From: Toshi Kani <toshi.kani@xxxxxxx>
Subject: mm/vmalloc: add interfaces to free unmapped page table
On architectures with CONFIG_HAVE_ARCH_HUGE_VMAP set, ioremap() may create
pud/pmd mappings. Kernel panic was observed on arm64 systems with
Cortex-A75 in the following steps as described by Hanjun Guo.
1. ioremap a 4K size, valid page table will build,
2. iounmap it, pte0 will set to 0;
3. ioremap the same address with 2M size, pgd/pmd is unchanged,
then set the a new value for pmd;
4. pte0 is leaked;
5. CPU may meet exception because the old pmd is still in TLB,
which will lead to kernel panic.
This panic is not reproducible on x86. INVLPG, called from iounmap,
purges all levels of entries associated with purged address on x86. x86
still has memory leak.
The patch changes the ioremap path to free unmapped page table(s) since
doing so in the unmap path has the following issues:
- The iounmap() path is shared with vunmap(). Since vmap() only
supports pte mappings, making vunmap() to free a pte page is an
overhead for regular vmap users as they do not need a pte page
freed up.
- Checking if all entries in a pte page are cleared in the unmap path
is racy, and serializing this check is expensive.
- The unmap path calls free_vmap_area_noflush() to do lazy TLB purges.
Clearing a pud/pmd entry before the lazy TLB purges needs extra TLB
purge.
Add two interfaces, pud_free_pmd_page() and pmd_free_pte_page(), which
clear a given pud/pmd entry and free up a page for the lower level
entries.
This patch implements their stub functions on x86 and arm64, which work as
workaround.
[akpm@xxxxxxxxxxxxxxxxxxxx: fix typo in pmd_free_pte_page() stub]
Link: http://lkml.kernel.org/r/20180314180155.19492-2-toshi.kani@xxxxxxx
Fixes: e61ce6ade404e ("mm: change ioremap to set up huge I/O mappings")
Reported-by: Lei Li <lious.lilei@xxxxxxxxxxxxx>
Signed-off-by: Toshi Kani <toshi.kani@xxxxxxx>
Cc: Catalin Marinas <catalin.marinas@xxxxxxx>
Cc: Wang Xuefeng <wxf.wang@xxxxxxxxxxxxx>
Cc: Will Deacon <will.deacon@xxxxxxx>
Cc: Hanjun Guo <guohanjun@xxxxxxxxxx>
Cc: Michal Hocko <mhocko@xxxxxxxx>
Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
Cc: Ingo Molnar <mingo@xxxxxxxxxx>
Cc: "H. Peter Anvin" <hpa@xxxxxxxxx>
Cc: Borislav Petkov <bp@xxxxxxx>
Cc: Matthew Wilcox <willy@xxxxxxxxxxxxx>
Cc: Chintan Pandya <cpandya@xxxxxxxxxxxxxx>
Cc: <stable@xxxxxxxxxxxxxxx>
Signed-off-by: Andrew Morton <>
---
arch/arm64/mm/mmu.c | 10 ++++++++++
arch/x86/mm/pgtable.c | 24 ++++++++++++++++++++++++
include/asm-generic/pgtable.h | 10 ++++++++++
lib/ioremap.c | 6 ++++--
4 files changed, 48 insertions(+), 2 deletions(-)
diff -puN arch/arm64/mm/mmu.c~mm-vmalloc-add-interfaces-to-free-unmapped-page-table arch/arm64/mm/mmu.c
--- a/arch/arm64/mm/mmu.c~mm-vmalloc-add-interfaces-to-free-unmapped-page-table
+++ a/arch/arm64/mm/mmu.c
@@ -972,3 +972,13 @@ int pmd_clear_huge(pmd_t *pmdp)
pmd_clear(pmdp);
return 1;
}
+
+int pud_free_pmd_page(pud_t *pud)
+{
+ return pud_none(*pud);
+}
+
+int pmd_free_pte_page(pmd_t *pmd)
+{
+ return pmd_none(*pmd);
+}
diff -puN arch/x86/mm/pgtable.c~mm-vmalloc-add-interfaces-to-free-unmapped-page-table arch/x86/mm/pgtable.c
--- a/arch/x86/mm/pgtable.c~mm-vmalloc-add-interfaces-to-free-unmapped-page-table
+++ a/arch/x86/mm/pgtable.c
@@ -702,4 +702,28 @@ int pmd_clear_huge(pmd_t *pmd)
return 0;
}
+
+/**
+ * pud_free_pmd_page - Clear pud entry and free pmd page.
+ * @pud: Pointer to a PUD.
+ *
+ * Context: The pud range has been unmaped and TLB purged.
+ * Return: 1 if clearing the entry succeeded. 0 otherwise.
+ */
+int pud_free_pmd_page(pud_t *pud)
+{
+ return pud_none(*pud);
+}
+
+/**
+ * pmd_free_pte_page - Clear pmd entry and free pte page.
+ * @pmd: Pointer to a PMD.
+ *
+ * Context: The pmd range has been unmaped and TLB purged.
+ * Return: 1 if clearing the entry succeeded. 0 otherwise.
+ */
+int pmd_free_pte_page(pmd_t *pmd)
+{
+ return pmd_none(*pmd);
+}
#endif /* CONFIG_HAVE_ARCH_HUGE_VMAP */
diff -puN include/asm-generic/pgtable.h~mm-vmalloc-add-interfaces-to-free-unmapped-page-table include/asm-generic/pgtable.h
--- a/include/asm-generic/pgtable.h~mm-vmalloc-add-interfaces-to-free-unmapped-page-table
+++ a/include/asm-generic/pgtable.h
@@ -983,6 +983,8 @@ int pud_set_huge(pud_t *pud, phys_addr_t
int pmd_set_huge(pmd_t *pmd, phys_addr_t addr, pgprot_t prot);
int pud_clear_huge(pud_t *pud);
int pmd_clear_huge(pmd_t *pmd);
+int pud_free_pmd_page(pud_t *pud);
+int pmd_free_pte_page(pmd_t *pmd);
#else /* !CONFIG_HAVE_ARCH_HUGE_VMAP */
static inline int p4d_set_huge(p4d_t *p4d, phys_addr_t addr, pgprot_t prot)
{
@@ -1008,6 +1010,14 @@ static inline int pmd_clear_huge(pmd_t *
{
return 0;
}
+static inline int pud_free_pmd_page(pud_t *pud)
+{
+ return 0;
+}
+static inline int pmd_free_pte_page(pmd_t *pmd)
+{
+ return 0;
+}
#endif /* CONFIG_HAVE_ARCH_HUGE_VMAP */
#ifndef __HAVE_ARCH_FLUSH_PMD_TLB_RANGE
diff -puN lib/ioremap.c~mm-vmalloc-add-interfaces-to-free-unmapped-page-table lib/ioremap.c
--- a/lib/ioremap.c~mm-vmalloc-add-interfaces-to-free-unmapped-page-table
+++ a/lib/ioremap.c
@@ -91,7 +91,8 @@ static inline int ioremap_pmd_range(pud_
if (ioremap_pmd_enabled() &&
((next - addr) == PMD_SIZE) &&
- IS_ALIGNED(phys_addr + addr, PMD_SIZE)) {
+ IS_ALIGNED(phys_addr + addr, PMD_SIZE) &&
+ pmd_free_pte_page(pmd)) {
if (pmd_set_huge(pmd, phys_addr + addr, prot))
continue;
}
@@ -117,7 +118,8 @@ static inline int ioremap_pud_range(p4d_
if (ioremap_pud_enabled() &&
((next - addr) == PUD_SIZE) &&
- IS_ALIGNED(phys_addr + addr, PUD_SIZE)) {
+ IS_ALIGNED(phys_addr + addr, PUD_SIZE) &&
+ pud_free_pmd_page(pud)) {
if (pud_set_huge(pud, phys_addr + addr, prot))
continue;
}
_