Date: Wed, 10 Apr 2013 20:09:32 +0000 From: Andrew Duane <aduane@juniper.net> To: Alfred Perlstein <bright@mu.org>, Benjamin Kaduk <kaduk@MIT.EDU> Cc: Wojciech Puchar <wojtek@wojtek.tensor.gdynia.pl>, Sebastian Feld <sebastian.n.feld@gmail.com>, "freebsd-hackers@freebsd.org" <freebsd-hackers@freebsd.org> Subject: RE: Multiple page size support on FreeBSD? Message-ID: <477C1270D3E5484DA2303CEBE274C9E1250B3DF7@CH1PRD0510MB392.namprd05.prod.outlook.com> In-Reply-To: <5165C4D7.3050308@mu.org> References: <CAHnbEGJuo9Jvskxaog0xLVM_LOse695b4E3fKae7YufOAVZuBg@mail.gmail.com> <alpine.GSO.1.10.1304072319300.9389@multics.mit.edu> <alpine.BSF.2.00.1304081238580.6013@wojtek.tensor.gdynia.pl> <201304101006.13960.jhb@freebsd.org> <alpine.BSF.2.00.1304101947470.19441@wojtek.tensor.gdynia.pl> <alpine.GSO.1.10.1304101438500.9389@multics.mit.edu> <5165C4D7.3050308@mu.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Like all "performance" items (especially VM), it depends on the hardware an= d the load. On systems with small TLBs it helps more than with large TLBs. = With software that needs access to lots of different areas the TLB gets mor= e traffic so large ones help more. The answer for your firefox browser box = with i386 is probably different from my compilation engine running MIPS, or= his web server running AMD. Back at Digital, we spent a lot of time trying to find the "one true answer= " to superpages, only to discover there wasn't one. We ended up with a comb= ination of automatic use from big allocations, a rarely used API to advise = for big TLBs, and some background work that coalesced when possible. =A0.................................... Andrew L. Duane Resident Architect - AT&T Technical Lead m=A0=A0=A0+1 603.770.7088 o +1 408.933.6944 (2-6944) skype: andrewlduane aduane@juniper.net -----Original Message----- From: owner-freebsd-hackers@freebsd.org [mailto:owner-freebsd-hackers@freeb= sd.org] On Behalf Of Alfred Perlstein Sent: Wednesday, April 10, 2013 4:00 PM To: Benjamin Kaduk Cc: Wojciech Puchar; Sebastian Feld; freebsd-hackers@freebsd.org Subject: Re: Multiple page size support on FreeBSD? On 4/10/13 11:42 AM, Benjamin Kaduk wrote: > On Wed, 10 Apr 2013, Wojciech Puchar wrote: > >>> How do your tests work? Do you examine PTEs directly to check for=20 >>> superpages or are you relying on the vm.pmap.pde sysctls? >> >> the later. >> >> anyway - algorithm described on list - that heuristics detects=20 >> consecutive page access doesn't really help the urgent case - RANDOM=20 >> access to large amount of memory. > > The algorithm is not a heuristic based on consecutive accesses,=20 > promotion occurs when the entire superpage's worth of memory has=20 > actually been accessed. If I remember correctly, the performance gain=20 > from superpages was only a few percent, so spending more time trying=20 > to decide when to use them would make the algorithm a net wash. > > You should really watch the talk I linked to if you're interested, it=20 > was quite interesting. > >> sequential access will get minimal improvement. >> >> IMHO the only way that really make sens is to add options to madvise=20 >> to give kernel information about usage. > > Maybe. It is cool that FreeBSD got this work via Alan Cox and the others that cont= ributed. I am wondering if it makes sense to have an explicit model. At one place, for a platform with high performance but a very small TLB, we= made it possible to explicitly request a large TLB for our process and it = made a BIG difference for performance. Sometimes being "general purpose" means that you can expose such low level = things for the user to tune instead of requiring them to fit the app to a h= euristic that may change. -Alfred > > -Ben Kaduk > _______________________________________________ > freebsd-hackers@freebsd.org mailing list=20 > http://lists.freebsd.org/mailman/listinfo/freebsd-hackers > To unsubscribe, send any mail to > "freebsd-hackers-unsubscribe@freebsd.org" > _______________________________________________ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/l= istinfo/freebsd-hackers To unsubscribe, send any mail to "freebsd-hackers-unsubscribe@freebsd.org"
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?477C1270D3E5484DA2303CEBE274C9E1250B3DF7>