testing/next (vm, gitlab)

[PATCH v1 12/18] tests/vm: allow us to take advantage of MTTCG

Posted by Alex Bennée 5 years, 7 months ago

We currently limit TCG guests to -smp 1 but now we have added some
aarch64 guests we can do better when running on x86_64 hardware.
Raise the limit for TCG guests when it is safe to do so.

Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
---
 tests/vm/basevm.py | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/tests/vm/basevm.py b/tests/vm/basevm.py
index 93859362606..dd96a6d4af6 100644
--- a/tests/vm/basevm.py
+++ b/tests/vm/basevm.py
@@ -553,6 +553,12 @@ def parse_args(vmcls):
     def get_default_jobs():
         if kvm_available(vmcls.arch):
             return multiprocessing.cpu_count() // 2
+        elif os.uname().machine == "x86_64" and \
+             vmcls.arch in ["aarch64", "x86_64", "i386"]:
+            # MTTCG is available on these arches and we can allow more cores.
+            # But only up to a reasonable limit. User can always override
+            # these limits with --jobs.
+            return min(multiprocessing.cpu_count() // 2, 8)
         else:
             return 1
 
-- 
2.20.1

Re: [PATCH v1 12/18] tests/vm: allow us to take advantage of MTTCG

Posted by Robert Foley 5 years, 7 months ago

Reviewed-by: Robert Foley <robert.foley@linaro.org>

On Mon, 22 Jun 2020 at 10:38, Alex Bennée <alex.bennee@linaro.org> wrote:
>
> We currently limit TCG guests to -smp 1 but now we have added some
> aarch64 guests we can do better when running on x86_64 hardware.
> Raise the limit for TCG guests when it is safe to do so.
>
> Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
> ---
>  tests/vm/basevm.py | 6 ++++++
>  1 file changed, 6 insertions(+)
>
> diff --git a/tests/vm/basevm.py b/tests/vm/basevm.py
> index 93859362606..dd96a6d4af6 100644
> --- a/tests/vm/basevm.py
> +++ b/tests/vm/basevm.py
> @@ -553,6 +553,12 @@ def parse_args(vmcls):
>      def get_default_jobs():
>          if kvm_available(vmcls.arch):
>              return multiprocessing.cpu_count() // 2
> +        elif os.uname().machine == "x86_64" and \
> +             vmcls.arch in ["aarch64", "x86_64", "i386"]:
> +            # MTTCG is available on these arches and we can allow more cores.
> +            # But only up to a reasonable limit. User can always override
> +            # these limits with --jobs.
> +            return min(multiprocessing.cpu_count() // 2, 8)
>          else:
>              return 1
>
> --
> 2.20.1
>
>

Re: [PATCH v1 12/18] tests/vm: allow us to take advantage of MTTCG

Posted by Richard Henderson 5 years, 7 months ago

On 6/22/20 7:31 AM, Alex Bennée wrote:
>          if kvm_available(vmcls.arch):
>              return multiprocessing.cpu_count() // 2
> +        elif os.uname().machine == "x86_64" and \
> +             vmcls.arch in ["aarch64", "x86_64", "i386"]:
> +            # MTTCG is available on these arches and we can allow more cores.
> +            # But only up to a reasonable limit. User can always override
> +            # these limits with --jobs.
> +            return min(multiprocessing.cpu_count() // 2, 8)
>          else:

And if multiprocessing.cpu_count() == 1?
Seems like we should add max(count, 1) as well.


r~

Re: [PATCH v1 12/18] tests/vm: allow us to take advantage of MTTCG

Posted by Alex Bennée 5 years, 7 months ago

Richard Henderson <richard.henderson@linaro.org> writes:

> On 6/22/20 7:31 AM, Alex Bennée wrote:
>>          if kvm_available(vmcls.arch):
>>              return multiprocessing.cpu_count() // 2
>> +        elif os.uname().machine == "x86_64" and \
>> +             vmcls.arch in ["aarch64", "x86_64", "i386"]:
>> +            # MTTCG is available on these arches and we can allow more cores.
>> +            # But only up to a reasonable limit. User can always override
>> +            # these limits with --jobs.
>> +            return min(multiprocessing.cpu_count() // 2, 8)
>>          else:
>
> And if multiprocessing.cpu_count() == 1?
> Seems like we should add max(count, 1) as well.

Or maybe?

   min(math.ceil(multiprocessing.cpu_count() / 2), 8)
>
>
> r~


-- 
Alex Bennée

Re: [PATCH v1 12/18] tests/vm: allow us to take advantage of MTTCG

Posted by Alex Bennée 5 years, 7 months ago

Richard Henderson <richard.henderson@linaro.org> writes:

> On 6/22/20 7:31 AM, Alex Bennée wrote:
>>          if kvm_available(vmcls.arch):
>>              return multiprocessing.cpu_count() // 2
>> +        elif os.uname().machine == "x86_64" and \
>> +             vmcls.arch in ["aarch64", "x86_64", "i386"]:
>> +            # MTTCG is available on these arches and we can allow more cores.
>> +            # But only up to a reasonable limit. User can always override
>> +            # these limits with --jobs.
>> +            return min(multiprocessing.cpu_count() // 2, 8)
>>          else:
>
> And if multiprocessing.cpu_count() == 1?
> Seems like we should add max(count, 1) as well.

As it also affects KVM:

    def get_default_jobs():
        if multiprocessing.cpu_count > 1:
            if kvm_available(vmcls.arch):
                return multiprocessing.cpu_count() // 2
            elif os.uname().machine == "x86_64" and \
                 vmcls.arch in ["aarch64", "x86_64", "i386"]:
                # MTTCG is available on these arches and we can allow more cores.
                # But only up to a reasonable limit. User can always override
                # these limits with --jobs.
                return min(multiprocessing.cpu_count() // 2, 8)
        else:
            return 1


>
>
> r~


-- 
Alex Bennée

Re: [PATCH v1 12/18] tests/vm: allow us to take advantage of MTTCG

Posted by Philippe Mathieu-Daudé 5 years, 7 months ago

On 6/29/20 4:41 PM, Alex Bennée wrote:
> 
> Richard Henderson <richard.henderson@linaro.org> writes:
> 
>> On 6/22/20 7:31 AM, Alex Bennée wrote:
>>>          if kvm_available(vmcls.arch):
>>>              return multiprocessing.cpu_count() // 2
>>> +        elif os.uname().machine == "x86_64" and \
>>> +             vmcls.arch in ["aarch64", "x86_64", "i386"]:
>>> +            # MTTCG is available on these arches and we can allow more cores.
>>> +            # But only up to a reasonable limit. User can always override
>>> +            # these limits with --jobs.
>>> +            return min(multiprocessing.cpu_count() // 2, 8)
>>>          else:
>>
>> And if multiprocessing.cpu_count() == 1?
>> Seems like we should add max(count, 1) as well.
> 
> As it also affects KVM:
> 
>     def get_default_jobs():
>         if multiprocessing.cpu_count > 1:
>             if kvm_available(vmcls.arch):
>                 return multiprocessing.cpu_count() // 2
>             elif os.uname().machine == "x86_64" and \
>                  vmcls.arch in ["aarch64", "x86_64", "i386"]:
>                 # MTTCG is available on these arches and we can allow more cores.
>                 # But only up to a reasonable limit. User can always override
>                 # these limits with --jobs.
>                 return min(multiprocessing.cpu_count() // 2, 8)
>         else:
>             return 1

For this hunk:
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>

> 
> 
>>
>>
>> r~
> 
>