Re: [PATCH] locking/osq: Fix ordering of node initialisation in osq_lock

2015-12-17 Thread Peter Zijlstra
On Thu, Dec 17, 2015 at 11:04:00AM -0800, Linus Torvalds wrote: > On Thu, Dec 17, 2015 at 8:05 AM, Peter Zijlstra wrote: > > > > Please consider this patch for 4.4. > > No problem, but just wanted to check that there's nothing else pending > in any locking tree? Just this one patch for now. Tha

Re: [PATCH] locking/osq: Fix ordering of node initialisation in osq_lock

2015-12-17 Thread Linus Torvalds
On Thu, Dec 17, 2015 at 8:05 AM, Peter Zijlstra wrote: > > Please consider this patch for 4.4. No problem, but just wanted to check that there's nothing else pending in any locking tree? Linus -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the bod

[PATCH] locking/osq: Fix ordering of node initialisation in osq_lock

2015-12-17 Thread Peter Zijlstra
Hi Linus, Please consider this patch for 4.4. --- Subject: locking/osq: Fix ordering of node initialisation in osq_lock From: Will Deacon Date: Fri, 11 Dec 2015 17:46:41 + The Cavium guys reported a soft lockup on their arm64 machine, caused by c55a6ffa6285 ("locking/osq: Relax atomic sema

Re: [PATCH] locking/osq: fix ordering of node initialisation in osq_lock

2015-12-17 Thread Will Deacon
On Wed, Dec 16, 2015 at 05:18:48PM -0800, David Daney wrote: > What is the status of this patch? It there a good likelihood that it will > make it into v4.4? > > If not, we should request that c55a6ffa6285 ("locking/osq: Relax atomic > semantics") be reverted for v4.4 I think Peter was going to

Re: [PATCH] locking/osq: fix ordering of node initialisation in osq_lock

2015-12-16 Thread David Daney
What is the status of this patch? It there a good likelihood that it will make it into v4.4? If not, we should request that c55a6ffa6285 ("locking/osq: Relax atomic semantics") be reverted for v4.4 David Daney On 12/11/2015 09:46 AM, Will Deacon wrote: The Cavium guys reported a soft loc

Re: [PATCH] locking/osq: fix ordering of node initialisation in osq_lock

2015-12-11 Thread David Daney
On 12/11/2015 09:46 AM, Will Deacon wrote: The Cavium guys reported a soft lockup on their arm64 machine, caused by c55a6ffa6285 ("locking/osq: Relax atomic semantics"): [ 68.909948] [] mutex_optimistic_spin+0x9c/0x1d0 [ 68.909951] [] __mutex_lock_slowpath+0x44/0x158 [ 68.909953] [] mutex_

[PATCH] locking/osq: fix ordering of node initialisation in osq_lock

2015-12-11 Thread Will Deacon
The Cavium guys reported a soft lockup on their arm64 machine, caused by c55a6ffa6285 ("locking/osq: Relax atomic semantics"): [ 68.909948] [] mutex_optimistic_spin+0x9c/0x1d0 [ 68.909951] [] __mutex_lock_slowpath+0x44/0x158 [ 68.909953] [] mutex_lock+0x54/0x58 [ 68.909956] [] kernfs_iop_p