Date: Fri, 18 Jul 2014 16:18:00 +0530 From: Kashyap Desai <kashyap.desai@avagotech.com> To: Alexander Motin <mav@freebsd.org> Cc: FreeBSD-scsi <freebsd-scsi@freebsd.org> Subject: RE: SSDs peformance on head/freebsd-10 stable using FIO Message-ID: <ad836df554a2b04482be624d319007cb@mail.gmail.com> In-Reply-To: <53C3A195.4020400@FreeBSD.org> References: <8fbe38cdad1e66717a9de7fdf63812c2@mail.gmail.com> <53BE8784.8060503@FreeBSD.org> <9f138f242e278476e5c542d695e58bc8@mail.gmail.com> <53BF1E6C.5030806@FreeBSD.org> <a4e86127552716ba989836fbcfc7676b@mail.gmail.com> <53C3A195.4020400@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
> -----Original Message----- > From: Alexander Motin [mailto:mavbsd@gmail.com] On Behalf Of Alexander > Motin > Sent: Monday, July 14, 2014 2:54 PM > To: Kashyap Desai > Cc: FreeBSD-scsi > Subject: Re: SSDs peformance on head/freebsd-10 stable using FIO > > On 14.07.2014 11:36, Kashyap Desai wrote: > > From: Alexander Motin [mailto:mavbsd@gmail.com] On Behalf Of > Alexander > >> First thing I noticed in this profile output is bunch of TLB shutdowns. > >> You can not reach reasonable performance from user-level without > >> having HBA support unmapped I/O. Both mps and mpr drivers support it, > >> but for some reason still not mrsas. Even at non-peak I/O rates on > >> multi-core system TLB shutdowns in such case can eat additional 30% > >> of CPU time. > > > > Thanks.! For this part, I can try In mrsas. Can you help me to > > understand what you mean by unmapped I/O ? > > That is a capability to work with data not mapped into the kernel virtual > address space, i.e. to work with physical addresses instead of virtual. > Main > prerequisite to support that is that driver should not try to access the > transferred data (because it can't do it for addresses not mapped to KVA). > If > that is true, then usually only minor modification is needed to teach the > driver to receive physical addresses from CAM. > > Looking on mps driver as example you may see PIM_UNMAPPED flag > reporting unmapped I/O support to CAM, and bus_dmamap_load_ccb() > helper function transparently doing all the physical address handling > magic. Thanks Motin.. I got the what is an issue here about UNMAPPED IO. Konstantin Belousov posted below performance result and there was also a changes in <mfi> to use Unmapped IO. https://kib.kiev.ua/kib/pgsql_perf.pdf I will work on this (will refer mpt driver ) and get back to FreeBSD Developers to seek any help.. ` Kashyap > > -- > Alexander Motin
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?ad836df554a2b04482be624d319007cb>