amd64_{get,set}_fsbase() are superfluous and unused; delete them.Delete unused/never-implemented sysarch defines and structs while here.ports check and ok naddy@ok deraadt@ mlarkin@
some Xr adjustment to catch up with MLINKS removal;
for some time now mandoc has not required MLINKS to functioncorrectly - logically complete that now by removing MLINKS from base;authors need only to ensure there is an entry in NAME for any funct
for some time now mandoc has not required MLINKS to functioncorrectly - logically complete that now by removing MLINKS from base;authors need only to ensure there is an entry in NAME for any function/util being added. MLINKS will still work, and remain for perl to easeupgrades;ok nicm (curses) bcook (ssl)ok schwarze, who provided a lot of feedback and assistanceok tb natano jung
show more ...
ansify amd64_iopl()
use .In rather than .Fd #include
remove useless casts (and the comments saying lint complains about them)
zap trailing newlines; "go for it" deraadt
remove never implemented amd64 get/set ioperm sysarch. ok guenther
Remove excessive sys/cdefs.h inclusionok guenther millert kettenis
tweak previous;
Clarify thread-localness in the wake of rfork's removal
Unrevert the FS.base diff: the issues were actually elsewhereAdditional testing by jasper@ and pea@
Revert bulk of the FS.base diff, as it causes issues on some machinesand the problem isn't obvious yet.
Add support for per-rthread base-offset for the %fs selector on amd64.Add pcb_fsbase to the PCB for tracking what the value for the threadis, and ci_cur_fsbase to struct cpu_info for tracking the C
Add support for per-rthread base-offset for the %fs selector on amd64.Add pcb_fsbase to the PCB for tracking what the value for the threadis, and ci_cur_fsbase to struct cpu_info for tracking the CPU's currentvalue for FS.base, then on return to user-space, skip the setting if theCPU has the right value already. Non-threaded processes without TLS leaveFS.base zero, which can be conveniently optimized: setting %fs zerosFS.base for fewer cycles than wrmsr.ok kettenis@
We do not do vm86 mode on amd64ok kettenis
First pass at removing clauses 3 and 4 from NetBSD licenses.Not sure what's more surprising: how long it took for NetBSD tocatch up to the rest of the BSDs (including UCB), or the amount ofcode t
First pass at removing clauses 3 and 4 from NetBSD licenses.Not sure what's more surprising: how long it took for NetBSD tocatch up to the rest of the BSDs (including UCB), or the amount ofcode that NetBSD has claimed for itself without attributing to theactual authors.OK deraadt@
Remove an arch-specific API that wasn't available in the first place, andwill be converted to a more generic one soon.ok deraadt, oga
- remove USER_LDT, it was never in a state where it would copile, nor willwe support i386-compat mode on amd64.agreed by beck@, dlg@, kettenis@ok deraadt@, tom@
convert to new .Dd format;
prefer `buses' to `busses' for the noun plural;
various improvements;based on -r1.22 from freebsd (i386 version anyway),submitted by andreas fuchs,improved by ru@freebsd,then again by myself and mickeyok mickey
slightly reword [EPERM] section; ok djm
amd64_iopl() needs securelevel <= 0, or machdep.allowapeture > 0just like i386_iopl()
typos;new sentence, new line;kill whitespace;
correct .Dt lines, and remove .Lb stuff;ok deraadt@
12