Re: [Xen-devel] [PATCH v2 07/48] xen/sched: move per cpu scheduler private data into struct sched_resource

From: Jan Beulich <jbeulich@suse.com>
To: Juergen Gross <jgross@suse.com>
Cc: "Tim Deegan" <tim@xen.org>,
	"Stefano Stabellini" <sstabellini@kernel.org>,
	"Wei Liu" <wl@xen.org>,
	"Konrad Rzeszutek Wilk" <konrad.wilk@oracle.com>,
	"George Dunlap" <George.Dunlap@eu.citrix.com>,
	"Andrew Cooper" <andrew.cooper3@citrix.com>,
	"Ian Jackson" <ian.jackson@eu.citrix.com>,
	"Robert VanVossen" <robert.vanvossen@dornerworks.com>,
	"Dario Faggioli" <dfaggioli@suse.com>,
	"Julien Grall" <julien.grall@arm.com>,
	"Josh Whitehead" <josh.whitehead@dornerworks.com>,
	"Meng Xu" <mengxu@cis.upenn.edu>,
	xen-devel@lists.xenproject.org,
	"Roger Pau Monné" <roger.pau@citrix.com>
Subject: Re: [Xen-devel] [PATCH v2 07/48] xen/sched: move per cpu scheduler private data into struct sched_resource
Date: Thu, 5 Sep 2019 09:38:21 +0200	[thread overview]
Message-ID: <3e4ed5b5-38e7-6ea7-e674-40d445a2826e@suse.com> (raw)
In-Reply-To: <ecad6cb4-c5e1-a280-f86f-3cc7255a0dd3@suse.com>

On 05.09.2019 09:13, Juergen Gross wrote:
> On 04.09.19 15:48, Jan Beulich wrote:
>> On 09.08.2019 16:57, Juergen Gross wrote:
>>> This prepares support of larger scheduling granularities, e.g. core
>>> scheduling.
>>>
>>> While at it move sched_has_urgent_vcpu() from include/asm-x86/cpuidle.h
>>> into schedule.c removing the need for including sched-if.h in
>>> cpuidle.h and multiple other C sources.
>>
>> Looking again, the #include-s in .c files should have been unnecessary
>> altogether (and dropping of them could, if it was a separate patch, go
>> in right away), because of ...
> 
> Patch sent.

Thanks. I'll go apply this in a minute.

>>> --- a/xen/include/asm-x86/cpuidle.h
>>> +++ b/xen/include/asm-x86/cpuidle.h
>>> @@ -4,7 +4,6 @@
>>>   #include <xen/cpuidle.h>
>>>   #include <xen/notifier.h>
>>>   #include <xen/sched.h>
>>> -#include <xen/sched-if.h>
>>
>> ... this.
>>
>>> @@ -27,14 +26,4 @@ void update_idle_stats(struct acpi_processor_power *,
>>>   void update_last_cx_stat(struct acpi_processor_power *,
>>>                            struct acpi_processor_cx *, uint64_t);
>>>   
>>> -/*
>>> - * vcpu is urgent if vcpu is polling event channel
>>> - *
>>> - * if urgent vcpu exists, CPU should not enter deep C state
>>> - */
>>> -static inline int sched_has_urgent_vcpu(void)
>>> -{
>>> -    return atomic_read(&this_cpu(schedule_data).urgent_count);
>>> -}
>>
>> And then, despite my previous ack, I'm slightly unhappy about this
>> very short function becoming an out-of-line one, when the only
>> users of it would preferably have as low latency as possible. I
>> don't suppose there's a way to keep it inline but still drop the
>> unwanted #include above?
> 
> The only alternatives I could think of are:
> 
> - Passing the "urgent" indicator via parameter to idle (you didn't
>    like that)
> 
> - Make urgent_count a plain percpu variable

Named sched_urgent_count or some such, this latter option doesn't
look too bad to me; not sure what the scheduler maintainers think
of this, though.

Jan

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel