qemu-devel
[Top][All Lists]
Advanced

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

Re: [Qemu-devel] [PATCH] msi: remove return code for msi_init()


From: Peter Xu
Subject: Re: [Qemu-devel] [PATCH] msi: remove return code for msi_init()
Date: Wed, 31 May 2017 16:26:51 +0800
User-agent: Mutt/1.5.24 (2015-08-30)

On Wed, May 31, 2017 at 03:40:14AM -0400, Paolo Bonzini wrote:
> 
> > > >>            I'd prefer to remove msi_nonbroken completely if we don't
> > > >> know where the problem is.
> > > > 
> > > > So you're proposing to (1) remove msi_nonbroken, (2) see which boards
> > > > burst into flames, and (3) fix them, or perhaps add a less wrong stop
> > > > gap msi_broken just for them?
> > > 
> > > Yes, adding back msi_broken is one "git revert" away.
> > 
> > Not sure whether I got the point here, but... Adding msi_broken is not
> > a "git revert"? Since there is no msi_broken before, only
> > msi_supported.
> 
> Not exactly a "git revert", but pretty close since the logic for error 
> reporting
> is the same and those are the 100 lines your patch removes.  The hard part is
> finding which boards need it.

Ok before I move on let's see whether this is what we want...

- firstly, find all machine types:

  pxdev:qemu [edu-fix]# grep -R ".parent = TYPE_MACHINE" * | wc
     49     196    2269

  so now we have 49 kinds of machines.

- rename msi_nonbroken into msi_broken, then:

  - x86/arm/spapr/s390 machines are the only ones that don't need to
    set msi_broken since they support MSI and have msi_nonbroken set,
    either in board init function or in irq chip init function

  - for all the rest of the machines, I should add "msi_broken" in its
    machine init() function.

Is this really what we want?

-- 
Peter Xu



reply via email to

[Prev in Thread] Current Thread [Next in Thread]