In article <409526B0.1000906@augustsson.net>,
	Lennart Augustsson <lennart@augustsson.net> writes:
> Yes, it's a 64K buffer.  The controller can deal with non-contigous
> pages and it's just laziness on my part that it doesn't.

Can you provide a pointer to that code?

> BUT, it used to work just fine allocating 64K.  I don't understand
> how memory can get so fragmented.

Is there a kernel function which flushes all non-dirty pages in the
buffer cache? The USB driver could use that in case the allocation
fails initially.

	Kind regards

-- 
Matthias Scheler                                  http://scheler.de/~matthias/