Message-ID: <63dd37f4.170a0220.9621b.27b3@mx.google.com>
Date:   Fri, 3 Feb 2023 16:36:04 +0000
From:   Kees Cook <keescook@chromium.org>
To:     Ard Biesheuvel <ardb@kernel.org>
Cc:     Catalin Marinas <catalin.marinas@arm.com>,
        Will Deacon <will@kernel.org>,
        Mark Rutland <mark.rutland@arm.com>,
        John Stultz <jstultz@google.com>,
        Yongqin Liu <yongqin.liu@linaro.org>,
        Sami Tolvanen <samitolvanen@google.com>,
        Yury Norov <yury.norov@gmail.com>,
        Andrey Konovalov <andreyknvl@gmail.com>,
        Marco Elver <elver@google.com>,
        linux-arm-kernel@lists.infradead.org, llvm@lists.linux.dev,
        Nathan Chancellor <nathan@kernel.org>,
        Nick Desaulniers <ndesaulniers@google.com>,
        Tom Rix <trix@redhat.com>,
        Alexander Potapenko <glider@google.com>,
        linux-kernel@vger.kernel.org, linux-hardening@vger.kernel.org
Subject: Re: [PATCH] arm64: Support Clang UBSAN trap codes for better
 reporting
References: <20230202223653.never.473-kees@kernel.org>
 <CAMj1kXHa=jD_YNxymjd8zQo2_QHgh104RcmCY3gZs3rp24X_Zg@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <CAMj1kXHa=jD_YNxymjd8zQo2_QHgh104RcmCY3gZs3rp24X_Zg@mail.gmail.com>
Precedence: bulk

On Fri, Feb 03, 2023 at 12:34:38PM +0100, Ard Biesheuvel wrote:
> [...]
> > +       switch (check_type) {
> > +       case ubsan_add_overflow:
> > +               return "UBSAN: addition overflow";
> 
> It would be nice if we could avoid duplicating this UBSAN: prefix 21x times.

Yeah, agreed. I didn't see a way around it without weirdness or loss of
details. die() takes a string not a format string, so without really
odd shenanigans (re-write a single string in memory?) this is the most
straight forward. One idea, though, is to drop all the strings that'll
never happen -- i.e. wrap individual case statements in the associated
CONFIG_UBSAN_foo. I already dropped several strings that were never
implemented in the kernel's UBSAN. I could be even more picky. I'll try
this in v2...

-- 
Kees Cook