Skip to contentSkip to conversation

x86/mm/pat: skip RWX verification until kernel text is set to read only

PATCH In review Mike Rapoport 4 participants 7 messages Sep 8 – Sep 9

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 x86/mm/pat: skip RWX verification until kernel text is set to read only

From: "Mike Rapoport (Microsoft)" <[email protected]>

Nathan Chancellor reports the following warning:

CPA detected W^X violation: 8000000000000123 -> 0000000000000123 range: 0xffffffffc0400000 - 0xffffffffc0400fff PFN 100e00

  WARNING: arch/x86/mm/pat/set_memory.c:722 at __change_page_attr_set_clr+0xde7/0x1290, CPU#0: swapper/0/0
  Modules linked in:
  CPU: 0 UID: 0 PID: 0 Comm: swapper/0 Not tainted 7.3.0-rc1-debug-00006-g453e78594434 #1 PREEMPT(full)  2950d432dd3910251071a66f3134fe0875432786
  Hardware name: ASUS System Product Name/PRIME Z590M-PLUS, BIOS 1801 12/26/2022
  RIP: 0010:__change_page_attr_set_clr+0xdff/0x1290
  Code: 80 7c 24 42 00 0f 85 3a 04 00 00 48 8d 3d 19 8d 79 02 49 89 d9 4c 89 e1 4c 89 d2 4c 89 f6 4d 8d 84 24 ff 0f 00 00 4c 89 14 24 <67> 48 0f b9 3a 4c 8b 14 24 48 8b 0d 81 44 bf 01 41 f6 c2 01
  RSP: 0000:ffffffff87003c60 EFLAGS: 00010246
  RAX: 0000000000000002 RBX: 0000000000100e00 RCX: ffffffffc0400000
  RDX: 0000000000000123 RSI: 8000000000000123 RDI: ffffffff872e50c0
  RBP: 8000000100e00123 R08: ffffffffc0400fff R09: 0000000000100e00
  R10: 0000000000000123 R11: 0000000000000001 R12: ffffffffc0400000
  R13: 0000000100e00123 R14: 8000000000000123 R15: ffffffff87003d58
  FS:  0000000000000000(0000) GS:ffff8ad1777a7000(0000) knlGS:0000000000000000
  CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
  CR2: ffff8ad0a4201000 CR3: 00000007e3022001 CR4: 0000000000770ef0
  PKRU: 55555554
  Call Trace:
   <TASK>
   ? _vm_unmap_aliases+0x219/0x280
   change_page_attr_set_clr+0x161/0x250
   ? events_sysfs_show+0x5d/0x80
   set_memory_x+0x39/0x50
   apply_retpolines+0x656/0x6d0
   ? events_sysfs_show+0x5d/0x80
   ? events_sysfs_show+0x6c/0x80
   ? events_sysfs_show+0x62/0x80
   alternative_instructions+0x3c/0xd0
   arch_cpu_finalize_init+0x130/0x190
   start_kernel+0x97d/0xa10
   x86_64_start_reservations+0x24/0x30
   x86_64_start_kernel+0xda/0xe0
   common_startup_64+0x13e/0x151
   </TASK>
  ---[ end trace 0000000000000000 ]---

The warning appears because commit 038176c21617f ("x86/mm/pat: fix effective RW computation in lookup_address_in_pgd_attr()") fixed the effective RW checked by verify_rwx() and it exposed that pages used for ITS trampolines temporarily have RWX permissions.

The permissions are updated in its_fini_core() after all the ITS trampolines are generated, but since verify_rwx() detects invalid transitions, it warns when its_alloc() makes RW memory executable.

At the time of alternatives patching the entire kernel text is mapped RWX, so the warning is bogus anyway.

Skip verification of W^X violations in verify_rwx() when they are triggered by transitions happening before the kernel text is remapped as read-only.

---
 arch/x86/mm/pat/set_memory.c | 4 ++++
 1 file changed, 4 insertions(+)
arch/x86/mm/pat/set_memory.c +4 -0
index 226660973d515..2eecb76703bb4 100644
@@ -708,6 +708,10 @@ static inline pgprot_t verify_rwx(pgprot_t old, pgprot_t new, unsigned long star
 	if (!(__supported_pte_mask & _PAGE_NX))
 		return new;
 
+	/* skip verification until kernel text is set to read only */
+	if (!kernel_set_to_readonly)
+		return new;
+
 	if (!((pgprot_val(old) ^ pgprot_val(new)) & (_PAGE_RW | _PAGE_NX)))
 		return new;
 

base-commit: 038176c21617fcc03ccc1ca43230ffedb712c047

Signature · 1 line
2.53.0
Tested-by on v1 1/1
#1 lore raw
Quoted 52 lines · Mike Rapoport

From: "Mike Rapoport (Microsoft)" <[email protected]>

Nathan Chancellor reports the following warning:

CPA detected W^X violation: 8000000000000123 -> 0000000000000123 range: 0xffffffffc0400000 - 0xffffffffc0400fff PFN 100e00

  WARNING: arch/x86/mm/pat/set_memory.c:722 at __change_page_attr_set_clr+0xde7/0x1290, CPU#0: swapper/0/0
  Modules linked in:
  CPU: 0 UID: 0 PID: 0 Comm: swapper/0 Not tainted 7.3.0-rc1-debug-00006-g453e78594434 #1 PREEMPT(full)  2950d432dd3910251071a66f3134fe0875432786
  Hardware name: ASUS System Product Name/PRIME Z590M-PLUS, BIOS 1801 12/26/2022
  RIP: 0010:__change_page_attr_set_clr+0xdff/0x1290
  Code: 80 7c 24 42 00 0f 85 3a 04 00 00 48 8d 3d 19 8d 79 02 49 89 d9 4c 89 e1 4c 89 d2 4c 89 f6 4d 8d 84 24 ff 0f 00 00 4c 89 14 24 <67> 48 0f b9 3a 4c 8b 14 24 48 8b 0d 81 44 bf 01 41 f6 c2 01
  RSP: 0000:ffffffff87003c60 EFLAGS: 00010246
  RAX: 0000000000000002 RBX: 0000000000100e00 RCX: ffffffffc0400000
  RDX: 0000000000000123 RSI: 8000000000000123 RDI: ffffffff872e50c0
  RBP: 8000000100e00123 R08: ffffffffc0400fff R09: 0000000000100e00
  R10: 0000000000000123 R11: 0000000000000001 R12: ffffffffc0400000
  R13: 0000000100e00123 R14: 8000000000000123 R15: ffffffff87003d58
  FS:  0000000000000000(0000) GS:ffff8ad1777a7000(0000) knlGS:0000000000000000
  CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
  CR2: ffff8ad0a4201000 CR3: 00000007e3022001 CR4: 0000000000770ef0
  PKRU: 55555554
  Call Trace:
   <TASK>
   ? _vm_unmap_aliases+0x219/0x280
   change_page_attr_set_clr+0x161/0x250
   ? events_sysfs_show+0x5d/0x80
   set_memory_x+0x39/0x50
   apply_retpolines+0x656/0x6d0
   ? events_sysfs_show+0x5d/0x80
   ? events_sysfs_show+0x6c/0x80
   ? events_sysfs_show+0x62/0x80
   alternative_instructions+0x3c/0xd0
   arch_cpu_finalize_init+0x130/0x190
   start_kernel+0x97d/0xa10
   x86_64_start_reservations+0x24/0x30
   x86_64_start_kernel+0xda/0xe0
   common_startup_64+0x13e/0x151
   </TASK>
  ---[ end trace 0000000000000000 ]---

The warning appears because commit 038176c21617f ("x86/mm/pat: fix effective RW computation in lookup_address_in_pgd_attr()") fixed the effective RW checked by verify_rwx() and it exposed that pages used for ITS trampolines temporarily have RWX permissions.

The permissions are updated in its_fini_core() after all the ITS trampolines are generated, but since verify_rwx() detects invalid transitions, it warns when its_alloc() makes RW memory executable.

At the time of alternatives patching the entire kernel text is mapped RWX, so the warning is bogus anyway.

Quoted 7 lines

Skip verification of W^X violations in verify_rwx() when they are triggered by transitions happening before the kernel text is remapped as read-only.

Quoted 24 lines
---
 arch/x86/mm/pat/set_memory.c | 4 ++++
 1 file changed, 4 insertions(+)
arch/x86/mm/pat/set_memory.c +4 -0
index 226660973d515..2eecb76703bb4 100644
@@ -708,6 +708,10 @@ static inline pgprot_t verify_rwx(pgprot_t old, pgprot_t new, unsigned long star
 	if (!(__supported_pte_mask & _PAGE_NX))
 		return new;
 
+	/* skip verification until kernel text is set to read only */
+	if (!kernel_set_to_readonly)
+		return new;
+
 	if (!((pgprot_val(old) ^ pgprot_val(new)) & (_PAGE_RW | _PAGE_NX)))
 		return new;
 

base-commit: 038176c21617fcc03ccc1ca43230ffedb712c047

Signature · 1 line
2.53.0
Signature · 2 lines
Cheers,
Nathan
#2 lore raw
In reply to Mike Rapoport

arch/x86/mm/pat/set_memory.c static inline pgprot_t verify_rwx(pgprot_t old, pgprot_t new, unsigned long star · line 713

Quoted 9 lines · Mike Rapoport
arch/x86/mm/pat/set_memory.c +3 -0
@@ -708,6 +708,10 @@ static inline pgprot_t verify_rwx(pgprot_t old, pgprot_t new, unsigned long star
 	if (!(__supported_pte_mask & _PAGE_NX))
 		return new;
 
+	/* skip verification until kernel text is set to read only */
+	if (!kernel_set_to_readonly)
+		return new;

If someone sets up a W+X mapping, such a mapping could persist until after boot and this would suppress the warning. Right?

Sure, you can _get_ checking with debug_checkwx(). But that's a debug option and it's kinda weird to shift the burden from an always-on thing like verify_rwx() to a literal debug option.

That said, I think you're completely on target for thinking that it's silly for verify_rwx() to even *try* to spew warnings in boot.

How about we add a kernel_strict_rwx() helper:

bool strict_kernel_rwx(void)

{
	return IS_ENABLED(CONFIG_STRICT_KERNEL_RWX) && rodata_enabled;
}

Have verify_rwx() check *that*. Also, make the DEBUG_WX functionality mandatory for STRICT_KERNEL_RWX (with appropriate renaming) so that it can be depended upon. I assume all the distros are turning DEBUG_WX on already (Ubuntu seems to).

Maybe just do the strict_kernel_rwx() to start and then circle back around to muck with making DEBUG_WX mandatory?

#3 lore raw
Quoted 21 lines · Dave Hansen

arch/x86/mm/pat/set_memory.c static inline pgprot_t verify_rwx(pgprot_t old, pgprot_t new, unsigned long star · line 713

Quoted 9 lines · Mike Rapoport
arch/x86/mm/pat/set_memory.c +3 -0
@@ -708,6 +708,10 @@ static inline pgprot_t verify_rwx(pgprot_t old, pgprot_t new, unsigned long star
 	if (!(__supported_pte_mask & _PAGE_NX))
 		return new;
 
+	/* skip verification until kernel text is set to read only */
+	if (!kernel_set_to_readonly)
+		return new;

If someone sets up a W+X mapping, such a mapping could persist until after boot and this would suppress the warning. Right?

Sure, you can _get_ checking with debug_checkwx(). But that's a debug option and it's kinda weird to shift the burden from an always-on thing like verify_rwx() to a literal debug option.

That said, I think you're completely on target for thinking that it's silly for verify_rwx() to even *try* to spew warnings in boot.

Quoted 8 lines

How about we add a kernel_strict_rwx() helper:

bool strict_kernel_rwx(void)

{
	return IS_ENABLED(CONFIG_STRICT_KERNEL_RWX) && rodata_enabled;
}

Have verify_rwx() check *that*.

Checking only that will bring the warning about ITS pages back :)

Both CONFIG_STRICT_KERNEL_RWX and rodata_enabled are set before the alternatives patching.

And rodata_enabled is kinda arm64 specific thingy :)

The check for (!kernel_set_to_readonly) has to stay to actually rule out the silly checks at boot and it covers your strict_kernel_rwx() because kernel_set_to_readonly is only set when strict_kernel_rwx() will be true.

Quoted 3 lines

Also, make the DEBUG_WX functionality mandatory for STRICT_KERNEL_RWX (with appropriate renaming) so that it can be depended upon. I assume all the distros are turning DEBUG_WX on already (Ubuntu seems to).

Just checked my Debian config, it does.

Quoted 2 lines

Maybe just do the strict_kernel_rwx() to start and then circle back around to muck with making DEBUG_WX mandatory?

I though about making DEBUG_WX mandatory right away, but that's surely not a oneliner, so I kept it for later. Maybe should have mentioned in the changelog, though.

Signature · 2 lines
Sincerely yours,
Mike.
#4 lore raw
Quoted 1 line · Mike Rapoport

On Tue, Sep 08, 2026 at 05:15:46PM -0700, Dave Hansen wrote:

...

Quoted 7 lines
Quoted 7 lines

How about we add a kernel_strict_rwx() helper:

bool strict_kernel_rwx(void)

{
	return IS_ENABLED(CONFIG_STRICT_KERNEL_RWX) && rodata_enabled;
}
Quoted 8 lines
Quoted 1 line

Have verify_rwx() check *that*.

Checking only that will bring the warning about ITS pages back :)

Both CONFIG_STRICT_KERNEL_RWX and rodata_enabled are set before the alternatives patching.

And rodata_enabled is kinda arm64 specific thingy :)

I think I was confusing rodata_enabled and kernel_set_to_readonly as I read through things.

But either way, I really don't like the idea of checking some random state bit. It's generally fragile.

Quoted 3 lines

The check for (!kernel_set_to_readonly) has to stay to actually rule out the silly checks at boot and it covers your strict_kernel_rwx() because kernel_set_to_readonly is only set when strict_kernel_rwx() will be true.

Could we do something slightly more generic?

mark_readonly() is awfully close to setting SYSTEM_RUNNING:

        mark_readonly();
        pti_finalize();
        system_state = SYSTEM_RUNNING;

What if we (eventually) did something like:

/*
* Get the kernel page tables ready to run userspace.
* There should be no page table manipulation between this and
* settings SYSTEM_RUNNING.
*/
void boot_finalize_page_tables(void)

{
	WARN_ON(system_state >= SYSTEM_RUNNING);

	mark_readonly();

	/*
         * Kernel mappings are now finalized - update
	 * the userspace page-able to finalize PTI.
	 */
	pti_finalize();

	/* Ensure the final kernel mappings have no W+X issues: */
	debug_checkwx();
}

and then verify_rwx() can just check system_state?

Quoted 4 lines
Quoted 3 lines

Also, make the DEBUG_WX functionality mandatory for STRICT_KERNEL_RWX (with appropriate renaming) so that it can be depended upon. I assume all the distros are turning DEBUG_WX on already (Ubuntu seems to).

Quoted 8 lines

Just checked my Debian config, it does.

Quoted 2 lines

Maybe just do the strict_kernel_rwx() to start and then circle back around to muck with making DEBUG_WX mandatory?

I though about making DEBUG_WX mandatory right away, but that's surely not a oneliner, so I kept it for later. Maybe should have mentioned in the changelog, though.

Yes, please. It would be nice to mention that the fix is a bit exposed without DEBUG_WX. It could honestly even pr_info_once() about it to make it less likely it somehow gets forgotten.

#5 lore raw
Quoted 29 lines · Dave Hansen
Quoted 1 line · Mike Rapoport

On Tue, Sep 08, 2026 at 05:15:46PM -0700, Dave Hansen wrote:

...

Quoted 7 lines
Quoted 7 lines

How about we add a kernel_strict_rwx() helper:

bool strict_kernel_rwx(void)

{
	return IS_ENABLED(CONFIG_STRICT_KERNEL_RWX) && rodata_enabled;
}
Quoted 8 lines
Quoted 1 line

Have verify_rwx() check *that*.

Checking only that will bring the warning about ITS pages back :)

Both CONFIG_STRICT_KERNEL_RWX and rodata_enabled are set before the alternatives patching.

And rodata_enabled is kinda arm64 specific thingy :)

I think I was confusing rodata_enabled and kernel_set_to_readonly as I read through things.

But either way, I really don't like the idea of checking some random state bit. It's generally fragile.

Quoted 3 lines

The check for (!kernel_set_to_readonly) has to stay to actually rule out the silly checks at boot and it covers your strict_kernel_rwx() because kernel_set_to_readonly is only set when strict_kernel_rwx() will be true.

Quoted 7 lines

Could we do something slightly more generic?

mark_readonly() is awfully close to setting SYSTEM_RUNNING:

        mark_readonly();
        pti_finalize();
        system_state = SYSTEM_RUNNING;

We could check SYSTEM_RUNNING, but OTOH using kernel_set_to_readonly is consistent with the existing checks in protect_rodata() and protect_kernel_text_ro() in the same file.

I think that it's better to use the consistent kernel_set_to_readonly now and flipping them all to SYSTEM_RUNNING ...

Quoted 1 line

What if we (eventually) did something like:

... when something like this materializes.

Quoted 21 lines

/*
* Get the kernel page tables ready to run userspace.
* There should be no page table manipulation between this and
* settings SYSTEM_RUNNING.
*/
void boot_finalize_page_tables(void)

{
	WARN_ON(system_state >= SYSTEM_RUNNING);

	mark_readonly();

	/*
         * Kernel mappings are now finalized - update
	 * the userspace page-able to finalize PTI.
	 */
	pti_finalize();

	/* Ensure the final kernel mappings have no W+X issues: */
	debug_checkwx();
}
Quoted 1 line

and then verify_rwx() can just check system_state?

and protect_rodata() and protect_kernel_text_ro() :)

Quoted 9 lines
Quoted 9 lines
Quoted 3 lines

Also, make the DEBUG_WX functionality mandatory for STRICT_KERNEL_RWX (with appropriate renaming) so that it can be depended upon. I assume all the distros are turning DEBUG_WX on already (Ubuntu seems to).

Just checked my Debian config, it does.

Quoted 2 lines

Maybe just do the strict_kernel_rwx() to start and then circle back around to muck with making DEBUG_WX mandatory?

Quoted 7 lines
Quoted 3 lines

I though about making DEBUG_WX mandatory right away, but that's surely not a oneliner, so I kept it for later. Maybe should have mentioned in the changelog, though.

Yes, please. It would be nice to mention that the fix is a bit exposed without DEBUG_WX. It could honestly even pr_info_once() about it to make it less likely it somehow gets forgotten.

I like the pr_info_once() idea, will add it in v2.

Signature · 2 lines
Sincerely yours,
Mike.
reviewer Tested-by on v1 1/1
#6 lore raw
In reply to Mike Rapoport
Quoted 31 lines · Mike Rapoport

From: "Mike Rapoport (Microsoft)" <[email protected]>

Nathan Chancellor reports the following warning:

CPA detected W^X violation: 8000000000000123 -> 0000000000000123 range: 0xffffffffc0400000 - 0xffffffffc0400fff PFN 100e00

  WARNING: arch/x86/mm/pat/set_memory.c:722 at __change_page_attr_set_clr+0xde7/0x1290, CPU#0: swapper/0/0
  Modules linked in:
  CPU: 0 UID: 0 PID: 0 Comm: swapper/0 Not tainted 7.3.0-rc1-debug-00006-g453e78594434 #1 PREEMPT(full)  2950d432dd3910251071a66f3134fe0875432786
  Hardware name: ASUS System Product Name/PRIME Z590M-PLUS, BIOS 1801 12/26/2022
  RIP: 0010:__change_page_attr_set_clr+0xdff/0x1290
  Code: 80 7c 24 42 00 0f 85 3a 04 00 00 48 8d 3d 19 8d 79 02 49 89 d9 4c 89 e1 4c 89 d2 4c 89 f6 4d 8d 84 24 ff 0f 00 00 4c 89 14 24 <67> 48 0f b9 3a 4c 8b 14 24 48 8b 0d 81 44 bf 01 41 f6 c2 01
  RSP: 0000:ffffffff87003c60 EFLAGS: 00010246
  RAX: 0000000000000002 RBX: 0000000000100e00 RCX: ffffffffc0400000
  RDX: 0000000000000123 RSI: 8000000000000123 RDI: ffffffff872e50c0
  RBP: 8000000100e00123 R08: ffffffffc0400fff R09: 0000000000100e00
  R10: 0000000000000123 R11: 0000000000000001 R12: ffffffffc0400000
  R13: 0000000100e00123 R14: 8000000000000123 R15: ffffffff87003d58
  FS:  0000000000000000(0000) GS:ffff8ad1777a7000(0000) knlGS:0000000000000000
  CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
  CR2: ffff8ad0a4201000 CR3: 00000007e3022001 CR4: 0000000000770ef0
  PKRU: 55555554
  Call Trace:
   <TASK>
   ? _vm_unmap_aliases+0x219/0x280
   change_page_attr_set_clr+0x161/0x250
   ? events_sysfs_show+0x5d/0x80
   set_memory_x+0x39/0x50
   apply_retpolines+0x656/0x6d0
   ? events_sysfs_show+0x5d/0x80
   ? events_sysfs_show+0x6c/0x80
   ? events_sysfs_show+0x62/0x80
Quoted 8 lines
   alternative_instructions+0x3c/0xd0
   arch_cpu_finalize_init+0x130/0x190
   start_kernel+0x97d/0xa10
   x86_64_start_reservations+0x24/0x30
   x86_64_start_kernel+0xda/0xe0
   common_startup_64+0x13e/0x151
   </TASK>
  ---[ end trace 0000000000000000 ]---

BPF CI has caught similar splats on linux-next [1]. The patch fixes it [2][3].

Thanks!

[1] https://github.com/kernel-patches/bpf/actions/runs/33918575854/job/101174705812
[2] https://github.com/kernel-patches/vmtest/pull/523
[3] https://github.com/kernel-patches/bpf/actions/runs/34279792157

Quoted 41 lines

The warning appears because commit 038176c21617f ("x86/mm/pat: fix effective RW computation in lookup_address_in_pgd_attr()") fixed the effective RW checked by verify_rwx() and it exposed that pages used for ITS trampolines temporarily have RWX permissions.

The permissions are updated in its_fini_core() after all the ITS trampolines are generated, but since verify_rwx() detects invalid transitions, it warns when its_alloc() makes RW memory executable.

At the time of alternatives patching the entire kernel text is mapped RWX, so the warning is bogus anyway.

Skip verification of W^X violations in verify_rwx() when they are triggered by transitions happening before the kernel text is remapped as read-only.

---
 arch/x86/mm/pat/set_memory.c | 4 ++++
 1 file changed, 4 insertions(+)
arch/x86/mm/pat/set_memory.c +4 -0
index 226660973d515..2eecb76703bb4 100644
@@ -708,6 +708,10 @@ static inline pgprot_t verify_rwx(pgprot_t old, pgprot_t new, unsigned long star
 	if (!(__supported_pte_mask & _PAGE_NX))
 		return new;
 
+	/* skip verification until kernel text is set to read only */
+	if (!kernel_set_to_readonly)
+		return new;
+
 	if (!((pgprot_val(old) ^ pgprot_val(new)) & (_PAGE_RW | _PAGE_NX)))
 		return new;
 

base-commit: 038176c21617fcc03ccc1ca43230ffedb712c047