Re: [RFC][PATCH 1/3] Freezer: Fix vfork problem

From: Oleg Nesterov <oleg@tv-sign.ru>
To: Aneesh Kumar <aneesh.kumar@gmail.com>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>,
	Srivatsa Vaddagiri <vatsa@in.ibm.com>,
	LKML <linux-kernel@vger.kernel.org>,
	Andrew Morton <akpm@osdl.org>, Pavel Machek <pavel@ucw.cz>,
	"Paul E. McKenney" <paulmck@linux.vnet.ibm.com>,
	Gautham R Shenoy <ego@in.ibm.com>
Subject: Re: [RFC][PATCH 1/3] Freezer: Fix vfork problem
Date: Tue, 27 Feb 2007 11:37:41 +0300	[thread overview]
Message-ID: <20070227083741.GA85@tv-sign.ru> (raw)
In-Reply-To: <cc723f590702262033k4fa6cbe0n8e327b8339c7a0e4@mail.gmail.com>

On 02/27, Aneesh Kumar wrote:
>
> >+static inline void freezer_do_not_count(void)
> >+{
> >+       current->flags |= PF_FREEZER_SKIP;
> >+}
> >+
> >+/*
> >+ * Tell the freezer to count this task as freezeable again and if it's a 
> >user
> >+ * space one, try to freeze it
> >+ */
> >+static inline void freezer_count(void)
> >+{
> >+       current->flags &= ~PF_FREEZER_SKIP;
> >+       if (current->mm)
> >+               try_to_freeze();
> >+}
> >+
> 
> 
> This means that we are not going to wait for the kernel thread
> (Parent) to freeze. I guess what vatsa suggested in previous mail is
> better.

Not only better, but correct. This is very simple: if we set PF_FREEZER_SKIP,
we must call try_to_freeze(), otherwise we confuse the freezer.

Oleg.