maximum MAXBSIZE

Wojciech Puchar wojtek at puchar.net
Thu Jan 9 19:42:45 UTC 2020


>
>       POLA (principle of least amazement).  I certainly don't need a MAXPHYS set
>       to 4MB on my desktop machine.  Not everyone using FreeBSD is running
>       servers with large amounts of memory and disk storage.
>
>       It's a trivial change if it's beneficial in a certain use scenario.  The
>       decision should be left up to the user.
> 
> 
> And if you change MAXPHYS, you'll also want to bump the insanely small runningbuf limits. And there may be a few other parameters

of course i did

> I've not reported that we set that are critical too :)
> 
> Warner
>  
>       >
>       > On Thu, 9 Jan 2020, Warner Losh wrote:
>       >
>       > > Netflix runs our entire network at MAXPHYS=8MB since we're doing huge reads off HDD.
>       > > Warner
>       > >
>       > >
>       > > On Thu, Jan 9, 2020 at 2:58 AM Wojciech Puchar <wojtek at puchar.net> wrote:
>       > >       2MB MAXPHYS was what i have set for over 3 years without problems.
>       > >
>       > >       On Wed, 8 Jan 2020, Conrad Meyer wrote:
>       > > 
>       > >       > Bufs are dynamically allocated from uma now, and perhaps a middle ground BSIZE is worth considering? Would
>       1MB and 2kB 
>       > >       bufs (1kB 
>       > >       > 32-bit) be awful?
>       > >       >
>       > >       > Cheers,
>       > >       > Conrad__
>       > >       >
>       > >       > On Wed, Jan 8, 2020 at 06:18 Konstantin Belousov <kostikbel at gmail.com> wrote:
>       > >       >__ __ __ __On Wed, Jan 08, 2020 at 02:52:57PM +0100, Wojciech Puchar wrote:
>       > >       >__ __ __ __> sorry i made a mistake - i change MAXPHYS not MAXBSIZE.
>       > >       >__ __ __ __>
>       > >       >__ __ __ __> 16MB works for now without problems
>       > >       >__ __ __ __MAXPHYS 16MB means that sizeof(struct buf) is around 32K (16K on 32bit).
>       > >       >
>       > >       >__ __ __ __>
>       > >       >__ __ __ __> On Wed, 8 Jan 2020, Gary Jennejohn wrote:
>       > >       >__ __ __ __>
>       > >       >__ __ __ __> > On Tue, 7 Jan 2020 22:47:54 +0000
>       > >       >__ __ __ __> > Rick Macklem <rmacklem at uoguelph.ca> wrote:
>       > >       >__ __ __ __> >
>       > >       >__ __ __ __> > > Hans Petter Selasky wrote:
>       > >       >__ __ __ __> > > > On 2020-01-07 22:12, Wojciech Puchar wrote:
>       > >       >__ __ __ __> > > > > default MAXBSIZE is 128kB. badly low for todays magnetic disks.
>       > >       >__ __ __ __> > > > >
>       > >       >__ __ __ __> > > > > i have it set to 2MB on all computers that have magnetic disks. Great
>       > >       >__ __ __ __> > > > > improvement with large files. especially when more than one are
>       > >       >__ __ __ __> > > > > read/wrote in parallel. And no problems experienced
>       > >       >__ __ __ __> > > > >
>       > >       >__ __ __ __> > > > > But for optimal performance MAXBSIZE should be transfered in few times
>       > >       >__ __ __ __> > > > > longer than average seek time. todays disk do 200-250MB/s so 2MB is
>       > >       >__ __ __ __> > > > > transfered below 10ms.
>       > >       >__ __ __ __> > > > >
>       > >       >__ __ __ __> > > > > 8-16MB seems like good choice. is there any reason not to set it that high?
>       > >       >__ __ __ __> > > >
>       > >       >__ __ __ __> > > > Old disk may not support it, especially USB 1.0/2.0 disks.
>       > >       >__ __ __ __> > > I also thought it was limited to MAXPHYS, but maybe I'm only thinking of the NFS
>       > >       >__ __ __ __> > > specific case?
>       > >       >__ __ __ __> > >
>       > >       >__ __ __ __> >
>       > >       >__ __ __ __> > There's a comment in param.h that it should not exceed MAXPHYS to be
>       > >       >__ __ __ __> > on the safe side.__ How old that comment is I can't say and that may
>       > >       >__ __ __ __> > not be the case today.
>       > >       >__ __ __ __> >
>       > >       >__ __ __ __> > MAXBSIZE is only 64KiB in my param.h.
>       > >       >__ __ __ __> >
>       > >       >__ __ __ __> > I have to agree with HPS.__ There are many old bridge-chips still in
>       > >       >__ __ __ __> > use and problems with a large MAXBSIZE might occur.__ It's certainly
>       > >       >__ __ __ __> > not uncommon to see capacity limitations - I have a docking station
>       > >       >__ __ __ __> > which can't see more than 3TB.
>       > >       >__ __ __ __> >
>       > >       >__ __ __ __> > --
>       > >       >__ __ __ __> > Gary Jennejohn
>       > >       >__ __ __ __> >
>       > >       >__ __ __ __> >
>       > >       >__ __ __ __> _______________________________________________
>       > >       >__ __ __ __> freebsd-hackers at freebsd.org mailing list
>       > >       >__ __ __ __> https://lists.freebsd.org/mailman/listinfo/freebsd-hackers
>       > >       >__ __ __ __> To unsubscribe, send any mail to "freebsd-hackers-unsubscribe at freebsd.org"
>       > >       >__ __ __ _________________________________________________
>       > >       >__ __ __ __freebsd-hackers at freebsd.org mailing list
>       > >       >__ __ __ __https://lists.freebsd.org/mailman/listinfo/freebsd-hackers
>       > >       >__ __ __ __To unsubscribe, send any mail to "freebsd-hackers-unsubscribe at freebsd.org"
>       > >       >
>       > >       >
>       > >       > 
>       > >       _______________________________________________
>       > >       freebsd-hackers at freebsd.org mailing list
>       > >       https://lists.freebsd.org/mailman/listinfo/freebsd-hackers
>       > >       To unsubscribe, send any mail to "freebsd-hackers-unsubscribe at freebsd.org"
>       > >
>       > >
>       > > 
>       > _______________________________________________
>       > freebsd-hackers at freebsd.org mailing list
>       > https://lists.freebsd.org/mailman/listinfo/freebsd-hackers
>       > To unsubscribe, send any mail to "freebsd-hackers-unsubscribe at freebsd.org"
> 
>
>       --
>       Gary Jennejohn
> 
> 
>


More information about the freebsd-hackers mailing list