Re: locking user memory and kiobuf

Tony E. Bennett (tbennett@nvidia.com)
Tue, 9 Nov 1999 11:42:34 -0500 (EST)


>> kiobuf sounds perfect for a problem we have. Currently I use mmap()
>> to map pages from __get_free_pages() into user space for dma. This is
>> not totally satisfactory, tho, in part because the resulting group of
>> pages is not mapped contiguously into kernel.

Stephen> kiobufs don't magically fix this. If you want physically contiguous
Stephen> pages, you need to allocate them contiguously in the first place.
Stephen> kiobufs are intended as holders of pages, independent of the source of
Stephen> those pages: you can map user space pages into a kiobuf or you can
Stephen> allocate inside the kernel and populate the kiobuf with those pages
Stephen> instead.

For my device, I don't need physically contiguous, it has scatter/gather dma.

Stephen> It is already in 2.3.

this is great; thanks

--tony

-
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/