On 2025-04-09 15:52:19 [+0200], Jan Kiszka wrote: > On second glance, Frederic's patch would probably also avoid the issue > we are seeing as it should bring PI to the CFS-throttled read-lock > holder (which is now a write-lock holder). > > But given how old that proposal is, I assume the performance impact was > even for the RT kernel too much, wasn't it? I don't remember any numbers and how bad things would get just the fear of it. But I guess this is workload related in terms how much does the RW lock improve the situation. RT then will multiple the CPU and reader resources to the point where it does not scale and present you the lockups. > Jan Sebastian
