Hello,
On Thu, Sep 12, 2024 at 01:59:16PM +0200, Morten Rasmussen wrote:
> On Fri, Sep 06, 2024 at 04:36:30PM +0800, Yicong Yang wrote:
> > On 2024/9/6 15:06, Morten Rasmussen wrote:
> > > Hi Yicong,
> > >
> > > On Thu, Sep 05, 2024 at 08:02:20PM +0800, Yicong Yang wrote:
> > >> On 2024/9/5 16:34
On Fri, Sep 06, 2024 at 04:36:30PM +0800, Yicong Yang wrote:
> On 2024/9/6 15:06, Morten Rasmussen wrote:
> > Hi Yicong,
> >
> > On Thu, Sep 05, 2024 at 08:02:20PM +0800, Yicong Yang wrote:
> >> On 2024/9/5 16:34, Pierre Gondois wrote:
> >>> Hello Yicong,
> >>>
> >>> If a platform has CPUs with:
>
Hi Yicong,
On Thu, Sep 05, 2024 at 08:02:20PM +0800, Yicong Yang wrote:
> On 2024/9/5 16:34, Pierre Gondois wrote:
> > Hello Yicong,
> >
> > If a platform has CPUs with:
> > - 1 thread
> > - X (!= 1) threads
> > Then I think that the asymmetry is not detected
>
> Ah ok, I only handle the case wh
On 2024/9/6 15:06, Morten Rasmussen wrote:
> Hi Yicong,
>
> On Thu, Sep 05, 2024 at 08:02:20PM +0800, Yicong Yang wrote:
>> On 2024/9/5 16:34, Pierre Gondois wrote:
>>> Hello Yicong,
>>>
>>> If a platform has CPUs with:
>>> - 1 thread
>>> - X (!= 1) threads
>>> Then I think that the asymmetry is n
On 2024/9/5 16:34, Pierre Gondois wrote:
> Hello Yicong,
>
Wondering if we can avoid this 2nd loop. Greg express the worries of
looping twice on large scale
system in v1. Maybe we could use the hetero_id and get the necessary
information in one loop, I need
further think
Hello Yicong,
Wondering if we can avoid this 2nd loop. Greg express the worries of looping
twice on large scale
system in v1. Maybe we could use the hetero_id and get the necessary
information in one loop, I need
further think.
I found this comments (not sure this is what you are refering to
On 2024/9/2 15:43, Pierre Gondois wrote:
> Hello Yicong
>
> On 8/30/24 11:35, Yicong Yang wrote:
>> On 2024/8/29 20:46, Pierre Gondois wrote:
>>> Hello Yicong,
>>>
>>> On 8/29/24 09:40, Yicong Yang wrote:
Hi Pierre,
On 2024/8/27 23:40, Pierre Gondois wrote:
> Hello Yicong,
>
Hello Yicong
On 8/30/24 11:35, Yicong Yang wrote:
On 2024/8/29 20:46, Pierre Gondois wrote:
Hello Yicong,
On 8/29/24 09:40, Yicong Yang wrote:
Hi Pierre,
On 2024/8/27 23:40, Pierre Gondois wrote:
Hello Yicong,
IIUC we are looking for the maximum number of threads a CPU can have
in the platf
On 2024/8/29 20:46, Pierre Gondois wrote:
> Hello Yicong,
>
> On 8/29/24 09:40, Yicong Yang wrote:
>> Hi Pierre,
>>
>> On 2024/8/27 23:40, Pierre Gondois wrote:
>>> Hello Yicong,
>>> IIUC we are looking for the maximum number of threads a CPU can have
>>> in the platform. But is is actually possib
Hello Yicong,
On 8/29/24 09:40, Yicong Yang wrote:
Hi Pierre,
On 2024/8/27 23:40, Pierre Gondois wrote:
Hello Yicong,
IIUC we are looking for the maximum number of threads a CPU can have
in the platform. But is is actually possible to have a platform with
CPUs not having the same number of thr
Hi Pierre,
On 2024/8/27 23:40, Pierre Gondois wrote:
> Hello Yicong,
> IIUC we are looking for the maximum number of threads a CPU can have
> in the platform. But is is actually possible to have a platform with
> CPUs not having the same number of threads ?
>
I was thinking of the heterogenous p
Hello Yicong,
IIUC we are looking for the maximum number of threads a CPU can have
in the platform. But is is actually possible to have a platform with
CPUs not having the same number of threads ?
For instance kernel/cpu.c::cpu_smt_num_threads_valid() will check
that the number of threads is eith
On 19/08/2024 09:03, Yicong Yang wrote:
> On 2024/8/16 23:55, Dietmar Eggemann wrote:
>> On 06/08/2024 10:53, Yicong Yang wrote:
>>> From: Yicong Yang
[...]
>> So the xarray contains one element for each core_id with the information
>> how often the core_id occurs? I assume you have to iterate o
On 2024/8/16 23:55, Dietmar Eggemann wrote:
> On 06/08/2024 10:53, Yicong Yang wrote:
>> From: Yicong Yang
>>
>> For ACPI we'll build the topology from PPTT and we cannot directly
>> get the SMT number of each core. Instead using a temporary xarray
>> to record the SMT number of each core when bui
On 06/08/2024 10:53, Yicong Yang wrote:
> From: Yicong Yang
>
> For ACPI we'll build the topology from PPTT and we cannot directly
> get the SMT number of each core. Instead using a temporary xarray
> to record the SMT number of each core when building the topology
> and we can know the largest S
From: Yicong Yang
For ACPI we'll build the topology from PPTT and we cannot directly
get the SMT number of each core. Instead using a temporary xarray
to record the SMT number of each core when building the topology
and we can know the largest SMT number in the system. Then we can
enable the supp
16 matches
Mail list logo
