Hello, On Wed, May 20, 2026 at 02:07:20PM +0800, Qiliang Yuan wrote: ... > This series introduces the "high" soft limit and associated task > throttling mechanism to the dmem cgroup v2 controller. > > The device memory (VRAM) management currently only supports hard limits > (max), which leads to immediate allocation failures when reached. This > can be disruptive for GPU-bound AI workloads. By introducing a soft > limit, we allow cgroups to exceed their quota temporarily while > applying backpressure via task throttling before the process returns > to user space. > > The mechanism is inspired by the memory cgroup's high limit: > - When usage > high, the task is marked with TIF_NOTIFY_RESUME. > - Upon returning to user space, it triggers a 100ms sleep. > - This provides a smoother over-subscription model for GPU resources.
I'm not sure about complicating dmem control model without implementing reclaim. What are we slowing them down for if the only recovery action is killing them? Thanks. -- tejun
