Skip to contentSkip to conversation

x86/mm/pti: remove dead logic during user pagetable population

PATCH Superseded Jike Song 5 participants 6 messages Jan 7, 2018 – Jan 8, 2018

Subsystems (3)
lore
Share & tools
lore

Jike Song

author
#0 lore raw

PATCH v2 1/1 x86/mm/pti: remove dead logic during user pagetable population

Look at one of the code snippets:

    162 if (pgd_none(*pgd)) {
    163         unsigned long new_p4d_page = __get_free_page(gfp);
    164         if (!new_p4d_page)
    165                 return NULL;
    166
    167         if (pgd_none(*pgd)) {
    168                 set_pgd(pgd, __pgd(_KERNPG_TABLE | __pa(new_p4d_page)));
    169                 new_p4d_page = 0;
    170         }
    171         if (new_p4d_page)
    172                 free_page(new_p4d_page);
    173 }

There can't be any difference between two pgd_none(*pgd) at L162 and L167, so it's always false at L171.

v2: add the commit message above.

Cc David Woodhouse <[email protected]>
Cc Alan Cox <[email protected]>
Cc Jiri Koshina <[email protected]>
Cc Linus Torvalds <[email protected]>
Cc Tim Chen <[email protected]>
Cc Andi Lutomirski <[email protected]>
Cc Andi Kleen <[email protected]>
Cc Peter Zijlstra <[email protected]>
Cc Paul Turner <[email protected]>
Cc Tom Lendacky <[email protected]>
Cc Greg KH <[email protected]>
Cc Dave Hansen <[email protected]>
Cc Kees Cook <[email protected]>
---
 arch/x86/mm/pti.c | 28 ++++------------------------
 1 file changed, 4 insertions(+), 24 deletions(-)
arch/x86/mm/pti.c +4 -24
index 43d4a4a29037..dc611d039bd5 100644
@@ -164,12 +164,7 @@ static p4d_t *pti_user_pagetable_walk_p4d(unsigned long address)
 		if (!new_p4d_page)
 			return NULL;
 
-		if (pgd_none(*pgd)) {
-			set_pgd(pgd, __pgd(_KERNPG_TABLE | __pa(new_p4d_page)));
-			new_p4d_page = 0;
-		}
-		if (new_p4d_page)
-			free_page(new_p4d_page);
+		set_pgd(pgd, __pgd(_KERNPG_TABLE | __pa(new_p4d_page)));
 	}
 	BUILD_BUG_ON(pgd_large(*pgd) != 0);
 
@@ -194,12 +189,7 @@ static pmd_t *pti_user_pagetable_walk_pmd(unsigned long address)
 		if (!new_pud_page)
 			return NULL;
 
-		if (p4d_none(*p4d)) {
-			set_p4d(p4d, __p4d(_KERNPG_TABLE | __pa(new_pud_page)));
-			new_pud_page = 0;
-		}
-		if (new_pud_page)
-			free_page(new_pud_page);
+		set_p4d(p4d, __p4d(_KERNPG_TABLE | __pa(new_pud_page)));
 	}
 
 	pud = pud_offset(p4d, address);
@@ -213,12 +203,7 @@ static pmd_t *pti_user_pagetable_walk_pmd(unsigned long address)
 		if (!new_pmd_page)
 			return NULL;
 
-		if (pud_none(*pud)) {
-			set_pud(pud, __pud(_KERNPG_TABLE | __pa(new_pmd_page)));
-			new_pmd_page = 0;
-		}
-		if (new_pmd_page)
-			free_page(new_pmd_page);
+		set_pud(pud, __pud(_KERNPG_TABLE | __pa(new_pmd_page)));
 	}
 
 	return pmd_offset(pud, address);
@@ -251,12 +236,7 @@ static __init pte_t *pti_user_pagetable_walk_pte(unsigned long address)
 		if (!new_pte_page)
 			return NULL;
 
-		if (pmd_none(*pmd)) {
-			set_pmd(pmd, __pmd(_KERNPG_TABLE | __pa(new_pte_page)));
-			new_pte_page = 0;
-		}
-		if (new_pte_page)
-			free_page(new_pte_page);
+		set_pmd(pmd, __pmd(_KERNPG_TABLE | __pa(new_pte_page)));
 	}
 
 	pte = pte_offset_kernel(pmd, address);
Signature · 1 line
2.14.3
maintainer
#1 lore raw
Quoted 15 lines · Jike Song

Look at one of the code snippets:

    162 if (pgd_none(*pgd)) {
    163         unsigned long new_p4d_page = __get_free_page(gfp);
    164         if (!new_p4d_page)
    165                 return NULL;
    166
    167         if (pgd_none(*pgd)) {
    168                 set_pgd(pgd, __pgd(_KERNPG_TABLE | __pa(new_p4d_page)));
    169                 new_p4d_page = 0;
    170         }
    171         if (new_p4d_page)
    172                 free_page(new_p4d_page);
    173 }
Quoted 2 lines

There can't be any difference between two pgd_none(*pgd) at L162 and L167, so it's always false at L171.

I think this is a remnant from the kaiser version which did this:

        if (pud_none(*pud)) {
                unsigned long new_pmd_page = __get_free_page(gfp);
                if (!new_pmd_page)
                        return NULL;
                spin_lock(&shadow_table_allocation_lock);
                if (pud_none(*pud))
                        set_pud(pud, __pud(_KERNPG_TABLE | __pa(new_pmd_page)));
                else
                        free_page(new_pmd_page);
                spin_unlock(&shadow_table_allocation_lock);
        }

I was wondering too, why the duplicated checks.

Which has this explanation about the need for the locking:

/*
* At runtime, the only things we map are some things for CPU
* hotplug, and stacks for new processes. No two CPUs will ever
* be populating the same addresses, so we only need to ensure
* that we protect between two CPUs trying to allocate and
* populate the same page table page.
*
* Only take this lock when doing a set_p[4um]d(), but it is not
* needed for doing a set_pte(). We assume that only the *owner*
* of a given allocation will be doing this for _their_
* allocation.
*
* This ensures that once a system has been running for a while
* and there have been stacks all over and these page tables
* are fully populated, there will be no further acquisitions of
* this lock.
*/
static DEFINE_SPINLOCK(shadow_table_allocation_lock);

Now I have my suspicions why that's not needed anymore upstream but I'd let tglx explain better.

Signature · 4 lines
Regards/Gruss,
    Boris.

Good mailing practices for 400: avoid top-posting and trim the reply.
#2 lore raw
Quoted 56 lines · Borislav Petkov
Quoted 15 lines · Jike Song

Look at one of the code snippets:

    162 if (pgd_none(*pgd)) {
    163         unsigned long new_p4d_page = __get_free_page(gfp);
    164         if (!new_p4d_page)
    165                 return NULL;
    166
    167         if (pgd_none(*pgd)) {
    168                 set_pgd(pgd, __pgd(_KERNPG_TABLE | __pa(new_p4d_page)));
    169                 new_p4d_page = 0;
    170         }
    171         if (new_p4d_page)
    172                 free_page(new_p4d_page);
    173 }
Quoted 2 lines

There can't be any difference between two pgd_none(*pgd) at L162 and L167, so it's always false at L171.

I think this is a remnant from the kaiser version which did this:

        if (pud_none(*pud)) {
                unsigned long new_pmd_page = __get_free_page(gfp);
                if (!new_pmd_page)
                        return NULL;
                spin_lock(&shadow_table_allocation_lock);
                if (pud_none(*pud))
                        set_pud(pud, __pud(_KERNPG_TABLE | __pa(new_pmd_page)));
                else
                        free_page(new_pmd_page);
                spin_unlock(&shadow_table_allocation_lock);
        }

I was wondering too, why the duplicated checks.

Which has this explanation about the need for the locking:

/*
* At runtime, the only things we map are some things for CPU
* hotplug, and stacks for new processes. No two CPUs will ever
* be populating the same addresses, so we only need to ensure
* that we protect between two CPUs trying to allocate and
* populate the same page table page.
*
* Only take this lock when doing a set_p[4um]d(), but it is not
* needed for doing a set_pte(). We assume that only the *owner*
* of a given allocation will be doing this for _their_
* allocation.
*
* This ensures that once a system has been running for a while
* and there have been stacks all over and these page tables
* are fully populated, there will be no further acquisitions of
* this lock.
*/
static DEFINE_SPINLOCK(shadow_table_allocation_lock);

Quoted 2 lines

Now I have my suspicions why that's not needed anymore upstream but I'd let tglx explain better.

We got rid of all that runtime mapping stuff and the functions are only called from pti_init(). So the locking and therefor the tests above are not needed anymore. While at it we should mark all those function __init.

Thanks,

tglx

#3 lore raw
Quoted 7 lines · Thomas Gleixner
Quoted 4 lines

static DEFINE_SPINLOCK(shadow_table_allocation_lock);

Now I have my suspicions why that's not needed anymore upstream but I'd let tglx explain better.

We got rid of all that runtime mapping stuff and the functions are only called from pti_init(). So the locking and therefor the tests above are not needed anymore. While at it we should mark all those function __init.

Yes, the double-test was part of an optimization where we attempted to avoid using a global spinlock in the fork() path. We would check for unallocated mid-level page tables without the lock. The lock was only taken it when we needed to *make* an entry to avoid collisions.

Now that it is all single-threaded, there is no chance of a collision, no need for a lock, and no need for the re-check.

^^ Just in case someone wants to include a bit more first-hand information about wtf that code was doing there.

author
#4 lore raw

PATCH v3 1/1 x86/mm/pti: remove dead logic during user pagetable population

In reply to Thomas Gleixner

Look at one of the code snippets:

162 if (pgd_none(*pgd)) {
163 unsigned long new_p4d_page = __get_free_page(gfp);
164 if (!new_p4d_page)
165 return NULL;
166
167 if (pgd_none(*pgd)) {
168 set_pgd(pgd, __pgd(_KERNPG_TABLE | __pa(new_p4d_page)));
169 new_p4d_page = 0;
170 }
171 if (new_p4d_page)
172 free_page(new_p4d_page);
173 }

There can't be any difference between two pgd_none(*pgd) at L162 and L167, so it's always false at L171.

To quote Dave Hansen:

Quoted 7 lines

Yes, the double-test was part of an optimization where we attempted to avoid using a global spinlock in the fork() path. We would check for unallocated mid-level page tables without the lock. The lock was only taken it when we needed to *make* an entry to avoid collisions.

Now that it is all single-threaded, there is no chance of a collision, no need for a lock, and no need for the re-check.

v3: mark functions __init; add first-hand explanation
v2: add commit message.

Cc David Woodhouse <[email protected]>
Cc Alan Cox <[email protected]>
Cc Jiri Koshina <[email protected]>
Cc Linus Torvalds <[email protected]>
Cc Tim Chen <[email protected]>
Cc Andi Lutomirski <[email protected]>
Cc Andi Kleen <[email protected]>
Cc Peter Zijlstra <[email protected]>
Cc Paul Turner <[email protected]>
Cc Tom Lendacky <[email protected]>
Cc Greg KH <[email protected]>
Cc Dave Hansen <[email protected]>
Cc Kees Cook <[email protected]>
Cc Borislav Petkov <[email protected]>
---
 arch/x86/mm/pti.c | 32 ++++++--------------------------
 1 file changed, 6 insertions(+), 26 deletions(-)
arch/x86/mm/pti.c +6 -26
index 43d4a4a29037..ce38f165489b 100644
@@ -149,7 +149,7 @@ pgd_t __pti_set_user_pgd(pgd_t *pgdp, pgd_t pgd)
  *
  * Returns a pointer to a P4D on success, or NULL on failure.
  */
-static p4d_t *pti_user_pagetable_walk_p4d(unsigned long address)
+static __init p4d_t *pti_user_pagetable_walk_p4d(unsigned long address)
 {
 	pgd_t *pgd = kernel_to_user_pgdp(pgd_offset_k(address));
 	gfp_t gfp = (GFP_KERNEL | __GFP_NOTRACK | __GFP_ZERO);
@@ -164,12 +164,7 @@ static p4d_t *pti_user_pagetable_walk_p4d(unsigned long address)
 		if (!new_p4d_page)
 			return NULL;
 
-		if (pgd_none(*pgd)) {
-			set_pgd(pgd, __pgd(_KERNPG_TABLE | __pa(new_p4d_page)));
-			new_p4d_page = 0;
-		}
-		if (new_p4d_page)
-			free_page(new_p4d_page);
+		set_pgd(pgd, __pgd(_KERNPG_TABLE | __pa(new_p4d_page)));
 	}
 	BUILD_BUG_ON(pgd_large(*pgd) != 0);
 
@@ -182,7 +177,7 @@ static p4d_t *pti_user_pagetable_walk_p4d(unsigned long address)
  *
  * Returns a pointer to a PMD on success, or NULL on failure.
  */
-static pmd_t *pti_user_pagetable_walk_pmd(unsigned long address)
+static __init pmd_t *pti_user_pagetable_walk_pmd(unsigned long address)
 {
 	gfp_t gfp = (GFP_KERNEL | __GFP_NOTRACK | __GFP_ZERO);
 	p4d_t *p4d = pti_user_pagetable_walk_p4d(address);
@@ -194,12 +189,7 @@ static pmd_t *pti_user_pagetable_walk_pmd(unsigned long address)
 		if (!new_pud_page)
 			return NULL;
 
-		if (p4d_none(*p4d)) {
-			set_p4d(p4d, __p4d(_KERNPG_TABLE | __pa(new_pud_page)));
-			new_pud_page = 0;
-		}
-		if (new_pud_page)
-			free_page(new_pud_page);
+		set_p4d(p4d, __p4d(_KERNPG_TABLE | __pa(new_pud_page)));
 	}
 
 	pud = pud_offset(p4d, address);
@@ -213,12 +203,7 @@ static pmd_t *pti_user_pagetable_walk_pmd(unsigned long address)
 		if (!new_pmd_page)
 			return NULL;
 
-		if (pud_none(*pud)) {
-			set_pud(pud, __pud(_KERNPG_TABLE | __pa(new_pmd_page)));
-			new_pmd_page = 0;
-		}
-		if (new_pmd_page)
-			free_page(new_pmd_page);
+		set_pud(pud, __pud(_KERNPG_TABLE | __pa(new_pmd_page)));
 	}
 
 	return pmd_offset(pud, address);
@@ -251,12 +236,7 @@ static __init pte_t *pti_user_pagetable_walk_pte(unsigned long address)
 		if (!new_pte_page)
 			return NULL;
 
-		if (pmd_none(*pmd)) {
-			set_pmd(pmd, __pmd(_KERNPG_TABLE | __pa(new_pte_page)));
-			new_pte_page = 0;
-		}
-		if (new_pte_page)
-			free_page(new_pte_page);
+		set_pmd(pmd, __pmd(_KERNPG_TABLE | __pa(new_pte_page)));
 	}
 
 	pte = pte_offset_kernel(pmd, address);
Signature · 1 line
2.14.3
#5 lore raw

[tip:x86/pti] x86/mm/pti: Remove dead logic in pti_user_pagetable_walk*()

x86/mm/pti: Remove dead logic in pti_user_pagetable_walk*()
Commit-ID:  8d56eff266f3e41a6c39926269c4c3f58f881a8e
Gitweb:     https://git.kernel.org/tip/8d56eff266f3e41a6c39926269c4c3f58f881a8e
Author:     Jike Song <[email protected]>
AuthorDate: Tue, 9 Jan 2018 00:03:41 +0800
Committer:  Thomas Gleixner <[email protected]>
CommitDate: Mon, 8 Jan 2018 17:42:13 +0100

x86/mm/pti: Remove dead logic in pti_user_pagetable_walk*()

The following code contains dead logic:

162 if (pgd_none(*pgd)) {
163 unsigned long new_p4d_page = __get_free_page(gfp);
164 if (!new_p4d_page)
165 return NULL;
166
167 if (pgd_none(*pgd)) {
168 set_pgd(pgd, __pgd(_KERNPG_TABLE | __pa(new_p4d_page)));
169 new_p4d_page = 0;
170 }
171 if (new_p4d_page)
172 free_page(new_p4d_page);
173 }

There can't be any difference between two pgd_none(*pgd) at L162 and L167, so it's always false at L171.

Dave Hansen explained:

Yes, the double-test was part of an optimization where we attempted to avoid using a global spinlock in the fork() path. We would check for unallocated mid-level page tables without the lock. The lock was only taken when we needed to *make* an entry to avoid collisions.

Now that it is all single-threaded, there is no chance of a collision, no need for a lock, and no need for the re-check.

As all these functions are only called during init, mark them __init as well.

Fixes 03f4424f348e ("x86/mm/pti: Add functions to clone kernel PMDs")
Cc Alan Cox <[email protected]>
Cc Andi Kleen <[email protected]>
Cc Tom Lendacky <[email protected]>
Cc Peter Zijlstra <[email protected]>
Cc Tim Chen <[email protected]>
Cc Jiri Koshina <[email protected]>
Cc Dave Hansen <[email protected]>
Cc Borislav Petkov <[email protected]>
Cc Kees Cook <[email protected]>
Cc Andi Lutomirski <[email protected]>
Cc Linus Torvalds <[email protected]>
Cc Greg KH <[email protected]>
Cc David Woodhouse <[email protected]>
Cc Paul Turner <[email protected]>
---
 arch/x86/mm/pti.c | 32 ++++++--------------------------
 1 file changed, 6 insertions(+), 26 deletions(-)
arch/x86/mm/pti.c +6 -26
index 43d4a4a..ce38f16 100644
@@ -149,7 +149,7 @@ pgd_t __pti_set_user_pgd(pgd_t *pgdp, pgd_t pgd)
  *
  * Returns a pointer to a P4D on success, or NULL on failure.
  */
-static p4d_t *pti_user_pagetable_walk_p4d(unsigned long address)
+static __init p4d_t *pti_user_pagetable_walk_p4d(unsigned long address)
 {
 	pgd_t *pgd = kernel_to_user_pgdp(pgd_offset_k(address));
 	gfp_t gfp = (GFP_KERNEL | __GFP_NOTRACK | __GFP_ZERO);
@@ -164,12 +164,7 @@ static p4d_t *pti_user_pagetable_walk_p4d(unsigned long address)
 		if (!new_p4d_page)
 			return NULL;
 
-		if (pgd_none(*pgd)) {
-			set_pgd(pgd, __pgd(_KERNPG_TABLE | __pa(new_p4d_page)));
-			new_p4d_page = 0;
-		}
-		if (new_p4d_page)
-			free_page(new_p4d_page);
+		set_pgd(pgd, __pgd(_KERNPG_TABLE | __pa(new_p4d_page)));
 	}
 	BUILD_BUG_ON(pgd_large(*pgd) != 0);
 
@@ -182,7 +177,7 @@ static p4d_t *pti_user_pagetable_walk_p4d(unsigned long address)
  *
  * Returns a pointer to a PMD on success, or NULL on failure.
  */
-static pmd_t *pti_user_pagetable_walk_pmd(unsigned long address)
+static __init pmd_t *pti_user_pagetable_walk_pmd(unsigned long address)
 {
 	gfp_t gfp = (GFP_KERNEL | __GFP_NOTRACK | __GFP_ZERO);
 	p4d_t *p4d = pti_user_pagetable_walk_p4d(address);
@@ -194,12 +189,7 @@ static pmd_t *pti_user_pagetable_walk_pmd(unsigned long address)
 		if (!new_pud_page)
 			return NULL;
 
-		if (p4d_none(*p4d)) {
-			set_p4d(p4d, __p4d(_KERNPG_TABLE | __pa(new_pud_page)));
-			new_pud_page = 0;
-		}
-		if (new_pud_page)
-			free_page(new_pud_page);
+		set_p4d(p4d, __p4d(_KERNPG_TABLE | __pa(new_pud_page)));
 	}
 
 	pud = pud_offset(p4d, address);
@@ -213,12 +203,7 @@ static pmd_t *pti_user_pagetable_walk_pmd(unsigned long address)
 		if (!new_pmd_page)
 			return NULL;
 
-		if (pud_none(*pud)) {
-			set_pud(pud, __pud(_KERNPG_TABLE | __pa(new_pmd_page)));
-			new_pmd_page = 0;
-		}
-		if (new_pmd_page)
-			free_page(new_pmd_page);
+		set_pud(pud, __pud(_KERNPG_TABLE | __pa(new_pmd_page)));
 	}
 
 	return pmd_offset(pud, address);
@@ -251,12 +236,7 @@ static __init pte_t *pti_user_pagetable_walk_pte(unsigned long address)
 		if (!new_pte_page)
 			return NULL;
 
-		if (pmd_none(*pmd)) {
-			set_pmd(pmd, __pmd(_KERNPG_TABLE | __pa(new_pte_page)));
-			new_pte_page = 0;
-		}
-		if (new_pte_page)
-			free_page(new_pte_page);
+		set_pmd(pmd, __pmd(_KERNPG_TABLE | __pa(new_pte_page)));
 	}
 
 	pte = pte_offset_kernel(pmd, address);