| commit | c61517cb5a7fe72ccbc8c8e9f51055d69ed14e88 | [log] [tgz] |
|---|---|---|
| author | David Benjamin <davidben@chromium.org> | Mon Dec 29 18:41:09 2014 -0500 |
| committer | Adam Langley <agl@google.com> | Wed Jan 14 21:52:36 2015 +0000 |
| tree | 22419f2eb74b60118db243047cad7445e1d41aff | |
| parent | 70bd80a2367fc00e579be6ab041048b8ec930b1f [diff] |
Define the error case's output in RSA_message_index_PKCS1_type_2. The use in s3_srvr.c doesn't care (it doesn't even have to be in bounds), but it's good to have the value be initialized and not a function of the input. (The old uninitialized case wasn't hit in s3_srvr.c because of the earlier bounds check.) Change-Id: Ib6b418b3c140aa564f8a46da3d34bb2b69f06195 Reviewed-on: https://boringssl-review.googlesource.com/2845 Reviewed-by: Adam Langley <agl@google.com>