[PATCH 0/5] Minor improvements to the compat vdso and sigpage

[PATCH 0/5] Minor improvements to the compat vdso and sigpage

[Will Deacon]

Hi all,

Here are a few small improvements to the compat vDSO and sigpage
implementation for arm64 which I cooked up when comparing the code with
the native 32-bit counterpart.

Cheers,

Will

--->8

Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Marc Zyngier <maz@kernel.org>
Cc: Vincenzo Frascino <vincenzo.frascino@arm.com>
Cc: Russell King <rmk+kernel@armlinux.org.uk>

Will Deacon (5):
  arm64: vdso: Use GFP_KERNEL for allocating compat vdso and signal
    pages
  arm64: vdso: Remove redundant calls to flush_dcache_page()
  arm64: compat: Allow signal page to be remapped
  arm64: vdso: Avoid ISB after reading from cntvct_el0
  arm64: compat: Poison the compat sigpage

 arch/arm64/include/asm/arch_timer.h        | 21 ---------------------
 arch/arm64/include/asm/barrier.h           | 19 +++++++++++++++++++
 arch/arm64/include/asm/vdso/gettimeofday.h |  6 +-----
 arch/arm64/kernel/vdso.c                   | 22 ++++++++++++++++------
 4 files changed, 36 insertions(+), 32 deletions(-)

-- 
2.31.0.rc2.261.g7f71774620-goog


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

[PATCH 1/5] arm64: vdso: Use GFP_KERNEL for allocating compat vdso and signal pages

[Will Deacon]

There's no need to allocate the compat vDSO and signal pages using
GFP_ATOMIC allocations, so use GFP_KERNEL instead.

Signed-off-by: Will Deacon <will@kernel.org>
---
 arch/arm64/kernel/vdso.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/arm64/kernel/vdso.c b/arch/arm64/kernel/vdso.c
index cee5d04ea9ad..2d057a4dc787 100644
--- a/arch/arm64/kernel/vdso.c
+++ b/arch/arm64/kernel/vdso.c
@@ -299,7 +299,7 @@ static int aarch32_alloc_kuser_vdso_page(void)
 	if (!IS_ENABLED(CONFIG_KUSER_HELPERS))
 		return 0;
 
-	vdso_page = get_zeroed_page(GFP_ATOMIC);
+	vdso_page = get_zeroed_page(GFP_KERNEL);
 	if (!vdso_page)
 		return -ENOMEM;
 
@@ -316,7 +316,7 @@ static int aarch32_alloc_sigpage(void)
 	int sigret_sz = __aarch32_sigret_code_end - __aarch32_sigret_code_start;
 	unsigned long sigpage;
 
-	sigpage = get_zeroed_page(GFP_ATOMIC);
+	sigpage = get_zeroed_page(GFP_KERNEL);
 	if (!sigpage)
 		return -ENOMEM;
 
-- 
2.31.0.rc2.261.g7f71774620-goog


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

[PATCH 2/5] arm64: vdso: Remove redundant calls to flush_dcache_page()

[Will Deacon]

flush_dcache_page() ensures that the 'PG_dcache_clean' flag for its
'page' argument is clear so that cache maintenance will be performed
if the page is mapped into userspace with execute permissions.

Newly allocated pages have this flag clear, so there is no need to call
flush_dcache_page() for the compat vdso or signal pages. Remove the
redundant calls.

Signed-off-by: Will Deacon <will@kernel.org>
---
 arch/arm64/kernel/vdso.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/arch/arm64/kernel/vdso.c b/arch/arm64/kernel/vdso.c
index 2d057a4dc787..421411981dc3 100644
--- a/arch/arm64/kernel/vdso.c
+++ b/arch/arm64/kernel/vdso.c
@@ -306,7 +306,6 @@ static int aarch32_alloc_kuser_vdso_page(void)
 	memcpy((void *)(vdso_page + 0x1000 - kuser_sz), __kuser_helper_start,
 	       kuser_sz);
 	aarch32_vectors_page = virt_to_page(vdso_page);
-	flush_dcache_page(aarch32_vectors_page);
 	return 0;
 }
 
@@ -322,7 +321,6 @@ static int aarch32_alloc_sigpage(void)
 
 	memcpy((void *)sigpage, __aarch32_sigret_code_start, sigret_sz);
 	aarch32_sig_page = virt_to_page(sigpage);
-	flush_dcache_page(aarch32_sig_page);
 	return 0;
 }
 
-- 
2.31.0.rc2.261.g7f71774620-goog


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

[PATCH 3/5] arm64: compat: Allow signal page to be remapped

[Will Deacon]

For compatability with 32-bit Arm, allow the compat signal page to be
remapped via mremap().

Signed-off-by: Will Deacon <will@kernel.org>
---
 arch/arm64/kernel/vdso.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/arch/arm64/kernel/vdso.c b/arch/arm64/kernel/vdso.c
index 421411981dc3..16bf0b46fb70 100644
--- a/arch/arm64/kernel/vdso.c
+++ b/arch/arm64/kernel/vdso.c
@@ -271,6 +271,14 @@ enum aarch32_map {
 static struct page *aarch32_vectors_page __ro_after_init;
 static struct page *aarch32_sig_page __ro_after_init;
 
+static int aarch32_sigpage_mremap(const struct vm_special_mapping *sm,
+				  struct vm_area_struct *new_vma)
+{
+	current->mm->context.sigpage = (void *)new_vma->vm_start;
+
+	return 0;
+}
+
 static struct vm_special_mapping aarch32_vdso_maps[] = {
 	[AA32_MAP_VECTORS] = {
 		.name	= "[vectors]", /* ABI */
@@ -279,6 +287,7 @@ static struct vm_special_mapping aarch32_vdso_maps[] = {
 	[AA32_MAP_SIGPAGE] = {
 		.name	= "[sigpage]", /* ABI */
 		.pages	= &aarch32_sig_page,
+		.mremap	= aarch32_sigpage_mremap,
 	},
 	[AA32_MAP_VVAR] = {
 		.name = "[vvar]",
-- 
2.31.0.rc2.261.g7f71774620-goog


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

[PATCH 4/5] arm64: vdso: Avoid ISB after reading from cntvct_el0

[Will Deacon]

We can avoid the expensive ISB instruction after reading the counter in
the vDSO gettime functions by creating a fake address hazard against a
dummy stack read, just like we do inside the kernel.

Signed-off-by: Will Deacon <will@kernel.org>
---
 arch/arm64/include/asm/arch_timer.h        | 21 ---------------------
 arch/arm64/include/asm/barrier.h           | 19 +++++++++++++++++++
 arch/arm64/include/asm/vdso/gettimeofday.h |  6 +-----
 3 files changed, 20 insertions(+), 26 deletions(-)

diff --git a/arch/arm64/include/asm/arch_timer.h b/arch/arm64/include/asm/arch_timer.h
index 9f0ec21d6327..88d20f04c64a 100644
--- a/arch/arm64/include/asm/arch_timer.h
+++ b/arch/arm64/include/asm/arch_timer.h
@@ -165,25 +165,6 @@ static inline void arch_timer_set_cntkctl(u32 cntkctl)
 	isb();
 }
 
-/*
- * Ensure that reads of the counter are treated the same as memory reads
- * for the purposes of ordering by subsequent memory barriers.
- *
- * This insanity brought to you by speculative system register reads,
- * out-of-order memory accesses, sequence locks and Thomas Gleixner.
- *
- * http://lists.infradead.org/pipermail/linux-arm-kernel/2019-February/631195.html
- */
-#define arch_counter_enforce_ordering(val) do {				\
-	u64 tmp, _val = (val);						\
-									\
-	asm volatile(							\
-	"	eor	%0, %1, %1\n"					\
-	"	add	%0, sp, %0\n"					\
-	"	ldr	xzr, [%0]"					\
-	: "=r" (tmp) : "r" (_val));					\
-} while (0)
-
 static __always_inline u64 __arch_counter_get_cntpct_stable(void)
 {
 	u64 cnt;
@@ -224,8 +205,6 @@ static __always_inline u64 __arch_counter_get_cntvct(void)
 	return cnt;
 }
 
-#undef arch_counter_enforce_ordering
-
 static inline int arch_timer_arch_init(void)
 {
 	return 0;
diff --git a/arch/arm64/include/asm/barrier.h b/arch/arm64/include/asm/barrier.h
index c3009b0e5239..37d891af8ea5 100644
--- a/arch/arm64/include/asm/barrier.h
+++ b/arch/arm64/include/asm/barrier.h
@@ -70,6 +70,25 @@ static inline unsigned long array_index_mask_nospec(unsigned long idx,
 	return mask;
 }
 
+/*
+ * Ensure that reads of the counter are treated the same as memory reads
+ * for the purposes of ordering by subsequent memory barriers.
+ *
+ * This insanity brought to you by speculative system register reads,
+ * out-of-order memory accesses, sequence locks and Thomas Gleixner.
+ *
+ * http://lists.infradead.org/pipermail/linux-arm-kernel/2019-February/631195.html
+ */
+#define arch_counter_enforce_ordering(val) do {				\
+	u64 tmp, _val = (val);						\
+									\
+	asm volatile(							\
+	"	eor	%0, %1, %1\n"					\
+	"	add	%0, sp, %0\n"					\
+	"	ldr	xzr, [%0]"					\
+	: "=r" (tmp) : "r" (_val));					\
+} while (0)
+
 #define __smp_mb()	dmb(ish)
 #define __smp_rmb()	dmb(ishld)
 #define __smp_wmb()	dmb(ishst)
diff --git a/arch/arm64/include/asm/vdso/gettimeofday.h b/arch/arm64/include/asm/vdso/gettimeofday.h
index 631ab1281633..4b4c0dac0e14 100644
--- a/arch/arm64/include/asm/vdso/gettimeofday.h
+++ b/arch/arm64/include/asm/vdso/gettimeofday.h
@@ -83,11 +83,7 @@ static __always_inline u64 __arch_get_hw_counter(s32 clock_mode,
 	 */
 	isb();
 	asm volatile("mrs %0, cntvct_el0" : "=r" (res) :: "memory");
-	/*
-	 * This isb() is required to prevent that the seq lock is
-	 * speculated.#
-	 */
-	isb();
+	arch_counter_enforce_ordering(res);
 
 	return res;
 }
-- 
2.31.0.rc2.261.g7f71774620-goog


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

[PATCH 5/5] arm64: compat: Poison the compat sigpage

[Will Deacon]

Commit 9c698bff66ab ("ARM: ensure the signal page contains defined contents")
poisoned the unused portions of the signal page for 32-bit Arm.

Implement the same poisoning for the compat signal page on arm64 rather
than using __GFP_ZERO.

Signed-off-by: Will Deacon <will@kernel.org>
---
 arch/arm64/kernel/vdso.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/arch/arm64/kernel/vdso.c b/arch/arm64/kernel/vdso.c
index 16bf0b46fb70..159b72a646ab 100644
--- a/arch/arm64/kernel/vdso.c
+++ b/arch/arm64/kernel/vdso.c
@@ -318,17 +318,20 @@ static int aarch32_alloc_kuser_vdso_page(void)
 	return 0;
 }
 
+#define COMPAT_SIGPAGE_POISON_WORD	0xe7fddef1
 static int aarch32_alloc_sigpage(void)
 {
 	extern char __aarch32_sigret_code_start[], __aarch32_sigret_code_end[];
 	int sigret_sz = __aarch32_sigret_code_end - __aarch32_sigret_code_start;
-	unsigned long sigpage;
+	__le32 poison = cpu_to_le32(COMPAT_SIGPAGE_POISON_WORD);
+	void *sigpage;
 
-	sigpage = get_zeroed_page(GFP_KERNEL);
+	sigpage = (void *)__get_free_page(GFP_KERNEL);
 	if (!sigpage)
 		return -ENOMEM;
 
-	memcpy((void *)sigpage, __aarch32_sigret_code_start, sigret_sz);
+	memset32(sigpage, (__force u32)poison, PAGE_SIZE / sizeof(poison));
+	memcpy(sigpage, __aarch32_sigret_code_start, sigret_sz);
 	aarch32_sig_page = virt_to_page(sigpage);
 	return 0;
 }
-- 
2.31.0.rc2.261.g7f71774620-goog


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH 1/5] arm64: vdso: Use GFP_KERNEL for allocating compat vdso and signal pages

[Vincenzo Frascino]

On 3/18/21 5:07 PM, Will Deacon wrote:
> There's no need to allocate the compat vDSO and signal pages using
> GFP_ATOMIC allocations, so use GFP_KERNEL instead.
> 
> Signed-off-by: Will Deacon <will@kernel.org>

Nit: We could explain why GFP_KERNEL is sufficient in the commit message.

Otherwise:

Reviewed-by: Vincenzo Frascino <vincenzo.frascino@arm.com>

> ---
>  arch/arm64/kernel/vdso.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/arch/arm64/kernel/vdso.c b/arch/arm64/kernel/vdso.c
> index cee5d04ea9ad..2d057a4dc787 100644
> --- a/arch/arm64/kernel/vdso.c
> +++ b/arch/arm64/kernel/vdso.c
> @@ -299,7 +299,7 @@ static int aarch32_alloc_kuser_vdso_page(void)
>  	if (!IS_ENABLED(CONFIG_KUSER_HELPERS))
>  		return 0;
>  
> -	vdso_page = get_zeroed_page(GFP_ATOMIC);
> +	vdso_page = get_zeroed_page(GFP_KERNEL);
>  	if (!vdso_page)
>  		return -ENOMEM;
>  
> @@ -316,7 +316,7 @@ static int aarch32_alloc_sigpage(void)
>  	int sigret_sz = __aarch32_sigret_code_end - __aarch32_sigret_code_start;
>  	unsigned long sigpage;
>  
> -	sigpage = get_zeroed_page(GFP_ATOMIC);
> +	sigpage = get_zeroed_page(GFP_KERNEL);
>  	if (!sigpage)
>  		return -ENOMEM;
>  
> 

-- 
Regards,
Vincenzo

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH 2/5] arm64: vdso: Remove redundant calls to flush_dcache_page()

[Vincenzo Frascino]

On 3/18/21 5:07 PM, Will Deacon wrote:
> flush_dcache_page() ensures that the 'PG_dcache_clean' flag for its
> 'page' argument is clear so that cache maintenance will be performed
> if the page is mapped into userspace with execute permissions.
> 
> Newly allocated pages have this flag clear, so there is no need to call
> flush_dcache_page() for the compat vdso or signal pages. Remove the
> redundant calls.
> 
> Signed-off-by: Will Deacon <will@kernel.org>

Reviewed-by: Vincenzo Frascino <vincenzo.frascino@arm.com>

> ---
>  arch/arm64/kernel/vdso.c | 2 --
>  1 file changed, 2 deletions(-)
> 
> diff --git a/arch/arm64/kernel/vdso.c b/arch/arm64/kernel/vdso.c
> index 2d057a4dc787..421411981dc3 100644
> --- a/arch/arm64/kernel/vdso.c
> +++ b/arch/arm64/kernel/vdso.c
> @@ -306,7 +306,6 @@ static int aarch32_alloc_kuser_vdso_page(void)
>  	memcpy((void *)(vdso_page + 0x1000 - kuser_sz), __kuser_helper_start,
>  	       kuser_sz);
>  	aarch32_vectors_page = virt_to_page(vdso_page);
> -	flush_dcache_page(aarch32_vectors_page);
>  	return 0;
>  }
>  
> @@ -322,7 +321,6 @@ static int aarch32_alloc_sigpage(void)
>  
>  	memcpy((void *)sigpage, __aarch32_sigret_code_start, sigret_sz);
>  	aarch32_sig_page = virt_to_page(sigpage);
> -	flush_dcache_page(aarch32_sig_page);
>  	return 0;
>  }
>  
> 

-- 
Regards,
Vincenzo

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH 3/5] arm64: compat: Allow signal page to be remapped

[Vincenzo Frascino]

On 3/18/21 5:07 PM, Will Deacon wrote:
> For compatability with 32-bit Arm, allow the compat signal page to be
> remapped via mremap().
> 
> Signed-off-by: Will Deacon <will@kernel.org>

Reviewed-by: Vincenzo Frascino <vincenzo.frascino@arm.com>

> ---
>  arch/arm64/kernel/vdso.c | 9 +++++++++
>  1 file changed, 9 insertions(+)
> 
> diff --git a/arch/arm64/kernel/vdso.c b/arch/arm64/kernel/vdso.c
> index 421411981dc3..16bf0b46fb70 100644
> --- a/arch/arm64/kernel/vdso.c
> +++ b/arch/arm64/kernel/vdso.c
> @@ -271,6 +271,14 @@ enum aarch32_map {
>  static struct page *aarch32_vectors_page __ro_after_init;
>  static struct page *aarch32_sig_page __ro_after_init;
>  
> +static int aarch32_sigpage_mremap(const struct vm_special_mapping *sm,
> +				  struct vm_area_struct *new_vma)
> +{
> +	current->mm->context.sigpage = (void *)new_vma->vm_start;
> +
> +	return 0;
> +}
> +
>  static struct vm_special_mapping aarch32_vdso_maps[] = {
>  	[AA32_MAP_VECTORS] = {
>  		.name	= "[vectors]", /* ABI */
> @@ -279,6 +287,7 @@ static struct vm_special_mapping aarch32_vdso_maps[] = {
>  	[AA32_MAP_SIGPAGE] = {
>  		.name	= "[sigpage]", /* ABI */
>  		.pages	= &aarch32_sig_page,
> +		.mremap	= aarch32_sigpage_mremap,
>  	},
>  	[AA32_MAP_VVAR] = {
>  		.name = "[vvar]",
> 

-- 
Regards,
Vincenzo

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH 4/5] arm64: vdso: Avoid ISB after reading from cntvct_el0

[Vincenzo Frascino]

On 3/18/21 5:07 PM, Will Deacon wrote:
> We can avoid the expensive ISB instruction after reading the counter in
> the vDSO gettime functions by creating a fake address hazard against a
> dummy stack read, just like we do inside the kernel.
> 
> Signed-off-by: Will Deacon <will@kernel.org>

Reviewed-by: Vincenzo Frascino <vincenzo.frascino@arm.com>

> ---
>  arch/arm64/include/asm/arch_timer.h        | 21 ---------------------
>  arch/arm64/include/asm/barrier.h           | 19 +++++++++++++++++++
>  arch/arm64/include/asm/vdso/gettimeofday.h |  6 +-----
>  3 files changed, 20 insertions(+), 26 deletions(-)
> 
> diff --git a/arch/arm64/include/asm/arch_timer.h b/arch/arm64/include/asm/arch_timer.h
> index 9f0ec21d6327..88d20f04c64a 100644
> --- a/arch/arm64/include/asm/arch_timer.h
> +++ b/arch/arm64/include/asm/arch_timer.h
> @@ -165,25 +165,6 @@ static inline void arch_timer_set_cntkctl(u32 cntkctl)
>  	isb();
>  }
>  
> -/*
> - * Ensure that reads of the counter are treated the same as memory reads
> - * for the purposes of ordering by subsequent memory barriers.
> - *
> - * This insanity brought to you by speculative system register reads,
> - * out-of-order memory accesses, sequence locks and Thomas Gleixner.
> - *
> - * http://lists.infradead.org/pipermail/linux-arm-kernel/2019-February/631195.html
> - */
> -#define arch_counter_enforce_ordering(val) do {				\
> -	u64 tmp, _val = (val);						\
> -									\
> -	asm volatile(							\
> -	"	eor	%0, %1, %1\n"					\
> -	"	add	%0, sp, %0\n"					\
> -	"	ldr	xzr, [%0]"					\
> -	: "=r" (tmp) : "r" (_val));					\
> -} while (0)
> -
>  static __always_inline u64 __arch_counter_get_cntpct_stable(void)
>  {
>  	u64 cnt;
> @@ -224,8 +205,6 @@ static __always_inline u64 __arch_counter_get_cntvct(void)
>  	return cnt;
>  }
>  
> -#undef arch_counter_enforce_ordering
> -
>  static inline int arch_timer_arch_init(void)
>  {
>  	return 0;
> diff --git a/arch/arm64/include/asm/barrier.h b/arch/arm64/include/asm/barrier.h
> index c3009b0e5239..37d891af8ea5 100644
> --- a/arch/arm64/include/asm/barrier.h
> +++ b/arch/arm64/include/asm/barrier.h
> @@ -70,6 +70,25 @@ static inline unsigned long array_index_mask_nospec(unsigned long idx,
>  	return mask;
>  }
>  
> +/*
> + * Ensure that reads of the counter are treated the same as memory reads
> + * for the purposes of ordering by subsequent memory barriers.
> + *
> + * This insanity brought to you by speculative system register reads,
> + * out-of-order memory accesses, sequence locks and Thomas Gleixner.
> + *
> + * http://lists.infradead.org/pipermail/linux-arm-kernel/2019-February/631195.html
> + */
> +#define arch_counter_enforce_ordering(val) do {				\
> +	u64 tmp, _val = (val);						\
> +									\
> +	asm volatile(							\
> +	"	eor	%0, %1, %1\n"					\
> +	"	add	%0, sp, %0\n"					\
> +	"	ldr	xzr, [%0]"					\
> +	: "=r" (tmp) : "r" (_val));					\
> +} while (0)
> +
>  #define __smp_mb()	dmb(ish)
>  #define __smp_rmb()	dmb(ishld)
>  #define __smp_wmb()	dmb(ishst)
> diff --git a/arch/arm64/include/asm/vdso/gettimeofday.h b/arch/arm64/include/asm/vdso/gettimeofday.h
> index 631ab1281633..4b4c0dac0e14 100644
> --- a/arch/arm64/include/asm/vdso/gettimeofday.h
> +++ b/arch/arm64/include/asm/vdso/gettimeofday.h
> @@ -83,11 +83,7 @@ static __always_inline u64 __arch_get_hw_counter(s32 clock_mode,
>  	 */
>  	isb();
>  	asm volatile("mrs %0, cntvct_el0" : "=r" (res) :: "memory");
> -	/*
> -	 * This isb() is required to prevent that the seq lock is
> -	 * speculated.#
> -	 */
> -	isb();
> +	arch_counter_enforce_ordering(res);
>  
>  	return res;
>  }
> 

-- 
Regards,
Vincenzo

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH 5/5] arm64: compat: Poison the compat sigpage

[Vincenzo Frascino]

On 3/18/21 5:07 PM, Will Deacon wrote:
> Commit 9c698bff66ab ("ARM: ensure the signal page contains defined contents")
> poisoned the unused portions of the signal page for 32-bit Arm.
> 
> Implement the same poisoning for the compat signal page on arm64 rather
> than using __GFP_ZERO.
> 
> Signed-off-by: Will Deacon <will@kernel.org>

Reviewed-by: Vincenzo Frascino <vincenzo.frascino@arm.com>

> ---
>  arch/arm64/kernel/vdso.c | 9 ++++++---
>  1 file changed, 6 insertions(+), 3 deletions(-)
> 
> diff --git a/arch/arm64/kernel/vdso.c b/arch/arm64/kernel/vdso.c
> index 16bf0b46fb70..159b72a646ab 100644
> --- a/arch/arm64/kernel/vdso.c
> +++ b/arch/arm64/kernel/vdso.c
> @@ -318,17 +318,20 @@ static int aarch32_alloc_kuser_vdso_page(void)
>  	return 0;
>  }
>  
> +#define COMPAT_SIGPAGE_POISON_WORD	0xe7fddef1
>  static int aarch32_alloc_sigpage(void)
>  {
>  	extern char __aarch32_sigret_code_start[], __aarch32_sigret_code_end[];
>  	int sigret_sz = __aarch32_sigret_code_end - __aarch32_sigret_code_start;
> -	unsigned long sigpage;
> +	__le32 poison = cpu_to_le32(COMPAT_SIGPAGE_POISON_WORD);
> +	void *sigpage;
>  
> -	sigpage = get_zeroed_page(GFP_KERNEL);
> +	sigpage = (void *)__get_free_page(GFP_KERNEL);
>  	if (!sigpage)
>  		return -ENOMEM;
>  
> -	memcpy((void *)sigpage, __aarch32_sigret_code_start, sigret_sz);
> +	memset32(sigpage, (__force u32)poison, PAGE_SIZE / sizeof(poison));
> +	memcpy(sigpage, __aarch32_sigret_code_start, sigret_sz);
>  	aarch32_sig_page = virt_to_page(sigpage);
>  	return 0;
>  }
> 

-- 
Regards,
Vincenzo

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH 0/5] Minor improvements to the compat vdso and sigpage

[Catalin Marinas]

On Thu, 18 Mar 2021 17:07:33 +0000, Will Deacon wrote:
> Here are a few small improvements to the compat vDSO and sigpage
> implementation for arm64 which I cooked up when comparing the code with
> the native 32-bit counterpart.
> 
> Cheers,
> 
> Will
> 
> [...]

Applied to arm64 (for-next/vdso), thanks!

[1/5] arm64: vdso: Use GFP_KERNEL for allocating compat vdso and signal pages
      https://git.kernel.org/arm64/c/7cd6ca1d7902
[2/5] arm64: vdso: Remove redundant calls to flush_dcache_page()
      https://git.kernel.org/arm64/c/e9be47eab1cd
[3/5] arm64: compat: Allow signal page to be remapped
      https://git.kernel.org/arm64/c/7adbf10e29c2
[4/5] arm64: vdso: Avoid ISB after reading from cntvct_el0
      https://git.kernel.org/arm64/c/77ec462536a1
[5/5] arm64: compat: Poison the compat sigpage
      https://git.kernel.org/arm64/c/6e554abd0700

-- 
Catalin


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH 1/5] arm64: vdso: Use GFP_KERNEL for allocating compat vdso and signal pages

[Catalin Marinas]

On Thu, Mar 18, 2021 at 06:03:44PM +0000, Vincenzo Frascino wrote:
> On 3/18/21 5:07 PM, Will Deacon wrote:
> > There's no need to allocate the compat vDSO and signal pages using
> > GFP_ATOMIC allocations, so use GFP_KERNEL instead.
> > 
> > Signed-off-by: Will Deacon <will@kernel.org>
> 
> Nit: We could explain why GFP_KERNEL is sufficient in the commit message.

Usually it's the other way around, document why GFP_ATOMIC is needed ;).

-- 
Catalin

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel