Re: [PATCH 1/3] sched: remove select_idle_core() for scalability

On Wed, May 30, 2018 at 03:08:21PM -0700, Subhra Mazumdar wrote: > I tested with FOLD+AGE+ONCE+PONIES+PONIES2 shift=0 vs baseline but see some > regression for hackbench and uperf: I'm not seeing a hackbench regression myself, but I let it run a whole lot of stuff over-night and I do indeed see

Re: [PATCH 1/3] sched: remove select_idle_core() for scalability

On Wed, May 30, 2018 at 03:08:21PM -0700, Subhra Mazumdar wrote: > I tested with FOLD+AGE+ONCE+PONIES+PONIES2 shift=0 vs baseline but see some > regression for hackbench and uperf: I'm not seeing a hackbench regression myself, but I let it run a whole lot of stuff over-night and I do indeed see

Re: [PATCH 1/3] sched: remove select_idle_core() for scalability

On 05/29/2018 02:36 PM, Peter Zijlstra wrote: On Wed, May 02, 2018 at 02:58:42PM -0700, Subhra Mazumdar wrote: I re-ran the test after fixing that bug but still get similar regressions for hackbench Hackbench process on 2 socket, 44 core and 88 threads Intel x86 machine (lower is better):

Re: [PATCH 1/3] sched: remove select_idle_core() for scalability

On 05/29/2018 02:36 PM, Peter Zijlstra wrote: On Wed, May 02, 2018 at 02:58:42PM -0700, Subhra Mazumdar wrote: I re-ran the test after fixing that bug but still get similar regressions for hackbench Hackbench process on 2 socket, 44 core and 88 threads Intel x86 machine (lower is better):

Re: [PATCH 1/3] sched: remove select_idle_core() for scalability

On Wed, May 02, 2018 at 02:58:42PM -0700, Subhra Mazumdar wrote: > I re-ran the test after fixing that bug but still get similar regressions > for hackbench > Hackbench process on 2 socket, 44 core and 88 threads Intel x86 machine > (lower is better): > groups  baseline   %stdev  patch %stdev

Re: [PATCH 1/3] sched: remove select_idle_core() for scalability

On Wed, May 02, 2018 at 02:58:42PM -0700, Subhra Mazumdar wrote: > I re-ran the test after fixing that bug but still get similar regressions > for hackbench > Hackbench process on 2 socket, 44 core and 88 threads Intel x86 machine > (lower is better): > groups  baseline   %stdev  patch %stdev

Re: [PATCH 1/3] sched: remove select_idle_core() for scalability

On 05/02/2018 02:58 PM, Subhra Mazumdar wrote: On 05/01/2018 11:03 AM, Peter Zijlstra wrote: On Mon, Apr 30, 2018 at 04:38:42PM -0700, Subhra Mazumdar wrote: I also noticed a possible bug later in the merge code. Shouldn't it be: if (busy < best_busy) { best_busy = busy;

Re: [PATCH 1/3] sched: remove select_idle_core() for scalability

On 05/02/2018 02:58 PM, Subhra Mazumdar wrote: On 05/01/2018 11:03 AM, Peter Zijlstra wrote: On Mon, Apr 30, 2018 at 04:38:42PM -0700, Subhra Mazumdar wrote: I also noticed a possible bug later in the merge code. Shouldn't it be: if (busy < best_busy) { best_busy = busy;

Re: [PATCH 1/3] sched: remove select_idle_core() for scalability

On 05/01/2018 11:03 AM, Peter Zijlstra wrote: On Mon, Apr 30, 2018 at 04:38:42PM -0700, Subhra Mazumdar wrote: I also noticed a possible bug later in the merge code. Shouldn't it be: if (busy < best_busy) {     best_busy = busy;     best_cpu = first_idle; } Uhh, quite. I did say

Re: [PATCH 1/3] sched: remove select_idle_core() for scalability

On 05/01/2018 11:03 AM, Peter Zijlstra wrote: On Mon, Apr 30, 2018 at 04:38:42PM -0700, Subhra Mazumdar wrote: I also noticed a possible bug later in the merge code. Shouldn't it be: if (busy < best_busy) {     best_busy = busy;     best_cpu = first_idle; } Uhh, quite. I did say

Re: [PATCH 1/3] sched: remove select_idle_core() for scalability

On Mon, Apr 30, 2018 at 04:38:42PM -0700, Subhra Mazumdar wrote: > I also noticed a possible bug later in the merge code. Shouldn't it be: > > if (busy < best_busy) { >     best_busy = busy; >     best_cpu = first_idle; > } Uhh, quite. I did say it was completely untested, but yes.. /me

Re: [PATCH 1/3] sched: remove select_idle_core() for scalability

On Mon, Apr 30, 2018 at 04:38:42PM -0700, Subhra Mazumdar wrote: > I also noticed a possible bug later in the merge code. Shouldn't it be: > > if (busy < best_busy) { >     best_busy = busy; >     best_cpu = first_idle; > } Uhh, quite. I did say it was completely untested, but yes.. /me

Re: [PATCH 1/3] sched: remove select_idle_core() for scalability

On 04/25/2018 10:49 AM, Peter Zijlstra wrote: On Tue, Apr 24, 2018 at 02:45:50PM -0700, Subhra Mazumdar wrote: So what you said makes sense in theory but is not borne out by real world results. This indicates that threads of these benchmarks care more about running immediately on any idle cpu

Re: [PATCH 1/3] sched: remove select_idle_core() for scalability

On 04/25/2018 10:49 AM, Peter Zijlstra wrote: On Tue, Apr 24, 2018 at 02:45:50PM -0700, Subhra Mazumdar wrote: So what you said makes sense in theory but is not borne out by real world results. This indicates that threads of these benchmarks care more about running immediately on any idle cpu

Re: [PATCH 1/3] sched: remove select_idle_core() for scalability

On Tue, Apr 24, 2018 at 02:45:50PM -0700, Subhra Mazumdar wrote: > So what you said makes sense in theory but is not borne out by real > world results. This indicates that threads of these benchmarks care more > about running immediately on any idle cpu rather than spending time to find > fully

Re: [PATCH 1/3] sched: remove select_idle_core() for scalability

On Tue, Apr 24, 2018 at 02:45:50PM -0700, Subhra Mazumdar wrote: > So what you said makes sense in theory but is not borne out by real > world results. This indicates that threads of these benchmarks care more > about running immediately on any idle cpu rather than spending time to find > fully

Re: [PATCH 1/3] sched: remove select_idle_core() for scalability

On 04/24/2018 05:46 AM, Peter Zijlstra wrote: On Mon, Apr 23, 2018 at 05:41:14PM -0700, subhra mazumdar wrote: select_idle_core() can potentially search all cpus to find the fully idle core even if there is one such core. Removing this is necessary to achieve scalability in the fast path. So

Re: [PATCH 1/3] sched: remove select_idle_core() for scalability

On 04/24/2018 05:46 AM, Peter Zijlstra wrote: On Mon, Apr 23, 2018 at 05:41:14PM -0700, subhra mazumdar wrote: select_idle_core() can potentially search all cpus to find the fully idle core even if there is one such core. Removing this is necessary to achieve scalability in the fast path. So

Re: [PATCH 1/3] sched: remove select_idle_core() for scalability

On Mon, Apr 23, 2018 at 05:41:14PM -0700, subhra mazumdar wrote: > select_idle_core() can potentially search all cpus to find the fully idle > core even if there is one such core. Removing this is necessary to achieve > scalability in the fast path. So this removes the whole core awareness from

Re: [PATCH 1/3] sched: remove select_idle_core() for scalability

On Mon, Apr 23, 2018 at 05:41:14PM -0700, subhra mazumdar wrote: > select_idle_core() can potentially search all cpus to find the fully idle > core even if there is one such core. Removing this is necessary to achieve > scalability in the fast path. So this removes the whole core awareness from

[PATCH 1/3] sched: remove select_idle_core() for scalability

select_idle_core() can potentially search all cpus to find the fully idle core even if there is one such core. Removing this is necessary to achieve scalability in the fast path. Signed-off-by: subhra mazumdar --- include/linux/sched/topology.h | 1 -

[PATCH 1/3] sched: remove select_idle_core() for scalability

select_idle_core() can potentially search all cpus to find the fully idle core even if there is one such core. Removing this is necessary to achieve scalability in the fast path. Signed-off-by: subhra mazumdar --- include/linux/sched/topology.h | 1 - kernel/sched/fair.c| 97