On Tue, Feb 25, 2014 at 06:09:20PM +0800, Hu Tao wrote:
> On Wed, Feb 19, 2014 at 10:03:13AM +0100, Paolo Bonzini wrote:
>
> <...>
>
> > > static int
> > > ram_backend_memory_init(HostMemoryBackend *backend, Error **errp)
> > > {
> > >+HostMemoryBackendRam *ram_backend = MEMORY_BACKEND_RAM(ba
Il 26/02/2014 15:25, Igor Mammedov ha scritto:
1: 'realize' approach I suggested
drawback is: assumption that all backends derived from HostMemoryBackend
will inherit NUMA controls even if backend shouldn't have
one (for example: fictional remote memor
On Wed, 26 Feb 2014 14:47:28 +0100
Paolo Bonzini wrote:
> Il 26/02/2014 14:43, Igor Mammedov ha scritto:
> > On Wed, 26 Feb 2014 13:45:38 +0100
> > Paolo Bonzini wrote:
> >
> >> Il 26/02/2014 13:31, Igor Mammedov ha scritto:
> > The problem is that some backends might not be handled the same
Il 26/02/2014 14:43, Igor Mammedov ha scritto:
On Wed, 26 Feb 2014 13:45:38 +0100
Paolo Bonzini wrote:
Il 26/02/2014 13:31, Igor Mammedov ha scritto:
The problem is that some backends might not be handled the same way.
For example, not all backends might produce a single void*/size_t pair
for
On Wed, 26 Feb 2014 13:45:38 +0100
Paolo Bonzini wrote:
> Il 26/02/2014 13:31, Igor Mammedov ha scritto:
> >> > The problem is that some backends might not be handled the same way.
> >> > For example, not all backends might produce a single void*/size_t pair
> >> > for the entire region. Think o
Il 26/02/2014 13:58, Marcelo Tosatti ha scritto:
I'd prefer to keep backends simple, with 1:1 mapping to memory regions.
I agree. However not all backends may have a mapping to a RAM
memory region. A composite backend could create a container memory
region whose children are other HostMemoryB
On Wed, Feb 26, 2014 at 01:45:38PM +0100, Paolo Bonzini wrote:
> Il 26/02/2014 13:31, Igor Mammedov ha scritto:
> >>> The problem is that some backends might not be handled the same way.
> >>> For example, not all backends might produce a single void*/size_t pair
> >>> for the entire region. Think
Il 26/02/2014 13:31, Igor Mammedov ha scritto:
> The problem is that some backends might not be handled the same way.
> For example, not all backends might produce a single void*/size_t pair
> for the entire region. Think of a "composite" backend that produces a
> large memory region from two sm
On Wed, 26 Feb 2014 11:33:30 +0100
Paolo Bonzini wrote:
> Il 26/02/2014 10:10, Igor Mammedov ha scritto:
> > if we assume that NUMA policies apply to every hostmem derived backend,
> > then we could realize() approach used by DEVICE. i.e.
> > set NUMA policies in hostmem.c:hostmemory_backend_memo
On Wed, 26 Feb 2014 16:59:44 +0800
Hu Tao wrote:
> On Wed, Feb 26, 2014 at 09:47:08AM +0100, Igor Mammedov wrote:
> > On Wed, 26 Feb 2014 13:00:03 +0800
> > Hu Tao wrote:
> >
> > > On Tue, Feb 25, 2014 at 03:15:53PM +0100, Paolo Bonzini wrote:
> > > > Il 25/02/2014 11:20, Hu Tao ha scritto:
> >
Il 26/02/2014 09:47, Igor Mammedov ha scritto:
I'm sorry that I've forgotten to mention prerequisite path
https://github.com/imammedo/qemu/commit/72a079b88e055fc690c8895a99ccbcce36f6fc1f
in stage tree https://github.com/imammedo/qemu/commits/memory-hotplug-v8
Thanks, I'll add this patch to the
Il 26/02/2014 10:10, Igor Mammedov ha scritto:
if we assume that NUMA policies apply to every hostmem derived backend,
then we could realize() approach used by DEVICE. i.e.
set NUMA policies in hostmem.c:hostmemory_backend_memory_init()
Add parent_complete field to ram-backend class and store th
On Wed, Feb 26, 2014 at 09:47:08AM +0100, Igor Mammedov wrote:
> On Wed, 26 Feb 2014 13:00:03 +0800
> Hu Tao wrote:
>
> > On Tue, Feb 25, 2014 at 03:15:53PM +0100, Paolo Bonzini wrote:
> > > Il 25/02/2014 11:20, Hu Tao ha scritto:
> > > >There is a problem that user_creatable_complete() is called
On Wed, 26 Feb 2014 13:57:06 +0800
Hu Tao wrote:
> On Wed, Feb 19, 2014 at 10:36:57AM +0100, Igor Mammedov wrote:
> > On Wed, 19 Feb 2014 10:03:13 +0100
> > Paolo Bonzini wrote:
> >
> > > 19/02/2014 08:54, Hu Tao ha scritto:
> > > > Thus makes user control how to allocate memory for ram backe
Il 26/02/2014 06:57, Hu Tao ha scritto:
If lines about memory polices are moved up to hostmem.c, the only thing
left in ram_backend_memory_init() is calling memory_region_init_ram() to
allocate memory. Then it comes a problem that when to apply memory
polices? Choices:
1. apply memory polices in
On Wed, 26 Feb 2014 13:00:03 +0800
Hu Tao wrote:
> On Tue, Feb 25, 2014 at 03:15:53PM +0100, Paolo Bonzini wrote:
> > Il 25/02/2014 11:20, Hu Tao ha scritto:
> > >There is a problem that user_creatable_complete() is called before
> > >adding object to "/objects" (see object_create()), which trigg
On Wed, Feb 19, 2014 at 10:36:57AM +0100, Igor Mammedov wrote:
> On Wed, 19 Feb 2014 10:03:13 +0100
> Paolo Bonzini wrote:
>
> > 19/02/2014 08:54, Hu Tao ha scritto:
> > > Thus makes user control how to allocate memory for ram backend.
> > >
> > > Signed-off-by: Hu Tao
> > > ---
> > > backend
On Tue, Feb 25, 2014 at 03:15:53PM +0100, Paolo Bonzini wrote:
> Il 25/02/2014 11:20, Hu Tao ha scritto:
> >There is a problem that user_creatable_complete() is called before
> >adding object to "/objects" (see object_create()), which triggers a
> >assert failure when calling object_get_canonical_p
Il 25/02/2014 11:20, Hu Tao ha scritto:
There is a problem that user_creatable_complete() is called before
adding object to "/objects" (see object_create()), which triggers a
assert failure when calling object_get_canonical_path() in
ram_backend_memory_init(). Any ideas?
You can call object_pro
On Wed, Feb 19, 2014 at 10:36:57AM +0100, Igor Mammedov wrote:
> On Wed, 19 Feb 2014 10:03:13 +0100
> Paolo Bonzini wrote:
>
> > 19/02/2014 08:54, Hu Tao ha scritto:
> > > Thus makes user control how to allocate memory for ram backend.
> > >
> > > Signed-off-by: Hu Tao
> > > ---
> > > backend
On Wed, Feb 19, 2014 at 10:03:13AM +0100, Paolo Bonzini wrote:
<...>
> > static int
> > ram_backend_memory_init(HostMemoryBackend *backend, Error **errp)
> > {
> >+HostMemoryBackendRam *ram_backend = MEMORY_BACKEND_RAM(backend);
> >+int mode = ram_backend->policy;
> >+void *p;
> >+
On Wed, 19 Feb 2014 10:03:13 +0100
Paolo Bonzini wrote:
> 19/02/2014 08:54, Hu Tao ha scritto:
> > Thus makes user control how to allocate memory for ram backend.
> >
> > Signed-off-by: Hu Tao
> > ---
> > backends/hostmem-ram.c | 158
> >
> >
19/02/2014 08:54, Hu Tao ha scritto:
Thus makes user control how to allocate memory for ram backend.
Signed-off-by: Hu Tao
---
backends/hostmem-ram.c | 158
include/sysemu/sysemu.h | 2 +
2 files changed, 160 insertions(+)
diff --git a/back
Thus makes user control how to allocate memory for ram backend.
Signed-off-by: Hu Tao
---
backends/hostmem-ram.c | 158
include/sysemu/sysemu.h | 2 +
2 files changed, 160 insertions(+)
diff --git a/backends/hostmem-ram.c b/backends/hostmem-ra
24 matches
Mail list logo
