Re: [PATCH 1/2] spapr: number of SMP sockets must be equal to NUMA nodes

2021-04-01 Thread Cédric Le Goater
On 4/1/21 4:59 AM, David Gibson wrote: > On Wed, Mar 31, 2021 at 05:18:45PM +0200, Cédric Le Goater wrote: >> On 3/31/21 2:57 AM, David Gibson wrote: >>> On Mon, Mar 29, 2021 at 03:32:37PM -0300, Daniel Henrique Barboza wrote: On 3/29/21 12:32 PM, Cédric Le Goater wrote: > On

Re: [PATCH 1/2] spapr: number of SMP sockets must be equal to NUMA nodes

2021-03-31 Thread David Gibson
On Wed, Mar 31, 2021 at 05:18:45PM +0200, Cédric Le Goater wrote: > On 3/31/21 2:57 AM, David Gibson wrote: > > On Mon, Mar 29, 2021 at 03:32:37PM -0300, Daniel Henrique Barboza wrote: > >> > >> > >> On 3/29/21 12:32 PM, Cédric Le Goater wrote: > >>> On 3/29/21 6:20 AM, David Gibson wrote: >

Re: [PATCH 1/2] spapr: number of SMP sockets must be equal to NUMA nodes

2021-03-31 Thread David Gibson
On Wed, Mar 31, 2021 at 05:22:39PM +0200, Cédric Le Goater wrote: > On 3/31/21 6:58 AM, Michael Ellerman wrote: > > David Gibson writes: > >> On Mon, Mar 29, 2021 at 03:32:37PM -0300, Daniel Henrique Barboza wrote: > > ... > >> > >>> We assign ibm,chip-id=0x0 to CPUs 0-3, but CPUs 2-3 are located

Re: [PATCH 1/2] spapr: number of SMP sockets must be equal to NUMA nodes

2021-03-31 Thread Cédric Le Goater
On 3/31/21 7:29 PM, Daniel Henrique Barboza wrote: > > On 3/31/21 12:18 PM, Cédric Le Goater wrote: >> On 3/31/21 2:57 AM, David Gibson wrote: >>> On Mon, Mar 29, 2021 at 03:32:37PM -0300, Daniel Henrique Barboza wrote: On 3/29/21 12:32 PM, Cédric Le Goater wrote: > On 3/29/21

Re: [PATCH 1/2] spapr: number of SMP sockets must be equal to NUMA nodes

2021-03-31 Thread Daniel Henrique Barboza
On 3/31/21 12:18 PM, Cédric Le Goater wrote: On 3/31/21 2:57 AM, David Gibson wrote: On Mon, Mar 29, 2021 at 03:32:37PM -0300, Daniel Henrique Barboza wrote: On 3/29/21 12:32 PM, Cédric Le Goater wrote: On 3/29/21 6:20 AM, David Gibson wrote: On Thu, Mar 25, 2021 at 09:56:04AM +0100,

Re: [PATCH 1/2] spapr: number of SMP sockets must be equal to NUMA nodes

2021-03-31 Thread Cédric Le Goater
On 3/31/21 6:58 AM, Michael Ellerman wrote: > David Gibson writes: >> On Mon, Mar 29, 2021 at 03:32:37PM -0300, Daniel Henrique Barboza wrote: > ... >> >>> We assign ibm,chip-id=0x0 to CPUs 0-3, but CPUs 2-3 are located in a >>> different NUMA node than 0-1. This would mean that the same socket

Re: [PATCH 1/2] spapr: number of SMP sockets must be equal to NUMA nodes

2021-03-31 Thread Cédric Le Goater
On 3/31/21 2:57 AM, David Gibson wrote: > On Mon, Mar 29, 2021 at 03:32:37PM -0300, Daniel Henrique Barboza wrote: >> >> >> On 3/29/21 12:32 PM, Cédric Le Goater wrote: >>> On 3/29/21 6:20 AM, David Gibson wrote: On Thu, Mar 25, 2021 at 09:56:04AM +0100, Cédric Le Goater wrote: > On

Re: [PATCH 1/2] spapr: number of SMP sockets must be equal to NUMA nodes

2021-03-30 Thread Michael Ellerman
David Gibson writes: > On Mon, Mar 29, 2021 at 03:32:37PM -0300, Daniel Henrique Barboza wrote: ... > >> We assign ibm,chip-id=0x0 to CPUs 0-3, but CPUs 2-3 are located in a >> different NUMA node than 0-1. This would mean that the same socket >> would belong to different NUMA nodes at the same

Re: [PATCH 1/2] spapr: number of SMP sockets must be equal to NUMA nodes

2021-03-30 Thread David Gibson
On Mon, Mar 29, 2021 at 03:32:37PM -0300, Daniel Henrique Barboza wrote: > > > On 3/29/21 12:32 PM, Cédric Le Goater wrote: > > On 3/29/21 6:20 AM, David Gibson wrote: > > > On Thu, Mar 25, 2021 at 09:56:04AM +0100, Cédric Le Goater wrote: > > > > On 3/25/21 3:10 AM, David Gibson wrote: > > > >

Re: [PATCH 1/2] spapr: number of SMP sockets must be equal to NUMA nodes

2021-03-30 Thread Daniel Henrique Barboza
On 3/29/21 8:51 PM, Igor Mammedov wrote: On Tue, 23 Mar 2021 12:03:58 +1100 David Gibson wrote: On Fri, Mar 19, 2021 at 03:34:52PM -0300, Daniel Henrique Barboza wrote: Kernel commit 4bce545903fa ("powerpc/topology: Update topology_core_cpumask") cause a regression in the pseries machine

Re: [PATCH 1/2] spapr: number of SMP sockets must be equal to NUMA nodes

2021-03-29 Thread Igor Mammedov
On Mon, 29 Mar 2021 15:32:37 -0300 Daniel Henrique Barboza wrote: > On 3/29/21 12:32 PM, Cédric Le Goater wrote: > > On 3/29/21 6:20 AM, David Gibson wrote: > >> On Thu, Mar 25, 2021 at 09:56:04AM +0100, Cédric Le Goater wrote: > >>> On 3/25/21 3:10 AM, David Gibson wrote: > On Tue,

Re: [PATCH 1/2] spapr: number of SMP sockets must be equal to NUMA nodes

2021-03-29 Thread Igor Mammedov
On Tue, 23 Mar 2021 12:03:58 +1100 David Gibson wrote: > On Fri, Mar 19, 2021 at 03:34:52PM -0300, Daniel Henrique Barboza wrote: > > Kernel commit 4bce545903fa ("powerpc/topology: Update > > topology_core_cpumask") cause a regression in the pseries machine when > > defining certain SMP

Re: [PATCH 1/2] spapr: number of SMP sockets must be equal to NUMA nodes

2021-03-29 Thread Daniel Henrique Barboza
On 3/29/21 12:32 PM, Cédric Le Goater wrote: On 3/29/21 6:20 AM, David Gibson wrote: On Thu, Mar 25, 2021 at 09:56:04AM +0100, Cédric Le Goater wrote: On 3/25/21 3:10 AM, David Gibson wrote: On Tue, Mar 23, 2021 at 02:21:33PM -0300, Daniel Henrique Barboza wrote: On 3/22/21 10:03 PM,

Re: [PATCH 1/2] spapr: number of SMP sockets must be equal to NUMA nodes

2021-03-29 Thread Cédric Le Goater
On 3/29/21 6:20 AM, David Gibson wrote: > On Thu, Mar 25, 2021 at 09:56:04AM +0100, Cédric Le Goater wrote: >> On 3/25/21 3:10 AM, David Gibson wrote: >>> On Tue, Mar 23, 2021 at 02:21:33PM -0300, Daniel Henrique Barboza wrote: On 3/22/21 10:03 PM, David Gibson wrote: > On Fri,

Re: [PATCH 1/2] spapr: number of SMP sockets must be equal to NUMA nodes

2021-03-28 Thread David Gibson
On Thu, Mar 25, 2021 at 09:56:04AM +0100, Cédric Le Goater wrote: > On 3/25/21 3:10 AM, David Gibson wrote: > > On Tue, Mar 23, 2021 at 02:21:33PM -0300, Daniel Henrique Barboza wrote: > >> > >> > >> On 3/22/21 10:03 PM, David Gibson wrote: > >>> On Fri, Mar 19, 2021 at 03:34:52PM -0300, Daniel

Re: [PATCH 1/2] spapr: number of SMP sockets must be equal to NUMA nodes

2021-03-25 Thread Daniel Henrique Barboza
On 3/25/21 5:56 AM, Cédric Le Goater wrote: On 3/25/21 3:10 AM, David Gibson wrote: On Tue, Mar 23, 2021 at 02:21:33PM -0300, Daniel Henrique Barboza wrote: On 3/22/21 10:03 PM, David Gibson wrote: On Fri, Mar 19, 2021 at 03:34:52PM -0300, Daniel Henrique Barboza wrote: Kernel commit

Re: [PATCH 1/2] spapr: number of SMP sockets must be equal to NUMA nodes

2021-03-25 Thread Cédric Le Goater
On 3/25/21 3:10 AM, David Gibson wrote: > On Tue, Mar 23, 2021 at 02:21:33PM -0300, Daniel Henrique Barboza wrote: >> >> >> On 3/22/21 10:03 PM, David Gibson wrote: >>> On Fri, Mar 19, 2021 at 03:34:52PM -0300, Daniel Henrique Barboza wrote: Kernel commit 4bce545903fa ("powerpc/topology:

Re: [PATCH 1/2] spapr: number of SMP sockets must be equal to NUMA nodes

2021-03-24 Thread David Gibson
On Tue, Mar 23, 2021 at 02:21:33PM -0300, Daniel Henrique Barboza wrote: > > > On 3/22/21 10:03 PM, David Gibson wrote: > > On Fri, Mar 19, 2021 at 03:34:52PM -0300, Daniel Henrique Barboza wrote: > > > Kernel commit 4bce545903fa ("powerpc/topology: Update > > > topology_core_cpumask") cause a

Re: [PATCH 1/2] spapr: number of SMP sockets must be equal to NUMA nodes

2021-03-23 Thread Daniel Henrique Barboza
On 3/22/21 10:03 PM, David Gibson wrote: On Fri, Mar 19, 2021 at 03:34:52PM -0300, Daniel Henrique Barboza wrote: Kernel commit 4bce545903fa ("powerpc/topology: Update topology_core_cpumask") cause a regression in the pseries machine when defining certain SMP topologies [1]. The reasoning

Re: [PATCH 1/2] spapr: number of SMP sockets must be equal to NUMA nodes

2021-03-22 Thread David Gibson
On Fri, Mar 19, 2021 at 03:34:52PM -0300, Daniel Henrique Barboza wrote: > Kernel commit 4bce545903fa ("powerpc/topology: Update > topology_core_cpumask") cause a regression in the pseries machine when > defining certain SMP topologies [1]. The reasoning behind the change is > explained in kernel

[PATCH 1/2] spapr: number of SMP sockets must be equal to NUMA nodes

2021-03-19 Thread Daniel Henrique Barboza
Kernel commit 4bce545903fa ("powerpc/topology: Update topology_core_cpumask") cause a regression in the pseries machine when defining certain SMP topologies [1]. The reasoning behind the change is explained in kernel commit 4ca234a9cbd7 ("powerpc/smp: Stop updating cpu_core_mask"). In short,