Messages in this thread | | | Date | Mon, 29 Jun 2009 13:21:23 +0300 | From | Avi Kivity <> | Subject | Re: [PATCHv2] kvm: remove in_range and switch to rwsem for iobus |
| |
On 06/29/2009 01:06 PM, Michael S. Tsirkin wrote: >>> 2. if I do it before in_range removal it's a lot of churn: >>> one of the reasons for code reorg is so that there are less >>> places to change locking. >>> >>> >>> >>> >> I don't think you really need to change anything. slots_lock is already >> taken (except where you modify the list). >> > > Are you sure about this? I don't understand the code well enough, so > this reuse of an apparently unrelated lock just makes me nervious. For > example what about emulate_instruction? It is sometimes called from > svm/vmx without slot lock ... >
vcpu context always has slots lock taken IIRC, except when in guest mode.
It's not an unrelated lock; slots lock locks memory hotplug, we extend it to lock mmio_bus and io_bus hotplug.
I'd really like to avoid a proliferation of locks.
>> How about this: >> >> 1. add slots_lock for write when modifying the list >> 2. change the api >> 3. drop kvm->lock >> >> ? >> > > Looks like I will just have to bite the bullet and switch to RCU. > >
You still need a lock to prevent concurrent modifications to mmio_bus (but can use kvm->lock for this).
-- error compiling committee.c: too many arguments to function
| |