From: Nadav Amit namit@vmware.com
commit 9bc4f28af75a91aea0ae383f50b0a430c4509303 upstream.
When page-table entries are set, the compiler might optimize their assignment by using multiple instructions to set the PTE. This might turn into a security hazard if the user somehow manages to use the interim PTE. L1TF does not make our lives easier, making even an interim non-present PTE a security hazard.
Using WRITE_ONCE() to set PTEs and friends should prevent this potential security hazard.
I skimmed the differences in the binary with and without this patch. The differences are (obviously) greater when CONFIG_PARAVIRT=n as more code optimizations are possible. For better and worse, the impact on the binary with this patch is pretty small. Skimming the code did not cause anything to jump out as a security hazard, but it seems that at least move_soft_dirty_pte() caused set_pte_at() to use multiple writes.
Signed-off-by: Nadav Amit namit@vmware.com Signed-off-by: Thomas Gleixner tglx@linutronix.de Acked-by: Peter Zijlstra (Intel) peterz@infradead.org Cc: Dave Hansen dave.hansen@linux.intel.com Cc: Andi Kleen ak@linux.intel.com Cc: Josh Poimboeuf jpoimboe@redhat.com Cc: Michal Hocko mhocko@suse.com Cc: Vlastimil Babka vbabka@suse.cz Cc: Sean Christopherson sean.j.christopherson@intel.com Cc: Andy Lutomirski luto@kernel.org Link: https://lkml.kernel.org/r/20180902181451.80520-1-namit@vmware.com [bwh: Backported to 4.14: - Drop changes in pmdp_establish() - 5-level paging is a compile-time option - Update both cases in native_set_pgd() - Adjust context] Signed-off-by: Ben Hutchings ben.hutchings@codethink.co.uk Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- arch/x86/include/asm/pgtable_64.h | 22 +++++++++++----------- arch/x86/mm/pgtable.c | 8 ++++---- 2 files changed, 15 insertions(+), 15 deletions(-)
--- a/arch/x86/include/asm/pgtable_64.h +++ b/arch/x86/include/asm/pgtable_64.h @@ -56,15 +56,15 @@ struct mm_struct; void set_pte_vaddr_p4d(p4d_t *p4d_page, unsigned long vaddr, pte_t new_pte); void set_pte_vaddr_pud(pud_t *pud_page, unsigned long vaddr, pte_t new_pte);
-static inline void native_pte_clear(struct mm_struct *mm, unsigned long addr, - pte_t *ptep) +static inline void native_set_pte(pte_t *ptep, pte_t pte) { - *ptep = native_make_pte(0); + WRITE_ONCE(*ptep, pte); }
-static inline void native_set_pte(pte_t *ptep, pte_t pte) +static inline void native_pte_clear(struct mm_struct *mm, unsigned long addr, + pte_t *ptep) { - *ptep = pte; + native_set_pte(ptep, native_make_pte(0)); }
static inline void native_set_pte_atomic(pte_t *ptep, pte_t pte) @@ -74,7 +74,7 @@ static inline void native_set_pte_atomic
static inline void native_set_pmd(pmd_t *pmdp, pmd_t pmd) { - *pmdp = pmd; + WRITE_ONCE(*pmdp, pmd); }
static inline void native_pmd_clear(pmd_t *pmd) @@ -110,7 +110,7 @@ static inline pmd_t native_pmdp_get_and_
static inline void native_set_pud(pud_t *pudp, pud_t pud) { - *pudp = pud; + WRITE_ONCE(*pudp, pud); }
static inline void native_pud_clear(pud_t *pud) @@ -220,9 +220,9 @@ static inline pgd_t pti_set_user_pgd(pgd static inline void native_set_p4d(p4d_t *p4dp, p4d_t p4d) { #if defined(CONFIG_PAGE_TABLE_ISOLATION) && !defined(CONFIG_X86_5LEVEL) - p4dp->pgd = pti_set_user_pgd(&p4dp->pgd, p4d.pgd); + WRITE_ONCE(p4dp->pgd, pti_set_user_pgd(&p4dp->pgd, p4d.pgd)); #else - *p4dp = p4d; + WRITE_ONCE(*p4dp, p4d); #endif }
@@ -238,9 +238,9 @@ static inline void native_p4d_clear(p4d_ static inline void native_set_pgd(pgd_t *pgdp, pgd_t pgd) { #ifdef CONFIG_PAGE_TABLE_ISOLATION - *pgdp = pti_set_user_pgd(pgdp, pgd); + WRITE_ONCE(*pgdp, pti_set_user_pgd(pgdp, pgd)); #else - *pgdp = pgd; + WRITE_ONCE(*pgdp, pgd); #endif }
--- a/arch/x86/mm/pgtable.c +++ b/arch/x86/mm/pgtable.c @@ -260,7 +260,7 @@ static void pgd_mop_up_pmds(struct mm_st if (pgd_val(pgd) != 0) { pmd_t *pmd = (pmd_t *)pgd_page_vaddr(pgd);
- pgdp[i] = native_make_pgd(0); + pgd_clear(&pgdp[i]);
paravirt_release_pmd(pgd_val(pgd) >> PAGE_SHIFT); pmd_free(mm, pmd); @@ -430,7 +430,7 @@ int ptep_set_access_flags(struct vm_area int changed = !pte_same(*ptep, entry);
if (changed && dirty) - *ptep = entry; + set_pte(ptep, entry);
return changed; } @@ -445,7 +445,7 @@ int pmdp_set_access_flags(struct vm_area VM_BUG_ON(address & ~HPAGE_PMD_MASK);
if (changed && dirty) { - *pmdp = entry; + set_pmd(pmdp, entry); /* * We had a write-protection fault here and changed the pmd * to to more permissive. No need to flush the TLB for that, @@ -465,7 +465,7 @@ int pudp_set_access_flags(struct vm_area VM_BUG_ON(address & ~HPAGE_PUD_MASK);
if (changed && dirty) { - *pudp = entry; + set_pud(pudp, entry); /* * We had a write-protection fault here and changed the pud * to to more permissive. No need to flush the TLB for that,