Re: [PATCH v3 1/4] hw/arm/virt: Consider SMP configuration in CPU topology

2022-04-03 Thread Gavin Shan
Hi Yanan, On 4/2/22 10:17 AM, wangyanan (Y) wrote: On 2022/3/23 15:24, Gavin Shan wrote: Currently, the SMP configuration isn't considered when the CPU topology is populated. In this case, it's impossible to provide the default CPU-to-NUMA mapping or association based on the socket ID of the gi

Re: [PATCH v3 1/4] hw/arm/virt: Consider SMP configuration in CPU topology

2022-04-03 Thread Gavin Shan
Hi Igor, On 3/30/22 9:18 PM, Igor Mammedov wrote: On Wed, 23 Mar 2022 15:24:35 +0800 Gavin Shan wrote: Currently, the SMP configuration isn't considered when the CPU topology is populated. In this case, it's impossible to provide the default CPU-to-NUMA mapping or association based on the soc

Re: [PATCH v3 1/4] hw/arm/virt: Consider SMP configuration in CPU topology

2022-04-03 Thread Gavin Shan
Hi Igor, On 3/30/22 8:50 PM, Igor Mammedov wrote: On Sat, 26 Mar 2022 02:49:59 +0800 Gavin Shan wrote: On 3/25/22 9:19 PM, Igor Mammedov wrote: On Wed, 23 Mar 2022 15:24:35 +0800 Gavin Shan wrote: Currently, the SMP configuration isn't considered when the CPU topology is populated. In this

Re: [PATCH v3 1/4] hw/arm/virt: Consider SMP configuration in CPU topology

2022-04-01 Thread wangyanan (Y)
On 2022/3/30 20:50, Igor Mammedov wrote: On Sat, 26 Mar 2022 02:49:59 +0800 Gavin Shan wrote: Hi Igor, On 3/25/22 9:19 PM, Igor Mammedov wrote: On Wed, 23 Mar 2022 15:24:35 +0800 Gavin Shan wrote: Currently, the SMP configuration isn't considered when the CPU topology is populated. In this

Re: [PATCH v3 1/4] hw/arm/virt: Consider SMP configuration in CPU topology

2022-04-01 Thread wangyanan (Y)
Hi Gavin, On 2022/3/23 15:24, Gavin Shan wrote: Currently, the SMP configuration isn't considered when the CPU topology is populated. In this case, it's impossible to provide the default CPU-to-NUMA mapping or association based on the socket ID of the given CPU. This takes account of SMP config

Re: [PATCH v3 1/4] hw/arm/virt: Consider SMP configuration in CPU topology

2022-03-30 Thread Igor Mammedov
On Wed, 23 Mar 2022 15:24:35 +0800 Gavin Shan wrote: > Currently, the SMP configuration isn't considered when the CPU > topology is populated. In this case, it's impossible to provide > the default CPU-to-NUMA mapping or association based on the socket > ID of the given CPU. > > This takes accou

Re: [PATCH v3 1/4] hw/arm/virt: Consider SMP configuration in CPU topology

2022-03-30 Thread Igor Mammedov
On Sat, 26 Mar 2022 02:49:59 +0800 Gavin Shan wrote: > Hi Igor, > > On 3/25/22 9:19 PM, Igor Mammedov wrote: > > On Wed, 23 Mar 2022 15:24:35 +0800 > > Gavin Shan wrote: > >> Currently, the SMP configuration isn't considered when the CPU > >> topology is populated. In this case, it's impossib

Re: [PATCH v3 1/4] hw/arm/virt: Consider SMP configuration in CPU topology

2022-03-25 Thread Gavin Shan
Hi Igor, On 3/25/22 9:19 PM, Igor Mammedov wrote: On Wed, 23 Mar 2022 15:24:35 +0800 Gavin Shan wrote: Currently, the SMP configuration isn't considered when the CPU topology is populated. In this case, it's impossible to provide the default CPU-to-NUMA mapping or association based on the sock

Re: [PATCH v3 1/4] hw/arm/virt: Consider SMP configuration in CPU topology

2022-03-25 Thread Igor Mammedov
On Wed, 23 Mar 2022 15:24:35 +0800 Gavin Shan wrote: > Currently, the SMP configuration isn't considered when the CPU > topology is populated. In this case, it's impossible to provide > the default CPU-to-NUMA mapping or association based on the socket > ID of the given CPU. > > This takes accou

[PATCH v3 1/4] hw/arm/virt: Consider SMP configuration in CPU topology

2022-03-23 Thread Gavin Shan
Currently, the SMP configuration isn't considered when the CPU topology is populated. In this case, it's impossible to provide the default CPU-to-NUMA mapping or association based on the socket ID of the given CPU. This takes account of SMP configuration when the CPU topology is populated. The die