Re: [PATCH v2] sched/fair: bump sd->max_newidle_lb_cost when newidle balance fails

[Dietmar Eggemann <dietmar.eggemann@arm.com>]

On 07.09.25 20:21, Ryan Roberts wrote:
> On 26/06/2025 15:39, Chris Mason wrote:

[...]

> I'm seeing a ~25% regression in requests per second for an nginx workload in 
> 6.17-rc4 compared with 6.16, when the number of simulated clients (threads) is 
> high (1000). Bisection led me to this patch. The workload is running on an 
> AmpereOne (arm64) system with 192 CPUs. FWIW, I don't see the regression on an 
> AWS Graviton3 system.

Can you look for any sched domain hierarchy differences between
AmpereOne and Grav3?

I assume Grav3 is bare-metal?

> I'm also seeing a 10% regression on the same system for a MySQL workload; but I 
> haven't yet bisected that one - I'll report back if that turns out to be due to 
> this too.

Is this the hammerdb test in which the SUT hosts the mysqld? Which
params (#Virtual Users, #Warehouses, ...) were you using?

> I saw that there was a regression raised against this patch by kernel test robot 
> for unixbench.throughput back in July, but it didn't look like it got resolved.

Can you run the unixbench shell1 test on AmpereOne to see if the
regression in nginx is related to this unixbench test?

[...]