cvs commit: src/sys/conf options.i386 src/sys/i386/i386machdep.c
mp_machdep.c pmap.c swtch.s src/sys/i386/include md_var.h pmap.h smp.h
src/sys/i386/isa apic_vector.s intr_machdep.h src/sys/kern kern_switch.c
kern_synch.c kern_thr.c kern_thread.c subr_witness.c ...
nate at root.org
Thu Apr 3 09:55:47 PST 2003
On Wed, 2 Apr 2003, Peter Wemm wrote:
> Nate Lawson wrote:
> > Shouldn't there be a "if (thread == curthread) return;" in the scheduling
> > code somewhere?
> There is now. :-)
> newtd = choosethread();
> if (td != newtd)
> cpu_switch(td, newtd); /* SHAZAM!! */
> #ifdef SWTCH_OPTIM_STATS
> sparc64's cpu_switch() already optimized this case. i386 didn't. I haven't
> checked alpha/ia64 yet, but they're still using the old no-args API.
The more things change...
"Probably 30% of the performance improvements came from fixing things in
the Sun kernel. I mean like, really, guys: If the current task doesn't
change, and it doesn't 80% of the time swtch is called, there's no need
to do a full context save and restore. Adding the two lines
jeq 6f restore of current proc is easy
just before the call to "resume" in sun3/vax.s:swtch got me a quick 70
KB/s performance increase but felt more like a bug fix than progress."
More information about the cvs-all