summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2019-09-29Fix histogram code to do an inclusive check for the upper bound toOtto Moerbeek
avoid an out-of-bound write for specific values and also check for oob writes in general; with input from kettenis; ok florian@ kn@
2019-09-29Continue SCSIDEBUG cleanup by nuking pointles SC_DEBUGN() #define.Kenneth R Westerback
2019-09-29oops .h file missingTheo de Raadt
2019-09-29group and sort single letter options; ok deraadtJason McIntyre
2016-09-05Import libcompiler_rt 3.9.0, LLVM's replacement for libgcc.Pascal Stumpf
This is the lib/builtin directory of the compiler-rt source tarball. comments/ok patrick@, ok kettenis@
2019-09-29zap trailing whitespace;Jason McIntyre
2019-09-29Move some logic inside scsi_show_xs() instead of surrounding theKenneth R Westerback
single invocation of the same. Clearer.
2019-09-29Tweak some whitespace to make autoindenter happy. Use consistent naming ↵Kenneth R Westerback
idiom for the debug functions.
2019-09-29turn the success paths of FCC registration into debug prints. TheTheo de Raadt
(unlikely) failure path remains noisy. discussed with claudio
2019-09-29Fix pasto.Mark Kettenis
2019-09-29annoying trailing whitespacesFlorian Obser
2019-09-29Replace dwge(4) with a new driver based on dwxe(4). There are manyMark Kettenis
similarities between the two and using a common approach helps fixing bugs. The new driver is better integrated with the device tree framework and is faster (mainly because the DMA engine is configured properly now). Tested on all currently supported variants of the hardware. ok jsg@, jmatthew@
2019-09-29Improve MSI support by parsing the "msi-map" attribute. Taken fromPatrick Wildt
arm64 to reduce the diff between the platforms. ok kettenis@
2019-09-29Add IPI support. Taken ftrom arm64.Mark Kettenis
ok patrick@
2019-09-29- remove specific cases for +a and +mx as the + prefix is handled earlierGilles Chehade
- support mx: notation diff from Quentin Rameau <quinq@fifth.space>
2019-09-29If a NULL or zero cofactor is passed to EC_GROUP_set_generator(),Theo Buehler
try to compute it using Hasse's bound. This works as long as the cofactor is small enough. Port of Brumley's fix for CVE-2019-1547 in OpenSSL 1.1.1 (old license) tests & ok inoguchi input & ok jsing commit 30c22fa8b1d840036b8e203585738df62a03cec8 Author: Billy Brumley <bbrumley@gmail.com> Date: Thu Sep 5 21:25:37 2019 +0300 [crypto/ec] for ECC parameters with NULL or zero cofactor, compute it The cofactor argument to EC_GROUP_set_generator is optional, and SCA mitigations for ECC currently use it. So the library currently falls back to very old SCA-vulnerable code if the cofactor is not present. This PR allows EC_GROUP_set_generator to compute the cofactor for all curves of cryptographic interest. Steering scalar multiplication to more SCA-robust code. This issue affects persisted private keys in explicit parameter form, where the (optional) cofactor field is zero or absent. It also affects curves not built-in to the library, but constructed programatically with explicit parameters, then calling EC_GROUP_set_generator with a nonsensical value (NULL, zero). The very old scalar multiplication code is known to be vulnerable to local uarch attacks, outside of the OpenSSL threat model. New results suggest the code path is also vulnerable to traditional wall clock timing attacks. CVE-2019-1547 Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org> Reviewed-by: Nicola Tuveri <nic.tuv@gmail.com> (Merged from https://github.com/openssl/openssl/pull/9781)
2019-09-29SRS uses base64 encoding for the checksum, however while this is ok when weGilles Chehade
only have MTA in the loop, some implementations like Dovecot's LMTP dislike finding '/' in an e-mail address. Since checksum is meant to be verified at the MX that generated the SRS encoding, use alternate rfc354 base64 encode, swapping '/' with '_' and '+' with '-'. ok eric@ millert@
2019-09-29Configure USB port power control if the power GPIO is defined in fdt.Visa Hankala
2019-09-29Add option to configure GPIO output select.Visa Hankala
2019-09-29Reserve a few GPIO_CONFIG bits for machine-dependent use.Visa Hankala
No objection from kettenis@
2019-09-28Make clang emit the proper path to our libcompiler_rt.a when asked to.Patrick Wildt
Change of behaviour in latest clang upgrade noticed by jsing@ during the Go port update, where --print-libgcc-file-name is being used which prints the compiler-rt path. ok kettenis@
2019-09-28syncTheo de Raadt
2019-09-28adjust for cn30xx -> oct driver renamingTheo de Raadt
2019-09-28rename the 'cn30xx' drivers to just 'oct', because digits inside devicesTheo de Raadt
drivers is gross. discussed with visa.
2019-09-28Add where missing and harmonize PATH to be _PATH_DEFPATH (without local norAntoine Jacoutot
X11R6). Suggested by tb@ ok deraadt@ tb@ millert@
2019-09-28It is better style if path[] buffers are PATH_MAX, rather thanTheo de Raadt
shorter. This subtly reduces the impact of truncation effects, as the truncation is less likely, and if it exists it may be detected directly by the system call with a proper error check. (this strange justification is sadly valid in a world where people moan oh so loudly about truncation but then don't audit and fix all instances) ok kettenis
2019-09-28Xr random 4 in a better wayTheo de Raadt
2019-09-28Update page a little, in particular try to describe the underlying mechanismsTheo de Raadt
in simple terms (mostly as a yardstick for others to be measured against): Entropy data stored previously is provided to the kernel during the boot sequence and used as inner-state of a stream cipher. High quality data is available immediately upon kernel startup. System activity (such as disk, network, and clock device interrupts), and hardware random generator output is collected, whitened with a crc and hash, then periodically folded together with stream cipher inner-state and outer- state to create a new inner state. Reads from all consumers (including the kernel itself, which makes many requests per second) are sliced from the same output stream, which carves the stream cipher output unpredictably and helps improve forward and backtracking protection beyond the strength of the stream cipher. some discussion with djm. There may be more updates.
2019-09-281) don't repeat the 256 / EIO commentaryTheo de Raadt
2) say that the data comes from the random(4) subsystem, so that curious people can go read up on how this works
2019-09-28Make compiler-rt's __clear_cache() work on 32-bit ARM by makingPatrick Wildt
it use sysarch(). From Josh Elsasser ok kettenis@
2019-09-28correctly handle read(2) return value.Eric Faurot
fix issue reported by Mikolaj Kucharski. ok martijn@ deraadt@
2019-09-28ensure dependency_libs and libdir are defined after parsing .la file.Sebastien Marie
problem raised with librsvg which use libtool-rust, and our libtool just die if deplib_list is undef. tested in bulk by sthen@ and naddy@ "modern" perl practice/syntax suggested by espie@ makes sense sthen@ ok espie@
2019-09-27Use consistent idiom/naming convention for the the #includeKenneth R Westerback
guards. Spelunkers using grep are easily confused.
2019-09-27fix gary zukav's "acceptance without proof" quote, which just appeared onJason McIntyre
my screen. i don;t have a copy of this book, and a brief search online shows conflicting punctuation, but i've chosen to replace the comma splice with a full stop, which is at least grammatically correct, and seems the most likely solution. whether this quote needs to be in two files escapes me...
2019-09-27fix the DH-GEX text in -a;Jason McIntyre
because this required a comma, i added a comma to the first part, for balance...
2019-09-27Add/tweak #endif comments to make spelunking via grep more rewarding.Kenneth R Westerback
2019-09-27Shuffle and consolidate SCSIDEBUG declarations into fewer sections inKenneth R Westerback
more intuitive locations.
2019-09-27Make umb(4) add a default route when configuring the interface. This routeClaudio Jeker
has a lower priority than wifi or wired LAN and so should only be used when no other interface is available. With this using umb(4) becomes less painful Now ifconfig umb0 up will be enough especially if unwind(8) is used to handle DNS requests. OK deraadt@ job@ benno@
2019-09-27Move the rde_send_pftable_commit() our of the prefix_update and withdrawClaudio Jeker
function and up higher in the call stack. While there also make sure that flushing announced networks does the pftable dance. This is now also using prefix_withdraw like most other code. Reshuffle rde_update_dispatch() a bit so that all returns are before the first update or withdraw call. After that the code always exits via the end of the function where the commit happens. OK benno@
2019-09-27Implement 'bgpctl show rib 8.8.8.8 or-shorter' to show all routes coveringClaudio Jeker
8.8.8.8. Additionally add 'or-longer' as an alias for 'all'. OK job@ sthen@
2019-09-27Implement F_SHORTER in the prefix lookup code for bgpctl. F_SHORTER willClaudio Jeker
match all prefixes that have a shorter prefixlen than the one in the request. It will print all routes which cover the specified prefix. OK job@ sthen@
2019-09-27Move the code to initialize the cluster-id from merge_config() toClaudio Jeker
parse_config(). The first is not called on startup which results in bgpd using 0.0.0.0 as cluster-id. Found and fix provided by Rivo Nurges (Rivo dot Nurges at smit dot ee) Thanks and OK claudio@
2019-09-27Test for empty result in expected bits. Remove CRs from log as theyDarren Tucker
confuse tools on some platforms. Re-enable the 3des-cbc test.
2019-09-27Use a better scale value in ksmn(4), this produces a slightly moreBryan Steele
stable result when idle and under some load. Tested by abieber@ on a R7 PRO 2700U, also by me on a R5 2500U MateBook D, and a R7 2700X desktop. Discussed with @nte@bsd.network. OK abieber@
2019-09-26For whatever reason it is fairly common that a ROA object references a CRLClaudio Jeker
that is not in the store. Put this paricular error under verbose logging. Agreed by deraadt@, job@
2019-09-26syncTheo de Raadt
2019-09-26Add short manual for ksmn(4) driver.Bryan Steele
2019-09-26Add a new driver for getting AMD CPU temperature sensor over the SMN.Bryan Steele
This currently supports Family 17h Zen/Zen+/Zen2 CPUs. The are still some issues with the scaler on certain models that can be fixed later. AMD Ryzen 2700X: hw.sensors.ksmn0.temp0=47.50 degC Tested by several people. "Make a move" deraadt@
2019-09-26regenBryan Steele
2019-09-26Add "AMD64 17h/3xh Root Complex" device.Bryan Steele
generated by cgit v1.2.3 (git 2.25.1) at 2025-01-24 08:26:49 +0000