Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1265506 > unrolled thread

Re: [PATCH v2 16/19] ARC: [plat-eznps] Use dedicated cpu_relax()

Started byPeter Zijlstra <peterz@infradead.org>
First post2015-11-09 11:10 +0100
Last post2015-11-09 14:00 +0100
Articles 5 — 3 participants

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Re: [PATCH v2 16/19] ARC: [plat-eznps] Use dedicated cpu_relax() Peter Zijlstra <peterz@infradead.org> - 2015-11-09 11:10 +0100
    Re: [PATCH v2 16/19] ARC: [plat-eznps] Use dedicated cpu_relax() Vineet Gupta <Vineet.Gupta1@synopsys.com> - 2015-11-09 11:30 +0100
      Re: [PATCH v2 16/19] ARC: [plat-eznps] Use dedicated cpu_relax() Peter Zijlstra <peterz@infradead.org> - 2015-11-09 11:50 +0100
        Re: [PATCH v2 16/19] ARC: [plat-eznps] Use dedicated cpu_relax() Vineet Gupta <vgupta@synopsys.com> - 2015-11-09 13:30 +0100
          Re: [PATCH v2 16/19] ARC: [plat-eznps] Use dedicated cpu_relax() Peter Zijlstra <peterz@infradead.org> - 2015-11-09 14:00 +0100

#1265506 — Re: [PATCH v2 16/19] ARC: [plat-eznps] Use dedicated cpu_relax()

FromPeter Zijlstra <peterz@infradead.org>
Date2015-11-09 11:10 +0100
SubjectRe: [PATCH v2 16/19] ARC: [plat-eznps] Use dedicated cpu_relax()
Message-ID<qsRyG-50w-31@gated-at.bofh.it>
On Sat, Nov 07, 2015 at 12:52:34PM +0200, Noam Camus wrote:
> diff --git a/arch/arc/include/asm/processor.h b/arch/arc/include/asm/processor.h
> index 7266ede..50f9bae 100644
> --- a/arch/arc/include/asm/processor.h
> +++ b/arch/arc/include/asm/processor.h
> @@ -58,12 +58,21 @@ struct task_struct;
>   * get optimised away by gcc
>   */
>  #ifdef CONFIG_SMP
> +#ifndef CONFIG_EZNPS_MTM_EXT
>  #define cpu_relax()	__asm__ __volatile__ ("" : : : "memory")
>  #else
> +#define cpu_relax()     \
> +	__asm__ __volatile__ (".word %0" : : "i"(CTOP_INST_SCHD_RW) : "memory")
> +#endif
> +#else
>  #define cpu_relax()	do { } while (0)

I'm fairly sure this is incorrect. Even on UP we expect cpu_relax() to
be a compiler barrier.

>  #endif
>  
> +#ifndef CONFIG_EZNPS_MTM_EXT
>  #define cpu_relax_lowlatency() cpu_relax()
> +#else
> +#define cpu_relax_lowlatency() barrier()
> +#endif

At which point you can unconditionally use that definition.

>  
>  #define copy_segments(tsk, mm)      do { } while (0)
>  #define release_segments(mm)        do { } while (0)
> -- 
> 1.7.1
> 
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [next] | [standalone]

#1265549

FromVineet Gupta <Vineet.Gupta1@synopsys.com>
Date2015-11-09 11:30 +0100
Message-ID<qsRS2-598-29@gated-at.bofh.it>
In reply to#1265506
On Monday 09 November 2015 03:35 PM, Peter Zijlstra wrote:
> On Sat, Nov 07, 2015 at 12:52:34PM +0200, Noam Camus wrote:
>> diff --git a/arch/arc/include/asm/processor.h b/arch/arc/include/asm/processor.h
>> index 7266ede..50f9bae 100644
>> --- a/arch/arc/include/asm/processor.h
>> +++ b/arch/arc/include/asm/processor.h
>> @@ -58,12 +58,21 @@ struct task_struct;
>>   * get optimised away by gcc
>>   */
>>  #ifdef CONFIG_SMP
>> +#ifndef CONFIG_EZNPS_MTM_EXT
>>  #define cpu_relax()	__asm__ __volatile__ ("" : : : "memory")
>>  #else
>> +#define cpu_relax()     \
>> +	__asm__ __volatile__ (".word %0" : : "i"(CTOP_INST_SCHD_RW) : "memory")
>> +#endif
>> +#else
>>  #define cpu_relax()	do { } while (0)
> I'm fairly sure this is incorrect. Even on UP we expect cpu_relax() to
> be a compiler barrier.

We discussed this a while back (why do https:/lkml.org/lkml/<year>/.... links work
psuedo randomly)

http://marc.info/?l=linux-kernel&m=140350765530113


>
>>  #endif
>>  
>> +#ifndef CONFIG_EZNPS_MTM_EXT
>>  #define cpu_relax_lowlatency() cpu_relax()
>> +#else
>> +#define cpu_relax_lowlatency() barrier()
>> +#endif
> At which point you can unconditionally use that definition.
>
>>  
>>  #define copy_segments(tsk, mm)      do { } while (0)
>>  #define release_segments(mm)        do { } while (0)
>> -- 
>> 1.7.1
>>
> _______________________________________________
> linux-snps-arc mailing list
> linux-snps-arc@lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-snps-arc
>

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1265557

FromPeter Zijlstra <peterz@infradead.org>
Date2015-11-09 11:50 +0100
Message-ID<qsSbn-5gy-15@gated-at.bofh.it>
In reply to#1265549
On Mon, Nov 09, 2015 at 10:22:27AM +0000, Vineet Gupta wrote:
> On Monday 09 November 2015 03:35 PM, Peter Zijlstra wrote:
> > On Sat, Nov 07, 2015 at 12:52:34PM +0200, Noam Camus wrote:
> >> diff --git a/arch/arc/include/asm/processor.h b/arch/arc/include/asm/processor.h
> >> index 7266ede..50f9bae 100644
> >> --- a/arch/arc/include/asm/processor.h
> >> +++ b/arch/arc/include/asm/processor.h
> >> @@ -58,12 +58,21 @@ struct task_struct;
> >>   * get optimised away by gcc
> >>   */
> >>  #ifdef CONFIG_SMP
> >> +#ifndef CONFIG_EZNPS_MTM_EXT
> >>  #define cpu_relax()	__asm__ __volatile__ ("" : : : "memory")
> >>  #else
> >> +#define cpu_relax()     \
> >> +	__asm__ __volatile__ (".word %0" : : "i"(CTOP_INST_SCHD_RW) : "memory")
> >> +#endif
> >> +#else
> >>  #define cpu_relax()	do { } while (0)
> > I'm fairly sure this is incorrect. Even on UP we expect cpu_relax() to
> > be a compiler barrier.
> 
> We discussed this a while back (why do https:/lkml.org/lkml/<year>/.... links work
> psuedo randomly)
> 
> http://marc.info/?l=linux-kernel&m=140350765530113

Hurm.. you have a better memory than me ;-)

So in general we assume cpu_relax() implies a barrier() and I think we
have loops like:

	while (!var)
		cpu_relax();

where var isn't volatile (or casted using READ_ONCE etc).

See for instance: kernel/time/timer.c:lock_timer_base() which has:

	for (;;) {
		u32 tf = timer->flags;

		if (!(tf & TIMER_MIGRATING)) {
		 ...
		}

		cpu_relax();
	}

So while TIMER_MIGRATING is set, it will only ever do regular loads,
which GCC is permitted to lift out if cpu_relax() is not a barrier.


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1265614

FromVineet Gupta <vgupta@synopsys.com>
Date2015-11-09 13:30 +0100
Message-ID<qsTKa-6ln-7@gated-at.bofh.it>
In reply to#1265557
On Monday 09 November 2015 04:15 PM, Peter Zijlstra wrote:
> On Mon, Nov 09, 2015 at 10:22:27AM +0000, Vineet Gupta wrote:
>> On Monday 09 November 2015 03:35 PM, Peter Zijlstra wrote:
>>> On Sat, Nov 07, 2015 at 12:52:34PM +0200, Noam Camus wrote:
>>>> diff --git a/arch/arc/include/asm/processor.h b/arch/arc/include/asm/processor.h
>>>> index 7266ede..50f9bae 100644
>>>> --- a/arch/arc/include/asm/processor.h
>>>> +++ b/arch/arc/include/asm/processor.h
>>>> @@ -58,12 +58,21 @@ struct task_struct;
>>>>   * get optimised away by gcc
>>>>   */
>>>>  #ifdef CONFIG_SMP
>>>> +#ifndef CONFIG_EZNPS_MTM_EXT
>>>>  #define cpu_relax()	__asm__ __volatile__ ("" : : : "memory")
>>>>  #else
>>>> +#define cpu_relax()     \
>>>> +	__asm__ __volatile__ (".word %0" : : "i"(CTOP_INST_SCHD_RW) : "memory")
>>>> +#endif
>>>> +#else
>>>>  #define cpu_relax()	do { } while (0)
>>> I'm fairly sure this is incorrect. Even on UP we expect cpu_relax() to
>>> be a compiler barrier.
>>
>> We discussed this a while back (why do https:/lkml.org/lkml/<year>/.... links work
>> psuedo randomly)
>>
>> http://marc.info/?l=linux-kernel&m=140350765530113
> 
> Hurm.. you have a better memory than me ;-)
> 
> So in general we assume cpu_relax() implies a barrier() and I think we
> have loops like:
> 
> 	while (!var)
> 		cpu_relax();
> 
> where var isn't volatile (or casted using READ_ONCE etc).
> 
> See for instance: kernel/time/timer.c:lock_timer_base() which has:
> 
> 	for (;;) {
> 		u32 tf = timer->flags;
> 
> 		if (!(tf & TIMER_MIGRATING)) {
> 		 ...
> 		}
> 
> 		cpu_relax();
> 	}
> 
> So while TIMER_MIGRATING is set, it will only ever do regular loads,
> which GCC is permitted to lift out if cpu_relax() is not a barrier.

I'll just bite the bullet and make it a compiler barrier and send Linus way in
4.4. Care to provide an Ack or some such.

-------------------->
From e29de8efa621b825891dcc744c84965b38f6b868 Mon Sep 17 00:00:00 2001
From: Vineet Gupta <vgupta@synopsys.com>
Date: Mon, 9 Nov 2015 17:48:34 +0530
Subject: [PATCH] ARC: cpu_relax() to be compiler barrier even for UP

cpu_relax() on ARC has been barrier only for SMP (and no-op for UP). Per
recent discussions, it is safer to make it a compiler barrier
unconditionally.

Link: http://lkml.kernel.org/r/53A7D3AA.9020100@synopsys.com
Cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
---
 arch/arc/include/asm/processor.h | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/arch/arc/include/asm/processor.h b/arch/arc/include/asm/processor.h
index 44545354e9e8..1d694c1ef6d6 100644
--- a/arch/arc/include/asm/processor.h
+++ b/arch/arc/include/asm/processor.h
@@ -57,11 +57,7 @@ struct task_struct;
  * A lot of busy-wait loops in SMP are based off of non-volatile data otherwise
  * get optimised away by gcc
  */
-#ifdef CONFIG_SMP
 #define cpu_relax()	__asm__ __volatile__ ("" : : : "memory")
-#else
-#define cpu_relax()	do { } while (0)
-#endif

 #define cpu_relax_lowlatency() cpu_relax()

-- 
1.9.1

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1265629

FromPeter Zijlstra <peterz@infradead.org>
Date2015-11-09 14:00 +0100
Message-ID<qsUdb-6vT-7@gated-at.bofh.it>
In reply to#1265614
On Mon, Nov 09, 2015 at 05:57:25PM +0530, Vineet Gupta wrote:
> From e29de8efa621b825891dcc744c84965b38f6b868 Mon Sep 17 00:00:00 2001
> From: Vineet Gupta <vgupta@synopsys.com>
> Date: Mon, 9 Nov 2015 17:48:34 +0530
> Subject: [PATCH] ARC: cpu_relax() to be compiler barrier even for UP
> 
> cpu_relax() on ARC has been barrier only for SMP (and no-op for UP). Per
> recent discussions, it is safer to make it a compiler barrier
> unconditionally.
> 
> Link: http://lkml.kernel.org/r/53A7D3AA.9020100@synopsys.com

Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>

> Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
> ---
>  arch/arc/include/asm/processor.h | 4 ----
>  1 file changed, 4 deletions(-)
> 
> diff --git a/arch/arc/include/asm/processor.h b/arch/arc/include/asm/processor.h
> index 44545354e9e8..1d694c1ef6d6 100644
> --- a/arch/arc/include/asm/processor.h
> +++ b/arch/arc/include/asm/processor.h
> @@ -57,11 +57,7 @@ struct task_struct;
>   * A lot of busy-wait loops in SMP are based off of non-volatile data otherwise
>   * get optimised away by gcc
>   */
> -#ifdef CONFIG_SMP
>  #define cpu_relax()	__asm__ __volatile__ ("" : : : "memory")
> -#else
> -#define cpu_relax()	do { } while (0)
> -#endif
> 
>  #define cpu_relax_lowlatency() cpu_relax()
> 
> -- 
> 1.9.1
> 
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web