Re: consistent_free re-revisited

On Thu, Sep 12, 2002 at 09:49:07AM +0200, Benjamin Herrenschmidt wrote:

quoted

quoted

I tend to hate anything that relies on in_interrupt() as they
are other contexts that will have in_interrupt() cleared but still
have the same limitations. Typically, anything on the VM path must
do either GFP_ATOMIC or GFP_NOIO allocations, wether it's running
at interrupt time or not.

The problem is that the atomic pool is limited, iirc.

Well, ATOMIC can fail, sure, but if you do GFP_KERNEL within a
VM code path, then be prepared for deadlocks.

Er, so you're objecting to existing GFP_KERNEL's then?

Or is it too early and I need more coffee? :)

--
Tom Rini (TR1265)
http://gate.crashing.org/~trini/

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/