Date: Fri, 15 Jan 2010 10:06:04 -0500 From: Attilio Rao <attilio@freebsd.org> To: John Baldwin <jhb@freebsd.org> Cc: FreeBSD Arch <arch@freebsd.org>, Ed Maste <emaste@freebsd.org> Subject: Re: [PATCH] Statclock aliasing by LAPIC Message-ID: <3bbf2fe11001150706y765159a2jbd37c7ae4cf378f0@mail.gmail.com> In-Reply-To: <200911301305.30572.jhb@freebsd.org> References: <3bbf2fe10911271542h2b179874qa0d9a4a7224dcb2f@mail.gmail.com> <200911301305.30572.jhb@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
2009/11/30 John Baldwin <jhb@freebsd.org>: > On Friday 27 November 2009 6:42:50 pm Attilio Rao wrote: >> Handling all the three clocks (hardclock, softclock, profclock) within >> the LAPIC can lead to aliasing for the softclock and profclock because >> hz is sized to fit mainly hardclock. The fashion to handle all of them >> within the LAPIC was introduced in 2005 and before than the softclock >> and profclock were supposed to be handled in the rtc. Right now, too, >> there is the necessary support to handle profclock and statclock in >> atrtc which gets enabled if the LAPIC signals it can't take in charge >> the three clocks. >> The proposed patch reverts the situation preferring the atrtc to >> handle the statclock and profclock (then a different source from the >> LAPIC) and then avoid the aliasing problem: >> > http://www.freebsd.org/~attilio/Sandvine/STABLE_8/statclock_aliasing/stat= clock_aliasing3.diff >> >> In this patch, lapic_setup_clock() has been changed in order to return >> a three-states variable which identified if the LAPIC got in charge >> all the three clocks, just the hardclock or none of them (the current >> situation is just none/all) and the rtc handling runs subsequently. >> A tunable as been added to force LAPI to get in charge all the three >> clocks if needed. >> In ia32 atrtc compiling is linked to atpic compiling, so a compile >> time flag has been added to check if atpic is not present and in case >> force LAPIC to take in charge all the three clocks (which is still >> better than the 'safe belt values' still present in the rtc code). >> >> Please note that statclock and profclock are widely used in our kernel >> (rusage is, for example, statclock driven) and fixing this would >> result in specific improvements (as a several-reported wrong CPU usage >> statistic in top). >> This bug has been found by Sandvine Incorporated. >> >> Reviews, comments and testing are welcome. > > Presumably in the RTC case lapic_timer_hz should always be hz and not som= e > multiple of hz. =C2=A0Also, did you check to make sure all the lock is pr= esent? =C2=A0I > think at one point I changed the locking for the RTC and/or ISA timer to = just > use critical_enter/exit so that UP machines running an SMP kernel wouldn'= t pay > the locking overhead since the code was only used on UP machines. =C2=A0I= t may very > well be that I only changed that in a development branch though and not i= n > HEAD. I still see clock_lock in place (and no particular critical section code in that paths) or you meant to say that the clock_lock doesn't still provide enough protection alone? BTW, you were right about the lapic_timer_hz (I forgot to revert to hz). There is an updated patch: http://www.freebsd.org/~attilio/Sandvine/STABLE_8/statclock_aliasing/statcl= ock_aliasing4.diff Thanks, Attilio --=20 Peace can only be achieved by understanding - A. Einstein
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3bbf2fe11001150706y765159a2jbd37c7ae4cf378f0>