Lines Matching +full:dma +full:- +full:requests

15 overhead of page reclaim. This may happen for opportunistic high-order
16 allocation requests that have order-0 fallback options. In such cases,
19 __GFP_IO allocation requests are made to prevent file system deadlocks.
21 In the absence of non sleepable allocation requests, it seems detrimental
26 That being said, the kernel should try to fulfill requests for direct
28 the dma pool, so as to keep the dma pool filled for dma requests (atomic
30 OTOH, if there is a lot of free dma pages, it is preferable to satisfy
31 regular memory requests by allocating one from the dma pool, instead
36 right ratio of dma and regular memory, it is quite possible that balancing
37 would not be done even when the dma zone was completely empty. 2.2 has
60 Note that if the size of the regular zone is huge compared to dma zone,
61 it becomes less significant to consider the free dma pages while
67 for non-sleepable allocations. Second, the HIGHMEM zone is also balanced,
76 probably because all allocation requests are coming from intr context
87 are per-zone fields, used to determine when a zone needs to be balanced. When
90 watermark[WMARK_HIGH]. When low_on_memory is set, page allocation requests will
99 1. Dynamic experience should influence balancing: number of failed requests
101 2. Implement a replace_with_highmem()-like replace_with_regular() to preserve
102    dma pages. (lkd@tantalophile.demon.co.uk)