Date:   Fri, 3 Dec 2021 13:37:56 +0100
From:   Marco Elver <elver@google.com>
To:     andrey.konovalov@linux.dev
Cc:     Alexander Potapenko <glider@google.com>,
        Vincenzo Frascino <vincenzo.frascino@arm.com>,
        Catalin Marinas <catalin.marinas@arm.com>,
        Peter Collingbourne <pcc@google.com>,
        Andrey Konovalov <andreyknvl@gmail.com>,
        Dmitry Vyukov <dvyukov@google.com>,
        Andrey Ryabinin <aryabinin@virtuozzo.com>,
        kasan-dev@googlegroups.com,
        Andrew Morton <akpm@linux-foundation.org>, linux-mm@kvack.org,
        Will Deacon <will@kernel.org>,
        linux-arm-kernel@lists.infradead.org,
        Evgenii Stepanov <eugenis@google.com>,
        linux-kernel@vger.kernel.org,
        Andrey Konovalov <andreyknvl@google.com>
Subject: Re: [PATCH 23/31] kasan, arm64: allow KASAN_VMALLOC with SW_TAGS
Message-ID: <YaoPpPAKi0/OZB2f@elver.google.com>
References: <cover.1638308023.git.andreyknvl@google.com>
 <f90dfb0c02598aab3ad1b5b6ea4a4104b14e099d.1638308023.git.andreyknvl@google.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <f90dfb0c02598aab3ad1b5b6ea4a4104b14e099d.1638308023.git.andreyknvl@google.com>
User-Agent: Mutt/2.0.5 (2021-01-21)
Precedence: bulk

On Tue, Nov 30, 2021 at 11:07PM +0100, andrey.konovalov@linux.dev wrote:
> From: Andrey Konovalov <andreyknvl@google.com>
> 
> vmalloc support for SW_TAGS KASAN is now complete.
> 
> Allow enabling CONFIG_KASAN_VMALLOC.
> 
> Signed-off-by: Andrey Konovalov <andreyknvl@google.com>

This change is small enough that I would have expected the
lib/Kconfig.kasan change to appear in "kasan, vmalloc: add vmalloc
support to SW_TAGS" because that sounds like it would fully unlock
core KASAN support.

However, the arm64 change could be in its own patch, since there may be
conflicts with arm64 tree or during backports, and only dropping that
may be ok.

I've been backporting too many patches lately, that I feel that would
help.

> ---
>  arch/arm64/Kconfig | 1 +
>  lib/Kconfig.kasan  | 2 +-
>  2 files changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
> index c4207cf9bb17..c05d7a06276f 100644
> --- a/arch/arm64/Kconfig
> +++ b/arch/arm64/Kconfig
> @@ -206,6 +206,7 @@ config ARM64
>  	select IRQ_DOMAIN
>  	select IRQ_FORCED_THREADING
>  	select KASAN_VMALLOC if KASAN_GENERIC
> +	select KASAN_VMALLOC if KASAN_SW_TAGS
>  	select MODULES_USE_ELF_RELA
>  	select NEED_DMA_MAP_STATE
>  	select NEED_SG_DMA_LENGTH
> diff --git a/lib/Kconfig.kasan b/lib/Kconfig.kasan
> index cdc842d090db..3f144a87f8a3 100644
> --- a/lib/Kconfig.kasan
> +++ b/lib/Kconfig.kasan
> @@ -179,7 +179,7 @@ config KASAN_TAGS_IDENTIFY
>  
>  config KASAN_VMALLOC
>  	bool "Back mappings in vmalloc space with real shadow memory"
> -	depends on KASAN_GENERIC && HAVE_ARCH_KASAN_VMALLOC
> +	depends on (KASAN_GENERIC || KASAN_SW_TAGS) && HAVE_ARCH_KASAN_VMALLOC
>  	help
>  	  By default, the shadow region for vmalloc space is the read-only
>  	  zero page. This means that KASAN cannot detect errors involving
> -- 
> 2.25.1
>