linux-kernel - Re: [PATCH] ACPICA: fix -Wfallthrough

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Fri, 13 Nov 2020 09:14:01 +0100
From:   Miguel Ojeda <miguel.ojeda.sandonis@...il.com>
To:     Nick Desaulniers <ndesaulniers@...gle.com>
Cc:     "Moore, Robert" <robert.moore@...el.com>,
        "Kaneda, Erik" <erik.kaneda@...el.com>,
        "Wysocki, Rafael J" <rafael.j.wysocki@...el.com>,
        "Gustavo A . R . Silva" <gustavoars@...nel.org>,
        "clang-built-linux@...glegroups.com" 
        <clang-built-linux@...glegroups.com>, Len Brown <lenb@...nel.org>,
        "linux-acpi@...r.kernel.org" <linux-acpi@...r.kernel.org>,
        "devel@...ica.org" <devel@...ica.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] ACPICA: fix -Wfallthrough

On Fri, Nov 13, 2020 at 1:09 AM Nick Desaulniers
<ndesaulniers@...gle.com> wrote:
>
> Thank you for the explicit diagnostics observed.  Something fishy is
> going on though, https://godbolt.org/z/Gbxbxa is how I expect MSVC to
> handle include/linux/compiler_attributes.h.
>
> The C preprocessor should make it such that MSVC never sees
> `__attribute__` or `__fallthrough__`; that it does begs the question.
> That would seem to imply that `#if __has_attribute(__fallthrough__)`
> somehow evaluates to true on MSVC, but my godbolt link shows it does
> not.
>
> Could the upstream ACPICA project be #define'ing something that could
> be altering this? (Or not #define'ing something?)
>
> Worst case, we could do as Joe Perches suggested and disable
> -Wfallthrough for drivers/acpi/acpica/.

I agree, something is fishy. MSVC has several flags for conformance
and extensions support, including two full C preprocessors in newer
versions; which means we might be missing something, but I don't see
how the code in compiler_attributes.h could be confusing MSVC even in
older non-conforming versions.

Cheers,
Miguel