Messages in this thread | | | Date | Mon, 25 Jan 1999 10:49:44 -0800 (PST) | From | Linus Torvalds <> | Subject | Re: MM deadlock [was: Re: arca-vm-8...] |
| |
On Mon, 25 Jan 1999, Stephen C. Tweedie wrote: > > Correct: I haven't been testing any of the networking stuff myself so it > has been a non-issue for any of my workloads here. Obviously any check > for this case would have to be outside the GFP_WAIT conditional, but it > does make sense to set low_on_memory there anyway.
In fact, I wonder if we shouldn't just get rid of the GFP_WAIT conditional in __get_free_pages(), and make all that unconditional, so that we track low memory situations correctly even for atomic network traffic - something that obviously is a GoodThing(tm) to do. Then we could just make sure that try_to_free_pages() returns immediately for anything that doesn't have GFP_WAIT set, and have all the kswapd logic there.
That would even get rid of a test in the common path.
Linus
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |