Turns out that older SPARC CPUs (like the UltraSPARC II) do report an exact
authorkettenis <kettenis@openbsd.org>
Tue, 28 Sep 2021 08:56:15 +0000 (08:56 +0000)
committerkettenis <kettenis@openbsd.org>
Tue, 28 Sep 2021 08:56:15 +0000 (08:56 +0000)
commit31e18b07b9f98cacc0ab5c092b28d4552a954be5
tree078bdacb08b56e6b7961c42de25d53cfd39d23cd
parentbf27995b5709fe3c496157249610d5207acd6484
Turns out that older SPARC CPUs (like the UltraSPARC II) do report an exact
fault address, while others (like the UltraSPARC T2) report an inexact
address.  Deal with this by using the EXPADDR_MASK on both the expected and
the reported fault address.

ok bluhm@
regress/sys/kern/signal/siginfo-fault/siginfo-fault.c