Re: [PATCH] MIPS: mm: Prevent a TLB shutdown on initial uniquification

[Thomas Bogendoerfer]

On Tue, Nov 11, 2025 at 03:41:50PM +0000, Maciej W. Rozycki wrote:
> On Tue, 11 Nov 2025, Thomas Bogendoerfer wrote:
> 
> > >  Can you please give it a try with your systems?
> > 
> > it's booting on my R4400 SGI Indy, but I see a lot of segmentation
> > faults during system start. If I comment out r4k_tlb_uniquify() every-
> > thing boots fine, which is kind of strange as there is a local_flush_tlb_all(),
> > which should leave the TLB in the same stage.... No idea why, yet.
> 
>  Can you try the diagnostic patch below, which is what I used to verify 
> this change, and report the entries produced?  Otherwise I wonder whether 
> I haven't missed a barrier somewhere.

Update on the issue: Your patch is good and the segmentation faults,
I'm seeing, have IMHO a different reason. Instead of removing the call
to r4k_tlb_uniquify() I've replaced the jal in the binary with a nop.
And the issue is still there with this patched kernel. I've seen
something similair on a R12k Octanes, which comes and goes probably
depeding on code layout. So far I wasn't able to nail this down :-(

Do you want to send a v2 of the patch ? I'm fine with the current version
for applying...

Thomas.

-- 
Crap can work. Given enough thrust pigs will fly, but it's not necessarily a
good idea.                                                [ RFC1925, 2.3 ]