Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
MIME-Version: 1.0
References: <20200301230537.2247550-1-nivedita@alum.mit.edu>
 <20200303221205.4048668-1-nivedita@alum.mit.edu> <20200303221205.4048668-6-nivedita@alum.mit.edu>
 <CAKv+Gu9LON5SeJ7UMTeHxP1OcSkz_eGGunpXx_csjh_fp1PhDA@mail.gmail.com> <20200303233457.GA154112@rani.riverdale.lan>
In-Reply-To: <20200303233457.GA154112@rani.riverdale.lan>
From:   Ard Biesheuvel <ardb@kernel.org>
Date:   Wed, 4 Mar 2020 08:30:20 +0100
Message-ID: <CAKv+Gu9GLik3VZBZg8VQKFBs3GtV+QB7KQSVfiKcCNtyYjW0XQ@mail.gmail.com>
Subject: Re: [PATCH v2 5/5] efi/x86: Don't relocate the kernel unless necessary
To:     Arvind Sankar <nivedita@alum.mit.edu>
Cc:     linux-efi <linux-efi@vger.kernel.org>,
        "the arch/x86 maintainers" <x86@kernel.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Wed, 4 Mar 2020 at 00:35, Arvind Sankar <nivedita@alum.mit.edu> wrote:
>
> On Wed, Mar 04, 2020 at 12:08:33AM +0100, Ard Biesheuvel wrote:
> > On Tue, 3 Mar 2020 at 23:12, Arvind Sankar <nivedita@alum.mit.edu> wrote:
> > >
> > > Add alignment slack to the PE image size, so that we can realign the
> > > decompression buffer within the space allocated for the image.
> > >
> > > Only relocate the kernel if it has been loaded at an unsuitable address:
> > > * Below LOAD_PHYSICAL_ADDR, or
> > > * Above 64T for 64-bit and 512MiB for 32-bit
> > >
> > > For 32-bit, the upper limit is conservative, but the exact limit can be
> > > difficult to calculate.
> > >
> >
> > Could we get rid of the call to efi_low_alloc_above() in
> > efi_relocate_kernel(), and just allocate top down with the right
> > alignment? I'd like to get rid of efi_low_alloc() et al if we can.
> >
>
> But we don't have a top-down allocator, do we? ALLOCATE_MAX_ADDRESS
> guarantees the maximum, but it doesn't guarantee that you'll be as high
> as possible.

Good point. We do have a top-down allocator in practice, but it is not
guaranteed by the API.