Re: [RFC][PATCH 5/6] x86/topo: Fix SNC topology mess

[Chen, Yu C]

On 3/3/2026 2:31 PM, Zhang Rui wrote:
> On Mon, 2026-03-02 at 17:43 +0100, Peter Zijlstra wrote:
> > nt On Sat, Feb 28, 2026 at 03:35:26PM +0800, Chen, Yu C wrote:
> >
> > > Here we check packages0 and 1, should we check all the packages?
> >
> > Might as well I suppose.
> >
> > Could you boot queue/x86/topo on an snc-3 machine to verify it
> > doesn't
> > explode with all the paranoia on?
>
> Hi, Peter,
>
> regarding slit_validate() in
> https://git.kernel.org/pub/scm/linux/kernel/git/peterz/queue.git/commit/?h=x86/topo&id=24ca94ac4b72803a7164b7ad84f06f0e9f0c49df
>
> I suppose we want to use
> 	WARN_ON_ONCE(!slit_cluster_symmetric(n))
> rather than
> 	WARN_ON_ONCE(slit_cluster_symmetric(n));

With Rui's fix, tested on GNR ,2 sockets, SNC3, NUMA distance symmetric 
platform,
it works as expected,
 smp: Brought up 6 nodes, 384 CPUs
  domain-0: span=0,192 level=SMT
  domain-1: span=0-31,192-223 level=MC
  domain-2: span=0-95,192-287 level=NUMA
    groups: 0:{ span=0-31,192-223 cap=65536 }, 32:{ span=32-63,224-255 
cap=65536 }, 64:{ span=64-95,256-287 cap=65536 }
  domain-3: span=0-383 level=NUMA
    groups: 0:{ span=0-95,192-287 cap=196608 }, 96:{ 
span=96-191,288-383 cap=196608 }

resctrl also detects SNC3
root@a4bf018d7604:/sys/fs/resctrl/mon_data# tree mon_L3_00/
mon_L3_00/
├── llc_occupancy
├── mbm_local_bytes
├── mbm_total_bytes
├── mon_sub_L3_00
│   ├── llc_occupancy
│   ├── mbm_local_bytes
│   └── mbm_total_bytes
├── mon_sub_L3_01
│   ├── llc_occupancy
│   ├── mbm_local_bytes
│   └── mbm_total_bytes
└── mon_sub_L3_02
    ├── llc_occupancy
    ├── mbm_local_bytes
    └── mbm_total_bytes

> So
>
> Tested-by: Zhang Rui <rui.zhang@xxxxxxxxx>

For this series,
Tested-by: Chen Yu <yu.c.chen@xxxxxxxxx>

thanks,
Chenyu