From mboxrd@z Thu Jan 1 00:00:00 1970 From: Nishanth Aravamudan Date: Wed, 13 Aug 2008 19:29:09 +0000 Subject: Re: [BISECTION RESULT] sched: revert cpu_clock to Message-Id: <20080813192909.GA31685@us.ibm.com> List-Id: References: <20080804194646.GA17390@us.ibm.com> <57C9024A16AD2D4C97DC78E552063EA30844B26C@orsmsx505.amr.corp.intel.com> <20080813162553.GA5720@elte.hu> In-Reply-To: <20080813162553.GA5720@elte.hu> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Ingo Molnar Cc: "Luck, Tony" , "linux-ia64@vger.kernel.org" , LKML On 13.08.2008 [18:25:53 +0200], Ingo Molnar wrote: > > * Luck, Tony wrote: > > > > I'm happy to test any better patches. > > > > Linus just took a patch from my tree that should solve this > > once and for all for ia64. > > > > commit 10617bbe84628eb18ab5f723d3ba35005adde143 > > Author: Tony Luck > > Date: Tue Aug 12 10:34:20 2008 -0700 > > > > [IA64] Ensure cpu0 can access per-cpu variables in early boot code > > > > > > When you get a chance please test with your config against this > > tree just to confirm that this is fixed. > > there's also Peter's other patch upstream since yesterday (and included > in -rc3) which should solve the same problem: > > | commit c1955a3d4762e7a9bf84035eb3c4886a900f0d15 > | Author: Peter Zijlstra > | Date: Mon Aug 11 08:59:03 2008 +0200 > | > | sched_clock: delay using sched_clock() > | > | Some arch's can't handle sched_clock() being called too early - delay > | this until sched_clock_init() has been called. > > so i'd be mighty surprised if this dual firepower of commits didnt at > least weaken this bug somewhat! ;-) Well Peter's patch was tested-by me, so I know it fixes the problem on the box I have access to, I'm verifying that -rc3 is ok. Tony, would you like me to test your patch in isolation (reverting c1955a3d4762e7a9bf84035eb3c4886a900f0d15) to see if it also fixes it? Thanks, Nish -- Nishanth Aravamudan IBM Linux Technology Center