Re: alloc_area_pte: page already exists

[Andrea Arcangeli <andrea@suse.de>]

On Thu, Aug 09, 2001 at 03:32:44PM +0200, Bjorn Wesen wrote:
> I'm trying to track down a problem which seems to be a race condition
> vmalloc page table delayed PTE copying", or "you must never
> call free_kiovec in an interrupt context" etc..)

The latter is certainly the case. The former could be the case, however
if you know you're running any free_kiovec from irqs remove it and try
again first.

BTW, you should also avoid all the kiobuf allocation in all the fast
paths, try to pre-allocate it in a slow path to deliver higher
performance. (shortly we'll split the bh/blocks array out of the kiobuf
in a data structure called kiobuf_io so the non-IO usage will avoid the
overhead of the bh/blocks arrays and it will become lighter but still
it won't be that light)

Andrea