Re: [PATCH v2 08/18] arm64: don't disable ADR_PREL_PG_HI21* with ARM64_ERRATUM_843419

From: Ard Biesheuvel <ard.biesheuvel@linaro.org>
To: Sami Tolvanen <samitolvanen@google.com>
Cc: Will Deacon <will.deacon@arm.com>,
	Mark Rutland <mark.rutland@arm.com>,
	Andi Kleen <ak@linux.intel.com>,
	Kees Cook <keescook@chromium.org>,
	linux-kbuild@vger.kernel.org,
	Nick Desaulniers <ndesaulniers@google.com>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	Greg Hackmann <ghackmann@google.com>,
	Masahiro Yamada <yamada.masahiro@socionext.com>,
	Michal Marek <michal.lkml@markovi.net>,
	Yury Norov <ynorov@caviumnetworks.com>,
	Alex Matveev <alxmtvv@gmail.com>,
	Matthias Kaehlcke <mka@chromium.org>,
	"linux-arm-kernel@lists.infradead.org" 
	<linux-arm-kernel@lists.infradead.org>,
	Maxim Kuvyrkov <maxim.kuvyrkov@linaro.org>
Subject: Re: [PATCH v2 08/18] arm64: don't disable ADR_PREL_PG_HI21* with ARM64_ERRATUM_843419
Date: Thu, 16 Nov 2017 22:25:46 +0000	[thread overview]
Message-ID: <CAKv+Gu8uWq+T2AJQvBqwb-hMeZgFJQUQozsr4Ufwwp2hheRXcQ@mail.gmail.com> (raw)
In-Reply-To: <CAKv+Gu_3R4SUC2hCUtEX_4ut9admtrBZS-S-mzQnQsoJoHPcBw@mail.gmail.com>

On 16 November 2017 at 22:14, Ard Biesheuvel <ard.biesheuvel@linaro.org> wrote:
> On 16 November 2017 at 21:37, Sami Tolvanen <samitolvanen@google.com> wrote:
>> On Thu, Nov 16, 2017 at 04:34:03PM +0000, Ard Biesheuvel wrote:
>>> You still have not explained to us how GOLD avoids the erratum.
>>
>> Sorry, I didn't realize you were asking that. If gold spots erratum
>> sequences, looks like it creates stubs to break them up:
>>
>>   https://sourceware.org/git/?p=binutils-gdb.git;a=blob;f=gold/aarch64.cc#l8396
>>
>> It also attempts to optimize the code by replacing adrps in these
>> sequences with adr where possible, but otherwise doesn't appear to
>> touch them:
>>
>>   https://sourceware.org/git/?p=binutils-gdb.git;a=blob;f=gold/aarch64.cc#l2053
>>
>
> OK, so my concern here is that this code probably only operates on
> fully linked binaries, and not partially linked object files like
> kernel modules. The same applies to ld.bfd, which is why we need to
> use the large module

*model* not module

> instead of a code model that may emit adrp
> instructions..
>
> What is preventing us from using the large model with clang? I know it
> uses movk/movz pairs rather than literals, but this shouldn't matter
> for modules, given that we support static ELF relocations.