linux_dsm_epyc7002/arch/x86/mm/highmem_32.c
Peter Zijlstra 20273941f2 mm: fix race in kunmap_atomic()
Christoph reported a nice splat which illustrated a race in the new stack
based kmap_atomic implementation.

The problem is that we pop our stack slot before we're completely done
resetting its state -- in particular clearing the PTE (sometimes that's
CONFIG_DEBUG_HIGHMEM).  If an interrupt happens before we actually clear
the PTE used for the last slot, that interrupt can reuse the slot in a
dirty state, which triggers a BUG in kmap_atomic().

Fix this by introducing kmap_atomic_idx() which reports the current slot
index without actually releasing it and use that to find the PTE and delay
the _pop() until after we're completely done.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Reported-by: Christoph Hellwig <hch@infradead.org>
Acked-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-10-27 18:03:05 -07:00

140 lines
3.3 KiB
C

#include <linux/highmem.h>
#include <linux/module.h>
#include <linux/swap.h> /* for totalram_pages */
void *kmap(struct page *page)
{
might_sleep();
if (!PageHighMem(page))
return page_address(page);
return kmap_high(page);
}
EXPORT_SYMBOL(kmap);
void kunmap(struct page *page)
{
if (in_interrupt())
BUG();
if (!PageHighMem(page))
return;
kunmap_high(page);
}
EXPORT_SYMBOL(kunmap);
/*
* kmap_atomic/kunmap_atomic is significantly faster than kmap/kunmap because
* no global lock is needed and because the kmap code must perform a global TLB
* invalidation when the kmap pool wraps.
*
* However when holding an atomic kmap it is not legal to sleep, so atomic
* kmaps are appropriate for short, tight code paths only.
*/
void *kmap_atomic_prot(struct page *page, pgprot_t prot)
{
unsigned long vaddr;
int idx, type;
/* even !CONFIG_PREEMPT needs this, for in_atomic in do_page_fault */
pagefault_disable();
if (!PageHighMem(page))
return page_address(page);
type = kmap_atomic_idx_push();
idx = type + KM_TYPE_NR*smp_processor_id();
vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx);
BUG_ON(!pte_none(*(kmap_pte-idx)));
set_pte(kmap_pte-idx, mk_pte(page, prot));
return (void *)vaddr;
}
EXPORT_SYMBOL(kmap_atomic_prot);
void *__kmap_atomic(struct page *page)
{
return kmap_atomic_prot(page, kmap_prot);
}
EXPORT_SYMBOL(__kmap_atomic);
/*
* This is the same as kmap_atomic() but can map memory that doesn't
* have a struct page associated with it.
*/
void *kmap_atomic_pfn(unsigned long pfn)
{
return kmap_atomic_prot_pfn(pfn, kmap_prot);
}
EXPORT_SYMBOL_GPL(kmap_atomic_pfn);
void __kunmap_atomic(void *kvaddr)
{
unsigned long vaddr = (unsigned long) kvaddr & PAGE_MASK;
if (vaddr >= __fix_to_virt(FIX_KMAP_END) &&
vaddr <= __fix_to_virt(FIX_KMAP_BEGIN)) {
int idx, type;
type = kmap_atomic_idx();
idx = type + KM_TYPE_NR * smp_processor_id();
#ifdef CONFIG_DEBUG_HIGHMEM
WARN_ON_ONCE(vaddr != __fix_to_virt(FIX_KMAP_BEGIN + idx));
#endif
/*
* Force other mappings to Oops if they'll try to access this
* pte without first remap it. Keeping stale mappings around
* is a bad idea also, in case the page changes cacheability
* attributes or becomes a protected page in a hypervisor.
*/
kpte_clear_flush(kmap_pte-idx, vaddr);
kmap_atomic_idx_pop();
}
#ifdef CONFIG_DEBUG_HIGHMEM
else {
BUG_ON(vaddr < PAGE_OFFSET);
BUG_ON(vaddr >= (unsigned long)high_memory);
}
#endif
pagefault_enable();
}
EXPORT_SYMBOL(__kunmap_atomic);
struct page *kmap_atomic_to_page(void *ptr)
{
unsigned long idx, vaddr = (unsigned long)ptr;
pte_t *pte;
if (vaddr < FIXADDR_START)
return virt_to_page(ptr);
idx = virt_to_fix(vaddr);
pte = kmap_pte - (idx - FIX_KMAP_BEGIN);
return pte_page(*pte);
}
EXPORT_SYMBOL(kmap_atomic_to_page);
void __init set_highmem_pages_init(void)
{
struct zone *zone;
int nid;
for_each_zone(zone) {
unsigned long zone_start_pfn, zone_end_pfn;
if (!is_highmem(zone))
continue;
zone_start_pfn = zone->zone_start_pfn;
zone_end_pfn = zone_start_pfn + zone->spanned_pages;
nid = zone_to_nid(zone);
printk(KERN_INFO "Initializing %s for node %d (%08lx:%08lx)\n",
zone->name, nid, zone_start_pfn, zone_end_pfn);
add_highpages_with_active_regions(nid, zone_start_pfn,
zone_end_pfn);
}
totalram_pages += totalhigh_pages;
}