Andi Kleen wrote:

quoted

You don't want to always have bad performance though, so you
could attempt to allocate if either the pointer is null _or_ it
points to the global structure?

Remember it's after a GFP_KERNEL OOM. If that fails most likely
you have deadlocked somewhere else already because Linux's handling
of that is so bad. Suboptimal readahead somewhere is the smallest
of your problems.

True. And in practice it may not even be able to happen at the
moment if the page allocator still doesn't fail small order allocs.

But I guess the dream one day is to robustly handle any OOM :\


Send instant messages to your online friends http://au.messenger.yahoo.com