Date: Tue, 18 Jan 2022 11:25:46 +0300 From: Dima Panov <fluffy@FreeBSD.org> To: Willem Jan Withagen <wjw@digiware.nl>, Tomoaki AOKI <junchoon@dec.sakura.ne.jp> Cc: Eugene Grosbein <eugen@grosbein.net>, stable@freebsd.org Subject: Re: Trying to boot a supermicro H8DMT board Message-ID: <b821a194-a87f-456e-8339-84d45ef9f8da@Canary> In-Reply-To: <8096cd7e-bc11-5fa7-cc96-6bcdf1278ffc@digiware.nl> References: <8ac447b6-eaaf-0a8f-da69-27db15dd6f55@digiware.nl> <2ec39eef-d2e2-c55e-b032-43de86e71a57@digiware.nl> <3d87a0b3-7bed-453b-df23-4a258ea46fbb@grosbein.net> <d8e6c746-3ec1-9c21-d5e7-44dc9600bb0b@digiware.nl> <802cf542-979d-b8e1-3f71-616b026eb852@grosbein.net> <48f57581-1f39-9f57-0e44-19c2c2bb3aeb@digiware.nl> <a0315a54-aefa-a3a3-2ac3-94d6e9410961@grosbein.net> <eac3dcef-9183-8fa5-b0de-b70650235960@digiware.nl> <78a47e83-a339-0c79-0ee0-9e55be80c78b@grosbein.net> <d0dbce19-f5e5-3ff0-99d6-55a9c94a4b48@digiware.nl> <2f49fd20-cb5a-5ccc-7f9b-0229bc8e14b1@grosbein.net> <86766549-be58-1125-867e-ae4c415e1bb4@digiware.nl> <7903a41f-94ba-2caf-9270-a1bd9582c600@grosbein.net> <229c3042-3297-7903-9778-9b55d5c3f998@digiware.nl> <71d1e25c-f1f6-2371-486e-2382d67a3fc5@grosbein.net> <c6588210-ac68-f081-f0a4-85669aa84eb3@digiware.nl> <9d73e9ba-af23-ea90-e5fa-cf3a04a8513b@grosbein.net> <7c5d9cc0-be85-c855-a294-71a93f2c5440@digiware.nl> <20220118021432.197aa1241d53b1cba6e8c562@dec.sakura.ne.jp> <8096cd7e-bc11-5fa7-cc96-6bcdf1278ffc@digiware.nl>
next in thread | previous in thread | raw e-mail | index | archive | help
--61e6798a_327b23c6_93a Content-Type: multipart/alternative; boundary="61e6798a_6b8b4567_93a" --61e6798a_6b8b4567_93a Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline Moin=21 As mobo manual says, u could use cpu1/1b, cpu1/1a, cpu2/1b, cpu2/1a due t= o hard dependency on pair interleaving support. Slots fills from far to close order from each cpu by pairs only -- Dima. (desktop, kde, x11, office, ports-secteam)=40=46reeBSD team (fluffy=40=46reeBSD.org, https://t.me/dima=5Fpanov) > On Monday, Jan 17, 2022 at 9:47 PM, Willem Jan Withagen <wjw=40digiware= .nl (mailto:wjw=40digiware.nl)> wrote: > On 17-1-2022 18:14, Tomoaki AOKI wrote: > > On Mon, 17 Jan 2022 15:04:16 +0100 > > Willem Jan Withagen <wjw=40digiware.nl> wrote: > > > > > On 17-1-2022 14:46, Eugene Grosbein wrote: > > > > 17.01.2022 20:24, Willem Jan Withagen wrote: > > > > > > > > > > Well, perform independent hardware (memory) testing with some= thing like memtest86+ > > > > > > and if it is all right, you show ask someone more knowledgeab= le. Maybe CC: arch=40freebsd.org > > > > > Perhaps should have done that when I started, but supplier assu= red me that > > > > > the they just retired the boards with out any issues. > > > > > Memtest86 found the faulty DIMM in 30 secs... > > > > > > > > > > Not sure if we could/want educate vm=5Fmem=5Finit() to actually= detect this. > > > > > It is still in the part where everthing is still running on the= first CPU. > > > > > Making things a bit easier to understand what is going on. > > > > > > > > > > Lets see if the box will run on 3 DIMMs for the rime being. > > > > > Then figure out with DMIdecode what we need expand again. > > > > Is it ECC memory or non-ECC=3F > > > > The kernel already have full memory testing performed at boot tim= e > > > > unless disabled with another loader knob: > > > > > > > > hw.memtest.tests=3D0 > > > > > > > > Try booting it with memory testing disabled and without hw.physme= m limitation. > > > > Maybe it will boot. > > > > > > > > With ECC, it could be hardware interrupt while kernel runs that t= est > > > > and wrong in-kernel processing of the interrupt. > > > Swapped the DIMM with 3 others, but still the same errors. > > > Then I changed DIMM slot, and the errors went away. > > > So definitely a hardware issue > > > > > > when booted =46reeBSD reported already only 12Gb in system ( there = are 4 > > > 4GB dimms) > > > Using 8Gb. DIMMs are ECC. > > > But then still it would only boot when mem set to 8G. > > > > > > Waiting for memtest to finish at least one pass. > > > Usually that will take quite some time. > > > > > > --WjW > > > > > > > > Not sure this is the case, but some motherboards have severe limitati= on > > about DIMM slot usage, if not fully used. > > > > =46or example, assuming slot No. are B0-0, 1, 2, 3 and B1-0, 1, 2, 3,= > > > > *Must use =22interleaved. If 4 in 8 slots are to be used, > > B0-0, B0-2, B1-0, B1-2 shall be used. > > (Some forced B0-1, B0-3, B1-1, B1-3, IIRC) > > > > *Must NOT use =22interleaved. > > B0-0, B0-1, B1-0, B1-1 shall be used. > > > > *Must NOT use B1 unless B0 is full of DIMs. > > B0-0. B0-1, B0-2, B0-3 shall be used. > > > > and so on, depending on motherboard vendor (at worst, per model.) > > Yup, I know... I used the board in the configuration I got it. > And its a DUAL processor board with 2 opterons. > The config works correct for the first Opteron (Called CPU1) > using slots: CPU1/DIMM1A and CPU1/DIMM1B > But on the second CPU I have to use the third slot.... > so using slots: CPU2/DIMM1B and CPU2/DIMM2B > > And my memtest86 has complete 1 full pass over 16G without errors. > So I'm guessing that the order is not majorly picky. > > But you are correct in noting this, so I will read up ont this in the > manual. > > Thanx, > --WjW > > --61e6798a_6b8b4567_93a Content-Type: text/html; charset="utf-8" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline <html xmlns=3D=22http://www.w3.org/1999/xhtml=22><head>; <title></title> <= meta name=3D=22viewport=22 content=3D=22width=3Ddevice-width, initial-sca= le=3D1.0, user-scalable=3Dno=22> </head> <body><img id=3D=22A8=4630934D1=46= 604=4637BC577051=46E521=46=46=22 alt=3D=22=22 width=3D=220px=22 src=3D=22= https://read-receipts.canarymail.io:8100/track/2BD3A57834=462C18772CBE819= A75674B4=5FA8=4630934D1=46604=4637BC577051=46E521=46=46.png=22 height=3D=22= 0px=22><div id=3D=22CanaryBody=22> <div> Moin=21</div><div><br></div><div= >As mobo manual says, u could use cpu1/1b, cpu1/1a, cpu2/1b, cpu2/1a due = to hard dependency on pair interleaving support.=C2=A0</div><div>Slots fi= lls from far to close order from each cpu by pairs only</div> <div><br></= div> </div> <div id=3D=22CanarySig=22> <div> <div style=3D=22font-family:= Helvetica;=22>--<br><div>Dima. (desktop, kde, x11, office, ports-secteam)= =40=46reeBSD team</div><div>(fluffy=40=46reeBSD.org, https://t.me/dima=5F= panov)</div></div> <div><br></div> </div> </div> <div id=3D=22CanaryDropb= ox=22> </div> <blockquote id=3D=22CanaryBlockquote=22> <div> <div>On Mond= ay, Jan 17, 2022 at 9:47 PM, Willem Jan Withagen <<a href=3D=22mailto:= wjw=40digiware.nl=22>wjw=40digiware.nl</a>> wrote:<br></div> <div>On 1= 7-1-2022 18:14, Tomoaki AOKI wrote: <br><blockquote type=3D=22cite=22>On = Mon, 17 Jan 2022 15:04:16 +0100 <br>Willem Jan Withagen <wjw=40digiwar= e.nl> wrote: <br> <br><blockquote type=3D=22cite=22>On 17-1-2022 14:46= , Eugene Grosbein wrote: <br><blockquote type=3D=22cite=22>17.01.2022 20:= 24, Willem Jan Withagen wrote: <br> <br><blockquote type=3D=22cite=22><bl= ockquote type=3D=22cite=22>Well, perform independent hardware (memory) te= sting with something like memtest86+ <br>and if it is all right, you show= ask someone more knowledgeable. Maybe CC: arch=40freebsd.org <br></block= quote>Perhaps should have done that when I started, but supplier assured = me that <br>the they just retired the boards with out any issues. <br>Mem= test86 found the faulty DIMM in 30 secs... <br> <br>Not sure if we could/= want educate vm=5Fmem=5Finit() to actually detect this. <br>It is still i= n the part where everthing is still running on the first CPU. <br>Making = things a bit easier to understand what is going on. <br> <br>Lets see if = the box will run on 3 DIMMs for the rime being. <br>Then figure out with = DMIdecode what we need expand again. <br></blockquote>Is it ECC memory or= non-ECC=3F <br>The kernel already have full memory testing performed at = boot time <br>unless disabled with another loader knob: <br> <br>hw.memte= st.tests=3D0 <br> <br>Try booting it with memory testing disabled and wit= hout hw.physmem limitation. <br>Maybe it will boot. <br> <br>With ECC, it= could be hardware interrupt while kernel runs that test <br>and wrong in= -kernel processing of the interrupt. <br></blockquote>Swapped the DIMM wi= th 3 others, but still the same errors. <br>Then I changed DIMM slot, and= the errors went away. <br>So definitely a hardware issue <br> <br>when b= ooted =46reeBSD reported already only 12Gb in system ( there are 4 <br>4G= B dimms) <br>Using 8Gb. DIMMs are ECC. <br>But then still it would only b= oot when mem set to 8G. <br> <br>Waiting for memtest to finish at least o= ne pass. <br>Usually that will take quite some time. <br> <br>--WjW <br> = <br> <br></blockquote>Not sure this is the case, but some motherboards ha= ve severe limitation <br>about DIMM slot usage, if not fully used. <br> <= br>=46or example, assuming slot No. are B0-0, 1, 2, 3 and B1-0, 1, 2, 3, = <br> <br>*Must use =22interleaved. If 4 in 8 slots are to be used, <br>B0= -0, B0-2, B1-0, B1-2 shall be used. <br>(Some forced B0-1, B0-3, B1-1, B1= -3, IIRC) <br> <br>*Must NOT use =22interleaved. <br>B0-0, B0-1, B1-0, B1= -1 shall be used. <br> <br>*Must NOT use B1 unless B0 is full of DIMs. <b= r>B0-0. B0-1, B0-2, B0-3 shall be used. <br> <br>and so on, depending on = motherboard vendor (at worst, per model.) <br></blockquote> <br>Yup, I kn= ow... I used the board in the configuration I got it. <br>And its a DUAL = processor board with 2 opterons. <br>The config works correct for the fir= st Opteron (Called CPU1) <br>using slots: CPU1/DIMM1A and CPU1/DIMM1B <br= >But on the second CPU I have to use the third slot.... <br>so using slot= s: CPU2/DIMM1B and CPU2/DIMM2B <br> <br>And my memtest86 has complete 1 f= ull pass over 16G without errors. <br>So I'm guessing that the order is n= ot majorly picky. <br> <br>But you are correct in noting this, so I will = read up ont this in the <br>manual. <br> <br>Thanx, <br>--WjW <br> <br> <= br></div> </div> </blockquote> </body></html> --61e6798a_6b8b4567_93a-- --61e6798a_327b23c6_93a Content-Type: application/pgp-signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: Canary PGP V3 iQJVBAABCgA/OBxEaW1hIFBhbm92IChGcmVlQlNELk9SRyBDb21taXR0ZXIpIDxm bHVmZnlARnJlZUJTRC5PUkc+BQJh5nmKAAoJEPuLoJ3VOY8p4p0QANZ8ZB2cnK2J YC5yOE2A/T6Mw3L+efPTzg83KYTDxeW4+MpDUzJv4ubmQDUF2R8qs4GPZXQmdtPW dvQ9Gjr909LQV2DiJo6jn9LXx+9yLuwDFZwp3P10p/hATakwQiwZxp2qrqQvVgBb 7oXgWe7Y379dnNc1LGg/GS4rbNQ3WCKLM6BMyAyKYXqWFvD5PDG5w4vIR9g7bszh L0q2oR7y3F2w4Icn7We7OmG0saZl7YQZ8GFdIiiKLUhc7R+kxJvRZErPUwECHei7 IDU5TXijPQg9Bc5JbJXcrFSxBL9XwGHKy/m9xw/OGFe0Gj4Qo9zsT0W46VT1smJS rF04EI5gowgB9XYI9YAaeGFEq7I8ywRBHQMABXKfJIq3ltyTY7m5JevDnSHKdPUG ccEdAQnHcM59z/J1fvwoebq5tKvMDre613fTIv4x3aFNi8O+sfpjIyME1EfcQ8p3 DEVdm7L94YBActUgQ6do3gs5MqV2rIhcnWrfXtsGhZWIx/53SIL4q7kRqsfJHdso 9qEu//nRT/sA6Cd1dj7xCJKtw53aB5UpZ86xhpgYUd2SkB/scV12A90UVY3DekBp Kj1WSllgXkVs3FVu3S5CQ4Qqb1XQS9sSvWLEdqLrI0PhYV8Iby25Ze2wTDcp8NAt yCjqHabd8AiYDyHs82KuAFdae91aIgvk =73Jv -----END PGP SIGNATURE----- --61e6798a_327b23c6_93a--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?b821a194-a87f-456e-8339-84d45ef9f8da>