Date: Thu, 12 Aug 2010 00:21:52 +0200 From: Bengt Ahlgren <bengta@sics.se> To: freebsd-stable@freebsd.org Subject: Re: GEOM label tasting of windows disk hangs on 8.1 Message-ID: <uh739uk7ptr.fsf@P142.sics.se> In-Reply-To: <uh7eie47s8o.fsf@P142.sics.se> (Bengt Ahlgren's message of "Wed\, 11 Aug 2010 23\:29\:43 %2B0200") References: <uh7k4nxz6vb.fsf@P142.sics.se> <AANLkTi=v3Nvu13CjEe36gU3fo4UgHcx9Wgok2phngKSo@mail.gmail.com> <uh7eie47s8o.fsf@P142.sics.se>
next in thread | previous in thread | raw e-mail | index | archive | help
Bengt Ahlgren <bengta@sics.se> writes: > Marius N=FCnnerich <marius@nuenneri.ch> writes: > >> On Wed, Aug 11, 2010 at 14:11, Bengt Ahlgren <bengta@sics.se> wrote: >>> Hi! >>> >>> When I was trying to install 8.1-REL from the dvd install media on a >>> ThinkPad X40 that has the original Windows XP partitions on its disk, >>> the geom label tasting seems to hang with the disk activity light on. >>> I just submitted this PR: >>> >>> http://www.freebsd.org/cgi/query-pr.cgi?pr=3D149523 >>> >>> 7.3-REL works just fine. >>> >>> Is this a known issue? >>> >>> I assume that wiping the mbr will fix it, but I will defer that if >>> additional debugging is needed. >> >> Could you please send a hexdump of the first sector of that device? > > Certainly: > > 00000000 eb 0e 03 1e 01 00 90 02 00 00 00 00 00 00 4e 50 |............= ..NP| > 00000010 fa 33 c0 bc 00 7a 8e d0 50 07 50 1f fb fc bf 00 |.3...z..P.P.= ....| > 00000020 08 be 00 7c b9 00 01 f3 a5 ea 2e 08 00 00 32 ed |...|........= ..2.| > 00000030 bb 00 06 be 02 08 8a 0c b8 01 02 ba 80 00 cd 13 |............= ....| > 00000040 b9 05 00 bb 00 14 51 8b f1 32 ed 8a 8c ff 05 81 |......Q..2..= ....| > 00000050 eb 00 02 b8 01 02 ba 80 00 cd 13 59 e2 e8 b3 00 |...........Y= ....| > 00000060 be 99 05 88 1c be 98 05 88 1c be 97 05 88 1c eb |............= ....| > 00000070 0a b3 01 be 98 05 88 1c e9 9d 00 e8 36 00 3c 01 |............= 6.<.| > 00000080 74 ef e8 5c 00 3c 01 74 e8 e8 db 04 be 22 06 80 |t..\.<.t....= ."..| > 00000090 3c 00 0f 84 b3 02 80 3c 01 0f 84 63 01 80 3c 02 |<......<...c= ..<.| > 000000a0 0f 84 86 01 be 05 08 0a 04 88 04 b1 01 bb 00 08 |............= ....| > 000000b0 e8 db 00 c3 be 00 06 e8 17 00 be 23 06 80 3c 00 |...........#= ..<.| > 000000c0 74 0c 3c 00 74 08 b0 02 e8 d9 ff b0 01 c3 b0 00 |t.<.t.......= ....| > 000000d0 c3 4e 32 c0 b9 00 02 8b d9 8a 10 32 c2 e2 f8 46 |.N2........2= ...F| > 000000e0 c3 b9 01 00 51 b8 00 02 f7 e1 05 00 08 8b f0 e8 |....Q.......= ....| > 000000f0 df ff 5e 56 8a 8c 05 06 80 f9 00 74 17 38 c1 75 |..^V.......t= .8.u| > 00000100 0a 59 41 51 83 f9 06 74 0b 75 da 59 b0 01 e8 93 |.YAQ...t.u.Y= ....| > 00000110 ff b0 01 c3 59 b0 00 c3 32 ed be 07 08 8a 0c b8 |....Y...2...= ....| > 00000120 01 02 bb 00 7c ba 80 00 cd 13 8b f3 e8 a2 ff be |....|.......= ....| > 00000130 06 08 8a 24 80 fc 00 74 30 38 c4 74 2c b0 08 e8 |...$...t08.t= ,...| > 00000140 62 ff be af 07 e8 f0 01 be 0e 06 32 ed 8a 0c 80 |b..........2= ....| > 00000150 c1 01 51 b9 80 3e e8 22 00 59 e2 f6 be 98 05 80 |..Q..>.".Y..= ....| > 00000160 3c 01 74 03 e8 99 00 cd 18 be be 09 bf be 7d b9 |<.t.........= ..}.| > 00000170 20 00 f3 a5 eb 00 ea 00 7c 00 00 50 e4 61 24 10 | .......|..P= .a$.| > 00000180 8a e0 e4 61 24 10 38 e0 74 f8 e2 f4 58 c3 32 ed |...a$.8.t...= X.2.| > 00000190 b8 01 03 ba 80 00 cd 13 c3 be 05 06 8a 04 24 0c |............= ..$.| > 000001a0 c0 e8 02 c3 be 22 06 88 0c e8 6a 01 c3 00 00 00 |....."....j.= ....| > 000001b0 00 00 00 00 00 00 00 00 cd cc cd cc 00 00 80 01 |............= ....| > 000001c0 01 00 0c ef ff ff 3f 00 00 00 31 0c 25 04 00 00 |......?...1.= %...| > 000001d0 c1 ff 12 ef ff ff 70 0c 25 04 90 46 83 00 00 00 |......p.%..F= ....| > 000001e0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |............= ....| > 000001f0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 55 aa |............= ..U.| > 00000200 Hmm, wiping this first sector does not make any difference, nor wiping the first 100. If you look at the PR, it seems like the disk IO never completes, which could indicate an interrupt problem. I therefore tried disabling the apic (set hint.apic.0.disabled=3D1 at the loader prompt). That solved the problem! Well, using the apic never really worked on these Thinkpads anyway, but there must be some change in 8.x since the 7.x and earlier install media never had any problems booting. Bengt
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?uh739uk7ptr.fsf>