From owner-freebsd-stable Thu May 16 7: 6:19 2002 Delivered-To: freebsd-stable@freebsd.org Received: from duke.cs.duke.edu (duke.cs.duke.edu [152.3.140.1]) by hub.freebsd.org (Postfix) with ESMTP id 7E98137B43C for ; Thu, 16 May 2002 06:35:54 -0700 (PDT) Received: from grasshopper.cs.duke.edu (grasshopper.cs.duke.edu [152.3.145.30]) by duke.cs.duke.edu (8.9.3/8.9.3) with ESMTP id JAA08537; Thu, 16 May 2002 09:35:50 -0400 (EDT) Received: (from gallatin@localhost) by grasshopper.cs.duke.edu (8.11.6/8.9.1) id g4GDZJH40147; Thu, 16 May 2002 09:35:19 -0400 (EDT) (envelope-from gallatin@cs.duke.edu) From: Andrew Gallatin MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <15587.46487.912732.71641@grasshopper.cs.duke.edu> Date: Thu, 16 May 2002 09:35:19 -0400 (EDT) To: Holger Kipp Cc: Phil Rosenthal , FreeBSD-STABLE Mailing List , Andrew Gallatin Subject: Re: ECC In-Reply-To: <3CE37562.63BFB3E8@alogis.com> References: <00a201c1fc66$23231510$0700a8c0@zoom> <3CE37562.63BFB3E8@alogis.com> X-Mailer: VM 6.75 under 21.1 (patch 12) "Channel Islands" XEmacs Lucid Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG I'm coming to the conclusion that the ecc port is broken for serverworks chipsets. I was hopeful, but since serverworks doesn't provide any technical docs for its chipset, its not terribly surprising. Sigh. There is some other code from Sun for its Cobalt appliances that suggests the ecc port may be totally misusing register 0x94. http://www.geocrawler.com/archives/3/35/2001/5/0/5870259/ (see cobalt_nmi()). However, this code looks very suspicious (lots of ">" where there should be ">>", etc). I plugged the code into ecc.c & I ran with it yesterday & it claimed I got several uncorrectable errors. I'm not sure if I beleive it. I think serverworks chipset support might be hopeless.. Drew Holger Kipp writes: > Phil Rosenthal wrote: > > > > > http://people.freebsd.org/~gallatin/ecc-0.14-freebsd.tgz > > > > I set this up on a Dell 1550 with 2GB of ram. > > > > I have this system set to MAXMEM 1800MB because the system panics (I guess > > 2GB support isnt really here yet) when I run my software on it with 2GB. > > > > here is what it says when I run it: > > ECC: monitor version 0.14 (Oct 10 2001) > > ECC: vendor 0x1166 > > ECC: device 0x8 > > ECC: Chipset ECC capability - ECC with hardware scrubber > > ECC: Current ECC mode - ECC detection and correction > > ECC: Bank Size Type ECC SBE MBE > > ECC: 3 1008M SDR Y 0 0 > > ECC: Total 1008M > > Me too, sort of: > Dell PowerEdge 2500. ECC-monitor 0.14 is only recognizing 1GB instead > of 2GB. See dmesg output below. Hardware available, me in front - willing > to test/debug. Directions needed, though ;-) > > Another hardware (currently complaining about 1 SBE, 1MBE) with 3x512M also > available - might be better suited for testing: > > ECC: monitor version 0.14 (Oct 10 2001) > ECC: vendor 0x1166 > ECC: device 0x9 > ECC: MBE Detected in DRAM row 6 > ECC: SBE Detected in DRAM row 6 > ECC: Chipset ECC capability - ECC with hardware scrubber > ECC: Current ECC mode - ECC with hardware scrubber > ECC: Bank Size Type ECC SBE MBE > ECC: 2 512M SDR Y 0 0 > ECC: 4 512M SDR Y 0 0 > ECC: 6 512M SDR Y 1 1 > ECC: Total 1536M > ECC: MBE Detected in DRAM row 6 > ECC: SBE Detected in DRAM row 6 > > How is the Bank # generated? This board only has 4 memory slots... > > Regards, > Holger > > output of dmesg (Dell 2500): > --- 8< ----------------------------------------------------------------- > FreeBSD 4.6-PRERELEASE #9: Mon May 6 13:49:01 CEST 2002 > root@majestix:/usr/obj/usr/src/sys/MAJESTIX > Timecounter "i8254" frequency 1193182 Hz > CPU: Pentium III/Pentium III Xeon/Celeron (997.46-MHz 686-class CPU) > Origin = "GenuineIntel" Id = 0x68a Stepping = 10 > Features=0x383fbff OV,PAT,PSE36,MMX,FXSR,SSE> > real memory = 2147418112 (2097088K bytes) > avail memory = 2088062976 (2039124K bytes) > Changing APIC ID for IO APIC #0 from 0 to 2 on chip > Changing APIC ID for IO APIC #1 from 0 to 3 on chip > Programming 16 pins in IOAPIC #0 > Programming 16 pins in IOAPIC #1 > FreeBSD/SMP: Multiprocessor motherboard > cpu0 (BSP): apic id: 1, version: 0x00040011, at 0xfee00000 > cpu1 (AP): apic id: 0, version: 0x00040011, at 0xfee00000 > io0 (APIC): apic id: 2, version: 0x000f0011, at 0xfec00000 > io1 (APIC): apic id: 3, version: 0x000f0011, at 0xfec01000 > Preloaded elf kernel "kernel" at 0xc038a000. > Pentium Pro MTRR support enabled > md0: Malloc disk > [...] > APIC_IO: routing 8254 via 8259 and IOAPIC #0 intpin 0 > SMP: AP CPU #1 Launched! > acd0: DVD-ROM at ata0-master PIO4 > Waiting 3 seconds for SCSI devices to settle > amrd0: on amr0 > amrd0: 17278MB (35385344 sectors) RAID 1 (optimal) > amrd1: on amr0 > amrd1: 104034MB (213061632 sectors) RAID 5 (optimal) > Mounting root from ufs:/dev/amrd0s1a > ECC: monitor version 0.14 (Oct 10 2001) > ECC: vendor 0x1166 > ECC: device 0x8 > ECC: Chipset ECC capability - ECC with hardware scrubber > ECC: Current ECC mode - ECC detection and correction > ECC: Bank Size Type ECC SBE MBE > ECC: 3 1024M SDR Y 0 0 > ECC: Total 1024M > ECC: unloaded. > > > -- > Holger Kipp, Dipl.-Math., Systemadministrator | alogis AG > Fon: +49 (0)30 / 43 65 8 - 114 | Berliner Strasse 26 > Fax: +49 (0)30 / 43 65 8 - 214 | D-13507 Berlin Tegel > email: holger.kipp@alogis.com | http://www.alogis.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message