OpenBSD cvs log

created 2019-05-08T06:45:48Z
begin 2018-02-28T00:00:00Z
end 2018-03-01T00:00:00Z
path src/sys
commits 8

date 2018-02-28T00:44:28Z
author jsg
files src/sys/dev/pci/drm/Attic/i915_drm.h log diff annotate
src/sys/dev/pci/drm/i915/i915_gem_execbuffer.c log diff annotate
message Correct a pointless longstanding difference in ioctl numbering between
linux and add back struct members.

Avoids diffs in inteldrm, libdrm and Mesa >= 17.2.

ok kettenis@

date 2018-02-28T06:19:28Z
author guenther
files src/sys/arch/amd64/amd64/vector.S log diff annotate
message It helps to spell 'GSBASE' with a 'G'. With the thinko the NMI and
double-fault handlers only worked when the GS.base was already the kernel
value (i.e., from inside the kernel). Retested in qemu for both positive
and negative cases ("if I put a bogus GS.base value there, does it crash
on NMI?")

date 2018-02-28T13:16:22Z
author kettenis
files src/sys/arch/arm/arm/locore.S log diff annotate
message Explicitly align svcstack and esym. Finally fixes the problem where
kernels wouldn't boot because the random order in which we happened to link
it would cause these symbols to be misaligned.

Thanks to Brandon Bergren for figuring out the problem.

date 2018-02-28T14:39:35Z
author stsp
files src/sys/dev/pci/if_iwm.c log diff annotate
src/sys/dev/pci/if_iwmreg.h log diff annotate
message Ignore 'fseq version mismatch' notifications from 8265 iwm(4) firmware.
Kills "iwm0: unhandled firmware response 0xff/0xb8000010 rx ring" dmesg spam.
Patch by jes@posteo via tech@, who found the corresponding change in Linux:

date 2018-02-28T14:56:46Z
author deraadt
files src/sys/arch/macppc/stand/tbxidata/bsd.tbxi log diff annotate
src/sys/conf/ log diff annotate
src/sys/sys/param.h log diff annotate
message move to 6.3-beta

date 2018-02-28T15:07:44Z
author deraadt
files src/sys/conf/ log diff annotate
message oops, skipped a step cranking to 6.3-beta

date 2018-02-28T18:47:33Z
author patrick
files src/sys/kern/init_main.c log diff annotate
message Revert the change that postpones CPUs until after mounthook activities.
This was needed to be able to use loadfirmware() to load the microcode
before letting the cores go. Now that the microcode is loaded earlier
we can restore the previous behaviour.

ok deraadt@

date 2018-02-28T23:28:05Z
author dlg
files src/sys/net/if_gif.c log diff annotate
message fix up matching of incoming traffic to interfaces.

i was assigning to ip dst to both the key src and dst, which meant
no traffic matched and it fell through the network stacks ipip

based on some excellent investigation by andreas bartelt.