FAILED: patch "[PATCH] arm64: Update PTE_RDONLY in set_pte_at() for PROT_NONE" failed to apply to 4.4-stable tree

PATCH [email protected] 2 participants 5 messages Apr 10, 2016 – Apr 18, 2016

Subsystems (2)

Stable branch arm64

lore

Share & tools

lore

[email protected]

author maintainer

2016-04-10 00:53 UTC #0 lore raw

The patch below does not apply to the 4.4-stable tree.
If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <[email protected]>.

thanks,

greg k-h

------------------ original commit in Linus's tree ------------------

Quoted 1 line

From fdc69e7df3cb24f18a93192641786e5b7ecd1dfe Mon Sep 17 00:00:00 2001

From: Catalin Marinas <[email protected]>
Date: Wed, 9 Mar 2016 16:31:29 +0000
Subject: [PATCH] arm64: Update PTE_RDONLY in set_pte_at() for PROT_NONE
permission

The set_pte_at() function must update the hardware PTE_RDONLY bit depending on the state of the PTE_WRITE and PTE_DIRTY bits of the given entry value. However, it currently only performs this for pte_valid() entries, ignoring PTE_PROT_NONE. The side-effect is that PROT_NONE mappings would not have the PTE_RDONLY bit set. Without CONFIG_ARM64_HW_AFDBM, this is not an issue since such PROT_NONE pages are not accessible anyway.

With commit 2f4b829c625e ("arm64: Add support for hardware updates of the access and dirty pte bits"), the ptep_set_wrprotect() function was re-written to cope with automatic hardware updates of the dirty state. As an optimisation, only PTE_RDONLY is checked to assess the "dirty" status. Since set_pte_at() does not set this bit for PROT_NONE mappings, such pages may be considered "dirty" as a result of ptep_set_wrprotect().

This patch updates the pte_valid() check to pte_present() in set_pte_at(). It also adds PTE_PROT_NONE to the swap entry bits comment.

Fixes 2f4b829c625e ("arm64: Add support for hardware updates of the access and dirty pte bits")

Signed-off-by Catalin Marinas <[email protected]>

Reported-by Ganapatrao Kulkarni <[email protected]>

Tested-by Ganapatrao Kulkarni <[email protected]>

Cc <[email protected]>

arch/arm64/include/asm/pgtable.h +2 -1

index 7c73b365fcfa..e308807105e2 100644
@@ -201,7 +201,7 @@ extern void __sync_icache_dcache(pte_t pteval, unsigned long addr);
 static inline void set_pte_at(struct mm_struct *mm, unsigned long addr,
 			      pte_t *ptep, pte_t pte)
 {
-	if (pte_valid(pte)) {
+	if (pte_present(pte)) {
 		if (pte_sw_dirty(pte) && pte_write(pte))
 			pte_val(pte) &= ~PTE_RDONLY;
 		else

@@ -626,6 +626,7 @@ extern pgd_t idmap_pg_dir[PTRS_PER_PGD];
  *	bits 0-1:	present (must be zero)
  *	bits 2-7:	swap type
  *	bits 8-57:	swap offset
+ *	bit  58:	PTE_PROT_NONE (must be zero)
  */
 #define __SWP_TYPE_SHIFT	2
 #define __SWP_TYPE_BITS		6

Catalin Marinas

maintainer

2016-04-11 14:48 UTC #1 lore raw

Hi Greg,

Quoted 33 lines · On Sun, Apr 10, 2016 at 01:53:35AM +0100, [email protected] wrote:

The patch below does not apply to the 4.4-stable tree.
If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <[email protected]>.
thanks,
greg k-h
------------------ original commit in Linus's tree ------------------
From fdc69e7df3cb24f18a93192641786e5b7ecd1dfe Mon Sep 17 00:00:00 2001
From: Catalin Marinas <[email protected]>
Date: Wed, 9 Mar 2016 16:31:29 +0000
Subject: [PATCH] arm64: Update PTE_RDONLY in set_pte_at() for PROT_NONE
permission
The set_pte_at() function must update the hardware PTE_RDONLY bit depending on the state of the PTE_WRITE and PTE_DIRTY bits of the given entry value. However, it currently only performs this for pte_valid() entries, ignoring PTE_PROT_NONE. The side-effect is that PROT_NONE mappings would not have the PTE_RDONLY bit set. Without CONFIG_ARM64_HW_AFDBM, this is not an issue since such PROT_NONE pages are not accessible anyway.
With commit 2f4b829c625e ("arm64: Add support for hardware updates of the access and dirty pte bits"), the ptep_set_wrprotect() function was re-written to cope with automatic hardware updates of the dirty state. As an optimisation, only PTE_RDONLY is checked to assess the "dirty" status. Since set_pte_at() does not set this bit for PROT_NONE mappings, such pages may be considered "dirty" as a result of ptep_set_wrprotect().

Quoted 8 lines

This patch updates the pte_valid() check to pte_present() in set_pte_at(). It also adds PTE_PROT_NONE to the swap entry bits comment.
Fixes 2f4b829c625e ("arm64: Add support for hardware updates of the access and dirty pte bits")
Signed-off-by Catalin Marinas <[email protected]>
Reported-by Ganapatrao Kulkarni <[email protected]>
Tested-by Ganapatrao Kulkarni <[email protected]>
Cc <[email protected]>

This patch requires another commit that was merged in 4.5. The actual "Cc: stable" line above should have been:

Cc <[email protected]> # 4.4.x: ac15bd63bbb2: arm64: Honour !PTE_WRITE in set_pte_at() for kernel mappings

Shall I re-send commit fdc69e7df3cb ("arm64: Update PTE_RDONLY in set_pte_at() for PROT_NONE permission") with the amended "Cc: stable" line or you're OK to cherry-pick the dependency?

Thanks.

Signature · 1 line

Catalin

[email protected]

author maintainer

2016-04-11 17:46 UTC #2 lore raw

Quoted 45 lines · Catalin Marinas

Hi Greg,
Quoted 33 lines · On Sun, Apr 10, 2016 at 01:53:35AM +0100, [email protected] wrote:
The patch below does not apply to the 4.4-stable tree.
If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <[email protected]>.
thanks,
greg k-h
------------------ original commit in Linus's tree ------------------
From fdc69e7df3cb24f18a93192641786e5b7ecd1dfe Mon Sep 17 00:00:00 2001
From: Catalin Marinas <[email protected]>
Date: Wed, 9 Mar 2016 16:31:29 +0000
Subject: [PATCH] arm64: Update PTE_RDONLY in set_pte_at() for PROT_NONE
permission
The set_pte_at() function must update the hardware PTE_RDONLY bit depending on the state of the PTE_WRITE and PTE_DIRTY bits of the given entry value. However, it currently only performs this for pte_valid() entries, ignoring PTE_PROT_NONE. The side-effect is that PROT_NONE mappings would not have the PTE_RDONLY bit set. Without CONFIG_ARM64_HW_AFDBM, this is not an issue since such PROT_NONE pages are not accessible anyway.
With commit 2f4b829c625e ("arm64: Add support for hardware updates of the access and dirty pte bits"), the ptep_set_wrprotect() function was re-written to cope with automatic hardware updates of the dirty state. As an optimisation, only PTE_RDONLY is checked to assess the "dirty" status. Since set_pte_at() does not set this bit for PROT_NONE mappings, such pages may be considered "dirty" as a result of ptep_set_wrprotect().
Quoted 8 lines
This patch updates the pte_valid() check to pte_present() in set_pte_at(). It also adds PTE_PROT_NONE to the swap entry bits comment.
Fixes 2f4b829c625e ("arm64: Add support for hardware updates of the access and dirty pte bits")
Signed-off-by Catalin Marinas <[email protected]>
Reported-by Ganapatrao Kulkarni <[email protected]>
Tested-by Ganapatrao Kulkarni <[email protected]>
Cc <[email protected]>

Quoted 8 lines

This patch requires another commit that was merged in 4.5. The actual "Cc: stable" line above should have been:
Cc <[email protected]> # 4.4.x: ac15bd63bbb2: arm64: Honour !PTE_WRITE in set_pte_at() for kernel mappings
Shall I re-send commit fdc69e7df3cb ("arm64: Update PTE_RDONLY in set_pte_at() for PROT_NONE permission") with the amended "Cc: stable" line or you're OK to cherry-pick the dependency?

I can cherry-pick, thanks.

greg k-h

[email protected]

author maintainer

2016-04-18 01:23 UTC #3 lore raw

Quoted 49 lines · On Mon, Apr 11, 2016 at 10:46:07AM -0700, [email protected] wrote:

Quoted 48 lines · Catalin Marinas
Hi Greg,
Quoted 33 lines · On Sun, Apr 10, 2016 at 01:53:35AM +0100, [email protected] wrote:
The patch below does not apply to the 4.4-stable tree.
If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <[email protected]>.
thanks,
greg k-h
------------------ original commit in Linus's tree ------------------
From fdc69e7df3cb24f18a93192641786e5b7ecd1dfe Mon Sep 17 00:00:00 2001
From: Catalin Marinas <[email protected]>
Date: Wed, 9 Mar 2016 16:31:29 +0000
Subject: [PATCH] arm64: Update PTE_RDONLY in set_pte_at() for PROT_NONE
permission
The set_pte_at() function must update the hardware PTE_RDONLY bit depending on the state of the PTE_WRITE and PTE_DIRTY bits of the given entry value. However, it currently only performs this for pte_valid() entries, ignoring PTE_PROT_NONE. The side-effect is that PROT_NONE mappings would not have the PTE_RDONLY bit set. Without CONFIG_ARM64_HW_AFDBM, this is not an issue since such PROT_NONE pages are not accessible anyway.
With commit 2f4b829c625e ("arm64: Add support for hardware updates of the access and dirty pte bits"), the ptep_set_wrprotect() function was re-written to cope with automatic hardware updates of the dirty state. As an optimisation, only PTE_RDONLY is checked to assess the "dirty" status. Since set_pte_at() does not set this bit for PROT_NONE mappings, such pages may be considered "dirty" as a result of ptep_set_wrprotect().
Quoted 8 lines
This patch updates the pte_valid() check to pte_present() in set_pte_at(). It also adds PTE_PROT_NONE to the swap entry bits comment.
Fixes 2f4b829c625e ("arm64: Add support for hardware updates of the access and dirty pte bits")
Signed-off-by Catalin Marinas <[email protected]>
Reported-by Ganapatrao Kulkarni <[email protected]>
Tested-by Ganapatrao Kulkarni <[email protected]>
Cc <[email protected]>
This patch requires another commit that was merged in 4.5. The actual "Cc: stable" line above should have been:

Quoted 7 lines

Quoted 5 lines
Cc <[email protected]> # 4.4.x: ac15bd63bbb2: arm64: Honour !PTE_WRITE in set_pte_at() for kernel mappings
Shall I re-send commit fdc69e7df3cb ("arm64: Update PTE_RDONLY in set_pte_at() for PROT_NONE permission") with the amended "Cc: stable" line or you're OK to cherry-pick the dependency?
I can cherry-pick, thanks.

Ok, I take it back, I need a backported version of fdc69e7df3cb for 4.4-stable as it didn't apply cleanly and I don't know the code at all to be able to do it myself by hand.

Can you provide that backport?

thanks,

greg k-h

Catalin Marinas

maintainer

2016-04-18 09:15 UTC #4 lore raw

Hi Greg,

Quoted 11 lines · On Mon, Apr 18, 2016 at 10:23:28AM +0900, [email protected] wrote:

Quoted 10 lines · On Mon, Apr 11, 2016 at 10:46:07AM -0700, [email protected] wrote:
Quoted 8 lines · Catalin Marinas
This patch requires another commit that was merged in 4.5. The actual "Cc: stable" line above should have been:
Cc <[email protected]> # 4.4.x: ac15bd63bbb2: arm64: Honour !PTE_WRITE in set_pte_at() for kernel mappings
Shall I re-send commit fdc69e7df3cb ("arm64: Update PTE_RDONLY in set_pte_at() for PROT_NONE permission") with the amended "Cc: stable" line or you're OK to cherry-pick the dependency?

Quoted 7 lines

Quoted 1 line
I can cherry-pick, thanks.
Ok, I take it back, I need a backported version of fdc69e7df3cb for 4.4-stable as it didn't apply cleanly and I don't know the code at all to be able to do it myself by hand.
Can you provide that backport?

As I mentioned in above, it applies cleanly once commit ac15bd63bbb2 ("arm64: Honour !PTE_WRITE in set_pte_at() for kernel mappings") is cherry-picked.

I'll send both commit ac15bd63bbb2 and fdc69e7df3cb as patches to [email protected].

Thanks.

Signature · 1 line

Catalin