Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 18 Jan 2013 13:58:10 +0200
From:      Andriy Gapon <avg@FreeBSD.org>
To:        Po-Li Soong <polis@spectralogic.com>
Cc:        Konstantin Belousov <kostikbel@gmail.com>, "stable@FreeBSD.org" <stable@FreeBSD.org>
Subject:   Re: zio_done panic on unadulterated FreeBSD Release 9.1
Message-ID:  <50F938D2.80304@FreeBSD.org>
In-Reply-To: <0C4D65F6A0FC9E4B95EA114508C7E0FE5F66E4F3@reactor.sldomain.com>
References:  <0C4D65F6A0FC9E4B95EA114508C7E0FE5F66DDB6@reactor.sldomain.com> <20130109234924.GM2561@kib.kiev.ua> <0C4D65F6A0FC9E4B95EA114508C7E0FE5F66E4F3@reactor.sldomain.com>

next in thread | previous in thread | raw e-mail | index | archive | help
on 11/01/2013 17:09 Po-Li Soong said the following:
> (kgdb) p/x *(struct vm_object *)0xffffffff81281580
[snip]
> 
> (kgdb)  p/x *(struct vm_page *)0xfffffe00cd733ab0
> $2 = {pageq = {tqe_next = 0x0, tqe_prev = 0xfffffe00c7e7d678}, listq = {
>     tqe_next = 0xfffffe00cd733b28, tqe_prev = 0xfffffe00cd7331d8},
>   left = 0xfffffe00c9b31c38, right = 0xfffffe00cd735c70,
>   object = 0xfffffffb81281580

So, the correct object pointer is 0xffffffff81281580, but vm_page has object =
0xfffffffb81281580.

0xffffffff81281580
0xfffffffb81281580
One bit flip (0x400000000).

Either there is some HW issue or we've got some code that arbitrarily flips bits
in vm_page_array (and perhaps beyond it).

-- 
Andriy Gapon



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?50F938D2.80304>