Date: Wed, 11 Nov 2015 21:21:56 +0700 From: Eugene Grosbein <eugen@grosbein.net> To: Konstantin Belousov <kostikbel@gmail.com> Cc: Jeff Roberson <jeff@freebsd.org>, benno <benno@freebsd.org>, "freebsd-mips@freebsd.org" <freebsd-mips@freebsd.org> Subject: Re: USB-related problem Message-ID: <56434F04.9010408@grosbein.net> In-Reply-To: <20151111120508.GZ2257@kib.kiev.ua> References: <CAJ-Vmo=0vOAq8db_GeLWmdXr7xJdzUh44ZZJrQ9vVdpvzT9hiQ@mail.gmail.com> <563F5630.2000407@grosbein.net> <563F6F6F.1010807@grosbein.net> <CAJ-Vmo=fPSi7yZO5Xjodg8HPtTLd44Y9Y_8qg4EgTGwEpHO10A@mail.gmail.com> <563F91A8.9080702@grosbein.net> <CAJ-VmomUvoUerMS20qQsQujcjULVA=_jaLp9Mh3fU1fEpdwzZA@mail.gmail.com> <CAJ-Vmo=eUUZ928KgQbyOi8EdDFSmxhjvDOyAMvfXsqwDbO96ng@mail.gmail.com> <5640C0FD.2040803@grosbein.net> <CAJ-Vmo=6mztfvvBd91LPO5H418K8vW=%2BLk=6V5Z_y5DHu7v7HA@mail.gmail.com> <5640F315.5020303@grosbein.net> <20151111120508.GZ2257@kib.kiev.ua>
next in thread | previous in thread | raw e-mail | index | archive | help
On 11.11.2015 19:05, Konstantin Belousov wrote: > On Tue, Nov 10, 2015 at 02:25:09AM +0700, Eugene Grosbein wrote: >> On 09.11.2015 22:53, Adrian Chadd wrote: >>> ok, let's get it to hang, then break into ddb and do 'show >>> allthreads', 'show alllocks', 'bufvars' with my vfs_bio.c patch, etc. >> >> I've re-read https://www.freebsd.org/doc/en/books/developers-handbook/kerneldebug-deadlocks.html >> and rebuilt kernel with additional debugging and reproduced the problem again to supply >> more vfs lock info: > > You have syncer running with the buffer arena mutex locked. This probably > means that you have DIAGNOSTIC option turned on, do you ? I do, I've just added it before latest series of my tests because of Handbook instuctions. The problem reproduces without DIAGNOSTICS too. > If yes, set debug.vmem_check to zero, otherwise it just eats your anemic CPU. > > Also, I am curious what is the value of vfs.unmapped_buf_allowed on > your platform. vfs.unmapped_buf_allowed: 0 I've set sysctl debug.vmem_check=0 and repeated the test. The problem reproduced without any changes in my trace.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?56434F04.9010408>