On Fri, Sep 14, 2012 at 01:03:13PM +0100, Attilio Rao wrote: > On 9/14/12, Konstantin Belousov <kostik...@gmail.com> wrote: > > On Thu, Sep 13, 2012 at 10:26:22PM +0000, Attilio Rao wrote: > >> Author: attilio > >> Date: Thu Sep 13 22:26:22 2012 > >> New Revision: 240475 > >> URL: http://svn.freebsd.org/changeset/base/240475 > >> > >> Log: > >> Remove all the checks on curthread != NULL with the exception of some > >> MD > >> trap checks (eg. printtrap()). > >> > >> Generally this check is not needed anymore, as there is not a > >> legitimate > >> case where curthread != NULL, after pcpu 0 area has been properly > >> initialized. > > I do not disagree with the patch, but I do with this statement. > > During the AP startup, there is indeed a window while curthread is NULL. > > Unfortunately, trying to panic there is worse then not panic. > > Yes and why do you disagree then? That is what I wanted to say -- > before pcpu->curthread is initialized you can get curthread == NULL.
I disagree with the statement that curthread is always non-NULL after pcpu 0 area has been initialized.
pgp3K4xyiGab5.pgp
Description: PGP signature
