summaryrefslogtreecommitdiff
path: root/src/lib/libcrypto (follow)
Commit message (Collapse)AuthorAgeFilesLines
* correct cases of code occuring directly after goto/break/returnjsg2014-04-151-1/+0
| | | | ok miod@ guenther@
* revert 1.2 and define OPENSSL_NO_EC_NISTP_64_GCC_128 again, it has a downsidesthen2014-04-142-0/+6
| | | | (breaks ssh ecdsa keys)
* Remove eng_cryptodev.c (merged into old-name hw_cryptodev.c); movemiod2014-04-144-1460/+38
| | | | | | | undo the move of crypto/engines/eng_padlock to engines/e_padlock. Requested by reyk@. Note that eng_padlock is not compiled in currently.
* used for windows dll builds onlymiod2014-04-142-4634/+0
|
* unifdef -U OPENBSD_DEV_CRYPTO and OPENSSL_OPENBSD_DEV_CRYPTO. This code was themiod2014-04-143-459/+0
| | | | | | early attempt at getting kernel-assisted crypto(4) used by libcrypto, before the engine API existed, and has been #if 0'd out for ages anyway. No API/ABI change.
* So the OpenSSL codebase does "get the time, add it as a random seed"deraadt2014-04-141-3/+0
| | | | | | | in a bunch of places inside the TLS engine, to try to keep entropy high. I wonder if their moto is "If you can't solve a problem, at least try to do it badly". ok miod
* remove auto-generated dependencies from the old unused build system, soderaadt2014-04-1456-5934/+0
| | | | | that it is easier to find code pieces. They are getting in the way. ok miod
* Remove the GMP engine: It was an experimental engine using libgmp asreyk2014-04-143-9/+0
| | | | | | | an alternative backend for BIGNUM calculations. It is PoC code that is not enabled in OpenSSL and probably not used by anymore. ok deraadt@
* Remove the CAPI engine: It is a backend for the Windows CryptoAPI andreyk2014-04-142-4/+0
| | | | | | could be maintained in an external package. "it should probably go" beck@
* replace PTR_SIZE_INT (only used for VMS) with uintptr_t, tweaks from jca@,sthen2014-04-144-54/+33
| | | | makes sense to beck@
* Remove the nuron engine. The static engine is not standalone and thereyk2014-04-143-5/+0
| | | | FPGA-based device is long obsolete.
* Remove the nCipher CHIL engine. It is not standalone and depends onreyk2014-04-144-26/+0
| | | | external libraries that aren't covered by the same license.
* Don't bother compiling files which end up containingmiod2014-04-145-1293/+2
| | | | | | #if OPENSSL_SYS_NOTYOURS <whole file> #endif
* two SHUTDOWN2 got away; noticed by beckderaadt2014-04-142-2/+4
|
* Use shutdown(s, SHUT_RDWR) instead of shutdown(s, 2).matthew2014-04-132-2/+2
| | | | ok beck deraadt
* Cope with the removal of openssl/symhacks.hderaadt2014-04-1314-22/+1
|
* Remove various horrible socket syscall wrappers, especially SHUTDOWN*deraadt2014-04-138-67/+67
| | | | | | which did shutdown + close, all nasty and surprising. Use the raw syscalls that everyone knows the behaviour of. ok beck matthew
* Flense a variety of windows support stuff, and a strange gettimeofday function.beck2014-04-131-12/+0
| | | | ok deraadt@
* symhacks for vms are no longer neededbeck2014-04-131-481/+0
| | | | ok deraadt@
* Remove vms support stuff.beck2014-04-133-153/+1
| | | | ok deraadt@
* Remove some stuff that isn't needed.beck2014-04-136-35/+35
| | | | ok miod@ deraadt@
* Remove the AEP engine: it is not standalone and doesn't seem to bereyk2014-04-133-5/+0
| | | | | | | | | relevant anymore. OpenSSL should have a better way to include 3rd party engines: either completely and free or external. But including a wrapper for a non-free wrapper in the code base does not make much sense and could also be provided by the vendor. ok deraadt@
* Remove the Atalla engine: It is not standalone and depends on externalreyk2014-04-133-5/+0
| | | | | | | | | non-free libraries. OpenSSL should have a better way to include 3rd party engines: either completely free or external. But including a wrapper for a non-free wrapper in the code base does not make much sense and could also be provided by the vendor. ok deraadt@
* Remove the cswift engine: it is not standalone and we don't have thereyk2014-04-133-7/+0
| | | | | | | | | | | | | hardware. The vendor_defns/cswift.h does not specify a copyright and theoretically defaults to the OpenSSL license, but it also mentions that it includes parts that have been "clipped" from CryptoSwift's proprietary headers. This file should better include an explicit copyright statement or mention OpenSSL's library instead of the ambiguous "Attribution notice". ok deraadt@
* Remove the "sureware" engine:reyk2014-04-133-5/+0
| | | | | | | | | | | | The vendor_defns/sureware.h file by Baltimore Technologies Ltd. has a copyright that does not grant rights! Vendor files should either include a compatible license in the copyright statement or use OpenSSL's defaults, but adding a copyright statement without any terms is not acceptable. It should not have been included in the first place. ok deraadt@
* Remove the Broadcom ubsec engine:reyk2014-04-134-6/+0
| | | | | | | | | | | | | The vendor_defns/hw_ubsec.h file has a copyright that does not grant rights! Vendor files should either include a compatible license in the copyright statement or use OpenSSL's defaults, but adding a copyright statement without any terms is not acceptable. It should not have been included in the first place. (The ubsec(4) kernel driver is not affected by this change) ok deraadt@
* Remove the IBM 4758 engine: we don't have this hardware and it is anreyk2014-04-134-6/+0
| | | | | | old PCI accelerator that was EOL'ed in 2005. ok deraadt@
* Don't edit on cvs. 23.1 as intended.miod2014-04-132-4/+4
|
* Do not install e_os.h, it is not an external header and should not be usedmiod2014-04-131-2/+1
| | | | by anything in userland.
* Move undocumented internal OPENSSL_DIR interface from libcrypto to libssl,miod2014-04-139-727/+2
| | | | | | so that libssl no longer need to access the non-external headers of libcrypto to build. No library bump, riding upon the recent update.
* Crank libcrypto minor (addition of BN_consttime_swap and CRYPTO_memcmp)miod2014-04-132-2/+2
| | | | and libssl major (ssl_check_clienthello_tlsext split into two functions)
* No longer define OPENSSL_NO_EC_NISTP_64_GCC_128 on amd64; requested by djmmiod2014-04-132-6/+0
|
* Merge conflicts; remove MacOS, Netware, OS/2, VMS and Windows build machinery.miod2014-04-13105-3260/+788
|
* This commit was generated by cvs2git to track changes on a CVS vendormiod2014-04-1367-276/+632
|\ | | | | branch.
| * Import OpenSSL 1.0.1gmiod2014-04-13148-568/+1532
| |
* | This commit was generated by cvs2git to track changes on a CVS vendormiod2014-04-1312-418/+1346
|\ \ | | | | | | branch.
| * | Import OpenSSL 1.0.1gmiod2014-04-1315-431/+1355
| | |
* | | This commit was generated by cvs2git to track changes on a CVS vendormiod2014-04-132-2/+8
|\ \ \ | | | | | | | | branch.
| * | | Import OpenSSL 1.0.1gmiod2014-04-1315-153/+507
| | | |
* | | | Move build machinery for libcrypto from libssl/crypto to libcrypto, as wellmiod2014-04-1137-0/+13153
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | as configuration files; split manpages and .pc files between libcrypto and libssl. No functional change, only there to make engineering easier, and libcrypto sources are still found in libssl/src/crypto at the moment. ok reyk@, also discussed with deraadt@ beck@ and the usual crypto suspects.
* | | | prepare manpages for new perl.espie2014-03-181-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Note that I missed two of these in the diff shown initially, thx to the atrocious Makefile rule... okay millert@, sthen@, basically
* | | | Reliability fix for SHA384 SSL/TLS ciphers on strict alignmentjca2013-12-191-1/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | architectures. ok miod@ djm@ Upstream patch: commit cdd1acd788020d2c525331da1712ada778f1373c Author: Andy Polyakov <appro@openssl.org> Date: Wed Dec 18 21:27:35 2013 +0100
* | | | Switch to using unhyphenated VIA padlock mnemonics. VIA abandoned thematthew2013-05-301-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | hyphen in their official programming guide sometime between 2003 and 2005, and Clang's integrated assembler does not support hyphenated mnemonics. ok jsg, deraadt
* | | | cherry pick bugfixes for http://www.openssl.org/news/secadv_20130205.txtmarkus2013-02-145-42/+216
| | | | | | | | | | | | | | | | | | | | from the openssl git (changes between openssl 1.0.1c and 1.0.1d). ok djm@
* | | | remove ACSS, crank libcrypto major; ok markus@ deraadt@djm2013-01-265-402/+0
| | | |
* | | | On amd64 OPENSSL_cpuid_setup and OPENSSL_ia32cap_P are now hidden so we don'tkettenis2012-10-314-9/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | have to go through the PLT/GOT to get at them anymore. In fact going through the GOT now fails since we no longer have a GOT entry for OPENSSL_ia32cap_P. Fixes the problem spotted by jasper@ and sthen@. Based on a diff from mikeb@ who did most of the actual work of tracking down the issue. ok millert@, mikeb@
* | | | Restore r1.10, lost during last update:miod2012-10-221-1/+1
| | | | | | | | | | | | | | | | Disable use of dladdr() on a.out arches, they do not provide it (yet);
* | | | import files that CVS missed; sighdjm2012-10-135-0/+749
| | | |
* | | | resolve conflictsdjm2012-10-13195-2795/+10581
| | | |
* | | | This commit was generated by cvs2git to track changes on a CVS vendordjm2012-10-1314-37/+149
|\| | | | | | | | | | | branch.
generated by cgit v1.2.3-55-g6feb (git 2.39.1) at 2025-04-26 06:05:45 +0000