On Mon, Sep 16, 2024 at 02:52:04PM -0300, Fabiano Rosas wrote:
> >>
> >> +/*
> >> + * A VM will at least require a few memslots to work, or it can even
> >> + * fail to boot. Make sure the supported value is always at least
> >> + * larger than what we will initially allocate.
>
Fabiano Rosas writes:
> Peter Xu writes:
>
>> Zhiyi reported an infinite loop issue in VFIO use case. The cause of that
>> was a separate discussion, however during that I found a regression of
>> dirty sync slowness when profiling.
>>
>> Each KVMMemoryListerner maintains an array of kvm memslo
Peter Xu writes:
> Zhiyi reported an infinite loop issue in VFIO use case. The cause of that
> was a separate discussion, however during that I found a regression of
> dirty sync slowness when profiling.
>
> Each KVMMemoryListerner maintains an array of kvm memslots. Currently it's
> statically
Zhiyi reported an infinite loop issue in VFIO use case. The cause of that
was a separate discussion, however during that I found a regression of
dirty sync slowness when profiling.
Each KVMMemoryListerner maintains an array of kvm memslots. Currently it's
statically allocated to be the max suppo
