summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2016-09-02syncStuart Henderson
2016-09-02provide a pool_setipl so tcpdump can compile this fileDavid Gwynne
spotted by deraadt@
2016-09-02proc.c tweaks: Rename proc_listento() to proc_accept() as it is theReyk Floeter
receiving side of proc_connect(). Move some code from main into proc_init(), the function is now called by parent and children, not just the parent and it is less copy + paste for other daemons. OK florian@
2016-09-02Pull in Chacha20 and Poly1305 source code as xform.o dependenciesMike Belopuhov
2016-09-02Pull in Chacha20 and Poly1305 source code as xform.o dependenciesMike Belopuhov
2016-09-02provide an implementation of red black trees using functionsDavid Gwynne
the main goal of this change is to reduce the amount of code that is generated as a result of using the macro implementation (RB_FOO) of red black trees. on amd64 we should get a few dozen kilobytes of code space back, and make red black trees more icache friendly at the same time. the new (RBT_FOO) implementation is modelled on the existing one, but has some minor api variations. generally you can replace RB_ with RBT_ and get most of the way to converting code. internally the red black tree functions all take an rb_type struct that describes the layout of the object wired into a tree (ie, the offset of the RBT_ENTRY inside a node), the comparison function, and an optional augment function. because the functions are supposed to be used for all types, they end up taking void * for the node pointers instead of specific types. the tree is operated on as pointers between the RBT_ENTRY structs instead of the nodes, which gave me some type safety when implementing the code (cos casts to/from void * dont ever fail, and continually calculating the offset of the rb entry is annoying). RBT_ENTRYs are turned into node pointers by prepending the offset stored in the rb_type struct before theyre given to the comparison function or returned to the caller. to provide type safety on top of this, RBT_PROTOTYPE generates static inline function wrappers that only take arguments of the right type, and implicitly provide the rb_type struct argument to the actual RBT functions. therefore the actual functions should never be called directly, all calls should go through the RBT_ wrappers. RBT_GENERATE is responsible for creating the rb_type struct used by these wrappers. notably it also generates a wrapper around the compare function so the user provided one must take the right types instead of void *. in terms of speed, this code is comparable to the macro implementation. eg, insertion is very slightly slower in microbenchmarks, but deletion appears to be significantly faster. this is possibly because of the aggressive inlining ive done inside the delete codepaths. the code is not yet wired into the kernel build. it also needs to be said that there have been several attempts before this to provide functions for at least some parts of the kernels red black trees. that work made this a lot easier. ok deraadt@ jung@ tedu@
2016-09-02Pass M_ZERO when allocating memory for "struct usbd_endpoint".Martin Pieuchot
These descriptors are filled with value parsed from untrusted USB descriptors and we don't want to left memory unitialized if an error occurs during the parsing.
2016-09-02Treat backoff_cutoff as a hard cutoff, not the midpoint in a rangeKenneth R Westerback
from .5 to 1.5 ofthe value. This is how the man page describes it and nobody can remember why it was not being used as a hard limit. ok benno@
2016-09-02for reporting changes, hex and decimal were swappedTheo de Raadt
noticed by pirofti
2016-09-02Adjust for the new default MODP groupMike Belopuhov
2016-09-02Cleanup usbd_fill_iface_data() to make it easier to check for badMartin Pieuchot
descriptors. No functionnal change.
2016-09-02Allow editing cpg in expert mode and align the cpg field properlyOtto Moerbeek
ok krw@
2016-09-02Remove obsolete DES-CBC testsMike Belopuhov
2016-09-02krb5 bits should rest in peaceGleydson Soares
OK deraadt
2016-09-01remove sparc supportTed Unangst
2016-08-31Crank minor due to API additionBob Beck
2016-08-31Avoid undefined-behavior right-shifting by a word-size # of bits.Brent Cook
Found with STACK, originally from OpenSSL, ok @beck
2016-08-31Bring in functions used by stunnel and exim from BoringSSL - this bringsBob Beck
in X509_check_host, X509_check_email, X509_check_ip, and X509_check_ip_asc, with some cleanup on the way in by myself and jsing@ ok bcook@
2016-08-31Fix some very unnecessary convoultion.Bob Beck
ok krw@
2016-08-27Be more strict when parsing TLS extensions.Joel Sing
Based on a diff from Kinichiro Inoguchi. ok beck@
2016-08-07Update the link for the getentropy(2) manual to man.openbsd.org/Theo Buehler
ok deraadt@
2016-08-05Do not *printf %s NULLTheo de Raadt
ok bcook
2016-07-31bump for LibreSSL 2.5.xBrent Cook
2016-07-18don't mix code and decls, ok tedu@Brent Cook
2016-07-17use memset to initialize the unionBrent Cook
2016-07-17remove unused OPENSSL_NO_OBJECT caseBrent Cook
ok tedu@
2016-07-17Initialize buffers before use, noted by Kinichiro Inoguchi.Brent Cook
ok beck@
2016-07-16Clean up OCSP_check_validity() a bit more.Bob Beck
- Return on first failure rather than continuing. - Don't compare times by comparing strings that possibly were not parsable as a time. ok deraadt@
2016-07-16Limit the support of the "backward compatible" ssl2 handshake to only beBob Beck
used if TLS 1.0 is enabled. Sugessted/discussed with jsing@ and bcook@. ok guenther@ sthen@
2016-07-10zero the read buffer after copying data to user so it doesn't linger.Ted Unangst
ok beck
2016-07-07call BN_init on temporaries to avoid use-before-set warningsBrent Cook
ok beck@
2016-07-05remove unneeded duplicate call - spotted by jsing@Bob Beck
2016-07-05On systems where we do not have BN_ULLONG defined (most 64-bit systems),Brent Cook
BN_mod_word() can return incorrect results if the supplied modulus is too big, so we need to fall back to BN_div_word. Now that BN_mod_word may fail, handle errors properly update the man page. Thanks to Brian Smith for pointing out these fixes from BoringSSL: https://boringssl.googlesource.com/boringssl/+/67cb49d045f04973ddba0f92fe8a8ad483c7da89 https://boringssl.googlesource.com/boringssl/+/44bedc348d9491e63c7ed1438db100a4b8a830be ok beck@
2016-07-05Add several fixes from OpenSSL to make OCSP work with intermediateBob Beck
certificates provided in the response. - makes our newly added ocsp regress test pass too.. ok bcook@
2016-06-30Tighten behavior of _rs_allocate failure for portable arc4random ↵Brent Cook
implementations. In the event of a failure in _rs_allocate for rsx, we still have a reference to freed memory for rs on return. Not a huge deal since we subsequently abort in _rs_init, but it looks strange on its own. ok deraadt@
2016-06-30Tighten behavior of _rs_allocate on Windows.Brent Cook
For Windows, we are simply using calloc, which has two annoyances: the memory has more permissions than needed by default, and it comes from the process heap, which looks like a memory leak since this memory is rightfully never freed. This switches _rs_alloc on Windows to use VirtualAlloc, which restricts the memory to READ|WRITE and keeps the memory out of the process heap. ok deraadt@
2016-06-30bump to 2.4.2Brent Cook
2016-06-30Remove flags for disabling constant-time operations.Brent Cook
This removes support for DSA_FLAG_NO_EXP_CONSTTIME, DH_FLAG_NO_EXP_CONSTTIME, and RSA_FLAG_NO_CONSTTIME flags, making all of these operations unconditionally constant-time. Based on the original patch by César Pereid. ok beck@
2016-06-25Fix from kinichiro.inoguchi@gmail.com to ensure that OCSP usesBob Beck
Generalized Time on requests as per RFC6960
2016-06-25Fix the ocsp code to actually check for errors when comparing time valuesBob Beck
which was not being done due to a lack of checking of the return code for X509_cmp_time. Ensure that we only compare GERNERALIZEDTIME values because this is what is specified by RFC6960. Issue reported, and fix provided by Kazuki Yamaguchi <k@rhe.jp> ok bcook@
2016-06-21Disable DSA_FLAG_NO_EXP_CONSTTIME, always enable constant-time behavior.Brent Cook
Improved patch from Cesar Pereida. See https://github.com/libressl-portable/openbsd/pull/61 for more details. ok beck@
2016-06-06Set BN_FLG_CONSTTIME on the correct variable. beck committed wrong fix.Ted Unangst
Mistake noted by Billy Brumley. Many thanks.
2016-06-06Correct a problem that prevents the DSA signing algorithm from runningBob Beck
in constant time even if the flag BN_FLG_CONSTTIME is set. This issue was reported by Cesar Pereida (Aalto University), Billy Brumley (Tampere University of Technology), and Yuval Yarom (The University of Adelaide and NICTA). The fix was developed by Cesar Pereida.
2016-06-06LibreSSL 2.4.1Brent Cook
2016-05-30deprecate internal use of EVP_[Cipher|Encrypt|Decrypt]_Final.Bob Beck
14 years ago these were changed in OpenSSL to be the same as the _ex functions. We use the _ex functions only internally to ensure it is obvious the ctx must be cleared. ok bcook@
2016-05-20Fix a short-read bug in the previous version of asn1_d2i_read_bioBrent Cook
The outer while() loop is missing, so we only read up to chunk_max bytes.
2016-05-11remove hppa64 port, which we never got going beyond broken single users.Theo de Raadt
hppa reverse-stack gives us a valuable test case, but most developers don't have a 2nd one to proceed further with this. ok kettenis
2016-05-04fix for integer overflow in encode and encrypt update functions.Ted Unangst
additionally, in EncodeUpdate, if the amount written would overflow, return 0 instead to prevent bugs in the caller. CVE-2016-2105 and CVE-2016-2106 from openssl.
2016-05-04fix a padding oracle in aesni cbc mac check. there must be enough dataTed Unangst
for both the mac and padding bytes. CVE-2016-2107 from openssl
2016-05-04internal only negative types should not be handled here.Ted Unangst
CVE-2016-2108 from openssl.
generated by cgit v1.2.3 (git 2.25.1) at 2024-12-02 08:30:26 +0000