linux-kernel - Re: [Question] Alignment requirement for readX() and writeX()

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:   Sat, 31 Jul 2021 09:51:05 +0800
From:   Boqun Feng <boqun.feng@...il.com>
To:     Arnd Bergmann <arnd@...db.de>
Cc:     linux-arch <linux-arch@...r.kernel.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        Miguel Ojeda <miguel.ojeda.sandonis@...il.com>,
        Wedson Almeida Filho <wedsonaf@...gle.com>,
        Gary Guo <gary@...yguo.net>, Hector Martin <marcan@...can.st>,
        Linus Walleij <linus.walleij@...aro.org>
Subject: Re: [Question] Alignment requirement for readX() and writeX()

On Fri, Jul 30, 2021 at 10:24:53PM +0200, Arnd Bergmann wrote:
> On Fri, Jul 30, 2021 at 7:31 PM Boqun Feng <boqun.feng@...il.com> wrote:
> > On Fri, Jul 30, 2021 at 06:58:30PM +0200, Arnd Bergmann wrote:
> >
> > If we want to check, I'd expect we do the checks inside
> > readX()/writeX(), for example, readl() could be implemented as:
> >
> >         #define readl(c)                                        \
> >         ({                                                      \
> >                 u32 __v;                                        \
> >                                                                 \
> >                 /* alignment checking */                        \
> >                 BUG_ON(c & (sizeof(__v) - 1));                  \
> >                 __v = readl_relaxed(c);                         \
> >                 __iormb(__v);                                   \
> >                 __v;                                            \
> >         })
> >
> > It's a runtime check, so if anyone hates it I can understand ;-)
> 
> Right, I really don't think that adds any value, this just replaces one
> oops message with a more different oops message, while adding
> some overhead.
> 

Agreed. I wasn't planning to propose this kind of checks for C code.
Just want to understand better on the alignment requirement of these
APIs. Thanks  ;-)

Regards,
Boqun

>         Arnd