Revert "x86/mm/gup: Switch GUP to the generic get_user_page_fast() implementation"

This reverts commit 2947ba054a4dabbd82848728d765346886050029.

Dan Williams reported dax-pmem kernel warnings with the following signature:

   WARNING: CPU: 8 PID: 245 at lib/percpu-refcount.c:155 percpu_ref_switch_to_atomic_rcu+0x1f5/0x200
   percpu ref (dax_pmem_percpu_release [dax_pmem]) <= 0 (0) after switching to atomic

... and bisected it to this commit, which suggests possible memory corruption
caused by the x86 fast-GUP conversion.

He also pointed out:

 "
  This is similar to the backtrace when we were not properly handling
  pud faults and was fixed with this commit: 220ced1676c4 "mm: fix
  get_user_pages() vs device-dax pud mappings"

  I've found some missing _devmap checks in the generic
  get_user_pages_fast() path, but this does not fix the regression
  [...]
 "

So given that there are known bugs, and a pretty robust looking bisection
points to this commit suggesting that are unknown bugs in the conversion
as well, revert it for the time being - we'll re-try in v4.13.

Reported-by: Dan Williams <dan.j.williams@intel.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Michal Hocko <mhocko@suse.cz>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Rik van Riel <riel@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: aneesh.kumar@linux.vnet.ibm.com
Cc: dann.frazier@canonical.com
Cc: dave.hansen@intel.com
Cc: steve.capper@linaro.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>

2 files changed, 497 insertions, 1 deletions

diff --git a/arch/x86/mm/Makefile b/arch/x86/mm/Makefile
index 0fbdcb64f9f8..96d2b847e09e 100644
--- a/arch/x86/mm/Makefile
+++ b/arch/x86/mm/Makefile
@@ -2,7 +2,7 @@
 KCOV_INSTRUMENT_tlb.o	:= n
 
 obj-y	:=  init.o init_$(BITS).o fault.o ioremap.o extable.o pageattr.o mmap.o \
-	    pat.o pgtable.o physaddr.o setup_nx.o tlb.o
+	    pat.o pgtable.o physaddr.o gup.o setup_nx.o tlb.o
 
 # Make sure __phys_addr has no stackprotector
 nostackp := $(call cc-option, -fno-stack-protector)
diff --git a/arch/x86/mm/gup.c b/arch/x86/mm/gup.c
new file mode 100644
index 000000000000..456dfdfd2249
--- /dev/null
+++ b/arch/x86/mm/gup.c
@@ -0,0 +1,496 @@
+/*
+ * Lockless get_user_pages_fast for x86
+ *
+ * Copyright (C) 2008 Nick Piggin
+ * Copyright (C) 2008 Novell Inc.
+ */
+#include <linux/sched.h>
+#include <linux/mm.h>
+#include <linux/vmstat.h>
+#include <linux/highmem.h>
+#include <linux/swap.h>
+#include <linux/memremap.h>
+
+#include <asm/mmu_context.h>
+#include <asm/pgtable.h>
+
+static inline pte_t gup_get_pte(pte_t *ptep)
+{
+#ifndef CONFIG_X86_PAE
+	return READ_ONCE(*ptep);
+#else
+	/*
+	 * With get_user_pages_fast, we walk down the pagetables without taking
+	 * any locks.  For this we would like to load the pointers atomically,
+	 * but that is not possible (without expensive cmpxchg8b) on PAE.  What
+	 * we do have is the guarantee that a pte will only either go from not
+	 * present to present, or present to not present or both -- it will not
+	 * switch to a completely different present page without a TLB flush in
+	 * between; something that we are blocking by holding interrupts off.
+	 *
+	 * Setting ptes from not present to present goes:
+	 * ptep->pte_high = h;
+	 * smp_wmb();
+	 * ptep->pte_low = l;
+	 *
+	 * And present to not present goes:
+	 * ptep->pte_low = 0;
+	 * smp_wmb();
+	 * ptep->pte_high = 0;
+	 *
+	 * We must ensure here that the load of pte_low sees l iff pte_high
+	 * sees h. We load pte_high *after* loading pte_low, which ensures we
+	 * don't see an older value of pte_high.  *Then* we recheck pte_low,
+	 * which ensures that we haven't picked up a changed pte high. We might
+	 * have got rubbish values from pte_low and pte_high, but we are
+	 * guaranteed that pte_low will not have the present bit set *unless*
+	 * it is 'l'. And get_user_pages_fast only operates on present ptes, so
+	 * we're safe.
+	 *
+	 * gup_get_pte should not be used or copied outside gup.c without being
+	 * very careful -- it does not atomically load the pte or anything that
+	 * is likely to be useful for you.
+	 */
+	pte_t pte;
+
+retry:
+	pte.pte_low = ptep->pte_low;
+	smp_rmb();
+	pte.pte_high = ptep->pte_high;
+	smp_rmb();
+	if (unlikely(pte.pte_low != ptep->pte_low))
+		goto retry;
+
+	return pte;
+#endif
+}
+
+static void undo_dev_pagemap(int *nr, int nr_start, struct page **pages)
+{
+	while ((*nr) - nr_start) {
+		struct page *page = pages[--(*nr)];
+
+		ClearPageReferenced(page);
+		put_page(page);
+	}
+}
+
+/*
+ * 'pteval' can come from a pte, pmd, pud or p4d.  We only check
+ * _PAGE_PRESENT, _PAGE_USER, and _PAGE_RW in here which are the
+ * same value on all 4 types.
+ */
+static inline int pte_allows_gup(unsigned long pteval, int write)
+{
+	unsigned long need_pte_bits = _PAGE_PRESENT|_PAGE_USER;
+
+	if (write)
+		need_pte_bits |= _PAGE_RW;
+
+	if ((pteval & need_pte_bits) != need_pte_bits)
+		return 0;
+
+	/* Check memory protection keys permissions. */
+	if (!__pkru_allows_pkey(pte_flags_pkey(pteval), write))
+		return 0;
+
+	return 1;
+}
+
+/*
+ * The performance critical leaf functions are made noinline otherwise gcc
+ * inlines everything into a single function which results in too much
+ * register pressure.
+ */
+static noinline int gup_pte_range(pmd_t pmd, unsigned long addr,
+		unsigned long end, int write, struct page **pages, int *nr)
+{
+	struct dev_pagemap *pgmap = NULL;
+	int nr_start = *nr, ret = 0;
+	pte_t *ptep, *ptem;
+
+	/*
+	 * Keep the original mapped PTE value (ptem) around since we
+	 * might increment ptep off the end of the page when finishing
+	 * our loop iteration.
+	 */
+	ptem = ptep = pte_offset_map(&pmd, addr);
+	do {
+		pte_t pte = gup_get_pte(ptep);
+		struct page *page;
+
+		/* Similar to the PMD case, NUMA hinting must take slow path */
+		if (pte_protnone(pte))
+			break;
+
+		if (!pte_allows_gup(pte_val(pte), write))
+			break;
+
+		if (pte_devmap(pte)) {
+			pgmap = get_dev_pagemap(pte_pfn(pte), pgmap);
+			if (unlikely(!pgmap)) {
+				undo_dev_pagemap(nr, nr_start, pages);
+				break;
+			}
+		} else if (pte_special(pte))
+			break;
+
+		VM_BUG_ON(!pfn_valid(pte_pfn(pte)));
+		page = pte_page(pte);
+		get_page(page);
+		put_dev_pagemap(pgmap);
+		SetPageReferenced(page);
+		pages[*nr] = page;
+		(*nr)++;
+
+	} while (ptep++, addr += PAGE_SIZE, addr != end);
+	if (addr == end)
+		ret = 1;
+	pte_unmap(ptem);
+
+	return ret;
+}
+
+static inline void get_head_page_multiple(struct page *page, int nr)
+{
+	VM_BUG_ON_PAGE(page != compound_head(page), page);
+	VM_BUG_ON_PAGE(page_count(page) == 0, page);
+	page_ref_add(page, nr);
+	SetPageReferenced(page);
+}
+
+static int __gup_device_huge(unsigned long pfn, unsigned long addr,
+		unsigned long end, struct page **pages, int *nr)
+{
+	int nr_start = *nr;
+	struct dev_pagemap *pgmap = NULL;
+
+	do {
+		struct page *page = pfn_to_page(pfn);
+
+		pgmap = get_dev_pagemap(pfn, pgmap);
+		if (unlikely(!pgmap)) {
+			undo_dev_pagemap(nr, nr_start, pages);
+			return 0;
+		}
+		SetPageReferenced(page);
+		pages[*nr] = page;
+		get_page(page);
+		put_dev_pagemap(pgmap);
+		(*nr)++;
+		pfn++;
+	} while (addr += PAGE_SIZE, addr != end);
+	return 1;
+}
+
+static int __gup_device_huge_pmd(pmd_t pmd, unsigned long addr,
+		unsigned long end, struct page **pages, int *nr)
+{
+	unsigned long fault_pfn;
+
+	fault_pfn = pmd_pfn(pmd) + ((addr & ~PMD_MASK) >> PAGE_SHIFT);
+	return __gup_device_huge(fault_pfn, addr, end, pages, nr);
+}
+
+static int __gup_device_huge_pud(pud_t pud, unsigned long addr,
+		unsigned long end, struct page **pages, int *nr)
+{
+	unsigned long fault_pfn;
+
+	fault_pfn = pud_pfn(pud) + ((addr & ~PUD_MASK) >> PAGE_SHIFT);
+	return __gup_device_huge(fault_pfn, addr, end, pages, nr);
+}
+
+static noinline int gup_huge_pmd(pmd_t pmd, unsigned long addr,
+		unsigned long end, int write, struct page **pages, int *nr)
+{
+	struct page *head, *page;
+	int refs;
+
+	if (!pte_allows_gup(pmd_val(pmd), write))
+		return 0;
+
+	VM_BUG_ON(!pfn_valid(pmd_pfn(pmd)));
+	if (pmd_devmap(pmd))
+		return __gup_device_huge_pmd(pmd, addr, end, pages, nr);
+
+	/* hugepages are never "special" */
+	VM_BUG_ON(pmd_flags(pmd) & _PAGE_SPECIAL);
+
+	refs = 0;
+	head = pmd_page(pmd);
+	page = head + ((addr & ~PMD_MASK) >> PAGE_SHIFT);
+	do {
+		VM_BUG_ON_PAGE(compound_head(page) != head, page);
+		pages[*nr] = page;
+		(*nr)++;
+		page++;
+		refs++;
+	} while (addr += PAGE_SIZE, addr != end);
+	get_head_page_multiple(head, refs);
+
+	return 1;
+}
+
+static int gup_pmd_range(pud_t pud, unsigned long addr, unsigned long end,
+		int write, struct page **pages, int *nr)
+{
+	unsigned long next;
+	pmd_t *pmdp;
+
+	pmdp = pmd_offset(&pud, addr);
+	do {
+		pmd_t pmd = *pmdp;
+
+		next = pmd_addr_end(addr, end);
+		if (pmd_none(pmd))
+			return 0;
+		if (unlikely(pmd_large(pmd) || !pmd_present(pmd))) {
+			/*
+			 * NUMA hinting faults need to be handled in the GUP
+			 * slowpath for accounting purposes and so that they
+			 * can be serialised against THP migration.
+			 */
+			if (pmd_protnone(pmd))
+				return 0;
+			if (!gup_huge_pmd(pmd, addr, next, write, pages, nr))
+				return 0;
+		} else {
+			if (!gup_pte_range(pmd, addr, next, write, pages, nr))
+				return 0;
+		}
+	} while (pmdp++, addr = next, addr != end);
+
+	return 1;
+}
+
+static noinline int gup_huge_pud(pud_t pud, unsigned long addr,
+		unsigned long end, int write, struct page **pages, int *nr)
+{
+	struct page *head, *page;
+	int refs;
+
+	if (!pte_allows_gup(pud_val(pud), write))
+		return 0;
+
+	VM_BUG_ON(!pfn_valid(pud_pfn(pud)));
+	if (pud_devmap(pud))
+		return __gup_device_huge_pud(pud, addr, end, pages, nr);
+
+	/* hugepages are never "special" */
+	VM_BUG_ON(pud_flags(pud) & _PAGE_SPECIAL);
+
+	refs = 0;
+	head = pud_page(pud);
+	page = head + ((addr & ~PUD_MASK) >> PAGE_SHIFT);
+	do {
+		VM_BUG_ON_PAGE(compound_head(page) != head, page);
+		pages[*nr] = page;
+		(*nr)++;
+		page++;
+		refs++;
+	} while (addr += PAGE_SIZE, addr != end);
+	get_head_page_multiple(head, refs);
+
+	return 1;
+}
+
+static int gup_pud_range(p4d_t p4d, unsigned long addr, unsigned long end,
+			int write, struct page **pages, int *nr)
+{
+	unsigned long next;
+	pud_t *pudp;
+
+	pudp = pud_offset(&p4d, addr);
+	do {
+		pud_t pud = *pudp;
+
+		next = pud_addr_end(addr, end);
+		if (pud_none(pud))
+			return 0;
+		if (unlikely(pud_large(pud))) {
+			if (!gup_huge_pud(pud, addr, next, write, pages, nr))
+				return 0;
+		} else {
+			if (!gup_pmd_range(pud, addr, next, write, pages, nr))
+				return 0;
+		}
+	} while (pudp++, addr = next, addr != end);
+
+	return 1;
+}
+
+static int gup_p4d_range(pgd_t pgd, unsigned long addr, unsigned long end,
+			int write, struct page **pages, int *nr)
+{
+	unsigned long next;
+	p4d_t *p4dp;
+
+	p4dp = p4d_offset(&pgd, addr);
+	do {
+		p4d_t p4d = *p4dp;
+
+		next = p4d_addr_end(addr, end);
+		if (p4d_none(p4d))
+			return 0;
+		BUILD_BUG_ON(p4d_large(p4d));
+		if (!gup_pud_range(p4d, addr, next, write, pages, nr))
+			return 0;
+	} while (p4dp++, addr = next, addr != end);
+
+	return 1;
+}
+
+/*
+ * Like get_user_pages_fast() except its IRQ-safe in that it won't fall
+ * back to the regular GUP.
+ */
+int __get_user_pages_fast(unsigned long start, int nr_pages, int write,
+			  struct page **pages)
+{
+	struct mm_struct *mm = current->mm;
+	unsigned long addr, len, end;
+	unsigned long next;
+	unsigned long flags;
+	pgd_t *pgdp;
+	int nr = 0;
+
+	start &= PAGE_MASK;
+	addr = start;
+	len = (unsigned long) nr_pages << PAGE_SHIFT;
+	end = start + len;
+	if (unlikely(!access_ok(write ? VERIFY_WRITE : VERIFY_READ,
+					(void __user *)start, len)))
+		return 0;
+
+	/*
+	 * XXX: batch / limit 'nr', to avoid large irq off latency
+	 * needs some instrumenting to determine the common sizes used by
+	 * important workloads (eg. DB2), and whether limiting the batch size
+	 * will decrease performance.
+	 *
+	 * It seems like we're in the clear for the moment. Direct-IO is
+	 * the main guy that batches up lots of get_user_pages, and even
+	 * they are limited to 64-at-a-time which is not so many.
+	 */
+	/*
+	 * This doesn't prevent pagetable teardown, but does prevent
+	 * the pagetables and pages from being freed on x86.
+	 *
+	 * So long as we atomically load page table pointers versus teardown
+	 * (which we do on x86, with the above PAE exception), we can follow the
+	 * address down to the the page and take a ref on it.
+	 */
+	local_irq_save(flags);
+	pgdp = pgd_offset(mm, addr);
+	do {
+		pgd_t pgd = *pgdp;
+
+		next = pgd_addr_end(addr, end);
+		if (pgd_none(pgd))
+			break;
+		if (!gup_p4d_range(pgd, addr, next, write, pages, &nr))
+			break;
+	} while (pgdp++, addr = next, addr != end);
+	local_irq_restore(flags);
+
+	return nr;
+}
+
+/**
+ * get_user_pages_fast() - pin user pages in memory
+ * @start:	starting user address
+ * @nr_pages:	number of pages from start to pin
+ * @write:	whether pages will be written to
+ * @pages:	array that receives pointers to the pages pinned.
+ * 		Should be at least nr_pages long.
+ *
+ * Attempt to pin user pages in memory without taking mm->mmap_sem.
+ * If not successful, it will fall back to taking the lock and
+ * calling get_user_pages().
+ *
+ * Returns number of pages pinned. This may be fewer than the number
+ * requested. If nr_pages is 0 or negative, returns 0. If no pages
+ * were pinned, returns -errno.
+ */
+int get_user_pages_fast(unsigned long start, int nr_pages, int write,
+			struct page **pages)
+{
+	struct mm_struct *mm = current->mm;
+	unsigned long addr, len, end;
+	unsigned long next;
+	pgd_t *pgdp;
+	int nr = 0;
+
+	start &= PAGE_MASK;
+	addr = start;
+	len = (unsigned long) nr_pages << PAGE_SHIFT;
+
+	end = start + len;
+	if (end < start)
+		goto slow_irqon;
+
+#ifdef CONFIG_X86_64
+	if (end >> __VIRTUAL_MASK_SHIFT)
+		goto slow_irqon;
+#endif
+
+	/*
+	 * XXX: batch / limit 'nr', to avoid large irq off latency
+	 * needs some instrumenting to determine the common sizes used by
+	 * important workloads (eg. DB2), and whether limiting the batch size
+	 * will decrease performance.
+	 *
+	 * It seems like we're in the clear for the moment. Direct-IO is
+	 * the main guy that batches up lots of get_user_pages, and even
+	 * they are limited to 64-at-a-time which is not so many.
+	 */
+	/*
+	 * This doesn't prevent pagetable teardown, but does prevent
+	 * the pagetables and pages from being freed on x86.
+	 *
+	 * So long as we atomically load page table pointers versus teardown
+	 * (which we do on x86, with the above PAE exception), we can follow the
+	 * address down to the the page and take a ref on it.
+	 */
+	local_irq_disable();
+	pgdp = pgd_offset(mm, addr);
+	do {
+		pgd_t pgd = *pgdp;
+
+		next = pgd_addr_end(addr, end);
+		if (pgd_none(pgd))
+			goto slow;
+		if (!gup_p4d_range(pgd, addr, next, write, pages, &nr))
+			goto slow;
+	} while (pgdp++, addr = next, addr != end);
+	local_irq_enable();
+
+	VM_BUG_ON(nr != (end - start) >> PAGE_SHIFT);
+	return nr;
+
+	{
+		int ret;
+
+slow:
+		local_irq_enable();
+slow_irqon:
+		/* Try to get the remaining pages with get_user_pages */
+		start += nr << PAGE_SHIFT;
+		pages += nr;
+
+		ret = get_user_pages_unlocked(start,
+					      (end - start) >> PAGE_SHIFT,
+					      pages, write ? FOLL_WRITE : 0);
+
+		/* Have to be a bit careful with return values */
+		if (nr > 0) {
+			if (ret < 0)
+				ret = nr;
+			else
+				ret += nr;
+		}
+
+		return ret;
+	}
+}