Re: Recurring INEQUIVALENT ALIASES issues and userland corruption/crashes

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 




> On 22 Mar 2022, at 20:29, John David Anglin <dave.anglin@xxxxxxxx> wrote:
> 
> On 2022-03-22 4:04 p.m., John David Anglin wrote:
>> On 2022-03-22 3:37 p.m., Sam James wrote:
>>> In our experience so far, there has been no good kernel version for us on this hardware.
>> What happens with the attached config?  My rp3440 runs okay with this config.
>>> 
>>>> I've seen this before but it's not occurring in my current builds for rp3440 and c8000.  I've been running for-next
>>>> changes on c8000 for several weeks.
>>>> 
>>> Yeah, I haven't seen this at all on my C8000 (or Gentoo's other HW, a C3600).
>> Architecturally, there is no difference between rp3440 and c8000 with respect to the cache and TLB.
>> 

Ah, thanks, I didn't realise this.

>> As I said, the inequivalent alias are a memory mapping issue. This might vary depending on memory size.
> 
> Config might make a difference as well.
> 
> In order to debug, one needs to find the circumstances that cause it.
> 

Understood & agreed. I'm going to try Helge's config w/ 5.10.x, then yours with 5.16.x, and try get some
more information on how to reproduce too. I'll report back.

Thanks for sharing your config. I can see how it might easily make a significant difference, especially
given it can even on x86/other more-popular platforms, let alone on PARISC where we can only test
far smaller combinations.

> Dave
> 
> --
> John David Anglin  dave.anglin@xxxxxxxx
> <config-5.16.16+.gz>

Best,
sam

Attachment: signature.asc
Description: Message signed with OpenPGP


[Index of Archives]     [Linux SoC]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]

  Powered by Linux