Re: [PATCH] smp: add a best_effort version of smp_call_function_many()

2021-04-20 Thread Luigi Rizzo
On Tue, Apr 20, 2021 at 3:33 PM Peter Zijlstra wrote: > > On Tue, Apr 20, 2021 at 12:41:08PM +0200, Luigi Rizzo wrote: > > On Tue, Apr 20, 2021 at 11:14 AM Peter Zijlstra > > wrote: ... > > My case too requires that the request is eventually handled, but with > > this non-blocking IPI the

Re: [PATCH] smp: add a best_effort version of smp_call_function_many()

2021-04-20 Thread Peter Zijlstra
On Tue, Apr 20, 2021 at 12:41:08PM +0200, Luigi Rizzo wrote: > On Tue, Apr 20, 2021 at 11:14 AM Peter Zijlstra wrote: > > We mostly try and avoid using this stuff wherever possible. Only when > > no other choice is left do we send IPIs. > > > > NOHZ_FULL already relies on this and gets massively

Re: [PATCH] smp: add a best_effort version of smp_call_function_many()

2021-04-20 Thread Luigi Rizzo
On Tue, Apr 20, 2021 at 11:14 AM Peter Zijlstra wrote: > > On Mon, Apr 19, 2021 at 11:07:08PM +0200, Luigi Rizzo wrote: > > On Mon, Apr 19, 2021 at 9:17 PM Peter Zijlstra wrote: > > > > > > On Mon, Apr 19, 2021 at 11:44:55AM -0700, Luigi Rizzo wrote: > > > > Regardless of the 'wait' argument,

Re: [PATCH] smp: add a best_effort version of smp_call_function_many()

2021-04-20 Thread Peter Zijlstra
On Mon, Apr 19, 2021 at 11:07:08PM +0200, Luigi Rizzo wrote: > On Mon, Apr 19, 2021 at 9:17 PM Peter Zijlstra wrote: > > > > On Mon, Apr 19, 2021 at 11:44:55AM -0700, Luigi Rizzo wrote: > > > Regardless of the 'wait' argument, smp_call_function_many() must spin > > > if any of the target CPUs

Re: [PATCH] smp: add a best_effort version of smp_call_function_many()

2021-04-19 Thread kernel test robot
Hi Luigi, Thank you for the patch! Perhaps something to improve: [auto build test WARNING on linux/master] [also build test WARNING on linus/master] [cannot apply to next-20210419] [If your patch is applied to the wrong git tree, kindly drop us a note. And when submitting patch, we suggest to

Re: [PATCH] smp: add a best_effort version of smp_call_function_many()

2021-04-19 Thread kernel test robot
Hi Luigi, Thank you for the patch! Perhaps something to improve: [auto build test WARNING on linux/master] [also build test WARNING on linus/master v5.12-rc8] [cannot apply to next-20210419] [If your patch is applied to the wrong git tree, kindly drop us a note. And when submitting patch, we

Re: [PATCH] smp: add a best_effort version of smp_call_function_many()

2021-04-19 Thread Luigi Rizzo
On Mon, Apr 19, 2021 at 9:17 PM Peter Zijlstra wrote: > > On Mon, Apr 19, 2021 at 11:44:55AM -0700, Luigi Rizzo wrote: > > Regardless of the 'wait' argument, smp_call_function_many() must spin > > if any of the target CPUs have their csd busy waiting to be processed > > for a previous call. This

Re: [PATCH] smp: add a best_effort version of smp_call_function_many()

2021-04-19 Thread Peter Zijlstra
On Mon, Apr 19, 2021 at 11:44:55AM -0700, Luigi Rizzo wrote: > Regardless of the 'wait' argument, smp_call_function_many() must spin > if any of the target CPUs have their csd busy waiting to be processed > for a previous call. This may cause high tail latencies e.g. when some > of the target CPUs

Re: [PATCH] smp: add a best_effort version of smp_call_function_many()

2021-04-19 Thread Randy Dunlap
Hi-- On 4/19/21 11:44 AM, Luigi Rizzo wrote: > > Signed-off-by: Luigi Rizzo > --- > include/linux/smp.h | 10 ++ > kernel/smp.c| 75 + > 2 files changed, 72 insertions(+), 13 deletions(-) > > diff --git a/kernel/smp.c b/kernel/smp.c >

[PATCH] smp: add a best_effort version of smp_call_function_many()

2021-04-19 Thread Luigi Rizzo
Regardless of the 'wait' argument, smp_call_function_many() must spin if any of the target CPUs have their csd busy waiting to be processed for a previous call. This may cause high tail latencies e.g. when some of the target CPUs are running functions that disable interrupts for a long time;