Re: please revert bogus patch to vmscan.c

[Hugh Dickins]

On Mon, 29 Oct 2001, Benjamin LaHaise wrote:
> Think of CPUs with tagged tlbs and lots of entries.  Or even a system that 
> only runs 1 threaded app.  Easily triggerable.  If people want to optimise 
> it, great.  But go for correctness first, please...

But was the original flush_tlb_page() fully correct?  In i386 it's
	if (vma->vm_mm == current->active_mm)
		__flush_tlb_one(addr);
without reference to whether vma->vm_mm is active on another cpu.

Hugh