linux-kernel - Re: [PATCH] MIPS: remove the obsolete RM7000 extended interrupts handler

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Thu, 1 Oct 2020 00:10:57 +0100 (BST)
From:   "Maciej W. Rozycki" <macro@...ux-mips.org>
To:     Thomas Bogendoerfer <tsbogend@...ha.franken.de>
cc:     Necip Fazil Yildiran <fazilyildiran@...il.com>,
        linux-mips@...r.kernel.org, linux-kernel@...r.kernel.org,
        paul@...zz.com, jeho@...utexas.edu
Subject: Re: [PATCH] MIPS: remove the obsolete RM7000 extended interrupts
 handler

On Wed, 30 Sep 2020, Thomas Bogendoerfer wrote:

> > > not yet;-) I have an RM7000 based system in my basement... I'm also
> > > not sure, if R7k O2s need that.
> > 
> >  One of the more exotic Malta daughtercards used that too I believe.  I'd 
> > have to check the docs.  Of course getting hands on such a daughtercard is 
> > another matter, but we continue having RM7000 enabled for the Malta.
> 
> I've checked interrupts on P6064 and doesn't use the "extented" interrupts.
> And since O2 uses the same CPU connector for all CPU boards I doubt it
> it either.

 Mind that interrupts can be asserted in the RM7k via an external write 
request too, and I'm fairly sure SGI hardware would use that rather than 
interrupt inputs given that R4000/R4400 SC/MC CPUs only have Int*(0) wired 
and all the remaining interrupts are only available via a processor write.  
Have you checked that?

> What about Malta ? 

 So this is the CoreBonito64 daughtercard, which can have either an RM5261 
or an RM7061 processor installed.  The Int*(9:6) inputs are not mentioned, 
which I take it means unconnected, and ExtRqst* has a pull-up and is 
otherwise wired to a debug connector only.  Which makes me conclude this 
feature cannot be used in a standard application with this card.

 NB there is another interrupt controller embedded in the Bonito64 system 
controller, which on the CoreBonito64, conversely, can actually be enabled 
with a jumper and used, but it's not usually either.

  Maciej