[PATCH v8 00/25] KVM: arm64: SMMUv3 driver for pKVM (trap and emulate)

Mostafa Saleh smostafa at google.com
Wed Sep 23 01:52:49 PDT 2026


On Tue, Sep 22, 2026 at 11:07:18AM -0700, Nicolin Chen wrote:
> On Tue, Sep 22, 2026 at 01:12:33PM +0000, Mostafa Saleh wrote:
> > Notes about Sashiko
> > ===================
> > I ran Sashiko locally and it was helpful in discovering problems in
> > the series. However, it still shows large number of critical and high
> > severity issues, I went through them and I believe they are false
> > positives, mainly because (in the order of frequently reported):
> > - It doesn't understand WARNs are fatal in the hypervisor.
> > - It doesn't understand that a malicious host can DoS the system and
> >   pKVM doesn't guarantee availability
> > - It doesn't understand the SMMUv3 spec and makes stuff up (eg. about
> >   CMD_SYC CS field it makes up an non-existent encoding or wrong
> >   semantics for the gbpa register)
> 
> Just my two cents here:
> 
> From my experience, making commit messages or inline notes slightly
> clearer against the findings that Sashiko reported could simply help
> it understand things better and mute most of the false positives.
> 

I try to do that also, it helps to a certain level, then the code
becomes too verbose and the comments will be trivial :/


> Otherwise, the number of the findings is a bit scaring; reviewers
> would have to go through the long list of findings to check whether
> each of them fits into one of the false-positive categories :-/
> 

My strategy is to reply to Sashiko on the list to save reviewers time.
But my main concern was the large number of false positives might
discourage reviewers thinking there will be a newer version with
fixes.

> > - It seems to look at one patch at a time and not the whole series, and
> >   as the series is written in a way to be bisectable that confuses it.
> 
> Yea, that happens too. Usually I would add a line at the end of the
> commit message "a subsequent change will do X", which could clarify.

There were a lot of those when I ran locally. However the one on
sashiko.dev does better.

Thanks,
Mostafa

> 
> Nicolin



More information about the linux-arm-kernel mailing list