All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jun Nie <jun.nie@linaro.org>
To: u-boot@lists.denx.de
Subject: [U-Boot] [PATCH] fit: skip signature verification if board request
Date: Wed, 18 Apr 2018 23:22:30 +0800	[thread overview]
Message-ID: <CABymUCO-gDBs-mvYcQNG9_RHXpXV1NRH5RYt3UMSb5QD8h6_ig@mail.gmail.com> (raw)
In-Reply-To: <20180417120648.hxyolrhkq4vqr2kj@qschulz>

2018-04-17 20:06 GMT+08:00 Quentin Schulz <quentin.schulz@bootlin.com>:
> Hi,
>
> On Wed, Apr 11, 2018 at 11:13:05PM +0800, Jun Nie wrote:
>> It may be unnecessary to check signature on unlocked board.
>> Get the hint from platform specific code to support secure boot
>> and non-secure boot with the same binary, so that boot is not
>> blocked if board is not locked and has no key for signature
>> verification.
>>
>
> Isn't it what the environment variable `verify` is made for?
>
> i.e. setting verify=no will skip checks and boot an image even though it
> isn't signed or hash/signature does not match.
>
> I may be missing some context here, so please ignore if it's not what
> you're after.

Thanks for pointing me for this. I check code and find that this variable
does not cover all signature verification cases, such as fit_image_verify().
There is no variable in SPL neither, I suppose.
>
> BTW, I saw that you were speaking of reading the lock fuse to decide
> whether to check the signature or not. I'd like to have at least a
> bypass option for this as it would be horribly tedious for
> debugging/development purposes. E.g. I want to be able to boot from an
> unverified U-Boot binary a signed (and checked) fitImage so that I can
> validate everything works as it should before locking down the
> bootloader.

For this case, it is OK if you do not detect lock fuse value and use default
weak function to indicate no skipping.

>
> Regards,
> Quentin
>
>> Signed-off-by: Jun Nie <jun.nie@linaro.org>
>> ---
>>  common/image-sig.c | 17 +++++++++++++++++
>>  1 file changed, 17 insertions(+)
>>
>> diff --git a/common/image-sig.c b/common/image-sig.c
>> index d9f712f..f3d1252 100644
>> --- a/common/image-sig.c
>> +++ b/common/image-sig.c
>> @@ -151,6 +151,11 @@ struct image_region *fit_region_make_list(const void *fit,
>>       return region;
>>  }
>>
>> +int __attribute__((weak)) fit_board_skip_sig_verification(void)
>> +{
>> +     return 0;
>> +}
>> +
>>  static int fit_image_setup_verify(struct image_sign_info *info,
>>               const void *fit, int noffset, int required_keynode,
>>               char **err_msgp)
>> @@ -188,6 +193,12 @@ int fit_image_check_sig(const void *fit, int noffset, const void *data,
>>       uint8_t *fit_value;
>>       int fit_value_len;
>>
>> +     /* Skip verification if board says that */
>> +     if (fit_board_skip_sig_verification()) {
>> +             printf("signature check skipped\n");
>> +             return 0;
>> +     }
>> +
>>       *err_msgp = NULL;
>>       if (fit_image_setup_verify(&info, fit, noffset, required_keynode,
>>                                  err_msgp))
>> @@ -438,6 +449,12 @@ int fit_config_verify_required_sigs(const void *fit, int conf_noffset,
>>       int noffset;
>>       int sig_node;
>>
>> +     /* Skip verification if board says that */
>> +     if (fit_board_skip_sig_verification()) {
>> +             printf("signature check skipped\n");
>> +             return 0;
>> +     }
>> +
>>       /* Work out what we need to verify */
>>       sig_node = fdt_subnode_offset(sig_blob, 0, FIT_SIG_NODENAME);
>>       if (sig_node < 0) {
>> --
>> 2.7.4
>>
>> _______________________________________________
>> U-Boot mailing list
>> U-Boot at lists.denx.de
>> https://lists.denx.de/listinfo/u-boot

      reply	other threads:[~2018-04-18 15:22 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-04-11 15:13 [U-Boot] [PATCH] fit: skip signature verification if board request Jun Nie
2018-04-12 13:53 ` Tom Rini
2018-04-13  9:58   ` Jun Nie
2018-04-12 17:24 ` Simon Glass
2018-04-13 10:05   ` Jun Nie
2018-04-16 19:06     ` Simon Glass
2018-04-18 15:34       ` Jun Nie
2018-04-17 12:06 ` Quentin Schulz
2018-04-18 15:22   ` Jun Nie [this message]

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=CABymUCO-gDBs-mvYcQNG9_RHXpXV1NRH5RYt3UMSb5QD8h6_ig@mail.gmail.com \
    --to=jun.nie@linaro.org \
    --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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.