wireguard-openbsd/sys/crypto, branch master WireGuard implementation for the OpenBSD kernel https://git.zx2c4.com/wireguard-openbsd/atom/sys/crypto?h=master 2021-03-12T10:22:46Z spelling 2021-03-12T10:22:46Z jsg jsg@openbsd.org 2021-03-12T10:22:46Z urn:sha1:84f3acfde2a7e7d22afae1b5e404e34b9403abc7 Add missing CVS tags. 2020-07-22T13:54:30Z tobhe tobhe@openbsd.org 2020-07-22T13:54:30Z urn:sha1:6a126883bc80051260ef4a80c3b9b17c95b29772 ok patrick@ don't try to expand alignment using __aligned(16) on local variables. 2020-06-27T15:58:47Z deraadt deraadt@openbsd.org 2020-06-27T15:58:47Z urn:sha1:034cea9fb2355e3cd2c7b6b3ad55b28b630dff96 that stops being normal C. the object type has to be sufficient, if it needed larger alignment such a type or a union should be used instead. (breaks landisk gcc, for one) kernel: use gettime(9)/getuptime(9) in lieu of time_second(9)/time_uptime(9) 2020-06-24T22:03:40Z cheloha cheloha@openbsd.org 2020-06-24T22:03:40Z urn:sha1:3209772dfcc3950dd5df01bc44eebf75e637511e time_second(9) and time_uptime(9) are widely used in the kernel to quickly get the system UTC or system uptime as a time_t. However, time_t is 64-bit everywhere, so it is not generally safe to use them on 32-bit platforms: you have a split-read problem if your hardware cannot perform atomic 64-bit reads. This patch replaces time_second(9) with gettime(9), a safer successor interface, throughout the kernel. Similarly, time_uptime(9) is replaced with getuptime(9). There is a performance cost on 32-bit platforms in exchange for eliminating the split-read problem: instead of two register reads you now have a lockless read loop to pull the values from the timehands. This is really not *too* bad in the grand scheme of things, but compared to what we were doing before it is several times slower. There is no performance cost on 64-bit (__LP64__) platforms. With input from visa@, dlg@, and tedu@. Several bugs squashed by visa@. ok kettenis@ add constructions for wireguard. 2020-06-21T10:50:01Z dlg dlg@openbsd.org 2020-06-21T10:50:01Z urn:sha1:76a075be08bd0affc816a095146a46cacfd345e7 from Matt Dunwoodie and Jason A. Donenfeld ok deraadt@ add hchacha20 for wireguard. 2020-06-21T10:46:49Z dlg dlg@openbsd.org 2020-06-21T10:46:49Z urn:sha1:09a3dd53443e850ce90c366150fff3c420b2cc57 i think we should turn the chacha code into an actual c file at some point to reduce duplication of object code, but that can happen later. from Matt Dunwoodie and Jason A. Donenfeld ok deraadt@ curve25519 for wireguard. 2020-06-21T10:43:37Z dlg dlg@openbsd.org 2020-06-21T10:43:37Z urn:sha1:2df0987dc7644d0c188262d2fe2e5f1a843c9ee0 via Matt Dunwoodie and Jason A. Donenfeld ok deraadt@ add blake2s for wireguard. 2020-06-21T10:34:37Z dlg dlg@openbsd.org 2020-06-21T10:34:37Z urn:sha1:9960953d15a1d2b1e5b61b541dba13adcff15f1d via Matt Dunwoodie and Jason A. Donenfeld ok deraadt@ rndvar.h not needed here 2020-05-29T01:22:53Z deraadt deraadt@openbsd.org 2020-05-29T01:22:53Z urn:sha1:416e7f624d7c7e736f2321f05d8daf48d6111744 idgen32(9): use time_uptime(9) for rekey timestamps, not time_second(9) 2020-04-15T03:56:14Z cheloha cheloha@openbsd.org 2020-04-15T03:56:14Z urn:sha1:14a8f15fd61273b1fbf3096a20c525920b20a4cd Timestamp contexts with a monotonic clock. time_second(9) can jump, stretching or truncating our rekey interval. time_uptime(9) will not jump around. ok djm@