Check trailing data when parsing ASN.1.

Check trailing data when parsing ASN.1.

Properly check that the entire buffer was consumed. d2i_* may process only a
prefix of its input. In addition, don't bother using a memory BIO when the
buffer can be parsed directly.

This aligns the NSS and OpenSSL port's behavior in most places:
SEC_QuickDERDecodeItem fails with SEC_ERROR_EXTRA_INPUT if there is excess data.
Add tests. Both for testing and to verify this is the NSS port's behavior.

For a PKCS #8 PrivateKeyInfo, NSS will silently accept trailing data. Fix
WebCrypto in NSS to align with the spec. RSAPrivateKey is left for a follow-up. (This includes an NSS roll to pick up a symbol export.)

BUG=430200
Committed: https://crrev.com/40af916dfb14ca12890e3a7cf8b50d8e62c69695
Cr-Commit-Position: refs/heads/master@{#303546}

[eroman, 2014-11-07 22:13:16]

eroman@chromium.org changed reviewers:
+ eroman@chromium.org

[eroman, 2014-11-07 22:13:17]

content/webcrypto LGTM

[davidben, 2014-11-07 22:58:07]

Patchset #7 (id:120001) has been deleted

[davidben, 2014-11-07 23:37:44]

davidben@chromium.org changed reviewers:
+ rsleevi@chromium.org

[davidben, 2014-11-07 23:37:45]

This depends on https://codereview.chromium.org/710743003/

Presuming this is actually the preferred way to peel off an element, when the
NSS CL lands, I'll update this to also do a roll.

https://codereview.chromium.org/685063007/diff/100001/content/child/webcrypto...
File content/child/webcrypto/nss/rsa_key_nss.cc (right):

https://codereview.chromium.org/685063007/diff/100001/content/child/webcrypto...
content/child/webcrypto/nss/rsa_key_nss.cc:656: SEC_ASN1_GET(SEC_AnyTemplate),
This seems silly, but I'm not familiar enough with NSS to know if there's a
better way to peel off an element.

https://codereview.chromium.org/685063007/diff/100001/crypto/rsa_private_key_...
File crypto/rsa_private_key_nss.cc (right):

https://codereview.chromium.org/685063007/diff/100001/crypto/rsa_private_key_...
crypto/rsa_private_key_nss.cc:290: // data.
Leaving this for a follow-up, so it can be easily reverted if existing consumers
get grumpy, since neither NSS nor OpenSSL ports check right now.

Follow-up here: https://codereview.chromium.org/713523004/

[Ryan Sleevi, 2014-11-08 00:04:36]

//net and //crypto LGTM

[davidben, 2014-11-10 20:24:43]

The CQ bit was checked by davidben@chromium.org

[commit-bot: I haz the power, 2014-11-10 20:25:34]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/685063007/160001

[commit-bot: I haz the power, 2014-11-10 21:03:11]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2014-11-10 21:03:12]

Try jobs failed on following builders:
  android_clang_dbg_recipe on tryserver.chromium.linux
(http://build.chromium.org/p/tryserver.chromium.linux/builders/android_clang_d...)

[davidben, 2014-11-10 23:51:40]

The CQ bit was checked by davidben@chromium.org

[commit-bot: I haz the power, 2014-11-10 23:53:18]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/685063007/160001

[commit-bot: I haz the power, 2014-11-11 00:13:20]

Committed patchset #8 (id:160001)

[commit-bot: I haz the power, 2014-11-11 00:14:17]

Patchset 8 (id:??) landed as
https://crrev.com/40af916dfb14ca12890e3a7cf8b50d8e62c69695
Cr-Commit-Position: refs/heads/master@{#303546}