() callbacks, which are now the preferred back-ends for when a user operates on a char device with read() and write() respectively.
--- rust/kernel/miscdevice.rs | 97 ++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 96 insertions(+), 1 deletion(-)
rust/kernel/miscdevice.rs +96 -1
@@ -14,13 +14,14 @@ error::{to_result, Error, Result, VTABLE_DEFAULT_ERROR}, ffi::{c_int, c_long, c_uint, c_ulong}, fs::File, + iov::{IovIterDest, IovIterSource}, mm::virt::VmaNew, prelude::*, seq_file::SeqFile, str::CStr, types::{ForeignOwnable, Opaque}, }; -use core::{marker::PhantomData , mem::MaybeUninit , pin::Pin }; +use core::{marker::PhantomData , mem::MaybeUninit , pin::Pin , ptr::NonNull };/// Options for creating a misc device. #[derive( Copy , Clone )]
@@ -136,6 +137,16 @@ fn mmap(build_error! (VTABLE_DEFAULT_ERROR ) } +/// Read from this miscdevice. +fn read_iter (_kiocb:Kiocb <'_,Self ::Ptr >, _iov: &mut IovIterDest <'_>) ->Result <usize > { +build_error! (VTABLE_DEFAULT_ERROR ) + } + +/// Write to this miscdevice. +fn write_iter (_kiocb:Kiocb <'_,Self ::Ptr >, _iov: &mut IovIterSource <'_>) ->Result <usize > { +build_error! (VTABLE_DEFAULT_ERROR ) + } +/// Handler for ioctls. /// /// The `cmd` argument is usually manipulated using the utilities in [`kernel::ioctl`].
@@ -177,6 +188,36 @@ fn show_fdinfo( } } +/// Wrapper for the kernel's `struct kiocb`. +/// +/// The type `T` represents the private data of the file. +pub struct Kiocb <'a,T > { + inner:NonNull <bindings::kiocb >, + _phantom:PhantomData <&'aT >, +} + +impl <'a,T :ForeignOwnable >Kiocb <'a,T > { +/// Get the Rust data that represents this misc device. +pub fn device (&self) -> <T as ForeignOwnable >::Borrowed <'a> { +// SAFETY: The `kiocb` lets us access the private data. +let private =unsafe { (*(*self.inner.as_ptr ()).ki_filp).private_data }; +// SAFETY: The kiocb has shared access to the private data. +unsafe { <T as ForeignOwnable >::borrow (private) } + } + +/// Gets the current value of `ki_pos`. +pub fn ki_pos (&self) ->i64 { +// SAFETY: The `kiocb` can access `ki_pos`. +unsafe { (*self.inner.as_ptr ()).ki_pos } + } + +/// Gets a mutable reference to the `ki_pos` field. +pub fn ki_pos_mut (&mut self) -> &mut i64 { +// SAFETY: The `kiocb` can access `ki_pos`. +unsafe { &mut (*self.inner.as_ptr ()).ki_pos } + } +} +/// A vtable for the file operations of a Rust miscdevice. struct MiscdeviceVTable <T :MiscDevice >(PhantomData <T >);
@@ -240,6 +281,50 @@ impl<T: MiscDevice> MiscdeviceVTable<T> {0 } +/// # Safety +/// +/// `kiocb` must be correspond to a valid file that is associated with a +/// `MiscDeviceRegistration<T>`. `iter` must be a valid `struct iov_iter` for writing. +unsafe extern "C" fn read_iter ( + kiocb: *mut bindings::kiocb , + iter: *mut bindings::iov_iter , + ) ->isize { +let kiocb =Kiocb { +// SAFETY: The read_iter call of a file is given a kiocb for that file. + inner:unsafe {NonNull ::new_unchecked (kiocb) }, + _phantom:PhantomData , + }; +// SAFETY: This is a valid `struct iov_iter` for writing. +let iov =unsafe {IovIterDest ::from_raw (iter) }; + +match T ::read_iter (kiocb, iov) { +Ok (res) => resas isize , +Err (err) => err.to_errno ()as isize , + } + } + +/// # Safety +/// +/// `kiocb` must be correspond to a valid file that is associated with a +/// `MiscDeviceRegistration<T>`. `iter` must be a valid `struct iov_iter` for writing. +unsafe extern "C" fn write_iter ( + kiocb: *mut bindings::kiocb , + iter: *mut bindings::iov_iter , + ) ->isize { +let kiocb =Kiocb { +// SAFETY: The read_iter call of a file is given a kiocb for that file. + inner:unsafe {NonNull ::new_unchecked (kiocb) }, + _phantom:PhantomData , + }; +// SAFETY: This is a valid `struct iov_iter` for reading. +let iov =unsafe {IovIterSource ::from_raw (iter) }; + +match T ::write_iter (kiocb, iov) { +Ok (res) => resas isize , +Err (err) => err.to_errno ()as isize , + } + } +/// # Safety /// /// `file` must be a valid file that is associated with a `MiscDeviceRegistration<T>`.
@@ -336,6 +421,16 @@ impl<T: MiscDevice> MiscdeviceVTable<T> { open: Some(Self::open), release: Some(Self::release), mmap: if T::HAS_MMAP { Some(Self::mmap) } else { None }, + read_iter: if T::HAS_READ_ITER { +Some (Self ::read_iter) + }else { +None + }, + write_iter: if T::HAS_WRITE_ITER { +Some (Self ::write_iter) + }else { +None + }, unlocked_ioctl: if T::HAS_IOCTL {Some (Self ::ioctl) }else {
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
Andreas Hindborg
reviewerRe: [PATCH v2 3/4] rust: miscdevice: Provide additional abstractions for iov_iter and kiocb structures
Quoted 49 lines · Alice Ryhl
These will be used for the
read_iter()andwrite_iter()callbacks, which are now the preferred back-ends for when a user operates on a char device withread()andwrite()respectively.Co-developed-by Lee Jones <[email protected]>Signed-off-by Lee Jones <[email protected]>Signed-off-by Alice Ryhl <[email protected]>--- rust/kernel/miscdevice.rs | 97 ++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 96 insertions(+), 1 deletion(-)rust/kernel/miscdevice.rs +12
-1@@ -14,13 +14,14 @@ error::{to_result, Error, Result, VTABLE_DEFAULT_ERROR}, ffi::{c_int, c_long, c_uint, c_ulong}, fs::File, + iov::{IovIterDest, IovIterSource}, mm::virt::VmaNew, prelude::*, seq_file::SeqFile, str::CStr, types::{ForeignOwnable, Opaque}, }; -use core::{marker::PhantomData , mem::MaybeUninit , pin::Pin }; +use core::{marker::PhantomData , mem::MaybeUninit , pin::Pin , ptr::NonNull };/// Options for creating a misc device. #[derive( Copy , Clone )] @@ -136,6 +137,16 @@ fn mmap(build_error! (VTABLE_DEFAULT_ERROR ) } +/// Read from this miscdevice. +fn read_iter (_kiocb:Kiocb <'_,Self ::Ptr >, _iov: &mut IovIterDest <'_>) ->Result <usize > { +build_error! (VTABLE_DEFAULT_ERROR ) + } + +/// Write to this miscdevice. +fn write_iter (_kiocb:Kiocb <'_,Self ::Ptr >, _iov: &mut IovIterSource <'_>) ->Result <usize > { +build_error! (VTABLE_DEFAULT_ERROR ) + } +/// Handler for ioctls. /// /// The `cmd` argument is usually manipulated using the utilities in [`kernel::ioctl`].
rust/kernel/miscdevice.rs fn show_fdinfo( · line 193
Quoted 6 lines
rust/kernel/miscdevice.rs +3
-0@@ -177,6 +188,36 @@ fn show_fdinfo( } } +/// Wrapper for the kernel's `struct kiocb`. +/// +/// The type `T` represents the private data of the file.
Could you give more context? Please describe the purpose for the type and intended use. Perhaps give an example that can be compile tested.
Best regards,
Andreas Hindborg
Alice Ryhl
reviewerRe: [PATCH v2 3/4] rust: miscdevice: Provide additional abstractions for iov_iter and kiocb structures
Quoted 7 lines · Andreas Hindborg
rust/kernel/miscdevice.rs fn show_fdinfo( · line 193
Quoted 3 lines · Alice Ryhl
rust/kernel/miscdevice.rs +3
-0+/// Wrapper for the kernel's `struct kiocb`. +/// +/// The type `T` represents the private data of the file. Could you give more context? Please describe the purpose for the type and intended use. Perhaps give an example that can be compile tested.
Right now, it's basically a (&mut i64, &T) tuple providing access to the file position and private data. That's it.
Alice
Andreas Hindborg
reviewerRe: [PATCH v2 3/4] rust: miscdevice: Provide additional abstractions for iov_iter and kiocb structures
Quoted 11 lines · Alice Ryhl
Quoted 7 lines · Andreas Hindborg
rust/kernel/miscdevice.rs fn show_fdinfo( · line 193
Quoted 3 lines · Alice Ryhl
rust/kernel/miscdevice.rs +3
-0+/// Wrapper for the kernel's `struct kiocb`. +/// +/// The type `T` represents the private data of the file. Could you give more context? Please describe the purpose for the type and intended use. Perhaps give an example that can be compile tested.
Right now, it's basically a
(&mut i64, &T)tuple providing access to the file position and private data. That's it.
Great! Let's get something along those lines in the docs.
Best regards,
Andreas Hindborg
Andreas Hindborg
reviewerRe: [PATCH v2 3/4] rust: miscdevice: Provide additional abstractions for iov_iter and kiocb structures
In reply to Alice RyhlQuoted 49 lines · Alice Ryhl
These will be used for the
read_iter()andwrite_iter()callbacks, which are now the preferred back-ends for when a user operates on a char device withread()andwrite()respectively.Co-developed-by Lee Jones <[email protected]>Signed-off-by Lee Jones <[email protected]>Signed-off-by Alice Ryhl <[email protected]>--- rust/kernel/miscdevice.rs | 97 ++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 96 insertions(+), 1 deletion(-)rust/kernel/miscdevice.rs +12
-1@@ -14,13 +14,14 @@ error::{to_result, Error, Result, VTABLE_DEFAULT_ERROR}, ffi::{c_int, c_long, c_uint, c_ulong}, fs::File, + iov::{IovIterDest, IovIterSource}, mm::virt::VmaNew, prelude::*, seq_file::SeqFile, str::CStr, types::{ForeignOwnable, Opaque}, }; -use core::{marker::PhantomData , mem::MaybeUninit , pin::Pin }; +use core::{marker::PhantomData , mem::MaybeUninit , pin::Pin , ptr::NonNull };/// Options for creating a misc device. #[derive( Copy , Clone )] @@ -136,6 +137,16 @@ fn mmap(build_error! (VTABLE_DEFAULT_ERROR ) } +/// Read from this miscdevice. +fn read_iter (_kiocb:Kiocb <'_,Self ::Ptr >, _iov: &mut IovIterDest <'_>) ->Result <usize > { +build_error! (VTABLE_DEFAULT_ERROR ) + } + +/// Write to this miscdevice. +fn write_iter (_kiocb:Kiocb <'_,Self ::Ptr >, _iov: &mut IovIterSource <'_>) ->Result <usize > { +build_error! (VTABLE_DEFAULT_ERROR ) + } +/// Handler for ioctls. /// /// The `cmd` argument is usually manipulated using the utilities in [`kernel::ioctl`].
rust/kernel/miscdevice.rs fn show_fdinfo( · line 197
Quoted 10 lines
rust/kernel/miscdevice.rs +7
-0@@ -177,6 +188,36 @@ fn show_fdinfo( } } +/// Wrapper for the kernel's `struct kiocb`. +/// +/// The type `T` represents the private data of the file. +pub struct Kiocb <'a,T > { + inner:NonNull <bindings::kiocb >, + _phantom:PhantomData <&'aT >, +}
Also, kiocb is not miscdevice specific. It should probably not live here.
Best regards,
Andreas Hindborg
Alice Ryhl
reviewerRe: [PATCH v2 3/4] rust: miscdevice: Provide additional abstractions for iov_iter and kiocb structures
Quoted 54 lines · Andreas Hindborg
Quoted 52 lines · Alice Ryhl
These will be used for the
read_iter()andwrite_iter()callbacks, which are now the preferred back-ends for when a user operates on a char device withread()andwrite()respectively.Co-developed-by Lee Jones <[email protected]>Signed-off-by Lee Jones <[email protected]>Signed-off-by Alice Ryhl <[email protected]>--- rust/kernel/miscdevice.rs | 97 ++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 96 insertions(+), 1 deletion(-)rust/kernel/miscdevice.rs +12
-1@@ -14,13 +14,14 @@ error::{to_result, Error, Result, VTABLE_DEFAULT_ERROR}, ffi::{c_int, c_long, c_uint, c_ulong}, fs::File, + iov::{IovIterDest, IovIterSource}, mm::virt::VmaNew, prelude::*, seq_file::SeqFile, str::CStr, types::{ForeignOwnable, Opaque}, }; -use core::{marker::PhantomData , mem::MaybeUninit , pin::Pin }; +use core::{marker::PhantomData , mem::MaybeUninit , pin::Pin , ptr::NonNull };/// Options for creating a misc device. #[derive( Copy , Clone )] @@ -136,6 +137,16 @@ fn mmap(build_error! (VTABLE_DEFAULT_ERROR ) } +/// Read from this miscdevice. +fn read_iter (_kiocb:Kiocb <'_,Self ::Ptr >, _iov: &mut IovIterDest <'_>) ->Result <usize > { +build_error! (VTABLE_DEFAULT_ERROR ) + } + +/// Write to this miscdevice. +fn write_iter (_kiocb:Kiocb <'_,Self ::Ptr >, _iov: &mut IovIterSource <'_>) ->Result <usize > { +build_error! (VTABLE_DEFAULT_ERROR ) + } +/// Handler for ioctls. /// /// The `cmd` argument is usually manipulated using the utilities in [`kernel::ioctl`]. @@ -177,6 +188,36 @@ fn show_fdinfo( } }
Quoted 9 lines
rust/kernel/miscdevice.rs fn show_fdinfo( · line 197
Quoted 7 lines
rust/kernel/miscdevice.rs +7
-0+/// Wrapper for the kernel's `struct kiocb`. +/// +/// The type `T` represents the private data of the file. +pub struct Kiocb <'a,T > { + inner:NonNull <bindings::kiocb >, + _phantom:PhantomData <&'aT >, +}Also,
kiocbis not miscdevice specific. It should probably not live here.
I can place it in rust/kernel/fs.rs, but this is an instance of the more general fact that miscdevice defines many things from file_operations that we should probably generalize in the future.
Alice
Andreas Hindborg
reviewerRe: [PATCH v2 3/4] rust: miscdevice: Provide additional abstractions for iov_iter and kiocb structures
Quoted 65 lines · Alice Ryhl
Quoted 54 lines · Andreas Hindborg
Quoted 52 lines · Alice Ryhl
These will be used for the
read_iter()andwrite_iter()callbacks, which are now the preferred back-ends for when a user operates on a char device withread()andwrite()respectively.Co-developed-by Lee Jones <[email protected]>Signed-off-by Lee Jones <[email protected]>Signed-off-by Alice Ryhl <[email protected]>--- rust/kernel/miscdevice.rs | 97 ++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 96 insertions(+), 1 deletion(-)rust/kernel/miscdevice.rs +12
-1@@ -14,13 +14,14 @@ error::{to_result, Error, Result, VTABLE_DEFAULT_ERROR}, ffi::{c_int, c_long, c_uint, c_ulong}, fs::File, + iov::{IovIterDest, IovIterSource}, mm::virt::VmaNew, prelude::*, seq_file::SeqFile, str::CStr, types::{ForeignOwnable, Opaque}, }; -use core::{marker::PhantomData , mem::MaybeUninit , pin::Pin }; +use core::{marker::PhantomData , mem::MaybeUninit , pin::Pin , ptr::NonNull };/// Options for creating a misc device. #[derive( Copy , Clone )] @@ -136,6 +137,16 @@ fn mmap(build_error! (VTABLE_DEFAULT_ERROR ) } +/// Read from this miscdevice. +fn read_iter (_kiocb:Kiocb <'_,Self ::Ptr >, _iov: &mut IovIterDest <'_>) ->Result <usize > { +build_error! (VTABLE_DEFAULT_ERROR ) + } + +/// Write to this miscdevice. +fn write_iter (_kiocb:Kiocb <'_,Self ::Ptr >, _iov: &mut IovIterSource <'_>) ->Result <usize > { +build_error! (VTABLE_DEFAULT_ERROR ) + } +/// Handler for ioctls. /// /// The `cmd` argument is usually manipulated using the utilities in [`kernel::ioctl`]. @@ -177,6 +188,36 @@ fn show_fdinfo( } }Quoted 9 lines
rust/kernel/miscdevice.rs fn show_fdinfo( · line 197
Quoted 7 lines
rust/kernel/miscdevice.rs +7
-0+/// Wrapper for the kernel's `struct kiocb`. +/// +/// The type `T` represents the private data of the file. +pub struct Kiocb <'a,T > { + inner:NonNull <bindings::kiocb >, + _phantom:PhantomData <&'aT >, +}Also,
kiocbis not miscdevice specific. It should probably not live here.
Quoted 3 lines
I can place it in
rust/kernel/fs.rs, but this is an instance of the more general fact that miscdevice defines many things fromfile_operationsthat we should probably generalize in the future.
I support moving Kiocb to fs now, rather than later.
Best regards,
Andreas Hindborg
Alice Ryhl
reviewerPATCH v2 4/4 samples: rust_misc_device: Expand the sample to support read()ing from userspace
In reply to Alice RyhlFrom: Lee Jones <[email protected]>
A userland application can now operate on the char device with read() in order to consume a locally held buffer. Memory for the buffer is to be provisioned and the buffer populated in its subsequently provided write() counterpart.
--- samples/rust/rust_misc_device.rs | 36 ++++++++++++++++++++++++++++++++++-- 1 file changed, 34 insertions(+), 2 deletions(-)
samples/rust/rust_misc_device.rs +34 -2
@@ -102,7 +102,8 @@ device::Device, fs::File, ioctl::{_IO, _IOC_SIZE, _IOR, _IOW}, - miscdevice::{MiscDevice, MiscDeviceOptions, MiscDeviceRegistration}, + iov::{IovIterDest, IovIterSource}, + miscdevice::{Kiocb, MiscDevice, MiscDeviceOptions, MiscDeviceRegistration}, new_mutex, prelude::*, sync::Mutex,
@@ -144,6 +145,7 @@ fn init(_module: &'static ThisModule) -> impl PinInit<Self, Error> {struct Inner { value:i32 , + buffer:KVVec <u8 >, }#[pin_data( PinnedDrop )]
@@ -165,7 +167,10 @@ fn open(_file: &File, misc: &MiscDeviceRegistration<Self>) -> Result<Pin<KBox<SeKBox ::try_pin_init (try_pin_init! {RustMiscDevice { - inner <-new_mutex! (Inner { value:0_i32 } ), + inner <-new_mutex! (Inner { + value:0_i32 , + buffer:KVVec ::new (), + }), dev: dev, } },
@@ -173,6 +178,33 @@ fn open(_file: &File, misc: &MiscDeviceRegistration<Self>) -> Result<Pin<KBox<Se ) } +fn read_iter (mut kiocb:Kiocb <'_,Self ::Ptr >, iov: &mut IovIterDest <'_>) ->Result <usize > { +let me = kiocb.device (); +dev_info! (me.dev,"Reading from Rust Misc Device Sample\n" ); + +let inner = me.inner.lock (); +// Read the buffer contents, taking the file position into account. +let read = iov.simple_read_from_buffer (kiocb.ki_pos_mut (), &inner.buffer)?; + +Ok (read) + } + +fn write_iter (mut kiocb:Kiocb <'_,Self ::Ptr >, iov: &mut IovIterSource <'_>) ->Result <usize > { +let me = kiocb.device (); +dev_info! (me.dev,"Writing to Rust Misc Device Sample\n" ); + +let mut inner = me.inner.lock (); + +// Replace buffer contents. + inner.buffer.clear (); +let len = iov.copy_from_iter_vec (&mut inner.buffer,GFP_KERNEL )?; + +// Set position to zero so that future `read` calls will see the new contents. + *kiocb.ki_pos_mut () =0 ; + +Ok (len) + } +fn ioctl (me:Pin <&RustMiscDevice >, _file: &File , cmd:u32 , arg:usize ) ->Result <isize > {dev_info! (me.dev,"IOCTLing Rust Misc Device Sample\n" );
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog