Message-ID: <d9b79f25-ae52-4cda-86d0-d4cde9338644@ryhl.io>
Date: Mon, 21 Oct 2024 21:49:17 +0200
From: Alice Ryhl <alice@...l.io>
To: John Hubbard <jhubbard@...dia.com>, Alice Ryhl <aliceryhl@...gle.com>
Cc: Miguel Ojeda <miguel.ojeda.sandonis@...il.com>,
 Miguel Ojeda <ojeda@...nel.org>, Andrew Morton <akpm@...ux-foundation.org>,
 "Matthew Wilcox (Oracle)" <willy@...radead.org>,
 Alex Gaynor <alex.gaynor@...il.com>, Boqun Feng <boqun.feng@...il.com>,
 Gary Guo <gary@...yguo.net>, Björn Roy Baron
 <bjorn3_gh@...tonmail.com>, Benno Lossin <benno.lossin@...ton.me>,
 Andreas Hindborg <a.hindborg@...nel.org>, Trevor Gross <tmgross@...ch.edu>,
 linux-mm@...ck.org, rust-for-linux@...r.kernel.org,
 linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2] rust: page: add Rust version of PAGE_ALIGN

On 10/21/24 9:34 PM, John Hubbard wrote:
> On 10/21/24 12:26 PM, Alice Ryhl wrote:
>> On Mon, Oct 21, 2024 at 9:09 PM John Hubbard <jhubbard@...dia.com> wrote:
>>>
>>> On 10/21/24 11:59 AM, Alice Ryhl wrote:
>>>> On 10/21/24 8:41 PM, John Hubbard wrote:
>>>>> On 10/21/24 11:37 AM, Miguel Ojeda wrote:
>>>>>> On Mon, Oct 21, 2024 at 8:35 PM John Hubbard <jhubbard@...dia.com> 
>>>>>> wrote:
>>>>>>>
>>>>>>> Is this another case of C and Rust using different words for 
>>>>>>> things??
>>>>>>> Wow. OK...
>>>>>>
>>>>>> I am not sure what you mean -- by BE I meant British English.
>>>>>>
>>>>>> See my other reply as well -- I just changed it anyway because Rust
>>>>>> apparently uses "parentheses".
>>>>>>
>>>>>
>>>>> Right. For spoken languages, that's simply preference, and I would not
>>>>> try to impose anything on anyone there.
>>>>>
>>>>> But in this case, at least for C (and, from reading my Rust book(s), I
>>>>> thought for Rust also), "parentheses" is a technical specification, 
>>>>> and
>>>>> we should prefer to be accurate:
>>>>>
>>>>>       parentheses: ()
>>>>>       brackets:    []
>>>>>
>>>>> Yes?
>>>> What word would you use to collectively talk about (), [], {}? In my 
>>>> native language they're all a kind of parenthesis.
>>>>
>>>
>>> Good question. I've never attempted that when discussing programming
>>> language details, because it hasn't come up, because it would be a
>>> programming error in C to use one in place of the other. And it is
>>> rare to refer to both cases in C.
>>>
>>> Rust so far seems to have the same distinction, although I am standing
>>> by to be corrected as necessary, there! :)
>>>
>>> At a higher level of abstraction, though, perhaps "grouping" is a good
>>> word.
>>
>> Rust macros can use different types of brackets. For example, the
>> `assert!(1 < 2)` macro uses round parenthesises, the `vec![1,2,3]`
>> macro uses square parenthesises, and the `thread_local! { ... }` macro
>> uses curly parenthesies. The round and square brackets are used for
>> expression-like things, and the curlies are used for things that
>> expand to top-level items such as global variables or functions.
>>
>> Macros cannot use any other delimiter than those three. So e.g. <>
>> wouldn't work.
> 
> That answers my implicit "are there any cases in which you would
> want to collectively refer to all three types of...bracket?", yes.
> 
> For the original point, though, we are not in a Rust macro. Is it
> actually allowable to use [] or {} here:
> 
> +    // Brackets around PAGE_SIZE-1 to avoid triggering overflow 
> sanitizers in the wrong cases.
> +    (addr + (PAGE_SIZE - 1)) & PAGE_MASK
> 
> ? Is that why you were not seeing a difference between saying "brackets"
> vs. "parentheses" there? If so, this would be yet another case of my
> Rust newbie-ness being inflicted on you. :)
You can use both () and {}, but you can only use brackets if you're 
European. ;)

Using {} to create a block works because a block evaluates to the value 
of the last expression in the block. It would be super weird to define a 
block here, though.

Alice

Powered by blists - more mailing lists