On 12/03/2019 20:39, Daniel Henrique Barboza wrote:
>
>
> On 3/12/19 1:05 AM, Alexey Kardashevskiy wrote:
>>
>> On 12/03/2019 09:15, Daniel Henrique Barboza wrote:
>>>
>>> On 3/7/19 10:29 PM, Alexey Kardashevskiy wrote:
On 08/03/2019 04:51, Piotr Jaroszynski wrote:
> On 3/7/19 7:39
On 3/12/19 1:05 AM, Alexey Kardashevskiy wrote:
On 12/03/2019 09:15, Daniel Henrique Barboza wrote:
On 3/7/19 10:29 PM, Alexey Kardashevskiy wrote:
On 08/03/2019 04:51, Piotr Jaroszynski wrote:
On 3/7/19 7:39 AM, Daniel Henrique Barboza wrote:
On 3/7/19 12:15 PM, Erik Skultety wrote:
On
On 12/03/2019 09:15, Daniel Henrique Barboza wrote:
>
>
> On 3/7/19 10:29 PM, Alexey Kardashevskiy wrote:
>>
>> On 08/03/2019 04:51, Piotr Jaroszynski wrote:
>>> On 3/7/19 7:39 AM, Daniel Henrique Barboza wrote:
On 3/7/19 12:15 PM, Erik Skultety wrote:
> On Tue, Mar 05, 2019 at
On 3/7/19 12:15 PM, Erik Skultety wrote:
On Tue, Mar 05, 2019 at 09:46:08AM -0300, Daniel Henrique Barboza wrote:
The NVLink2 support in QEMU implements the detection of NVLink2
capable devices by verfying the attributes of the VFIO mem region
QEMU allocates for the NVIDIA GPUs. To properly
On 3/7/19 12:39 PM, Daniel Henrique Barboza wrote:
On 3/7/19 12:15 PM, Erik Skultety wrote:
On Tue, Mar 05, 2019 at 09:46:08AM -0300, Daniel Henrique Barboza wrote:
The NVLink2 support in QEMU implements the detection of NVLink2
capable devices by verfying the attributes of the VFIO mem
On 3/7/19 10:29 PM, Alexey Kardashevskiy wrote:
On 08/03/2019 04:51, Piotr Jaroszynski wrote:
On 3/7/19 7:39 AM, Daniel Henrique Barboza wrote:
On 3/7/19 12:15 PM, Erik Skultety wrote:
On Tue, Mar 05, 2019 at 09:46:08AM -0300, Daniel Henrique Barboza wrote:
The NVLink2 support in QEMU
...
> >
> > Sorry for the delay in responding. The problem is that all the V100 GPUs
> > support NVLink, but it may or may not be connected up. This is detected
> > at runtime during GPU initialization, which seems like much too heavy of
> > an operation to perform as part of passthrough
On 08/03/2019 04:51, Piotr Jaroszynski wrote:
> On 3/7/19 7:39 AM, Daniel Henrique Barboza wrote:
>>
>>
>> On 3/7/19 12:15 PM, Erik Skultety wrote:
>>> On Tue, Mar 05, 2019 at 09:46:08AM -0300, Daniel Henrique Barboza wrote:
The NVLink2 support in QEMU implements the detection of NVLink2
On 3/7/19 7:39 AM, Daniel Henrique Barboza wrote:
>
>
> On 3/7/19 12:15 PM, Erik Skultety wrote:
>> On Tue, Mar 05, 2019 at 09:46:08AM -0300, Daniel Henrique Barboza wrote:
>>> The NVLink2 support in QEMU implements the detection of NVLink2
>>> capable devices by verfying the attributes of the
On 3/7/19 12:15 PM, Erik Skultety wrote:
On Tue, Mar 05, 2019 at 09:46:08AM -0300, Daniel Henrique Barboza wrote:
The NVLink2 support in QEMU implements the detection of NVLink2
capable devices by verfying the attributes of the VFIO mem region
QEMU allocates for the NVIDIA GPUs. To properly
On Tue, Mar 05, 2019 at 09:46:08AM -0300, Daniel Henrique Barboza wrote:
> The NVLink2 support in QEMU implements the detection of NVLink2
> capable devices by verfying the attributes of the VFIO mem region
> QEMU allocates for the NVIDIA GPUs. To properly allocate an
> adequate amount of memLock,
The NVLink2 support in QEMU implements the detection of NVLink2
capable devices by verfying the attributes of the VFIO mem region
QEMU allocates for the NVIDIA GPUs. To properly allocate an
adequate amount of memLock, Libvirt needs this information before
a QEMU instance is even created.
An
12 matches
Mail list logo