aboutsummaryrefslogtreecommitdiffstats
path: root/CHANGES
Commit message (Collapse)AuthorAgeFilesLines
* Fix PKCS#12 key generation bug.Dr. Stephen Henson2001-03-181-0/+6
|
* The change on handling shared libraries was never applied inRichard Levitte2001-03-151-0/+15
| | | | 0.9.6a-dev...
* More err_data memory leaksBodo Möller2001-03-151-1/+2
|
* that was useless - still fails with GCCUlf Möller2001-03-131-4/+0
|
* Alpha workaround. This is a lot slower!Ulf Möller2001-03-131-0/+4
|
* fix memory leak in err.cBodo Möller2001-03-121-0/+3
|
* ssl23_peekBodo Möller2001-03-081-0/+4
|
* Note the rand_win.c changeUlf Möller2001-03-081-0/+3
|
* SSLv2 session reuse bugfix from main development branch.Richard Levitte2001-03-051-0/+3
|
* Fix bug in copy_email() which would notDr. Stephen Henson2001-03-011-0/+4
| | | | find emailAddress at start of subject name.
* improved bignum test as in 0.9.7.Ulf Möller2001-02-271-0/+3
| | | | | We need this to find out if the bignum failures on Irix and Alpha are caused by new 0.9.7 code or just aren't triggered in the 0.9.6 test suite.
* Backfit a bugfix from 0.9.7-dev to 0.9.6-stable. init() and finish()Geoff Thorpe2001-02-241-0/+6
| | | | | | handlers were previously getting called before (and after, respectively) the "ex_data" structures - this meant init() had very little that it could initialise, and finish() had very little it could cleanup.
* Stop PKCS7_verify() core dumping with unknown publicDr. Stephen Henson2001-02-241-0/+5
| | | | | key algorithms and leaking if the signature verify fails.
* Fix BN_[pseudo_]rand: 'mask' must be used even if top=-1.Bodo Möller2001-02-201-0/+3
| | | | Mention BN_[pseudo_]rand with top=-1 in CHANGES.
* OPENSSL_issetugid() as in the main branch.Ulf Möller2001-02-191-0/+4
|
* Memory leak checking bugfixes for multi-threading.Bodo Möller2001-02-191-0/+13
|
* Move entry to match chronologic ordering.Lutz Jänicke2001-02-151-3/+3
|
* Add '-rand' option to s_server and s_client.Lutz Jänicke2001-02-151-0/+3
|
* IRIX bugfixUlf Möller2001-02-141-0/+4
|
* Make X509_NAME produce correct encoding when empty.Dr. Stephen Henson2001-02-121-0/+4
|
* Workaround for libsafe "error".Dr. Stephen Henson2001-02-121-0/+5
|
* DSA fix from main branch.Ulf Möller2001-02-071-0/+4
|
* EBCDIC bug fix from main branch.Ulf Möller2001-02-071-0/+3
|
* Avoid coredumps for CONF_get_...(NULL, ...)Bodo Möller2001-02-061-0/+9
|
* 0.9.6a will not be release in Y2K. :-)Richard Levitte2001-02-051-1/+1
|
* Tolerate some "variations" used in someDr. Stephen Henson2001-02-011-0/+4
| | | | | | | | | | certificates. One is a valid CA which has no basicConstraints but does have certSign keyUsage. Other is S/MIME signer with nonRepudiation but no digitalSignature.
* Transport from development branch.Richard Levitte2001-01-301-0/+7
|
* Backported manual pages from 0.9.7.Lutz Jänicke2001-01-281-0/+3
|
* New ASN1 macros which will encode an empty SEQUENCE OF.Dr. Stephen Henson2001-01-281-0/+5
| | | | | | | | | | Fix CRL encoders to encode empty SEQUENCE OF. The old code was breaking CRL signatures. Note: it is best to add new macros because changing the old ones could break other code which expects that behaviour. None of this is needed with the new ASN1 code anyway...
* Zero the premaster secret after deriving the master secret in DHDr. Stephen Henson2001-01-251-0/+4
| | | | ciphersuites.
* EVP_add_digest_alias additions to SSL_library_initBodo Möller2001-01-231-0/+7
|
* Irix fix as in main branchUlf Möller2001-01-231-0/+3
|
* remove newlineUlf Möller2001-01-211-1/+1
|
* config bug fixes from the main branch.Ulf Möller2001-01-211-0/+3
|
* Fix openssl passwd -1Bodo Möller2001-01-191-0/+3
|
* Fix PKCS#12 PBE routines to cope with passwordsDr. Stephen Henson2001-01-141-0/+6
| | | | from PEM callbacks which are not null terminated.
* Fix C code generate by 'openssl dsaparam -C'.Bodo Möller2001-01-101-0/+4
|
* Fix uni2asc() so it can properly convert zero lengthDr. Stephen Henson2001-01-101-0/+4
| | | | | unicode strings. Certain PKCS#12 files contain these in BMPStrings and it used to crash on them.
* Finish SSL_peek/SSL_pending fixes.Bodo Möller2000-12-261-0/+1
|
* Fix SSL_peek and SSL_pending.Bodo Möller2000-12-251-5/+13
|
* Include CRYPTO_mem_leaks deadlock fix.Bodo Möller2000-12-201-0/+3
|
* Don't hold CRYPTO_LOCK_RSA during time-consuming operations.Bodo Möller2000-12-191-3/+3
|
* Obtain lock CRYPTO_LOCK_RSA before creating BN_MONT_CTXBodo Möller2000-12-181-0/+9
| | | | | | structures and setting rsa->_method_mod_{n,p,q}. Submitted by: "Reddie, Steven" <Steven.Reddie@ca.com>
* Increase wbuf by one byte to fix the bug reported byBodo Möller2000-12-181-0/+5
| | | | | Eric Day <eday@concentric.net> to openssl-dev@openssl.org, Message-ID: <20001218013437.A5526@concentric.net>
* The first step towards a SSL_peek fix.Bodo Möller2000-12-141-2/+3
| | | | | | | | | | The main thing to verify about these changes is that nothing at all has changed, as far as behaviour is concerned (except that some SSLerr() invocations now have a different function code): SSL_read (ssl2_read, ssl3_read) behaves exactly as before, and SSL_peek refuses to do any work exactly as before. But now the functions actually doing the work have a 'peek' flag, so it should be easy to change them to behave accordingly.
* fix for Borland CUlf Möller2000-12-011-0/+3
|
* Fix BN_rshift.Bodo Möller2000-11-301-0/+3
|
* Store verify_result with sessions to avoid potential security hole.Lutz Jänicke2000-11-291-0/+6
| | | | For the server side this was already done one year ago :-(
* Disable SSL_peek.Bodo Möller2000-11-281-0/+6
|
* SSL_CTX-related fixes.Bodo Möller2000-11-081-0/+4
|
generated by cgit v1.2.3 (git 2.39.1) at 2024-08-23 02:40:04 +0000