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

13 overhead of page reclaim. This may happen for opportunistic high-order
14 allocation requests that have order-0 fallback options. In such cases,
17 __GFP_IO allocation requests are made to prevent file system deadlocks.
19 In the absence of non sleepable allocation requests, it seems detrimental
24 That being said, the kernel should try to fulfill requests for direct
26 the dma pool, so as to keep the dma pool filled for dma requests (atomic
28 OTOH, if there is a lot of free dma pages, it is preferable to satisfy
29 regular memory requests by allocating one from the dma pool, instead
34 right ratio of dma and regular memory, it is quite possible that balancing
35 would not be done even when the dma zone was completely empty. 2.2 has
58 Note that if the size of the regular zone is huge compared to dma zone,
59 it becomes less significant to consider the free dma pages while
65 for non-sleepable allocations. Second, the HIGHMEM zone is also balanced,
74 probably because all allocation requests are coming from intr context
85 are per-zone fields, used to determine when a zone needs to be balanced. When
88 watermark[WMARK_HIGH]. When low_on_memory is set, page allocation requests will
97 1. Dynamic experience should influence balancing: number of failed requests
99 2. Implement a replace_with_highmem()-like replace_with_regular() to preserve
100    dma pages. ([email protected])