Re: [RFC PATCH 0/4] Scheduler time slice extension

[Mathieu Desnoyers <mathieu.desnoyers@efficios.com>]

On 2024-12-09 15:36, Prakash Sangappa wrote:
> 
> 
>> On Nov 14, 2024, at 11:41 AM, Prakash Sangappa <prakash.sangappa@oracle.com> wrote:
>>
>>
>>
>>> On Nov 14, 2024, at 2:28 AM, Peter Zijlstra <peterz@infradead.org> wrote:
>>>
>>> On Wed, Nov 13, 2024 at 08:10:52PM +0000, Prakash Sangappa wrote:
>>>>
>>>>
>>>>> On Nov 13, 2024, at 11:36 AM, Mathieu Desnoyers <mathieu.desnoyers@efficios.com> wrote:
>>>>>
>>>>> On 2024-11-13 13:50, Peter Zijlstra wrote:
>>>>>> On Wed, Nov 13, 2024 at 12:01:22AM +0000, Prakash Sangappa wrote:
>>>>>>> This patch set implements the above mentioned 50us extension time as posted
>>>>>>> by Peter. But instead of using restartable sequences as API to set the flag
>>>>>>> to request the extension, this patch proposes a new API with use of a per
>>>>>>> thread shared structure implementation described below. This shared structure
>>>>>>> is accessible in both users pace and kernel. The user thread will set the
>>>>>>> flag in this shared structure to request execution time extension.
>>>>>> But why -- we already have rseq, glibc uses it by default. Why add yet
>>>>>> another thing?
>>>>>
>>>>> Indeed, what I'm not seeing in this RFC patch series cover letter is an
>>>>> explanation that justifies adding yet another per-thread memory area
>>>>> shared between kernel and userspace when we have extensible rseq
>>>>> already.
>>>>
>>>> It mainly provides pinned memory, can be useful for  future use cases
>>>> where updating user memory in kernel context can be fast or needs to
>>>> avoid pagefaults.
>>>
>>> 'might be useful' it not good enough a justification. Also, I don't
>>> think you actually need this.
>>
>> Will get back with database benchmark results using rseq API for scheduler time extension.
> 
> Sorry about the delay in response.
> 
> Here are the database swingbench numbers   - includes results with use of rseq API.
> 
> Test results:
> =========
> Test system 2 socket AMD Genoa
> 
>   	Swingbench - standard database benchmark
> 		Cached(database files on tmpfs) run, with 1000 clients.
> 
> 		Baseline(Without Sched time extension):  99K SQL exec/sec
> 
> 		With Sched time extension:
> 				Shared structure API use: 		153K SQL exec/sec  (Previously reported)
> 			55% improvement in throughput.
> 
> 				Restartable sequences API use:	147K SQL exec/sec
>   			48% improvement in throughput
> 
> While both show good performance benefit with scheduler time extension,
> there is a 7% difference in throughput between Shared structure & Restartable sequences API.
> Use of shared structure is faster.

Can you share the code for both test cases ? And do you have relevant
perf profile showing where time is spent ?

Thanks,

Mathieu

> 
> 
> 
>>
>>>
>>> See:
>>>
>>> https://lkml.kernel.org/r/20220113233940.3608440-4-posk@google.com
>>>
>>> for a more elaborate scheme.
>>>
>>>>> Peter, was there anything fundamentally wrong with your approach based
>>>>> on rseq ? https://lore.kernel.org/lkml/20231030132949.GA38123@noisy.programming.kicks-ass.net
>>>>>
>>>>> The main thing I wonder is whether loading the rseq delay resched flag
>>>>> on return to userspace is too late in your patch. Also, I'm not sure it is
>>>>> realistic to require that no system calls should be done within time extension
>>>>> slice. If we have this scenario:
>>>>
>>>> I am also not sure if we need to prevent system calls in this scenario.
>>>> Was that restriction mainly because of restartable sequence API implements it?
>>>
>>> No, the whole premise of delaying resched was because people think that
>>> syscalls are too slow. If you do not think this, then you shouldn't be
>>> using this.
>>
>> Agree.
>>
>> Thanks,
>> -Prakash
> 
> 

-- 
Mathieu Desnoyers
EfficiOS Inc.
https://www.efficios.com