Date: Sun, 28 Sep 2014 14:41:08 +0200 From: "Ronald Klop" <ronald-lists@klop.ws> To: freebsd-arm@freebsd.org, "Mattia Rossi" <mattia.rossi.mate@gmail.com> Subject: Re: Random Kernel Panic on Dreamplug (FS related) Message-ID: <op.xmwdiud6kndu52@82-171-231-144.ip.telfort.nl> In-Reply-To: <5427DB19.4000304@gmail.com> References: <542559BC.7090100@gmail.com> <op.xmt8hvspkndu52@pinky.home> <5427DB19.4000304@gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, 28 Sep 2014 11:55:37 +0200, Mattia Rossi <mattia.rossi.mate@gmail.com> wrote: > On 27/09/14 10:57, Ronald Klop wrote: >> On Fri, 26 Sep 2014 14:19:08 +0200, Mattia Rossi >> <mattia.rossi.mailinglists@gmail.com> wrote: >> >>> This might be part of the weird FFS issues the Dreamplug has and >>> no-one knows why they're happening. >> >> I don't know if it is related, but my Sheevaplug also has issues with >> ffs while running 11-CURRENT. The fs gets corrupted or something. Which >> is not fixed by fsck. Every run of fsck finds more unlinked files and >> removes them. Also files which are stable on the fs since installation >> like /lib/*. > I've had a discussion with Ian Lepore about this, and he actually found > out that fsck is broken. >> This ffs corruption + panic mostly happened while installing ports on >> the first day of operation. >> >> But, the day before yesterday I compiled with gcc again instead of >> clang and it seems to run stable for 2 days now. > This is interesting though.. How do I switch to gcc as compiler for > world? I'd like to test that, and to see if it's really a clang problem. I put this in the environment. export WITH_GCC=yes export WITH_GNUCXX=yes export WITH_GCC_BOOTSTRAP=yes export WITHOUT_CLANG=yes export WITHOUT_CLANG_IS_CC=yes export WITHOUT_CLANG_BOOTSTRAP=yes Got that from this email: http://lists.freebsd.org/pipermail/freebsd-arm/2014-August/009001.html Regards, Ronald. >> >> Unfortunately I don't have the backtraces of the crashes anymore. >> >> NB: running 11-CURRENT from usb-stick with ports mounted via nfs. > Well, it's pretty much the same I have, only that the dreamplug has an > internal sd card which is attached via usb. I have ports on nfs as well. >> >> Ronald. >> >> >>> The panic occurred while running nsd-control reload (which should >>> simply re-read a config file from disk). I was previously editing >>> files without issues. >>> >>> Result is the following: >>> >>> vm_fault(0xc10a0000, d0238000, 2, 0) -> 2 >>> Fatal kernel mode data abort: 'Permission Fault (P)' >>> trapframe: 0xde019898 >>> FSR=0000000f, FAR=d0238120, spsr=20000013 >>> r0 =d0238120, r1 =00000e60, r2 =00000000, r3 =00000000 >>> r4 =00000120, r5 =00000000, r6 =c3f3f6c0, r7 =00001000 >>> r8 =c443e880, r9 =00000000, r10=c3d69000, r11=de019a20 >>> r12=d0238120, ssp=de0198e8, slr=c0d53828, pc =c0de521c >>> >>> [ thread pid 21116 tid 100073 ] >>> Stopped at memset+0x48: undge 0xa0cc20f8 >>> db> >>> db> bt >>> Tracing pid 21116 tid 100073 td 0xc3e97000 >>> db_trace_self() at db_trace_self >>> pc = 0xc0dd5418 lr = 0xc094f8a8 (db_hex2dec+0x490) >>> sp = 0xde0195a0 fp = 0xde0195b8 >>> r10 = 0xc0f5e8c8 >>> db_hex2dec() at db_hex2dec+0x490 >>> pc = 0xc094f8a8 lr = 0xc094f260 (db_command_loop+0x300) >>> sp = 0xde0195c0 fp = 0xde019660 >>> r4 = 0x00000000 r5 = 0x00000000 >>> r6 = 0x00000000 >>> db_command_loop() at db_command_loop+0x300 >>> pc = 0xc094f260 lr = 0xc094efb0 (db_command_loop+0x50) >>> sp = 0xde019668 fp = 0xde019678 >>> r4 = 0xc0e2dfe4 r5 = 0xc0e4402e >>> r6 = 0xc0f5e8b4 r7 = 0xc0ef62b8 >>> r8 = 0xc0f52754 r9 = 0xc0f52750 >>> r10 = 0xc3e97000 >>> db_command_loop() at db_command_loop+0x50 >>> pc = 0xc094efb0 lr = 0xc09519ec (X_db_symbol_values+0x250) >>> sp = 0xde019680 fp = 0xde0197a0 >>> r4 = 0x00000000 r5 = 0xc0f5e8c0 >>> r6 = 0xc0f52778 >>> X_db_symbol_values() at X_db_symbol_values+0x250 >>> pc = 0xc09519ec lr = 0xc0b37b08 (kdb_trap+0xc4) >>> sp = 0xde0197a8 fp = 0xde0197c8 >>> r4 = 0x00000000 r5 = 0x0000000f >>> r6 = 0xc0f52778 r7 = 0xc0ef62b8 >>> kdb_trap() at kdb_trap+0xc4 >>> pc = 0xc0b37b08 lr = 0xc0de7c60 (data_abort_handler+0x7f8) >>> sp = 0xde0197d0 fp = 0xde0197e8 >>> r4 = 0xde019898 r5 = 0x0000000f >>> r6 = 0x600000d3 r7 = 0xd0238120 >>> r8 = 0x00000000 r9 = 0xc0f648d4 >>> r10 = 0xc3e97000 >>> data_abort_handler() at data_abort_handler+0x7f8 >>> pc = 0xc0de7c60 lr = 0xc0de7a28 (data_abort_handler+0x5c0) >>> sp = 0xde0197f0 fp = 0xde019890 >>> r4 = 0xc10a0000 r5 = 0x00000013 >>> r6 = 0xde019eb0 r7 = 0x00000002 >>> data_abort_handler() at data_abort_handler+0x5c0 >>> pc = 0xc0de7a28 lr = 0xc0dd711c (exception_exit) >>> sp = 0xde019898 fp = 0xde019a20 >>> r4 = 0xffffffff r5 = 0xffff1004 >>> r6 = 0xc3f3f6c0 r7 = 0x00001000 >>> r8 = 0xc443e880 r9 = 0x00000000 >>> r10 = 0xc3d69000 >>> exception_exit() at exception_exit >>> pc = 0xc0dd711c lr = 0xc0d53828 (ffs_truncate+0xaa8) >>> sp = 0xde0198e8 fp = 0xde019a20 >>> r0 = 0xd0238120 r1 = 0x00000e60 >>> r2 = 0x00000000 r3 = 0x00000000 >>> r4 = 0x00000120 r5 = 0x00000000 >>> r6 = 0xc3f3f6c0 r7 = 0x00001000 >>> r8 = 0xc443e880 r9 = 0x00000000 >>> r10 = 0xc3d69000 r12 = 0xd0238120 >>> memset() at memset+0x48 >>> pc = 0xc0de521c lr = 0xc0d53828 (ffs_truncate+0xaa8) >>> sp = 0xde0198e8 fp = 0xde019a20 >>> Unwind failure (no registers changed) >>> >>> The sad thing is, that with fsck broken for the dreamplug, I have to >>> re-format the disk, reinstall everything and recreate the config files >>> which I didn't manage to copy to a safe place beforehand :-( >>> >>> Before I do that I'll leave the system in debugging mode for a few >>> days, in case someone can help and needs some more information. >>> >>> Cheers, >>> >>> Mat >>> _______________________________________________ >>> freebsd-arm@freebsd.org mailing list >>> http://lists.freebsd.org/mailman/listinfo/freebsd-arm >>> To unsubscribe, send any mail to "freebsd-arm-unsubscribe@freebsd.org" >> _______________________________________________ >> freebsd-arm@freebsd.org mailing list >> http://lists.freebsd.org/mailman/listinfo/freebsd-arm >> To unsubscribe, send any mail to "freebsd-arm-unsubscribe@freebsd.org"
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?op.xmwdiud6kndu52>