Commit 4150d3f5 authored Sep 11, 2007 by Andrew Morton Committed by Linus Torvalds Sep 11, 2007

revert "highmem: catch illegal nesting"



Revert

	commit 656dad31
	Author: Ingo Molnar <mingo@elte.hu>
	Date:   Sat Feb 10 01:46:36 2007 -0800

	[PATCH] highmem: catch illegal nesting

	Catch illegally nested kmap_atomic()s even if the page that is mapped by
	the 'inner' instance is from lowmem.

	This avoids spuriously zapped kmap-atomic ptes and turns hard to find
	crashes into clear asserts at the bug site.

Problem is, a get_zeroed_page(GFP_KERNEL) from interrupt context will trigger
this check if non-irq code on this CPU holds a KM_USER0 mapping.  But that
get_zeroed_page() will never be altering the kmap slot anyway due to the
GFP_KERNEL.

Cc: Christoph Lameter <clameter@sgi.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

parent 298a5df4

arch/i386/mm/highmem.c

+3 −4

Original line number	Original line	Diff line number	Diff line
	@@ -34,13 +34,12 @@ void kmap_atomic_prot(struct page page, enum km_type type, pgprot_t prot)
	/* even !CONFIG_PREEMPT needs this, for in_atomic in do_page_fault */		/* even !CONFIG_PREEMPT needs this, for in_atomic in do_page_fault */
	pagefault_disable();		pagefault_disable();

	idx = type + KM_TYPE_NR*smp_processor_id();
	BUG_ON(!pte_none(*(kmap_pte-idx)));

	if (!PageHighMem(page))		if (!PageHighMem(page))
	return page_address(page);		return page_address(page);

			idx = type + KM_TYPE_NR*smp_processor_id();
	vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx);		vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx);
			BUG_ON(!pte_none(*(kmap_pte-idx)));
	set_pte(kmap_pte-idx, mk_pte(page, prot));		set_pte(kmap_pte-idx, mk_pte(page, prot));
	arch_flush_lazy_mmu_mode();		arch_flush_lazy_mmu_mode();

Admin message