Re: [PATCH] smp, ipi: Speed up IPI handling by invoking the callbacks in reverse order

2014-06-06 Thread Srivatsa S. Bhat
On 06/05/2014 12:56 PM, Peter Zijlstra wrote: > On Thu, Jun 05, 2014 at 01:37:25AM +0530, Srivatsa S. Bhat wrote: >> On 06/05/2014 01:17 AM, Peter Zijlstra wrote: >>> On Thu, Jun 05, 2014 at 01:09:35AM +0530, Srivatsa S. Bhat wrote: The current implementation of lockless list (llist) has a dra

Re: [PATCH] smp, ipi: Speed up IPI handling by invoking the callbacks in reverse order

2014-06-05 Thread Peter Zijlstra
On Thu, Jun 05, 2014 at 01:37:25AM +0530, Srivatsa S. Bhat wrote: > On 06/05/2014 01:17 AM, Peter Zijlstra wrote: > > On Thu, Jun 05, 2014 at 01:09:35AM +0530, Srivatsa S. Bhat wrote: > >> The current implementation of lockless list (llist) has a drawback: if we > >> want to traverse the list in FI

Re: [PATCH] smp, ipi: Speed up IPI handling by invoking the callbacks in reverse order

2014-06-04 Thread Srivatsa S. Bhat
On 06/05/2014 01:17 AM, Peter Zijlstra wrote: > On Thu, Jun 05, 2014 at 01:09:35AM +0530, Srivatsa S. Bhat wrote: >> The current implementation of lockless list (llist) has a drawback: if we >> want to traverse the list in FIFO order (oldest to newest), we need to >> reverse the list first (and thi

Re: [PATCH] smp, ipi: Speed up IPI handling by invoking the callbacks in reverse order

2014-06-04 Thread Peter Zijlstra
On Thu, Jun 05, 2014 at 01:09:35AM +0530, Srivatsa S. Bhat wrote: > The current implementation of lockless list (llist) has a drawback: if we > want to traverse the list in FIFO order (oldest to newest), we need to > reverse the list first (and this can be expensive if the list is large, > since th

[PATCH] smp, ipi: Speed up IPI handling by invoking the callbacks in reverse order

2014-06-04 Thread Srivatsa S. Bhat
The current implementation of lockless list (llist) has a drawback: if we want to traverse the list in FIFO order (oldest to newest), we need to reverse the list first (and this can be expensive if the list is large, since this is an O(n) operation). However, for callbacks that are queued using sm