Scott Tsai wrote:
> I do have a newbie question, when exactly would vrtio have to handle
> concurrent access from multiple threads?
> My current reading of the code suggests:
> 1. when CONFIG_IOTHREAD is true
> 2. when CONFIG_KVM is true and the guest machine has multiple CPUs
It's enough to have
On Thu, Nov 12, 2009 at 02:37:26AM +0800, Scott Tsai wrote:
> On Thu, Nov 12, 2009 at 2:09 AM, Michael S. Tsirkin wrote:
> >> I do have a newbie question, when exactly would vrtio have to handle
> >> concurrent access from multiple threads?
> >> My current reading of the code suggests:
> >> 1. whe
On Thu, Nov 12, 2009 at 2:09 AM, Michael S. Tsirkin wrote:
>> I do have a newbie question, when exactly would vrtio have to handle
>> concurrent access from multiple threads?
>> My current reading of the code suggests:
>> 1. when CONFIG_IOTHREAD is true
>> 2. when CONFIG_KVM is true and the guest
On Thu, Nov 12, 2009 at 01:18:11AM +0800, Scott Tsai wrote:
> On Wed, Nov 11, 2009 at 10:08 PM, Michael S. Tsirkin wrote:
> > On Wed, Nov 11, 2009 at 01:45:35PM +, Paul Brook wrote:
> >> If you don't need real barriers, then why does the kvm code have them?
> >
> > We need real barriers but AF
On Thu, Nov 12, 2009 at 01:18:11AM +0800, Scott Tsai wrote:
> On Wed, Nov 11, 2009 at 10:08 PM, Michael S. Tsirkin wrote:
> > On Wed, Nov 11, 2009 at 01:45:35PM +, Paul Brook wrote:
> >> If you don't need real barriers, then why does the kvm code have them?
> >
> > We need real barriers but AF
On Wed, Nov 11, 2009 at 10:08 PM, Michael S. Tsirkin wrote:
> On Wed, Nov 11, 2009 at 01:45:35PM +, Paul Brook wrote:
>> If you don't need real barriers, then why does the kvm code have them?
>
> We need real barriers but AFAIK kvm does not have them :(
> IOW: virtio is currently broken with k
On Wednesday 11 November 2009, Michael S. Tsirkin wrote:
> On Wed, Nov 11, 2009 at 02:16:00PM +, Paul Brook wrote:
> > On Wednesday 11 November 2009, Michael S. Tsirkin wrote:
> > > On Wed, Nov 11, 2009 at 01:45:35PM +, Paul Brook wrote:
> > > > If you don't need real barriers, then why doe
On Wed, Nov 11, 2009 at 02:16:00PM +, Paul Brook wrote:
> On Wednesday 11 November 2009, Michael S. Tsirkin wrote:
> > On Wed, Nov 11, 2009 at 01:45:35PM +, Paul Brook wrote:
> > > If you don't need real barriers, then why does the kvm code have them?
> >
> > We need real barriers but AFAI
On Wednesday 11 November 2009, Michael S. Tsirkin wrote:
> On Wed, Nov 11, 2009 at 01:45:35PM +, Paul Brook wrote:
> > If you don't need real barriers, then why does the kvm code have them?
>
> We need real barriers but AFAIK kvm does not have them :(
> IOW: virtio is currently broken with kvm
On Wed, Nov 11, 2009 at 01:45:35PM +, Paul Brook wrote:
> If you don't need real barriers, then why does the kvm code have them?
We need real barriers but AFAIK kvm does not have them :(
IOW: virtio is currently broken with kvm, and my patch did
not fix this. The comment that I added says as m
> > > > > wmb must be at least a compiler barrier, even without SMP.
> > > >
> > > > Why?
> > >
> > > Because virtio code might run on a separate thread from guest.
> > > If compiler reorders writes, guest might see inconsistent data.
> >
> > If you've got threads running in parallel (which may be
On Wed, Nov 11, 2009 at 01:01:03PM +, Paul Brook wrote:
> A "compiler memory barrier" provides absolutely no guarantees in a
> multithreaded environment.
Claims including words "absolutely no" have absolutely no
chance to be correct.
--
MST
On Wed, Nov 11, 2009 at 01:01:03PM +, Paul Brook wrote:
> On Wednesday 11 November 2009, Michael S. Tsirkin wrote:
> > On Wed, Nov 11, 2009 at 01:34:12AM +, Paul Brook wrote:
> > > On Monday 26 October 2009, Michael S. Tsirkin wrote:
> > > > wmb must be at least a compiler barrier, even wit
On Wednesday 11 November 2009, Michael S. Tsirkin wrote:
> On Wed, Nov 11, 2009 at 01:34:12AM +, Paul Brook wrote:
> > On Monday 26 October 2009, Michael S. Tsirkin wrote:
> > > wmb must be at least a compiler barrier, even without SMP.
> >
> > Why?
>
> Because virtio code might run on a separ
On Wed, Nov 11, 2009 at 01:34:12AM +, Paul Brook wrote:
> On Monday 26 October 2009, Michael S. Tsirkin wrote:
> > wmb must be at least a compiler barrier, even without SMP.
>
> Why?
Because virtio code might run on a separate thread from guest.
If compiler reorders writes, guest might see in
15 matches
Mail list logo
