Re: [PATCH RESEND 1/1] lib/Kconfig: remove DEBUG_PER_CPU_MAPS dependency for CPUMASK_OFFSTACK

[Libo Chen <libo.chen@oracle.com>]

On 4/13/22 13:52, Arnd Bergmann wrote:
> On Wed, Apr 13, 2022 at 9:28 PM Libo Chen <libo.chen@oracle.com> wrote:
>> On 4/13/22 08:41, Randy Dunlap wrote:
>>> On 4/12/22 23:56, Libo Chen wrote:
>>>>> --- a/lib/Kconfig
>>>>> +++ b/lib/Kconfig
>>>>> @@ -511,7 +511,8 @@ config CHECK_SIGNATURE
>>>>>         bool
>>>>>       config CPUMASK_OFFSTACK
>>>>> -    bool "Force CPU masks off stack" if DEBUG_PER_CPU_MAPS
>>>>> +    bool "Force CPU masks off stack"
>>>>> +    depends on DEBUG_PER_CPU_MAPS
>>>> This forces every arch to enable DEBUG_PER_CPU_MAPS if they want to enable CPUMASK_OFFSTACK, it's even stronger than "if". My whole argument is CPUMASK_OFFSTACK should be enable/disabled independent of DEBUG_PER_CPU_MASK
>>>>>         help
>>>>>           Use dynamic allocation for cpumask_var_t, instead of putting
>>>>>           them on the stack.  This is a bit more expensive, but avoids
>>>>>
>>>>>
>>>>> As I said earlier, the "if" on the "bool" line just controls the prompt message.
>>>>> This patch make CPUMASK_OFFSTACK require DEBUG_PER_CPU_MAPS -- which might be overkill.
>>>>>
>>>> Okay I understand now "if" on the "boot" is not a dependency and it only controls the prompt message, then the question is why we cannot enable CPUMASK_OFFSTACK without DEBUG_PER_CPU_MAPS if it only controls prompt message? Is it not the behavior we expect?
>>> Yes, it is. I don't know that the problem is...
>> Masahiro explained that CPUMASK_OFFSTACK can only be configured by
>> options not users if DEBUG_PER_CPU_MASK is not enabled. This doesn't
>> seem to be what we want.
> I think the correct way to do it is to follow x86 and powerpc, and tying
> CPUMASK_OFFSTACK to "large" values of CONFIG_NR_CPUS.
> For smaller values of NR_CPUS, the onstack masks are obviously
> cheaper, we just need to decide what the cut-off point is.
I agree. It appears enabling CPUMASK_OFFSTACK breaks kernel builds on 
some architectures such as parisc and nios2 as reported by kernel test 
robot. Maybe it makes sense to use DEBUG_PER_CPU_MAPS as some kind of 
guard on CPUMASK_OFFSTACK.
> In x86, the onstack masks can be selected for normal SMP builds with
> up to 512 CPUs, while CONFIG_MAXSMP=y raises the limit to 8192
> CPUs while selecting CPUMASK_OFFSTACK.
> PowerPC does it the other way round, selecting CPUMASK_OFFSTACK
> implicitly whenever NR_CPUS is set to 8192 or more.
>
> I think we can easily do the same as powerpc on arm64. With the
I am leaning more towards x86's way because even NR_CPUS=160 is too 
expensive for 4-core arm64 VMs according to apachebench. I highly doubt 
that there is a good cut-off point to make everybody happy (or not unhappy).
> ApacheBench test you cite in the patch description, what is the
> value of NR_CPUS at which you start seeing a noticeable
> benefit for offstack masks? Can you do the same test for
> NR_CPUS=1024 or 2048?
As mentioned above, a good cut-off point moves depends on the actual 
number of CPUs. But yeah I can do the same test for 1024 or even smaller 
NR_CPUs values on the same 64-core arm64 VM setup.

Libo

>
>             Arnd