On 11/26/24 09:37, Lazar, Lijo wrote:
On 11/26/2024 7:50 PM, Mario Limonciello wrote:
Based on https://gitlab.freedesktop.org/drm/amd/-/issues/3781
What if this callback takes care only of suspend case and leaves the
hibernate case to dpm_prepare callback?
Then hibernate would fail under
On 11/26/2024 7:50 PM, Mario Limonciello wrote:
Based on https://gitlab.freedesktop.org/drm/amd/-/issues/3781
What if this callback takes care only of suspend case and leaves the
hibernate case to dpm_prepare callback?
>>>
>>> Then hibernate would fail under memory pressure.
Based on https://gitlab.freedesktop.org/drm/amd/-/issues/3781
What if this callback takes care only of suspend case and leaves the
hibernate case to dpm_prepare callback?
Then hibernate would fail under memory pressure.
My take is this failure with hibernate is a userspace problem (whether
use
On 11/25/2024 9:46 PM, Mario Limonciello wrote:
>
>
> On 11/25/24 08:59, Lazar, Lijo wrote:
>>
>>
>> On 11/25/2024 12:30 AM, Mario Limonciello wrote:
>>> From: Mario Limonciello
>>>
>>> As part of the suspend sequence VRAM needs to be evicted on dGPUs.
>>> In order to make suspend/resume more
On 11/25/24 08:59, Lazar, Lijo wrote:
On 11/25/2024 12:30 AM, Mario Limonciello wrote:
From: Mario Limonciello
As part of the suspend sequence VRAM needs to be evicted on dGPUs.
In order to make suspend/resume more reliable we moved this into
the pmops prepare() callback so that the suspe
On 11/25/2024 12:30 AM, Mario Limonciello wrote:
> From: Mario Limonciello
>
> As part of the suspend sequence VRAM needs to be evicted on dGPUs.
> In order to make suspend/resume more reliable we moved this into
> the pmops prepare() callback so that the suspend sequence would fail
> but the
From: Mario Limonciello
As part of the suspend sequence VRAM needs to be evicted on dGPUs.
In order to make suspend/resume more reliable we moved this into
the pmops prepare() callback so that the suspend sequence would fail
but the system could remain operational under high memory usage suspend.
