Date: Sat, 13 Nov 2004 11:24:13 +0100 From: =?ISO-8859-1?Q?S=F8ren_Schmidt?= <sos@DeepCore.dk> To: Scott Long <scottl@freebsd.org> Cc: Garance A Drosihn <drosih@rpi.edu> Subject: Re: 5.3-RELEASE: WARNING - WRITE_DMA interrupt timout Message-ID: <4195E0CD.40608@DeepCore.dk> In-Reply-To: <4195DD31.10201@freebsd.org> References: <25539.1100339599@critter.freebsd.dk> <4195DB3E.2040807@DeepCore.dk> <4195DD31.10201@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Scott Long wrote: > S=F8ren Schmidt wrote: >=20 >> Poul-Henning Kamp wrote: >> >>> In message <4195D903.2090801@DeepCore.dk>,=20 >>> =3D?ISO-8859-1?Q?S=3DF8ren_Schmidt?=3D wri >>> tes: >>> >>>> Zoltan Frombach wrote: >>>> >>>>> This is still an issue for me. Please read this post of mine: >>>>> http://lists.freebsd.org/pipermail/freebsd-stable/2004-November/009= 420.html=20 >>>>> >>>>> >>>>> Can anyone help? I would gladely install test patches to track this= =20 >>>>> problem down. My system is 5.3-R. And the WRITE_DMA warning happens= =20 >>>>> at least twice a day, it is so predictable. With thanks, >>>> >>>> >>>> >>>> Hmmm, that warning is issued from ATA when requests has been=20 >>>> returned to the systems bio_taskqueue but the system hasn't finished= =20 >>>> them within the timeout. Now this is an indication of the system=20 >>>> being unresponsive already at that point, or at least that was the=20 >>>> idea. >>>> It has nothing to do with a bad drive, since the interrupt was seen = >>>> the drive has finished the request it was asked, its the layers=20 >>>> above ATA that doesn't respond to the request beeing returned as=20 >>>> finished. >>> >>> It is not really the task of the ata driver to fail requests at that >>> time. How long is the timeout anyway ? >> >> Oh, ATA doesn't fail them, it just yells that the request hasn't been = >> finished yet by the upper layers, it doesn't do anything to the reques= t. >> >> Timeout is 5 secs, which is a pretty long time in this context IMHO.. >> >> However, if it can take that long time to get data pushed up the=20 >> chain, it might also explain some of the reduced I/O performance=20 >> reported ? >> > I'm able to get 11,000+ transactions/sec and achieve PCI line rate with= > the aac driver and a good version of an aac controller. aac is a block= =20 > driver just like ata, the only differnece is that it uses a fast=20 > interrupt handler to acknowledge the hardware, and batch-processes the > completions in a mpsafe (!Giant) taskqueue. I don't see any performanc= e > problems with this, and in fact it performs quite a bit better than > Linux and even Windows. I see _no_ taskqueue stalls or other strange > timing or synchronization problems. Well, with sufficiently fast ATA hardware you can easily get that many=20 transactions through ATA as well, so there is nothing performance wise=20 hindering that at all. --=20 -S=F8ren
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4195E0CD.40608>