Re: [PATCH 6/8] KVM: PPC: Add support for multiple-TCE hcalls

On 11.07.2013, at 15:13, Alexey Kardashevskiy wrote:

On 07/11/2013 10:58 PM, Benjamin Herrenschmidt wrote:

quoted

On Thu, 2013-07-11 at 14:51 +0200, Alexander Graf wrote:

quoted

I don't like bloat usually. But Alexey even had an #ifdef DEBUG in
there to selectively disable in-kernel handling of multi-TCE. Not
calling ENABLE_CAP would give him exactly that without ugly #ifdefs =

in

quoted

quoted

the kernel.

=20
I don't see much point in disabling it... but ok, if that's a =

valuable

quoted

feature, then shoot some VM level ENABLE_CAP (please don't iterate =

all

quoted

VCPUs, that's gross).

=20
No use for me whatsoever as I only want to disable real more handlers =

and

keep virtual mode handlers enabled (sometime, for debug only) and this
capability is not about that - I can easily just not enable it in QEMU =

with

the exactly the same effect.
=20
So please, fellas, decide whether I should iterate vcpu's or add =

ENABLE_CAP

per KVM. Thanks.

Thinking hard about this it might actually be ok to not have an =
ENABLE_CAP for this, if kernel code always works properly, because from =
the guest's point of view nothing changes - it either gets handled by =
kernel or by user space. And user space either handles it or doesn't, so =
it's ok.

Just leave it out this time. But be very weary of adding new features =
without an ENABLE_CAP switch. They might be guest visible changes.


Alex