Date: Mon, 03 Jan 2005 20:58:50 +0100 From: =?ISO-8859-1?Q?S=F8ren_Schmidt?= <sos@DeepCore.dk> To: Nate Lawson <nate@root.org> Cc: FreeBSD Current <freebsd-current@freebsd.org> Subject: Re: ATA rman performance enhancement Message-ID: <41D9A3FA.20204@DeepCore.dk> In-Reply-To: <41D9A2CF.30704@root.org> References: <41D984A5.7010408@root.org> <41D99FBD.8070500@DeepCore.dk> <41D9A2CF.30704@root.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Nate Lawson wrote: > Søren Schmidt wrote: > >> Nate Lawson wrote: >> >>> While doing some benchmarking of other code, I noticed that there >>> were a lot of calls to rman_get_bustag/handle(). They weren't taking >>> up much actual time since they're pretty lightweight but seemed to be >>> unnecessary. >>> >>> I worked up the attached diff and benchmarked it. There are about >>> 1000 calls a second to the rman routines without this patch and >>> essentially none with it. It makes about a 1% difference in >>> throughput under some IO loads. It is only for non-Promise or >>> non-SII controllers right now since I didn't extend the >>> initialization step to more than ata-pci.c. The same approach could >>> be used for the other INW/OUTW calls as well but they're not in the >>> fast path. I think it may make more of a difference with small reads. >> >> I had something semilar to this once back when, but since HW got lots >> faster I couldn't measure it anymore, but maybe things has changed... >> >> Anyhow it needs to be applied to ata-isa.c ata-card.c ata-cbus.c etc >> to not break anything at least. > > Yes, I agree. I limited the change only to the IDX macros that were > used in the fast path although it could apply to all of them. I *must* be applied to those files, otherwise the handle & tag fields are not initialized when using !pci based devices with you patch. >> I'll think about it and eventually do something about it in ATA-mkIII >> if it really is mesureable again.. > > It is a minimal difference but with small transfers on a fast drive with > a low-powered CPU, it's measurable. I made the same change a while back > in acpi_timer.c since the resource is only allocated once but accessed > very often and in a low-latency context. Right, but I'd like to get the much bigger fishes catched first before doing micro optims, thats what ATA-mkIII is all about in the first place. However I'll put it back on the TODO list as something to look into when the time comes... -- -Søren
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?41D9A3FA.20204>