Check ktrstart() error in doktrace()

This diff makes doktrace() check the outcome of ktrstart() and skip tracing if the trace file header cannot be written. OK? Index: kern/kern_ktrace.c === RCS file: src/sys/kern/kern_ktrace.c,v retrieving revision 1.101 diff -u -p -r1

Fix tsleep(9) during execve(2)

The recent change of initializing sls_sig = 0 in sleep_setup() (r1.164 of kern_synch.c) has introduced a regression with execve(2). execve(2) sets the current process in single thread mode for replacing the program image. In this mode, sleep_setup_signal() cancels a sleep, making tsleep(9) with PCA

ospf6d: bring ospf6d closer to ospfd

Biggest chunk is rework of rde_asext_get()/rde_asext_put(). Also change get_net_link() and get_rtr_link() to work like ospfd couterpart. Index: rde.c === RCS file: /cvs/src/usr.sbin/ospf6d/rde.c,v retrieving revision 1.84 diff -u -p -

Add C11 static_assert macro to assert.h

Taken from FreeBSD. I removed a comment to the effect that C's _Static_assert, if implemented as a macro, would not work well with C++ due to the possibility of unshielded commas appearing in template parameter lists, explaining why *all* C++ standards are excluded rather than just C++11 and later

EFI bootloader fix

Some versions of U-Boot do not include a "media device path" node in the boot device path, even though the OpenBSD bootloader has been loaded from an MS-DOS partition. The UEFI specification isn't very clear whether such a media device path node is required or not. Unfortunately our current EFI b