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


Groups > linux.kernel > #1612552 > unrolled thread

Re: spin_lock behavior with ARM64 big.Little/HMP

Started byVikram Mulukutla <markivx@codeaurora.org>
First post2017-03-30 06:20 +0200
Last post2017-03-30 12:30 +0200
Articles 2 — 2 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: spin_lock behavior with ARM64 big.Little/HMP Vikram Mulukutla <markivx@codeaurora.org> - 2017-03-30 06:20 +0200
    Re: spin_lock behavior with ARM64 big.Little/HMP Sudeep Holla <sudeep.holla@arm.com> - 2017-03-30 12:30 +0200

#1612552 — Re: spin_lock behavior with ARM64 big.Little/HMP

FromVikram Mulukutla <markivx@codeaurora.org>
Date2017-03-30 06:20 +0200
SubjectRe: spin_lock behavior with ARM64 big.Little/HMP
Message-ID<tqzJ0-HD-7@gated-at.bofh.it>
Hi Sudeep,

> 
> Interesting. Just curious if this is r0p0/p1 A53 ? If so, is the errata
> 819472 enabled ?

Sorry for bringing this up after the loo-ong delay, but I've been 
assured that the A53 involved is > r0p1. I've also confirmed this 
problem on multiple internal platforms, and I'm pretty sure that it 
occurs on any b.L out there today. Also, we found the same problematic 
lock design used in the workqueue code in the kernel, causing the same 
livelock. It's very very rare and requires a perfect set of 
circumstances.

If it would help I can provide a unit test if you folks would be 
generous enough to test it on the latest Juno or something b.L that's 
also upstream.

Thanks,
Vikram

[toc] | [next] | [standalone]


#1612891

FromSudeep Holla <sudeep.holla@arm.com>
Date2017-03-30 12:30 +0200
Message-ID<tqFv4-4UX-5@gated-at.bofh.it>
In reply to#1612552

On 30/03/17 05:12, Vikram Mulukutla wrote:
> 
> Hi Sudeep,
> 
>>
>> Interesting. Just curious if this is r0p0/p1 A53 ? If so, is the errata
>> 819472 enabled ?
> 
> Sorry for bringing this up after the loo-ong delay, but I've been
> assured that the A53 involved is > r0p1. I've also confirmed this
> problem on multiple internal platforms, and I'm pretty sure that it
> occurs on any b.L out there today. Also, we found the same problematic
> lock design used in the workqueue code in the kernel, causing the same
> livelock. It's very very rare and requires a perfect set of circumstances.
> 
> If it would help I can provide a unit test if you folks would be
> generous enough to test it on the latest Juno or something b.L that's
> also upstream.
> 

Sure, please do share the unit test. I will give that a try on Juno.

-- 
Regards,
Sudeep

[toc] | [prev] | [standalone]


Back to top | Article view | linux.kernel


csiph-web