Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
References: <20210414184604.23473-1-ojeda@kernel.org> <20210414184604.23473-5-ojeda@kernel.org>
 <YHmTWEAS/QjX++w4@hirez.programming.kicks-ass.net> <CAHk-=wh_zb=K1B-N8mgHmSZDqTLgOm711NRXbTX_OwFAzDYg0Q@mail.gmail.com>
 <CANiq72nx7ngazsH7sZgc=HeU0cNj45F9+-rwQb7AkdYsRCmRbQ@mail.gmail.com> <YHnS92ZKZ4tRWTiA@zeniv-ca.linux.org.uk>
In-Reply-To: <YHnS92ZKZ4tRWTiA@zeniv-ca.linux.org.uk>
From:   Miguel Ojeda <miguel.ojeda.sandonis@gmail.com>
Date:   Fri, 16 Apr 2021 20:57:07 +0200
Message-ID: <CANiq72=RLf0FiuLVL-ZeLFp9P2LxTymbzhXoyQGG=tvUY_J-Sg@mail.gmail.com>
Subject: Re: [PATCH 04/13] Kbuild: Rust support
To:     Al Viro <viro@zeniv.linux.org.uk>
Cc:     Linus Torvalds <torvalds@linux-foundation.org>,
        Peter Zijlstra <peterz@infradead.org>,
        Miguel Ojeda <ojeda@kernel.org>,
        Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        rust-for-linux@vger.kernel.org,
        Linux Kbuild mailing list <linux-kbuild@vger.kernel.org>,
        "open list:DOCUMENTATION" <linux-doc@vger.kernel.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        Alex Gaynor <alex.gaynor@gmail.com>,
        Geoffrey Thomas <geofft@ldpreload.com>,
        Finn Behrens <me@kloenk.de>,
        Adam Bratschi-Kaye <ark.email@gmail.com>,
        Wedson Almeida Filho <wedsonaf@google.com>,
        Michael Ellerman <mpe@ellerman.id.au>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Fri, Apr 16, 2021 at 8:10 PM Al Viro <viro@zeniv.linux.org.uk> wrote:
>
> How well would ? operator fit that pattern?  _If_ it's just a syntax sugar
> along the lines of "if argument matches Err(_), return Err(_)", the types
> shouldn't be an issue, but that might need some fun with releasing resources,
> etc.  If it's something more elaborate... details, please.

Yes, it is just syntax sugar -- it doesn't introduce any power to the language.

It was introduced because it is a very common pattern when using the
`Result` and `Option` enums. In fact, before it existed, it was just a
simple macro that you could also implement yourself.

For instance, given `Foo` and `Bar` types that need RAII cleanup of
some kind (let's say `kill_foo()` and `kill_bar()`):

    fn foo() -> KernelResult<Foo> {
        if black_box() {
            return Err(EINVAL);
        }

        // something that gets you a `Foo`
        let foo = ...;

        Ok(foo)
    }

    fn bar() -> KernelResult<Bar> {
        let p = foo()?;

        // something that gets you a `Bar`, possibly using the `p`
        let bar = ...;

        Ok(bar)
    }

This reduces to (full example at https://godbolt.org/z/hjTxd3oP1):

    bar:
            push    rbx
            mov     ebx, 1
            call    qword ptr [rip + black_box@GOTPCREL]
            test    al, al
            jne     .LBB2_2
            call    qword ptr [rip + kill_foo@GOTPCREL]
            xor     ebx, ebx
    .LBB2_2:
            mov     eax, ebx
            mov     edx, -1234
            pop     rbx
            ret

You can see `bar()` calls `black_box()`. If it failed, it returns the
EINVAL. Otherwise, it cleans up the `foo` automatically and returns
the successful `bar`.

Cheers,
Miguel