the problem with just waiting for memory is that the pagedaemon won't do
anything unless there's actually a shortage of free pages.  we would have to
modify the pagedaemon to be able to keep freeing more pages even though there
no reason to do so as far as the VM system is concerned.

if a driver needs to allocate contiguous physical memory, it would be best
to arrange to do that early in the boot process, before other processes
have a chance to fragment memory.  running a utility which allocates the
memory as the first rc.d script would probably be good enough.

if we really need to be able to run this later as well, we could have another
utility which just allocates and touches a bunch of memory and then exits.
this would create enough free memory that the first utility should be able
to get the pages it needs.  or we could combine these two utility into one.
but it seems much better to me to do this in userland than in the kernel.

-Chuck


On Tue, Sep 18, 2001 at 01:37:09PM +0200, Frank van der Linden wrote:
> Matthias Drochner pointed out a problem that became apparent while
> using the AGP code. The AGP code uses bus_dmamem_alloc to allocate
> physical pages that are addressable through the AGP (PCI) bus.
> 
> bus_dmamem_alloc uses (for the i386 port, but I'm sure on other ports
> as well), uvm_pglistalloc to allocate pages in the suited ranges
> of memory. Normally, bus_dmamem_alloc is only used by drivers at
> config time, when there's still plenty of memory available, so it
> will never fail. However, the AGP code will call bus_dmamem_alloc
> when triggered by a request via an ioctl call (it's hard to
> change this, unless you want to allocate the entire AGP aperture
> upfront, which can be 256M in size..). At this time, not much
> memory may be available, because UBC or process images may be using
> a lot of it.
> 
> Naturally, the AGP code passes BUS_DMA_WAITOK to bus_dmamem_alloc,
> but it still fails. This is because uvm_pglistalloc ignores the
> waitok flag altogether, and bailing out if:
> 
>         /* Are there even any free pages? */
>         if (uvmexp.free <= (uvmexp.reserve_pagedaemon + uvmexp.reserve_kernel))
>                 goto out;
> 
> The first thing that comes to mind is that it should simply call uvm_wait(),
> grab all the pages that match its constraints, and repeat this step until
> it has them all. Naturally, a port must make sure that memory ranges that
> may not be wanted for bus_dma are preferred above those that are, otherwise
> it may never get the pages it wants.
> 
> Does that sound like a reasonable idea?
> 
> - Frank
> 
> -- 
> Frank van der Linden                           fvdl@wasabisystems.com
> ======================================================================
> Quality NetBSD CDs, Support & Service.   http://www.wasabisystems.com/