Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 1 Jun 2009 17:34:06 -0400
From:      John Baldwin <jhb@freebsd.org>
To:        src-committers@freebsd.org
Cc:        svn-src-head@freebsd.org, svn-src-all@freebsd.org
Subject:   Re: svn commit: r193272 - in head/sys: dev/cxgb/ulp/iw_cxgb kern netgraph netgraph/bluetooth/socket netinet netsmb nfsclient nfsserver rpc sys
Message-ID:  <200906011734.06721.jhb@freebsd.org>
In-Reply-To: <200906012117.n51LH3F8092452@svn.freebsd.org>
References:  <200906012117.n51LH3F8092452@svn.freebsd.org>

next in thread | previous in thread | raw e-mail | index | archive | help
On Monday 01 June 2009 5:17:03 pm John Baldwin wrote:
> Author: jhb
> Date: Mon Jun  1 21:17:03 2009
> New Revision: 193272
> URL: http://svn.freebsd.org/changeset/base/193272
> 
> Log:
>   Rework socket upcalls to close some races with setup/teardown of upcalls.
>   - The various RPC client upcalls currently drop the socket buffer lock
>     while invoking soreceive() as a temporary band-aid.  The plan for
>     the future is to add a new flag to allow soreceive() to be called with
>     the socket buffer locked.

Hopefully once this last bit is done, the various panics people have seen with 
the new NFS code in 8 will be fixed as they all seem to be due to races 
between socket upcall teardown and socket upcall invocations.

-- 
John Baldwin



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200906011734.06721.jhb>