linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Danilo Krummrich" <dakr@kernel.org>
To: "Daniel Almeida" <daniel.almeida@collabora.com>
Cc: <abdiel.janulgue@gmail.com>, <robin.murphy@arm.com>,
	<a.hindborg@kernel.org>, <ojeda@kernel.org>,
	<alex.gaynor@gmail.com>, <boqun.feng@gmail.com>,
	<gary@garyguo.net>, <bjorn3_gh@protonmail.com>,
	<lossin@kernel.org>, <aliceryhl@google.com>, <tmgross@umich.edu>,
	<bhelgaas@google.com>, <kwilczynski@kernel.org>,
	<gregkh@linuxfoundation.org>, <rafael@kernel.org>,
	<rust-for-linux@vger.kernel.org>, <linux-pci@vger.kernel.org>,
	<linux-kernel@vger.kernel.org>
Subject: Re: [PATCH v2 2/5] rust: dma: add DMA addressing capabilities
Date: Wed, 16 Jul 2025 23:13:11 +0200	[thread overview]
Message-ID: <DBDSFO1LGSYM.VFQKKLN6BX3H@kernel.org> (raw)
In-Reply-To: <DBDP0BJW9VAZ.5KRU4V4288R8@kernel.org>

On Wed Jul 16, 2025 at 8:32 PM CEST, Danilo Krummrich wrote:
> On Wed Jul 16, 2025 at 7:55 PM CEST, Danilo Krummrich wrote:
>> On Wed Jul 16, 2025 at 7:32 PM CEST, Daniel Almeida wrote:
>>> Hi Danilo,
>>>
>>>> +    #[inline]
>>>> +    pub const fn new(n: usize) -> Result<Self> {
>>>> +        Ok(Self(match n {
>>>> +            0 => 0,
>>>> +            1..=64 => u64::MAX >> (64 - n),
>>>> +            _ => return Err(EINVAL),
>>>> +        }))
>>>> +    }
>>>> +
>>>
>>> Isn’t this equivalent to genmask_u64(0..=n) ? See [0].
>>
>> Instead of the match this can use genmask_checked_u64() and convert the Option
>> to a Result, once genmask is upstream.
>>
>>> You should also get a compile-time failure if n is out of bounds by default using
>>> genmask.
>>
>> No, we can't use genmask_u64(), `n` is not guaranteed to be known at compile
>> time, so we'd need to use genmask_checked_u64().
>>
>> Of course, we could have a separate DmaMask constructor, e.g. with a const
>> generic -- not sure that's worth though.
>
> On the other hand, it doesn't hurt. Guess I will add another constructor with a
> const generic. :)

diff --git a/rust/kernel/dma.rs b/rust/kernel/dma.rs
index afd3ba538e3c..ad69ef316295 100644
--- a/rust/kernel/dma.rs
+++ b/rust/kernel/dma.rs
@@ -81,25 +81,6 @@ unsafe fn dma_set_mask_and_coherent(&self, mask: DmaMask) -> Result {
 /// never exceed the bit width of `u64`.
 ///
 /// This is the Rust equivalent of the C macro `DMA_BIT_MASK()`.
-///
-/// # Examples
-///
-/// ```
-/// use kernel::dma::DmaMask;
-///
-/// let mask0 = DmaMask::new(0)?;
-/// assert_eq!(mask0.value(), 0);
-///
-/// let mask1 = DmaMask::new(1)?;
-/// assert_eq!(mask1.value(), 0b1);
-///
-/// let mask64 = DmaMask::new(64)?;
-/// assert_eq!(mask64.value(), u64::MAX);
-///
-/// let mask_overflow = DmaMask::new(100);
-/// assert!(mask_overflow.is_err());
-/// # Ok::<(), Error>(())
-/// ```
 #[derive(Debug, Clone, Copy, PartialEq, Eq)]
 pub struct DmaMask(u64);

@@ -108,8 +89,27 @@ impl DmaMask {
     ///
     /// For `n <= 64`, sets exactly the lowest `n` bits.
     /// For `n > 64`, returns [`EINVAL`].
+    ///
+    /// # Examples
+    ///
+    /// ```
+    /// use kernel::dma::DmaMask;
+    ///
+    /// let mask0 = DmaMask::try_new(0)?;
+    /// assert_eq!(mask0.value(), 0);
+    ///
+    /// let mask1 = DmaMask::try_new(1)?;
+    /// assert_eq!(mask1.value(), 0b1);
+    ///
+    /// let mask64 = DmaMask::try_new(64)?;
+    /// assert_eq!(mask64.value(), u64::MAX);
+    ///
+    /// let mask_overflow = DmaMask::try_new(100);
+    /// assert!(mask_overflow.is_err());
+    /// # Ok::<(), Error>(())
+    /// ```
     #[inline]
-    pub const fn new(n: usize) -> Result<Self> {
+    pub const fn try_new(n: u32) -> Result<Self> {
         Ok(Self(match n {
             0 => 0,
             1..=64 => u64::MAX >> (64 - n),
@@ -117,6 +117,38 @@ pub const fn new(n: usize) -> Result<Self> {
         }))
     }

+    /// Constructs a `DmaMask` with the lowest `n` bits set to `1`.
+    ///
+    /// For `n <= 64`, sets exactly the lowest `n` bits.
+    /// For `n > 64`, results in a build error.
+    ///
+    /// # Examples
+    ///
+    /// ```
+    /// use kernel::dma::DmaMask;
+    /// use kernel::bits::genmask_u64;
+    ///
+    /// let mask0 = DmaMask::new::<0>();
+    /// assert_eq!(mask0.value(), 0);
+    ///
+    /// let mask1 = DmaMask::new::<1>();
+    /// assert_eq!(mask1.value(), 0b1);
+    ///
+    /// let mask64 = DmaMask::new::<64>();
+    /// assert_eq!(mask64.value(), u64::MAX);
+    ///
+    /// // Build failure.
+    /// // let mask_overflow = DmaMask::new::<100>();
+    /// ```
+    #[inline(always)]
+    pub const fn new<const N: u32>() -> Self {
+        let Ok(mask) = Self::try_new(N) else {
+            build_error!("Invalid DMA Mask.");
+        };
+
+        mask
+    }
+
     /// Returns the underlying `u64` bitmask value.
     #[inline]
     pub const fn value(&self) -> u64 {
diff --git a/samples/rust/rust_dma.rs b/samples/rust/rust_dma.rs
index 9422ac68c139..c5e7cce68654 100644
--- a/samples/rust/rust_dma.rs
+++ b/samples/rust/rust_dma.rs
@@ -58,7 +58,7 @@ impl pci::Driver for DmaSampleDriver {
     fn probe(pdev: &pci::Device<Core>, _info: &Self::IdInfo) -> Result<Pin<KBox<Self>>> {
         dev_info!(pdev.as_ref(), "Probe DMA test driver.\n");

-        let mask = DmaMask::new(64)?;
+        let mask = DmaMask::new::<64>();

         // SAFETY: There are no concurrent calls to DMA allocation and mapping primitives.
         unsafe { pdev.dma_set_mask_and_coherent(mask)? };


  reply	other threads:[~2025-07-16 21:13 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-07-16 15:02 [PATCH v2 0/5] dma::Device trait and DMA mask Danilo Krummrich
2025-07-16 15:02 ` [PATCH v2 1/5] rust: dma: implement `dma::Device` trait Danilo Krummrich
2025-07-16 15:02 ` [PATCH v2 2/5] rust: dma: add DMA addressing capabilities Danilo Krummrich
2025-07-16 17:32   ` Daniel Almeida
2025-07-16 17:41     ` Daniel Almeida
2025-07-16 17:55     ` Danilo Krummrich
2025-07-16 18:32       ` Danilo Krummrich
2025-07-16 21:13         ` Danilo Krummrich [this message]
2025-07-16 22:19           ` Daniel Almeida
2025-07-16 22:32             ` Danilo Krummrich
2025-07-16 15:02 ` [PATCH v2 3/5] rust: pci: implement the `dma::Device` trait Danilo Krummrich
2025-07-16 15:02 ` [PATCH v2 4/5] rust: platform: " Danilo Krummrich
2025-07-16 15:02 ` [PATCH v2 5/5] rust: samples: dma: set DMA mask Danilo Krummrich
2025-07-16 17:28 ` [PATCH v2 0/5] dma::Device trait and " Greg KH
2025-07-16 17:34 ` Daniel Almeida
2025-07-20 14:33 ` Danilo Krummrich

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=DBDSFO1LGSYM.VFQKKLN6BX3H@kernel.org \
    --to=dakr@kernel.org \
    --cc=a.hindborg@kernel.org \
    --cc=abdiel.janulgue@gmail.com \
    --cc=alex.gaynor@gmail.com \
    --cc=aliceryhl@google.com \
    --cc=bhelgaas@google.com \
    --cc=bjorn3_gh@protonmail.com \
    --cc=boqun.feng@gmail.com \
    --cc=daniel.almeida@collabora.com \
    --cc=gary@garyguo.net \
    --cc=gregkh@linuxfoundation.org \
    --cc=kwilczynski@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=lossin@kernel.org \
    --cc=ojeda@kernel.org \
    --cc=rafael@kernel.org \
    --cc=robin.murphy@arm.com \
    --cc=rust-for-linux@vger.kernel.org \
    --cc=tmgross@umich.edu \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).