public inbox for u-boot@lists.denx.de
 help / color / mirror / Atom feed
From: "Heiko Stübner" <heiko@sntech.de>
To: u-boot@lists.denx.de
Subject: [PATCH v2 5/7] spl: fit: enable signing a generated u-boot.itb
Date: Thu, 30 Apr 2020 14:18:27 +0200	[thread overview]
Message-ID: <2027730.If5eCpfMFM@diego> (raw)
In-Reply-To: <28fdf752-0e2a-8161-fc6f-508ff308bde7@rock-chips.com>

Hi Kever,

Am Donnerstag, 30. April 2020, 11:03:38 CEST schrieb Kever Yang:
> This patch will cause build fail on sandbox_spl_defconfig:
> 
> dtc: option requires an argument -- 'p'

sandbox_spl is confusing on first glance, it enables the spl_fit-options
but does not define any fit sources.

But I also found a general issue with my code below, and by fixing that
one sandbox_spl also gets happy again.

> On 2020/4/21 ??8:23, Heiko Stuebner wrote:
> > From: Heiko Stuebner <heiko.stuebner@theobroma-systems.com>
> >
> > With SPL_FIT_SIGNATURE enabled we will likely want a generated
> > u-boot.itb to be signed and the key stores so that the spl can
> > reach it.
> >
> > So add a SPL_FIT_SIGNATURE_KEY_DIR option and suitable hooks
> > into the Makefile to have mkimage sign the .itb and store the
> > used key into the spl dtb file.
> >
> > The added dependencies should make sure that the u-boot.itb
> > gets generated before the spl-binary gets build, so that there
> > is the necessary space for the key to get included.
> >
> > Signed-off-by: Heiko Stuebner <heiko.stuebner@theobroma-systems.com>
> > Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
> > ---
> >   Kconfig  |  8 ++++++++
> >   Makefile | 11 ++++++++++-
> >   2 files changed, 18 insertions(+), 1 deletion(-)
> >
> > diff --git a/Kconfig b/Kconfig
> > index 4051746319..15a783a67d 100644
> > --- a/Kconfig
> > +++ b/Kconfig
> > @@ -451,6 +451,14 @@ config SPL_FIT_SIGNATURE
> >   	select SPL_RSA_VERIFY
> >   	select IMAGE_SIGN_INFO
> >   
> > +config SPL_FIT_SIGNATURE_KEY_DIR
> > +	string "key directory for signing U-Boot FIT image"
> > +	depends on SPL_FIT_SIGNATURE
> > +	default "keys"
> > +	help
> > +	  The directory to give to mkimage to retrieve keys from when
> > +	  generating a signed U-Boot FIT image.
> > +
> >   config SPL_LOAD_FIT
> >   	bool "Enable SPL loading U-Boot as a FIT (basic fitImage features)"
> >   	select SPL_FIT
> > diff --git a/Makefile b/Makefile
> > index 26307fd4a6..8e7a7cb50e 100644
> > --- a/Makefile
> > +++ b/Makefile
> > @@ -1394,6 +1394,14 @@ MKIMAGEFLAGS_u-boot.itb =
> >   else
> >   MKIMAGEFLAGS_u-boot.itb = -E
> >   endif
> > +ifdef CONFIG_SPL_FIT_SIGNATURE
> > +ifdef CONFIG_SPL_OF_CONTROL
> > +MKIMAGEFLAGS_u-boot.itb += -K dts/dt-spl.dtb -r
> > +ifneq ($(CONFIG_SPL_FIT_SIGNATURE_KEY_DIR),"")
> > +MKIMAGEFLAGS_u-boot.itb += -k $(CONFIG_SPL_FIT_SIGNATURE_KEY_DIR)
> > +endif
> > +endif
> > +endif
> >   
> >   u-boot.itb: u-boot-nodtb.bin \
> >   		$(if $(CONFIG_OF_SEPARATE)$(CONFIG_OF_EMBED)$(CONFIG_OF_HOSTFILE),dts/dt.dtb) \
> > @@ -1913,7 +1921,8 @@ spl/u-boot-spl.bin: spl/u-boot-spl
> >   
> >   spl/u-boot-spl: tools prepare \
> >   		$(if $(CONFIG_OF_SEPARATE)$(CONFIG_OF_EMBED)$(CONFIG_SPL_OF_PLATDATA),dts/dt.dtb) \
> > -		$(if $(CONFIG_OF_SEPARATE)$(CONFIG_OF_EMBED)$(CONFIG_TPL_OF_PLATDATA),dts/dt.dtb)
> > +		$(if $(CONFIG_OF_SEPARATE)$(CONFIG_OF_EMBED)$(CONFIG_TPL_OF_PLATDATA),dts/dt.dtb) \
> > +		$(if $(CONFIG_SPL_FIT_GENERATOR),u-boot.itb FORCE)

I now realized that this is the wrong check ... i.e. it only checks for
SPL_FIT_GENERATOR but that is a string so always defined if SPL_LOAD_FIT
is enabled ... also this doesn't take into account SPL_FIT_SOURCE, so the
way to go seems to be to check against $U_BOOT_ITS and
CONFIG_SPL_FIT_SIGNATZRE instead which gets defined if a suitable fit
source is available.


Background for this dependency is that the signature must be done before
the spl-binary gets build, because mkimage for the .itb needs to write the
key to the spl dtb.


I'll send an updated patch as a reply to this mail.


Heiko

  reply	other threads:[~2020-04-30 12:18 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-04-21  0:23 [PATCH v2 0/7] rockchip: make it possible to sign the u-boot.itb Heiko Stuebner
2020-04-21  0:23 ` [PATCH v2 1/7] spl: fit: select SPL_HASH_SUPPORT for SPL_FIT_SIGNATURE Heiko Stuebner
2020-04-28 13:45   ` Kever Yang
2020-04-21  0:23 ` [PATCH v2 2/7] spl: fit: select SPL_CRYPTO_SUPPORT " Heiko Stuebner
2020-04-28 13:46   ` Kever Yang
2020-04-21  0:23 ` [PATCH v2 3/7] lib: rsa: distinguish between tpl and spl for CONFIG_RSA_VERIFY Heiko Stuebner
2020-04-28 13:47   ` Kever Yang
2020-04-21  0:23 ` [PATCH v2 4/7] mkimage: fit_image: handle multiple errors when writing signatures Heiko Stuebner
2020-04-28 13:48   ` Kever Yang
2020-04-21  0:23 ` [PATCH v2 5/7] spl: fit: enable signing a generated u-boot.itb Heiko Stuebner
2020-04-28 13:48   ` Kever Yang
2020-04-30  9:03   ` Kever Yang
2020-04-30 12:18     ` Heiko Stübner [this message]
2020-04-30 12:32     ` [PATCH v2.1 " Heiko Stuebner
2020-04-21  0:23 ` [PATCH v2 6/7] spl: fit: add Kconfig option to specify key-hint for fit_generator Heiko Stuebner
2020-04-21 17:37   ` Simon Glass
2020-04-28 13:53   ` Kever Yang
2020-04-21  0:23 ` [PATCH v2 7/7] rockchip: make_fit_atf: add signature handling Heiko Stuebner
2020-04-21 17:37   ` Simon Glass
2020-04-28 13:53   ` Kever Yang
2020-05-01 10:32   ` Kever Yang
2020-05-04  0:33     ` Heiko Stübner
2020-05-06  8:55     ` Heiko Stübner

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=2027730.If5eCpfMFM@diego \
    --to=heiko@sntech.de \
    --cc=u-boot@lists.denx.de \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox