wireguard-openbsd, branch jd/simplify-queueingWireGuard implementation for the OpenBSD kernelhttps://git.zx2c4.com/wireguard-openbsd/atom/?h=jd%2Fsimplify-queueing2020-06-22T23:44:05ZSimplify if_output to avoid queueing2020-06-22T23:44:05ZJason A. DonenfeldJason@zx2c4.com2020-06-22T23:44:05Zurn:sha1:4caf19369cb6ee8750c01de2904d8be3b9fbd3b7
fq_codel is done by copying the mbuf flow hash for the
final egress, but not at the encapsulation stage.
The minimum frame size for the ELFv2 ABI is 32 bytes. Using this value2020-06-22T18:49:36Zketteniskettenis@openbsd.org2020-06-22T18:49:36Zurn:sha1:4d943fd61ab7ebc3390510da6f072b3fbd6ff841
instead of 48 in the trap setup functions makes it possible to get rid
of the magic offset in proc_trampoline.
Suggested by gkoehler@
inittodr(9): introduce dedicated flag to enable writes from resettodr(9)2020-06-22T18:25:57Zchelohacheloha@openbsd.org2020-06-22T18:25:57Zurn:sha1:bf2d84a07eb2512358d130f21cd3f3c03b4c2d28
We don't want resettodr(9) to write the RTC until inittodr(9) has
actually run. Until inittodr(9) calls tc_setclock() the system UTC
clock will contain a meaningless value and there's no sense in
overwriting a good value with a value we know is nonsense.
This is not an uncommon problem if you're debugging a problem in early
boot, e.g. a panic that occurs prior to inittodr(9).
Currently we use the following logic in resettodr(9) to inhibit writes:
if (time_second == 1)
return;
... this is too magical.
A better way to accomplish the same thing is to introduce a dedicated
flag set from inittodr(9). Hence, "inittodr_done".
Suggested by visa@.
ok kettenis@
Remove unused variable2020-06-22T18:18:20Zdenisdenis@openbsd.org2020-06-22T18:18:20Zurn:sha1:7e30acb46a662278fca2affbc6dfe466a120ae52Enable interrupts when they were enabled before the trap (except when we2020-06-22T18:15:50Zketteniskettenis@openbsd.org2020-06-22T18:15:50Zurn:sha1:be00ebe64ffda37c7ac803f330b1c897f12b55ea
are handling "genuine" interrupts like the decrementer and hypervisor
virtualization interrupts).
Handle data storage and data segment interrupts from userland as well.2020-06-22T18:03:22Zketteniskettenis@openbsd.org2020-06-22T18:03:22Zurn:sha1:c0e8dfc26859acc5c80a7cd731a9232aad396d94Make return-to-user and kernel re-entry work. This adds a per-pmap SLB2020-06-22T16:58:20Zketteniskettenis@openbsd.org2020-06-22T16:58:20Zurn:sha1:37d56e14ef1a0db6f4d11db0ecc4e6ce6dbee7df
cache. We might want to turn that in a per-proc cache at some point, but
this gets us to the point where we can sucessfully have init(1) do its
first system call.
Fix previous commit; pasted instructed in the wrong spot.2020-06-22T16:47:10Zketteniskettenis@openbsd.org2020-06-22T16:47:10Zurn:sha1:ebbbb1399a0803cd282701c7145a2416733aed07Remove AMRR/MiRA support from iwx(4). Tx rate selection happens in firmware.2020-06-22T16:27:37Zstspstsp@openbsd.org2020-06-22T16:27:37Zurn:sha1:9fd0d4312dbee62e0609dec3aa6abf88be4c76ceRemove unused code for parsing non-volatile memory (NVM) from iwx(4).2020-06-22T16:25:55Zstspstsp@openbsd.org2020-06-22T16:25:55Zurn:sha1:f2a6a16817b3f0e0697b8be84536199913088b48