linux-kernel - [PATCH v3 11/17] x86: pgtable: move pagetable_dtor() to __tlb_remove

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <0dc5a3bf5a692e24379c1d3b879a6d4396f0dbbd.1734945104.git.zhengqi.arch@bytedance.com>
Date: Mon, 23 Dec 2024 17:40:57 +0800
From: Qi Zheng <zhengqi.arch@...edance.com>
To: peterz@...radead.org,
	agordeev@...ux.ibm.com,
	kevin.brodsky@....com,
	tglx@...utronix.de,
	david@...hat.com,
	jannh@...gle.com,
	hughd@...gle.com,
	yuzhao@...gle.com,
	willy@...radead.org,
	muchun.song@...ux.dev,
	vbabka@...nel.org,
	lorenzo.stoakes@...cle.com,
	akpm@...ux-foundation.org,
	rientjes@...gle.com,
	vishal.moola@...il.com,
	arnd@...db.de,
	will@...nel.org,
	aneesh.kumar@...nel.org,
	npiggin@...il.com,
	dave.hansen@...ux.intel.com,
	rppt@...nel.org,
	ryan.roberts@....com
Cc: linux-mm@...ck.org,
	linux-arm-kernel@...ts.infradead.org,
	linuxppc-dev@...ts.ozlabs.org,
	linux-riscv@...ts.infradead.org,
	linux-s390@...r.kernel.org,
	sparclinux@...r.kernel.org,
	linux-kernel@...r.kernel.org,
	x86@...nel.org,
	linux-arch@...r.kernel.org,
	linux-csky@...r.kernel.org,
	linux-hexagon@...r.kernel.org,
	loongarch@...ts.linux.dev,
	linux-m68k@...ts.linux-m68k.org,
	linux-mips@...r.kernel.org,
	linux-openrisc@...r.kernel.org,
	linux-sh@...r.kernel.org,
	linux-um@...ts.infradead.org,
	Qi Zheng <zhengqi.arch@...edance.com>
Subject: [PATCH v3 11/17] x86: pgtable: move pagetable_dtor() to __tlb_remove_table()

Move pagetable_dtor() to __tlb_remove_table(), so that ptlock and page
table pages can be freed together (regardless of whether RCU is used).
This prevents the use-after-free problem where the ptlock is freed
immediately but the page table pages is freed later via RCU.

Page tables shouldn't have swap cache, so use pagetable_free() instead of
free_page_and_swap_cache() to free page table pages.

Signed-off-by: Qi Zheng <zhengqi.arch@...edance.com>
Suggested-by: Peter Zijlstra (Intel) <peterz@...radead.org>
Cc: x86@...nel.org
---
 arch/x86/include/asm/tlb.h | 17 ++++++++++-------
 arch/x86/kernel/paravirt.c |  1 +
 arch/x86/mm/pgtable.c      | 12 ++----------
 3 files changed, 13 insertions(+), 17 deletions(-)

diff --git a/arch/x86/include/asm/tlb.h b/arch/x86/include/asm/tlb.h
index 73f0786181cc9..f64730be5ad67 100644
--- a/arch/x86/include/asm/tlb.h
+++ b/arch/x86/include/asm/tlb.h
@@ -31,24 +31,27 @@ static inline void tlb_flush(struct mmu_gather *tlb)
  */
 static inline void __tlb_remove_table(void *table)
 {
-	free_page_and_swap_cache(table);
+	struct ptdesc *ptdesc = (struct ptdesc *)table;
+
+	pagetable_dtor(ptdesc);
+	pagetable_free(ptdesc);
 }
 
 #ifdef CONFIG_PT_RECLAIM
 static inline void __tlb_remove_table_one_rcu(struct rcu_head *head)
 {
-	struct page *page;
+	struct ptdesc *ptdesc;
 
-	page = container_of(head, struct page, rcu_head);
-	put_page(page);
+	ptdesc = container_of(head, struct ptdesc, pt_rcu_head);
+	__tlb_remove_table(ptdesc);
 }
 
 static inline void __tlb_remove_table_one(void *table)
 {
-	struct page *page;
+	struct ptdesc *ptdesc;
 
-	page = table;
-	call_rcu(&page->rcu_head, __tlb_remove_table_one_rcu);
+	ptdesc = table;
+	call_rcu(&ptdesc->pt_rcu_head, __tlb_remove_table_one_rcu);
 }
 #define __tlb_remove_table_one __tlb_remove_table_one
 #endif /* CONFIG_PT_RECLAIM */
diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c
index 7bdcf152778c0..46d5d325483b0 100644
--- a/arch/x86/kernel/paravirt.c
+++ b/arch/x86/kernel/paravirt.c
@@ -62,6 +62,7 @@ void __init native_pv_lock_init(void)
 #ifndef CONFIG_PT_RECLAIM
 static void native_tlb_remove_table(struct mmu_gather *tlb, void *table)
 {
+	pagetable_dtor(table);
 	tlb_remove_page(tlb, table);
 }
 #else
diff --git a/arch/x86/mm/pgtable.c b/arch/x86/mm/pgtable.c
index a6cd9660e29ec..a0b0e501ba663 100644
--- a/arch/x86/mm/pgtable.c
+++ b/arch/x86/mm/pgtable.c
@@ -23,6 +23,7 @@ EXPORT_SYMBOL(physical_mask);
 static inline
 void paravirt_tlb_remove_table(struct mmu_gather *tlb, void *table)
 {
+	pagetable_dtor(table);
 	tlb_remove_page(tlb, table);
 }
 #else
@@ -60,7 +61,6 @@ early_param("userpte", setup_userpte);
 
 void ___pte_free_tlb(struct mmu_gather *tlb, struct page *pte)
 {
-	pagetable_dtor(page_ptdesc(pte));
 	paravirt_release_pte(page_to_pfn(pte));
 	paravirt_tlb_remove_table(tlb, pte);
 }
@@ -68,7 +68,6 @@ void ___pte_free_tlb(struct mmu_gather *tlb, struct page *pte)
 #if CONFIG_PGTABLE_LEVELS > 2
 void ___pmd_free_tlb(struct mmu_gather *tlb, pmd_t *pmd)
 {
-	struct ptdesc *ptdesc = virt_to_ptdesc(pmd);
 	paravirt_release_pmd(__pa(pmd) >> PAGE_SHIFT);
 	/*
 	 * NOTE! For PAE, any changes to the top page-directory-pointer-table
@@ -77,16 +76,12 @@ void ___pmd_free_tlb(struct mmu_gather *tlb, pmd_t *pmd)
 #ifdef CONFIG_X86_PAE
 	tlb->need_flush_all = 1;
 #endif
-	pagetable_dtor(ptdesc);
-	paravirt_tlb_remove_table(tlb, ptdesc_page(ptdesc));
+	paravirt_tlb_remove_table(tlb, virt_to_page(pmd));
 }
 
 #if CONFIG_PGTABLE_LEVELS > 3
 void ___pud_free_tlb(struct mmu_gather *tlb, pud_t *pud)
 {
-	struct ptdesc *ptdesc = virt_to_ptdesc(pud);
-
-	pagetable_dtor(ptdesc);
 	paravirt_release_pud(__pa(pud) >> PAGE_SHIFT);
 	paravirt_tlb_remove_table(tlb, virt_to_page(pud));
 }
@@ -94,9 +89,6 @@ void ___pud_free_tlb(struct mmu_gather *tlb, pud_t *pud)
 #if CONFIG_PGTABLE_LEVELS > 4
 void ___p4d_free_tlb(struct mmu_gather *tlb, p4d_t *p4d)
 {
-	struct ptdesc *ptdesc = virt_to_ptdesc(p4d);
-
-	pagetable_dtor(ptdesc);
 	paravirt_release_p4d(__pa(p4d) >> PAGE_SHIFT);
 	paravirt_tlb_remove_table(tlb, virt_to_page(p4d));
 }
-- 
2.20.1