summaryrefslogtreecommitdiff
path: root/lib/libcrypto
AgeCommit message (Collapse)Author
2014-08-28preserve errno value on success.Brent Cook
If getrandom returns a temporary failure, make sure errno is not polluted when it succeeds. Thanks to deraadt@ for pointing it out.
2014-08-16only build the getrandom path if SYS_getrandom is defined.bcook
like the sysctl path
2014-08-16getrandom(2) support for getentropy_linuxbcook
This enables support for the new getrandom(2) syscall in Linux 3.17. If the call exists and fails, return a failure in getentropy(2) emulation as well. This adds a EINTR check in case the urandom pool is not initialized. Tested on Fedora Rawhide with 3.17rc0 and Ubuntu 14.04 ok deraadt@
2014-08-14fixed overrid(d)en typoTobias Stoeckmann
millert@ and jmc@ agree that "overriden" is wrong
2014-08-13munmap correct object in (extremely unlikely, and effectively terminal)Theo de Raadt
case of failing to map the 2nd object. found by Paul Maurers
2014-08-11Guard RSA / RC4-5 ASM when NO_ASM is not definedbcook
Most assembly blocks remain inactive if OPENSSL_NO_ASM is not defined, only enabling inline assembly, but the RSA / RC4-5 blocks (used only in amd64 systems) turn on implicitly. Guard these two as well. This simplifies enabling just inline ASM in portable, no effective change in OpenBSD.
2014-07-28Remove SRP code. It contains a bug (this should not surprise anyone), butTed Unangst
the details are under embargo. The original plan was to wait for the embargo to lift, but we've been waiting for quite some time, and there's no indication of when or even if it will end. No sense in dragging this out any longer. The SRP code has never been enabled in OpenBSD, though I understand it is in use by some other people. However, in light of this and other issues, we're officially saying SRP is outside the scope of libressl. (For now.)
2014-07-22better match proposed syscall apibcook
2014-07-21protect sysctl path with SYS__sysctl instead; from enh@google, ok bcookTheo de Raadt
2014-07-21Use explicit_bzero() instead of memset() on buffers going out of scope.Philip Guenther
Also, zero the SHA256 context. suggested by "eric" in a comment on an opensslrampage.org post ok miod@ deraadt@
2014-07-21cast from void * before math; enh@googleTheo de Raadt
2014-07-20Move more OS-specific functionality to arc4random.h headers.bcook
Move <sys/mman.h> and raise(SIGKILL) calls to OS-specific headers. On OpenBSD, move thread_private.h as well to arc4random.h. On Windows, use TerminateProcess on getentropy failure. ok deraadt@
2014-07-20initial win32 ARC4_LOCK/UNLOCK implementation.bcook
It may make sense to later replace this with a Critical Section later. ok guenther@
2014-07-20Demonstrate how new linux getrandom() will be called, at least untilTheo de Raadt
it shows up in libraries. Even the system call is probably not finalized. Bit dissapointed it has turned out to be a descriptor-less read() with EINVAL and EINTR error conditions, but we can work with it.
2014-07-19remove disabled main hook; we use phdr now; ok bcookTheo de Raadt
2014-07-19tab loveTheo de Raadt
2014-07-19Move _ARC4_ATFORK handlers from thread_private.h in portable.bcook
2014-07-19move _ARC4_LOCK/UNLOCK primitives from thread_private into OS-specific modulesbcook
2014-07-19fixup typosbcook
2014-07-19Change _rs_allocate so it can combine the two regions (rs and rsx)Theo de Raadt
into one if a system has an awesome getentropy(). In that case it is valid to totally throw away the rsx state in the child. If the getentropy() is not very good and has a lazy reseed operation, this combining is a bad idea, and the reseed should probably continue to use the "something old, something new" mix. _rs_allocate() can accomodate either method, but not on the fly. ok matthew
2014-07-18Cleanup portable arc4random fork detection code:Matthew Dempsky
1. Use "len" parameter instead of sizeof(*rs). 2. Simplify the atfork handler to be strictly async signal safe by simply writing to a global volatile sig_atomic_t object, and then checking for this in _rs_forkdetect(). (Idea from discussions with Szabolcs Nagy and Rich Felker.) 3. Use memset(rs, 0, sizeof(*rs)) to match OpenBSD's MAP_INHERIT_ZERO fork semantics to avoid any skew in behavior across platforms. ok deraadt
2014-07-18Seperate arc4random's os-dependent parts into static inline functions,Theo de Raadt
making it much easier for libressl -portable to fill in the gaps. ok bcook beck
2014-07-16Only call getauxval(3) if HAVE_GETAUXVAL is defined. Fixes build on olderMark Kettenis
Linux (such as Ubuntu 12.04LTS) that don't have it yet. Seems the AT_XXX defines are pulled in by <link.h> now. ok beck@
2014-07-13Use dl_iterate_phdr() to iterate over the segments and throw the addressesTheo de Raadt
into the hash; hoping the system has some ASLR or PIE. This replaces and substantially improves upon &main which proved problematic with some picky linkers. Work with kettenis, testing by beck
2014-07-13Provide a link to the canonical API specification.Theo de Raadt
ok beck
2014-07-13Take away the use of the address of main as a source of entropy. CausesBob Beck
distractions to people testing and seeing link errors in some setups. This will come back in another form ok deraadt@
2014-07-12more MLINKSMiod Vallat
2014-07-12getentropy on Windows. It compiles but has not been thoroughly tested yet.wouter
OK: beck@
2014-07-12Remove signed/unsigned warning, statement before declaration andwouter
add a function to use function pointers that does not take sizeof(fptr). OK beck@
2014-07-12Remove private_{Camellia,RC4}_set_key FIPS indirection tentacles, as has beenMiod Vallat
done for other symmetric algorithms recently.
2014-07-12more MLINKsMiod Vallat
2014-07-12guard inclusion of sys/sysctl.h so we can detect at compile time andBob Beck
keep linux distros happy that don't have it. ok bcook@
2014-07-12remove gratuitous differences, ok beckTheo de Raadt
2014-07-12remove gratuitous differences, ok beck bcookTheo de Raadt
2014-07-12Solaris uses a symbolic link for /dev/urandom which harms best practice ofBob Beck
using O_NOFOLLOW - cope with it as best as possible by trying two different paths. - written by deraadt@ and kettenis@
2014-07-12odds are that some ABI change occured today, no matter how careful everyoneTheo de Raadt
is
2014-07-11add comment about format requirementsBob Beck
ok miod@
2014-07-11Huge documentation update for libcrypto and libssl, mostly from Matt Caswell,Miod Vallat
Jeff Trawick, Jean-Paul Calderone, Michal Bozon, Jeffrey Walton and Rich Salz, via OpenSSL trunk (with some parts not applying to us, such as SSLv2 support, at least partially removed).
2014-07-11i'm a dumbdumb. fix build.Ted Unangst
2014-07-11move all the feature settings to a common header.Ted Unangst
probably ok beck jsing miod
2014-07-11Fix incorrect duplicate mlinksBob Beck
ok bcook@
2014-07-11Correct incorrect mlinksBob Beck
ok bcook@
2014-07-11Fix incorrect mlinksBob Beck
ok bcook@
2014-07-11Make sure we leave OPENSSL_NO_PSK in the conf files so thingsBob Beck
can know... ok jsing@
2014-07-09Update to match the current state of crypto/rand, and remove MLINKS forMiod Vallat
RAND_event and RAND_screen.
2014-07-09remove unused variables getentropy for OS Xbcook
ok beck@
2014-07-09RSA_NULL used to be a compile option allowing the RSA interfaces to beMiod Vallat
compiled-in, with nonfunctional code, to be able to cope with the RSA patent. However, we don't use this option, and the RSA patent has expired more than 10 years ago, so just drop this piece.
2014-07-08Remove undocumented _des_crypt() interface and its companion header file,Miod Vallat
which had never been installed, so it's unlikely something ever used this in the last 15~20 years. ok deraadt@ jsing@ beck@
2014-07-08getentropy for osx and solaris. will be needed for a portable releaseBob Beck
2014-07-08fix oops, accidental delete.. darn copying of files between machinesBob Beck
generated by cgit v1.2.3 (git 2.25.1) at 2024-12-27 22:31:50 +0000