Re: [Qemu-devel] [RFC] Removing brdv_read()s from device init functions

2012-07-13 Thread Markus Armbruster
Peter Crosthwaite peter.crosthwa...@petalogix.com writes: Hi All, This RFC comes from the recent discussion Re coroutines and the block layer - the current topic of disucussion there has shifted to bdrv_read() from device init, so rather than continuing the discussion as a tangent to the

Re: [Qemu-devel] [RFC] Removing brdv_read()s from device init functions

2012-07-13 Thread Peter Crosthwaite
On Fri, Jul 13, 2012 at 6:33 PM, Markus Armbruster arm...@redhat.com wrote: Peter Crosthwaite peter.crosthwa...@petalogix.com writes: Hi All, This RFC comes from the recent discussion Re coroutines and the block layer - the current topic of disucussion there has shifted to bdrv_read() from

Re: [Qemu-devel] [RFC] Removing brdv_read()s from device init functions

2012-07-13 Thread Kevin Wolf
Am 13.07.2012 11:18, schrieb Peter Crosthwaite: On Fri, Jul 13, 2012 at 6:33 PM, Markus Armbruster arm...@redhat.com wrote: Peter Crosthwaite peter.crosthwa...@petalogix.com writes: One policy suggested is to ban bdrv_read() from init period and require devices to Lazy init. That is, on the

[Qemu-devel] [RFC] Removing brdv_read()s from device init functions

2012-07-02 Thread Peter Crosthwaite
Hi All, This RFC comes from the recent discussion Re coroutines and the block layer - the current topic of disucussion there has shifted to bdrv_read() from device init, so rather than continuing the discussion as a tangent to the unrelated original topic I'm recreating the thread. So anyways,