Do you have any specific workload where you have observed such OOMs and that represent a regression?
Can you open a new bug for such regressions?
As far as I was able to investigate, bringing this commit would imply bringing a whole set which introduce and change should_compact_retry. At first, it was introduced to prevent other OOMs as well, but I would rather see such a problem with the updated Xenial kernel before we can bring a whole bunch of updates.
Hi, Mr. Grünbichler.
Do you have any specific workload where you have observed such OOMs and that represent a regression?
Can you open a new bug for such regressions?
As far as I was able to investigate, bringing this commit would imply bringing a whole set which introduce and change should_ compact_ retry. At first, it was introduced to prevent other OOMs as well, but I would rather see such a problem with the updated Xenial kernel before we can bring a whole bunch of updates.
Thank you very much.
Thadeu Cascardo.