[R] help.start() caused high CPU loading

[Jinsong Zhao]

On 2020/4/27 16:54, Abby Spurdle wrote:
> What's the benchmark?
I don't have any benchmark about this problem.

> 
> I (still) have Windows 10 x64.
> Opening the task manager, causes the CPU to jump to over 90%.

It will drop to about 0.5% or below very quickly.

> 
> In regards to the CPU usage remaining high:
> (1) Are you running R in a standard way?

TRUE. I closed all applications, and run the R using Rgui on Windows 10 x64.

> (2) If (1) true, is R idle (i.e. waiting for user input)?

TRUE.

> (3) If (2) true, how much CPU is R using?

0%.

> (4) If (3) is zero, what processes are using the CPU the most?

In such case, almost every processes is using the CPU in a very low 
percentage, i.e., 1% or below. I don't recognized those processes.

Now, I invoke help.start(), about 25% of CPU loading is used by R. Quit 
from R, the CPU loading go back to idle.

> 
> If (1) is false, it's not R's fault.
> If (1) is true and (2) is false, then I don't know, that would be interesting...
> If (1) and (2) are true and (3) is high, then that would be a bug in R.
> 
> If (4) is processes other than R, it's also not R's fault.
> But if you provide information, someone here maybe able to offer advise...
> 

Thanks for those suggestions.

Best,
Jinsong

> 
> On Mon, Apr 27, 2020 at 7:51 PM Jinsong Zhao <jszhao using yeah.net> wrote:
>>
>> Hi there,
>>
>> After calling help.start(), I noticed that CPU loading is up to about
>> 25% (my CPU is core i5 with 4 core). In the out html page, if I click
>> "Packages", the CPU loading may be up to 75% or much higher. After
>> closing the browser, the CPU loading does not drop.
>>
>> If I just use ?plot to open a specific help page, the CPU loading is not
>> change dramatically. However, if I click the package name (base or
>> graphics package in this example) in the help page to open the index
>> page of the corresponding package, then the CPU loading will increase to
>> more than 75%.
>>
>> I do not know if this phenomenon could be reproduced... And I do not
>> know how to trace the detailed information.
>>
>> Best,
>> Jinsonng
>>
>>   > sessionInfo()
>> R version 4.0.0 (2020-04-24)
>> Platform: x86_64-w64-mingw32/x64 (64-bit)
>> Running under: Windows 10 x64 (build 18363)
>>
>> Matrix products: default
>>
>> locale:
>> [1] LC_COLLATE=Chinese (Simplified)_China.936
>> [2] LC_CTYPE=Chinese (Simplified)_China.936
>> [3] LC_MONETARY=Chinese (Simplified)_China.936
>> [4] LC_NUMERIC=C
>> [5] LC_TIME=Chinese (Simplified)_China.936
>>
>> attached base packages:
>> [1] stats     graphics  grDevices utils     datasets  methods   base
>>
>> loaded via a namespace (and not attached):
>> [1] compiler_4.0.0 tools_4.0.0