diff options
author | Lutz Jänicke <jaenicke@openssl.org> | 2007-09-20 07:39:15 +0000 |
---|---|---|
committer | Lutz Jänicke <jaenicke@openssl.org> | 2007-09-20 07:39:15 +0000 |
commit | 9ce3ee47ba21fd19859cf2953ee7e951f53e7da7 (patch) | |
tree | 2ab6a6e8a8becab508d73c22166e03a2e581a0cc /FAQ | |
parent | 2e3fd543370c6aa82d7c8bc57cdf5b58248efbfc (diff) |
Add FAQ entry on how to get rid of Valgrind warnings.
PR: 521
Diffstat (limited to 'FAQ')
-rw-r--r-- | FAQ | 15 |
1 files changed, 15 insertions, 0 deletions
@@ -66,6 +66,7 @@ OpenSSL - Frequently Asked Questions * Why doesn't my server application receive a client certificate? * Why does compilation fail due to an undefined symbol NID_uniqueIdentifier? * I think I've detected a memory leak, is this a bug? +* Why does Valgrind complain about the use of uninitialized data? =============================================================================== @@ -894,5 +895,19 @@ thread-safe): ERR_free_strings(), EVP_cleanup() and CRYPTO_cleanup_all_ex_data(). +* Why does Valgrind complain about the use of uninitialized data? + +OpenSSL does internally call its own PRNG routines to retrieve random +numbers. It so does with uninitialed buffer contents. The buffer +contents is mixed into the entropy pool so that it technically does +not matter whether the buffer is initialized at this point or not. +Valgrind (and other test tools) will complain whatsoever. When +using Valgrind, make sure to use an OpenSSL library that has been +compiled with the PEDANTIC macro being defined (-DPEDANTIC) to +get rid of these warnings. Compling with -DPURIFY will help as well. + +The PEDANTIC macro was added in OpenSSL 0.9.8f. + + =============================================================================== |