Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <20230908104458.775740-1-songshuaishuai@tinylab.org>
In-Reply-To: <20230908104458.775740-1-songshuaishuai@tinylab.org>
From:   Rob Herring <robh+dt@kernel.org>
Date:   Fri, 8 Sep 2023 09:58:02 -0500
Message-ID: <CAL_JsqJFWTWg9gmO7P8d5-W9VeMtmazhCsQW121LDAWLMAHEnQ@mail.gmail.com>
Subject: Re: [PATCH] of/kexec: Omit kaslr-seed when kexecing with nokaslr
To:     Song Shuai <songshuaishuai@tinylab.org>
Cc:     frowand.list@gmail.com, devicetree@vger.kernel.org,
        linux-kernel@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Fri, Sep 8, 2023 at 5:45=E2=80=AFAM Song Shuai <songshuaishuai@tinylab.o=
rg> wrote:
>
> The chosen node always adds the kaslr-seed property which is needless
> for the kernels kexeced with the cmdline contained "nokaslr".
>
> So omit the kaslr-seed when kexecing with nokaslr.

The kexec'ed kernel can ignore bootargs with CONFIG_CMDLINE_FORCE. So
you don't know if kaslr-seed will be used or not.

> Signed-off-by: Song Shuai <songshuaishuai@tinylab.org>
> ---
>  drivers/of/kexec.c | 15 ++++++++++++---
>  1 file changed, 12 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/of/kexec.c b/drivers/of/kexec.c
> index f26d2ba8a371..c0d53b10cb70 100644
> --- a/drivers/of/kexec.c
> +++ b/drivers/of/kexec.c
> @@ -19,6 +19,7 @@
>  #include <linux/random.h>
>  #include <linux/slab.h>
>  #include <linux/types.h>
> +#include <linux/string.h>
>
>  #define RNG_SEED_SIZE          128
>
> @@ -263,6 +264,14 @@ static inline int setup_ima_buffer(const struct kima=
ge *image, void *fdt,
>  }
>  #endif /* CONFIG_IMA_KEXEC */
>
> +static bool is_nokaslr_cmdline(const char *cmdline)
> +{
> +       char *str;
> +
> +       str =3D strstr(cmdline, "nokaslr");
> +       return str =3D=3D cmdline || (str > cmdline && *(str - 1) =3D=3D =
' ');

Yet another copy of the same code. That's a clue to refactor into a
common function.

Rob