Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 25 Oct 2012 11:54:08 +0300
From:      Andriy Gapon <avg@FreeBSD.org>
To:        Poul-Henning Kamp <phk@phk.freebsd.dk>
Cc:        freebsd-geom@FreeBSD.org
Subject:   Re: geom access method and g_topology_lock
Message-ID:  <5088FE30.2030903@FreeBSD.org>
In-Reply-To: <70960.1351154478@critter.freebsd.dk>
References:  <5088E0E0.2080307@FreeBSD.org> <28185.1351150111@critter.freebsd.dk> <5088FA15.30205@FreeBSD.org> <70960.1351154478@critter.freebsd.dk>

next in thread | previous in thread | raw e-mail | index | archive | help
on 25/10/2012 11:41 Poul-Henning Kamp said the following:
> --------
> In message <5088FA15.30205@FreeBSD.org>, Andriy Gapon writes:
> 
>> Both pieces look sane and non-racy (because of g_topology_assert) until we allow
>> g_access (some geom access method, rather) to drop the topology lock.
> 
> You lost me there.
> 
> g_access() cannot do its job without holding the topology lock.
> 
> 

Will it help if I repeat my original questions:

Is that bad if a geom's access method drops and re-acquires g_topology_lock
while doing some internal stuff?
Is that allowed at all?

To clarify, if that's needed, by "geom's access method" I meant the 'access'
member with g_access_t type in struct g_geom.

-- 
Andriy Gapon



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?5088FE30.2030903>