buildable state after the yamt-idlelwp merge.
Struct cpu_info now must contain ci_want_resched, so get rid of the
global want_resched.
The cpu_need_resched macro now takes two arguments.
We need to declare proc_trampoline() from for the new m68k vm_machdep.c.
The fields l_back and l_forw no longer exists in struct lwp, so excise
their definition from genassym.cf.
Remove the use of the apparently now obsolete m68k/proc_subr.s.
from proc_trampoline to match the other ports).
A DIAGNOSTIC kernel will now boot and run. LOCKDEBUG still doesn't work yet.
Also, my amiga no longer loses time.
Use return value from cpu_switchto for previous lwp in lwp_trampoline.
Also shamelessly steal a comment from uwe written for sparc that explains
all this.
Thanks to tnn, mrg, and uwe for helping to debug this.
- in lwp_trampoline, oldlwp is already in %o0. don't save or the wrongly
saved copy from cpu_lwp_fork(). as repeatedly pointed out my uwe.
- in cpu_idle use retl not ret. as pointed out by martin.
thanks to uwe/martin/tnn for all the real debugging work.
- use macros to deal with pathnames in userspace, when veriexec is used.
- reorder the veriexec_ call arguments for consistency.
With help from elad@ finding the last bug.
Using movq would copy pcb_flags along as well, which would cause traps
because bad bits would be written into %cr0 a few instructions later.
This wouldn't happen for native NetBSD processes (no pcb_flags set), but
it would happen for Linux processes.
to the beginning so they can still be accessed in assymbly code using 8 bit
offsets. The addition of a big structure in cpu_data moved these outside
the range of the addq instruction. Gas should have complained that the
immediate value was too large, but just silently truncated it. I can now
run multiprocessor again.
Actually, we do need separate "no references in file server" and
"noref + inactive" flags if we wish to correctly support unix open
file semantics and optimize away pre-reclaim cache flushes. So,
add PNODE_DYING which stands for norefs + inactive.
"noref + inactive" flags if we wish to correctly support unix open
file semantics and optimize away pre-reclaim cache flushes. So,
add PNODE_DYING which stands for norefs + inactive.
curcpu()->ci_want_resched is now required, so get rid of the
want_resched global. Also, cpu_need_resched() now takes two arguments.
The l_next and l_back fields in struct lwp no longer exists, so delete
references to it from genassym.cf.
Also, m68k/m68k/proc_subr.s is no longer used, says mhitch, so remove its use.
This brings the atari BOOT kernel back to a buildable state.
- don't use void * for pointer arithmetic
- don't try to modify const parameters
A kernel with 'options TCP_SIGNATURE' works as well as it ever did, now.
(ie, clunky, but passable)
the struct tcphdr * argument of tcp_dooptions(). RFC2385 support
(options TCP_SIGNATURE) needs to modify the header during options
processing, and this revision broke it.
OK yamt@.
the kernel it has 0 references to the node in question. In other
words, this can be used to avoid inactive(), or, if the file server
does not implement inactive, prompt reclaim for removed nodes.
