FAILED: patch "[PATCH] arm64: errata: Add Cortex-A725 erratum 3821522 workaround" failed to apply to 6.1-stable tree

FAILED: patch "[PATCH] arm64: errata: Add Cortex-A725 erratum 3821522 workaround" failed to apply to 6.1-stable tree

[gregkh]

The patch below does not apply to the 6.1-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 <stable@vger.kernel.org>.

To reproduce the conflict and resubmit, you may use the following commands:

git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-6.1.y
git checkout FETCH_HEAD
git cherry-pick -x 45f7304679875eb000d67e194090ee7abd05c89c
# <resolve conflicts, build, test, etc.>
git commit -s
git send-email --to '<stable@vger.kernel.org>' --in-reply-to '2026100502-outtakes-dragonfly-ead9@gregkh' --subject-prefix 'PATCH 6.1.y' 'HEAD^..'

Possible dependencies:



thanks,

greg k-h

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

From 45f7304679875eb000d67e194090ee7abd05c89c Mon Sep 17 00:00:00 2001
From: Beata Michalska <beata.michalska@arm.com>
Date: Thu, 1 Oct 2026 11:00:32 +0200
Subject: [PATCH] arm64: errata: Add Cortex-A725 erratum 3821522 workaround

Cortex-A725 erratum 3821522 affects the CNT_CYCLES event, which can
incur a significant increment error when a CPU enters and subsequently
exits WFE or WFI, and may no longer track the system counter
frequency.

The AMEVCNTR01_EL0 counter is used as the AMU constant counter for
frequency invariance and CPPC FFH feedback counters. Wire the affected
Cortex-A725 range into the shared broken AMU constant-counter capability
so the affected counter is treated as unavailable by returning zero in
the AMU counter paths. This prevents the broken counter from being used
as a reference source.

The erratum can also affect PMUv3 users of the CNT_CYCLES event,
but this workaround intentionally does not change PMU event handling.
Hiding or rejecting the PMU event from the erratum code would change
the perf-visible PMU event interface, including raw event selection,
and would need a separate PMU-specific approach rather than being
folded into the AMU reference-counter workaround.

Cc: stable@vger.kernel.org
Signed-off-by: Beata Michalska <beata.michalska@arm.com>
Reviewed-by: Vladimir Murzin <vladimir.murzin@arm.com>
Signed-off-by: Will Deacon <will@kernel.org>

diff --git a/Documentation/arch/arm64/silicon-errata.rst b/Documentation/arch/arm64/silicon-errata.rst
index ac3248b9f2f3..99a1eb4b833d 100644
--- a/Documentation/arch/arm64/silicon-errata.rst
+++ b/Documentation/arch/arm64/silicon-errata.rst
@@ -174,6 +174,8 @@ stable kernels.
 +----------------+-----------------+-----------------+-----------------------------+
 | ARM            | Cortex-A725     | #3456106        | ARM64_ERRATUM_3194386       |
 +----------------+-----------------+-----------------+-----------------------------+
+| ARM            | Cortex-A725     | #3821522        | ARM64_ERRATUM_3821522       |
++----------------+-----------------+-----------------+-----------------------------+
 | ARM            | Cortex-X1       | #1502854        | N/A                         |
 +----------------+-----------------+-----------------+-----------------------------+
 | ARM            | Cortex-X1       | #3324344        | ARM64_ERRATUM_3194386       |
diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
index 3b6c052d46a1..cab741a695f5 100644
--- a/arch/arm64/Kconfig
+++ b/arch/arm64/Kconfig
@@ -1081,6 +1081,28 @@ config ARM64_ERRATUM_2645198
 
 	  If unsure, say Y.
 
+config ARM64_ERRATUM_3821522
+	bool "Cortex-A725: 3821522: workaround for possible CNT_CYCLES increment error due to WFE/WFI"
+	depends on ARM64_AMU_EXTN
+	select ARM64_WORKAROUND_BROKEN_AMU_CONSTCNT
+	default y
+	help
+	  This option adds the workaround for ARM Cortex-A725 erratum 3821522.
+
+	  On affected A725 cores, the CNT_CYCLES event may incur a significant
+	  increment error when entering and subsequently exiting WFx.
+	  As a result, the CNT_CYCLES may diverge from the system counter
+	  frequency at which it is expected to increment.
+	  This renders the AMU counter AMEVCNTR01, that implements CNT_CYCLES,
+	  being unreliable and unsuitable for use.
+
+	  Since there is no hardware workaround, reads of the affected CNT_CYCLES
+	  counter return 0 in the relevant paths. This causes users of the counter
+	  to treat it as unavailable and is functionally equivalent to firmware
+	  disabling the affected counter.
+
+	  If unsure, say Y.
+
 config ARM64_WORKAROUND_SPECULATIVE_UNPRIV_LOAD
 	bool
 
diff --git a/arch/arm64/kernel/cpu_errata.c b/arch/arm64/kernel/cpu_errata.c
index dbd70728a9bd..e0c09402540c 100644
--- a/arch/arm64/kernel/cpu_errata.c
+++ b/arch/arm64/kernel/cpu_errata.c
@@ -389,6 +389,10 @@ static const struct midr_range workaround_amu_constcnt_list[] = {
 #ifdef CONFIG_ARM64_ERRATUM_2457168
 	/* Cortex-A510 r0p0-r1p1 */
 	MIDR_RANGE(MIDR_CORTEX_A510, 0, 0, 1, 1),
+#endif
+#ifdef CONFIG_ARM64_ERRATUM_3821522
+	/* Cortex-A725 r0p0 - r0p2 */
+	MIDR_RANGE(MIDR_CORTEX_A725, 0, 0, 0, 2),
 #endif
 	{}
 };
diff --git a/arch/arm64/kernel/topology.c b/arch/arm64/kernel/topology.c
index 214d153cba6e..806373aac3db 100644
--- a/arch/arm64/kernel/topology.c
+++ b/arch/arm64/kernel/topology.c
@@ -395,10 +395,11 @@ static void cpu_read_corecnt(void *val)
 static void cpu_read_constcnt(void *val)
 {
 	/*
-	 * Return 0 if the current CPU is affected by erratum 2457168. A value
-	 * of 0 is also returned if the current CPU does not support AMUs or if
-	 * the counter is disabled. A return value of 0 at counter read is
-	 * properly handled as an error case by the users of the counter.
+	 * Return 0 if the current CPU is affected by a HW erratum.
+	 * A value of 0 is also returned if the current CPU does not
+	 * support AMUs or if the counter is disabled. A return
+	 * value of 0 at counter read is properly handled as an error
+	 * case by the users of the counter.
 	 */
 	*(u64 *)val = this_cpu_has_cap(ARM64_WORKAROUND_BROKEN_AMU_CONSTCNT) ?
 		      0UL : read_constcnt();

[PATCH 6.1.y] arm64: errata: Add Cortex-A725 erratum 3821522 workaround

[Sasha Levin]

From: Beata Michalska <beata.michalska@arm.com>

[ Upstream commit 45f7304679875eb000d67e194090ee7abd05c89c ]

Cortex-A725 erratum 3821522 affects the CNT_CYCLES event, which can
incur a significant increment error when a CPU enters and subsequently
exits WFE or WFI, and may no longer track the system counter
frequency.

The AMEVCNTR01_EL0 counter is used as the AMU constant counter for
frequency invariance and CPPC FFH feedback counters. Wire the affected
Cortex-A725 range into the shared broken AMU constant-counter capability
so the affected counter is treated as unavailable by returning zero in
the AMU counter paths. This prevents the broken counter from being used
as a reference source.

The erratum can also affect PMUv3 users of the CNT_CYCLES event,
but this workaround intentionally does not change PMU event handling.
Hiding or rejecting the PMU event from the erratum code would change
the perf-visible PMU event interface, including raw event selection,
and would need a separate PMU-specific approach rather than being
folded into the AMU reference-counter workaround.

Cc: stable@vger.kernel.org
Signed-off-by: Beata Michalska <beata.michalska@arm.com>
Reviewed-by: Vladimir Murzin <vladimir.murzin@arm.com>
Signed-off-by: Will Deacon <will@kernel.org>
[ sashal: Reused ARM64_WORKAROUND_2457168 in place of the unavailable generic AMU capability. ]
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 Documentation/arm64/silicon-errata.rst |  2 ++
 arch/arm64/Kconfig                     | 21 +++++++++++++++++++++
 arch/arm64/kernel/cpu_errata.c         | 24 +++++++++++++++++++-----
 arch/arm64/kernel/topology.c           |  9 +++++----
 4 files changed, 47 insertions(+), 9 deletions(-)

diff --git a/Documentation/arm64/silicon-errata.rst b/Documentation/arm64/silicon-errata.rst
index fe7b90ac52d7b..74ae9e4f637a7 100644
--- a/Documentation/arm64/silicon-errata.rst
+++ b/Documentation/arm64/silicon-errata.rst
@@ -155,6 +155,8 @@ stable kernels.
 +----------------+-----------------+-----------------+-----------------------------+
 | ARM            | Cortex-A725     | #3456106        | ARM64_ERRATUM_3194386       |
 +----------------+-----------------+-----------------+-----------------------------+
+| ARM            | Cortex-A725     | #3821522        | ARM64_ERRATUM_3821522       |
++----------------+-----------------+-----------------+-----------------------------+
 | ARM            | Cortex-X1       | #3324344        | ARM64_ERRATUM_3194386       |
 +----------------+-----------------+-----------------+-----------------------------+
 | ARM            | Cortex-X1       | #4193791        | ARM64_ERRATUM_4118414       |
diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
index 9cb8867eb1205..314a9ac22be54 100644
--- a/arch/arm64/Kconfig
+++ b/arch/arm64/Kconfig
@@ -983,6 +983,27 @@ config ARM64_ERRATUM_2457168
 
 	  If unsure, say Y.
 
+config ARM64_ERRATUM_3821522
+	bool "Cortex-A725: 3821522: workaround for possible CNT_CYCLES increment error due to WFE/WFI"
+	depends on ARM64_AMU_EXTN
+	default y
+	help
+	  This option adds the workaround for ARM Cortex-A725 erratum 3821522.
+
+	  On affected A725 cores, the CNT_CYCLES event may incur a significant
+	  increment error when entering and subsequently exiting WFx.
+	  As a result, the CNT_CYCLES may diverge from the system counter
+	  frequency at which it is expected to increment.
+	  This renders the AMU counter AMEVCNTR01, that implements CNT_CYCLES,
+	  being unreliable and unsuitable for use.
+
+	  Since there is no hardware workaround, reads of the affected CNT_CYCLES
+	  counter return 0 in the relevant paths. This causes users of the counter
+	  to treat it as unavailable and is functionally equivalent to firmware
+	  disabling the affected counter.
+
+	  If unsure, say Y.
+
 config ARM64_WORKAROUND_SPECULATIVE_UNPRIV_LOAD
 	bool
 
diff --git a/arch/arm64/kernel/cpu_errata.c b/arch/arm64/kernel/cpu_errata.c
index cff85a371c353..fbc80f64f3eeb 100644
--- a/arch/arm64/kernel/cpu_errata.c
+++ b/arch/arm64/kernel/cpu_errata.c
@@ -275,6 +275,21 @@ static const struct arm64_cpu_capabilities arm64_repeat_tlbi_list[] = {
 };
 #endif
 
+#if defined(CONFIG_ARM64_ERRATUM_2457168) || \
+	defined(CONFIG_ARM64_ERRATUM_3821522)
+static const struct midr_range workaround_amu_constcnt_list[] = {
+#ifdef CONFIG_ARM64_ERRATUM_2457168
+	/* Cortex-A510 r0p0-r1p1 */
+	MIDR_RANGE(MIDR_CORTEX_A510, 0, 0, 1, 1),
+#endif
+#ifdef CONFIG_ARM64_ERRATUM_3821522
+	/* Cortex-A725 r0p0 - r0p2 */
+	MIDR_RANGE(MIDR_CORTEX_A725, 0, 0, 0, 2),
+#endif
+	{}
+};
+#endif
+
 #ifdef CONFIG_CAVIUM_ERRATUM_23154
 static const struct midr_range cavium_erratum_23154_cpus[] = {
 	MIDR_ALL_VERSIONS(MIDR_THUNDERX),
@@ -737,14 +752,13 @@ const struct arm64_cpu_capabilities arm64_errata[] = {
 		ERRATA_MIDR_REV_RANGE(MIDR_CORTEX_A510, 0, 0, 2)
 	},
 #endif
-#ifdef CONFIG_ARM64_ERRATUM_2457168
+#if defined(CONFIG_ARM64_ERRATUM_2457168) || \
+	defined(CONFIG_ARM64_ERRATUM_3821522)
 	{
-		.desc = "ARM erratum 2457168",
+		.desc = "Broken AMU AMEVCNTR01 (const counter)",
 		.capability = ARM64_WORKAROUND_2457168,
 		.type = ARM64_CPUCAP_WEAK_LOCAL_CPU_FEATURE,
-
-		/* Cortex-A510 r0p0-r1p1 */
-		CAP_MIDR_RANGE(MIDR_CORTEX_A510, 0, 0, 1, 1)
+		CAP_MIDR_RANGE_LIST(workaround_amu_constcnt_list)
 	},
 #endif
 #ifdef CONFIG_ARM64_ERRATUM_2038923
diff --git a/arch/arm64/kernel/topology.c b/arch/arm64/kernel/topology.c
index 817d788cd8666..575d0f9a76870 100644
--- a/arch/arm64/kernel/topology.c
+++ b/arch/arm64/kernel/topology.c
@@ -268,10 +268,11 @@ static void cpu_read_corecnt(void *val)
 static void cpu_read_constcnt(void *val)
 {
 	/*
-	 * Return 0 if the current CPU is affected by erratum 2457168. A value
-	 * of 0 is also returned if the current CPU does not support AMUs or if
-	 * the counter is disabled. A return value of 0 at counter read is
-	 * properly handled as an error case by the users of the counter.
+	 * Return 0 if the current CPU is affected by a HW erratum.
+	 * A value of 0 is also returned if the current CPU does not
+	 * support AMUs or if the counter is disabled. A return
+	 * value of 0 at counter read is properly handled as an error
+	 * case by the users of the counter.
 	 */
 	*(u64 *)val = this_cpu_has_cap(ARM64_WORKAROUND_2457168) ?
 		      0UL : read_constcnt();
-- 
2.53.0

Re: [PATCH 6.1.y] arm64: errata: Add Cortex-A725 erratum 3821522 workaround

[Beata Michalska]

This one is also missing/mixing things.
Do let me know if you want me to make the backport.

---
BR
Beata
On Tue, Oct 06, 2026 at 04:31:46AM -0400, Sasha Levin wrote:
> From: Beata Michalska <beata.michalska@arm.com>
> 
> [ Upstream commit 45f7304679875eb000d67e194090ee7abd05c89c ]
> 
> Cortex-A725 erratum 3821522 affects the CNT_CYCLES event, which can
> incur a significant increment error when a CPU enters and subsequently
> exits WFE or WFI, and may no longer track the system counter
> frequency.
> 
> The AMEVCNTR01_EL0 counter is used as the AMU constant counter for
> frequency invariance and CPPC FFH feedback counters. Wire the affected
> Cortex-A725 range into the shared broken AMU constant-counter capability
> so the affected counter is treated as unavailable by returning zero in
> the AMU counter paths. This prevents the broken counter from being used
> as a reference source.
> 
> The erratum can also affect PMUv3 users of the CNT_CYCLES event,
> but this workaround intentionally does not change PMU event handling.
> Hiding or rejecting the PMU event from the erratum code would change
> the perf-visible PMU event interface, including raw event selection,
> and would need a separate PMU-specific approach rather than being
> folded into the AMU reference-counter workaround.
> 
> Cc: stable@vger.kernel.org
> Signed-off-by: Beata Michalska <beata.michalska@arm.com>
> Reviewed-by: Vladimir Murzin <vladimir.murzin@arm.com>
> Signed-off-by: Will Deacon <will@kernel.org>
> [ sashal: Reused ARM64_WORKAROUND_2457168 in place of the unavailable generic AMU capability. ]
> Signed-off-by: Sasha Levin <sashal@kernel.org>
> ---
>  Documentation/arm64/silicon-errata.rst |  2 ++
>  arch/arm64/Kconfig                     | 21 +++++++++++++++++++++
>  arch/arm64/kernel/cpu_errata.c         | 24 +++++++++++++++++++-----
>  arch/arm64/kernel/topology.c           |  9 +++++----
>  4 files changed, 47 insertions(+), 9 deletions(-)
> 
> diff --git a/Documentation/arm64/silicon-errata.rst b/Documentation/arm64/silicon-errata.rst
> index fe7b90ac52d7b..74ae9e4f637a7 100644
> --- a/Documentation/arm64/silicon-errata.rst
> +++ b/Documentation/arm64/silicon-errata.rst
> @@ -155,6 +155,8 @@ stable kernels.
>  +----------------+-----------------+-----------------+-----------------------------+
>  | ARM            | Cortex-A725     | #3456106        | ARM64_ERRATUM_3194386       |
>  +----------------+-----------------+-----------------+-----------------------------+
> +| ARM            | Cortex-A725     | #3821522        | ARM64_ERRATUM_3821522       |
> ++----------------+-----------------+-----------------+-----------------------------+
>  | ARM            | Cortex-X1       | #3324344        | ARM64_ERRATUM_3194386       |
>  +----------------+-----------------+-----------------+-----------------------------+
>  | ARM            | Cortex-X1       | #4193791        | ARM64_ERRATUM_4118414       |
> diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
> index 9cb8867eb1205..314a9ac22be54 100644
> --- a/arch/arm64/Kconfig
> +++ b/arch/arm64/Kconfig
> @@ -983,6 +983,27 @@ config ARM64_ERRATUM_2457168
>  
>  	  If unsure, say Y.
>  
> +config ARM64_ERRATUM_3821522
> +	bool "Cortex-A725: 3821522: workaround for possible CNT_CYCLES increment error due to WFE/WFI"
> +	depends on ARM64_AMU_EXTN
> +	default y
> +	help
> +	  This option adds the workaround for ARM Cortex-A725 erratum 3821522.
> +
> +	  On affected A725 cores, the CNT_CYCLES event may incur a significant
> +	  increment error when entering and subsequently exiting WFx.
> +	  As a result, the CNT_CYCLES may diverge from the system counter
> +	  frequency at which it is expected to increment.
> +	  This renders the AMU counter AMEVCNTR01, that implements CNT_CYCLES,
> +	  being unreliable and unsuitable for use.
> +
> +	  Since there is no hardware workaround, reads of the affected CNT_CYCLES
> +	  counter return 0 in the relevant paths. This causes users of the counter
> +	  to treat it as unavailable and is functionally equivalent to firmware
> +	  disabling the affected counter.
> +
> +	  If unsure, say Y.
> +
>  config ARM64_WORKAROUND_SPECULATIVE_UNPRIV_LOAD
>  	bool
>  
> diff --git a/arch/arm64/kernel/cpu_errata.c b/arch/arm64/kernel/cpu_errata.c
> index cff85a371c353..fbc80f64f3eeb 100644
> --- a/arch/arm64/kernel/cpu_errata.c
> +++ b/arch/arm64/kernel/cpu_errata.c
> @@ -275,6 +275,21 @@ static const struct arm64_cpu_capabilities arm64_repeat_tlbi_list[] = {
>  };
>  #endif
>  
> +#if defined(CONFIG_ARM64_ERRATUM_2457168) || \
> +	defined(CONFIG_ARM64_ERRATUM_3821522)
> +static const struct midr_range workaround_amu_constcnt_list[] = {
> +#ifdef CONFIG_ARM64_ERRATUM_2457168
> +	/* Cortex-A510 r0p0-r1p1 */
> +	MIDR_RANGE(MIDR_CORTEX_A510, 0, 0, 1, 1),
> +#endif
> +#ifdef CONFIG_ARM64_ERRATUM_3821522
> +	/* Cortex-A725 r0p0 - r0p2 */
> +	MIDR_RANGE(MIDR_CORTEX_A725, 0, 0, 0, 2),
> +#endif
> +	{}
> +};
> +#endif
> +
>  #ifdef CONFIG_CAVIUM_ERRATUM_23154
>  static const struct midr_range cavium_erratum_23154_cpus[] = {
>  	MIDR_ALL_VERSIONS(MIDR_THUNDERX),
> @@ -737,14 +752,13 @@ const struct arm64_cpu_capabilities arm64_errata[] = {
>  		ERRATA_MIDR_REV_RANGE(MIDR_CORTEX_A510, 0, 0, 2)
>  	},
>  #endif
> -#ifdef CONFIG_ARM64_ERRATUM_2457168
> +#if defined(CONFIG_ARM64_ERRATUM_2457168) || \
> +	defined(CONFIG_ARM64_ERRATUM_3821522)
>  	{
> -		.desc = "ARM erratum 2457168",
> +		.desc = "Broken AMU AMEVCNTR01 (const counter)",
>  		.capability = ARM64_WORKAROUND_2457168,
>  		.type = ARM64_CPUCAP_WEAK_LOCAL_CPU_FEATURE,
> -
> -		/* Cortex-A510 r0p0-r1p1 */
> -		CAP_MIDR_RANGE(MIDR_CORTEX_A510, 0, 0, 1, 1)
> +		CAP_MIDR_RANGE_LIST(workaround_amu_constcnt_list)
>  	},
>  #endif
>  #ifdef CONFIG_ARM64_ERRATUM_2038923
> diff --git a/arch/arm64/kernel/topology.c b/arch/arm64/kernel/topology.c
> index 817d788cd8666..575d0f9a76870 100644
> --- a/arch/arm64/kernel/topology.c
> +++ b/arch/arm64/kernel/topology.c
> @@ -268,10 +268,11 @@ static void cpu_read_corecnt(void *val)
>  static void cpu_read_constcnt(void *val)
>  {
>  	/*
> -	 * Return 0 if the current CPU is affected by erratum 2457168. A value
> -	 * of 0 is also returned if the current CPU does not support AMUs or if
> -	 * the counter is disabled. A return value of 0 at counter read is
> -	 * properly handled as an error case by the users of the counter.
> +	 * Return 0 if the current CPU is affected by a HW erratum.
> +	 * A value of 0 is also returned if the current CPU does not
> +	 * support AMUs or if the counter is disabled. A return
> +	 * value of 0 at counter read is properly handled as an error
> +	 * case by the users of the counter.
>  	 */
>  	*(u64 *)val = this_cpu_has_cap(ARM64_WORKAROUND_2457168) ?
>  		      0UL : read_constcnt();
> -- 
> 2.53.0
> 

[PATCH 6.1.y v2 1/2] arm64: errata: Factor out broken AMU const counter cap

[Beata Michalska]

Move the workaround from the erratum 2457168-specific cpucap to a generic
broken AMU constant-counter one. This keeps the existing Cortex-A510
handling unchanged while allowing other errata with similar AMU constant
counter issue to share the capability bit and call sites.

Signed-off-by: Beata Michalska <beata.michalska@arm.com>
Reviewed-by: Vladimir Murzin <vladimir.murzin@arm.com>
Signed-off-by: Will Deacon <will@kernel.org>
(cherry picked from commit 573371caf7aea8582212da81dcead2c3f48fac4b)

Skipped comment updated on amu_read_core_const_ctrs, as that
functionality is not available here. Otherwise unmodified.
Signed-off-by: Beata Michalska <beata.michalska@arm.com>
---
 arch/arm64/Kconfig             |  8 ++++++++
 arch/arm64/kernel/cpu_errata.c | 20 ++++++++++++++------
 arch/arm64/kernel/cpufeature.c |  2 +-
 arch/arm64/kernel/topology.c   |  2 +-
 arch/arm64/tools/cpucaps       |  2 +-
 5 files changed, 25 insertions(+), 9 deletions(-)

diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
index 9cb8867eb120..370bde2b2b07 100644
--- a/arch/arm64/Kconfig
+++ b/arch/arm64/Kconfig
@@ -966,9 +966,17 @@ config ARM64_ERRATUM_1902691
 
 	  If unsure, say Y.
 
+config ARM64_WORKAROUND_BROKEN_AMU_CONSTCNT
+	bool
+	# Targeting errata affecting AMEVCNTR01, the AMU constant counter.
+	# On affected CPUs the counter may not increment at the expected rate,
+	# making it unreliable. As a result, AMU users should treat it as
+	# unavailable rather than relying on misleading counter values.
+
 config ARM64_ERRATUM_2457168
 	bool "Cortex-A510: 2457168: workaround for AMEVCNTR01 incrementing incorrectly"
 	depends on ARM64_AMU_EXTN
+	select ARM64_WORKAROUND_BROKEN_AMU_CONSTCNT
 	default y
 	help
 	  This option adds the workaround for ARM Cortex-A510 erratum 2457168.
diff --git a/arch/arm64/kernel/cpu_errata.c b/arch/arm64/kernel/cpu_errata.c
index cff85a371c35..518e3b6877b5 100644
--- a/arch/arm64/kernel/cpu_errata.c
+++ b/arch/arm64/kernel/cpu_errata.c
@@ -275,6 +275,16 @@ static const struct arm64_cpu_capabilities arm64_repeat_tlbi_list[] = {
 };
 #endif
 
+#ifdef CONFIG_ARM64_WORKAROUND_BROKEN_AMU_CONSTCNT
+static const struct midr_range workaround_amu_constcnt_list[] = {
+#ifdef CONFIG_ARM64_ERRATUM_2457168
+	/* Cortex-A510 r0p0-r1p1 */
+	MIDR_RANGE(MIDR_CORTEX_A510, 0, 0, 1, 1),
+#endif
+	{}
+};
+#endif /* CONFIG_ARM64_WORKAROUND_BROKEN_AMU_CONSTCNT */
+
 #ifdef CONFIG_CAVIUM_ERRATUM_23154
 static const struct midr_range cavium_erratum_23154_cpus[] = {
 	MIDR_ALL_VERSIONS(MIDR_THUNDERX),
@@ -737,14 +747,12 @@ const struct arm64_cpu_capabilities arm64_errata[] = {
 		ERRATA_MIDR_REV_RANGE(MIDR_CORTEX_A510, 0, 0, 2)
 	},
 #endif
-#ifdef CONFIG_ARM64_ERRATUM_2457168
+#ifdef CONFIG_ARM64_WORKAROUND_BROKEN_AMU_CONSTCNT
 	{
-		.desc = "ARM erratum 2457168",
-		.capability = ARM64_WORKAROUND_2457168,
+		.desc = "Broken AMU AMEVCNTR01 (const counter)",
+		.capability = ARM64_WORKAROUND_BROKEN_AMU_CONSTCNT,
 		.type = ARM64_CPUCAP_WEAK_LOCAL_CPU_FEATURE,
-
-		/* Cortex-A510 r0p0-r1p1 */
-		CAP_MIDR_RANGE(MIDR_CORTEX_A510, 0, 0, 1, 1)
+		CAP_MIDR_RANGE_LIST(workaround_amu_constcnt_list)
 	},
 #endif
 #ifdef CONFIG_ARM64_ERRATUM_2038923
diff --git a/arch/arm64/kernel/cpufeature.c b/arch/arm64/kernel/cpufeature.c
index 547c47358133..467737615bf3 100644
--- a/arch/arm64/kernel/cpufeature.c
+++ b/arch/arm64/kernel/cpufeature.c
@@ -1907,7 +1907,7 @@ static void cpu_amu_enable(struct arm64_cpu_capabilities const *cap)
 		cpumask_set_cpu(smp_processor_id(), &amu_cpus);
 
 		/* 0 reference values signal broken/disabled counters */
-		if (!this_cpu_has_cap(ARM64_WORKAROUND_2457168))
+		if (!this_cpu_has_cap(ARM64_WORKAROUND_BROKEN_AMU_CONSTCNT))
 			update_freq_counters_refs();
 	}
 }
diff --git a/arch/arm64/kernel/topology.c b/arch/arm64/kernel/topology.c
index 817d788cd866..d61d1afdee15 100644
--- a/arch/arm64/kernel/topology.c
+++ b/arch/arm64/kernel/topology.c
@@ -273,7 +273,7 @@ static void cpu_read_constcnt(void *val)
 	 * the counter is disabled. A return value of 0 at counter read is
 	 * properly handled as an error case by the users of the counter.
 	 */
-	*(u64 *)val = this_cpu_has_cap(ARM64_WORKAROUND_2457168) ?
+	*(u64 *)val = this_cpu_has_cap(ARM64_WORKAROUND_BROKEN_AMU_CONSTCNT) ?
 		      0UL : read_constcnt();
 }
 
diff --git a/arch/arm64/tools/cpucaps b/arch/arm64/tools/cpucaps
index f5c9a4f5f722..f609e8d93937 100644
--- a/arch/arm64/tools/cpucaps
+++ b/arch/arm64/tools/cpucaps
@@ -69,7 +69,6 @@ WORKAROUND_1902691
 WORKAROUND_2038923
 WORKAROUND_2064142
 WORKAROUND_2077057
-WORKAROUND_2457168
 WORKAROUND_2658417
 WORKAROUND_AMPERE_AC03_CPU_38
 WORKAROUND_TRBE_OVERWRITE_FILL_MODE
@@ -88,3 +87,4 @@ WORKAROUND_REPEAT_TLBI
 WORKAROUND_SPECULATIVE_AT
 WORKAROUND_SPECULATIVE_SSBS
 WORKAROUND_SPECULATIVE_UNPRIV_LOAD
+WORKAROUND_BROKEN_AMU_CONSTCNT
-- 
2.43.0

[PATCH 6.1.y v2 2/2] arm64: errata: Add Cortex-A725 erratum 3821522 workaround

[Beata Michalska]

Cortex-A725 erratum 3821522 affects the CNT_CYCLES event, which can
incur a significant increment error when a CPU enters and subsequently
exits WFE or WFI, and may no longer track the system counter
frequency.

The AMEVCNTR01_EL0 counter is used as the AMU constant counter for
frequency invariance and CPPC FFH feedback counters. Wire the affected
Cortex-A725 range into the shared broken AMU constant-counter capability
so the affected counter is treated as unavailable by returning zero in
the AMU counter paths. This prevents the broken counter from being used
as a reference source.

The erratum can also affect PMUv3 users of the CNT_CYCLES event,
but this workaround intentionally does not change PMU event handling.
Hiding or rejecting the PMU event from the erratum code would change
the perf-visible PMU event interface, including raw event selection,
and would need a separate PMU-specific approach rather than being
folded into the AMU reference-counter workaround.

Cc: stable@vger.kernel.org
Signed-off-by: Beata Michalska <beata.michalska@arm.com>
Reviewed-by: Vladimir Murzin <vladimir.murzin@arm.com>
Signed-off-by: Will Deacon <will@kernel.org>
(cherry picked from commit 45f7304679875eb000d67e194090ee7abd05c89c)
Signed-off-by: Beata Michalska <beata.michalska@arm.com>
---
 Documentation/arm64/silicon-errata.rst |  2 ++
 arch/arm64/Kconfig                     | 22 ++++++++++++++++++++++
 arch/arm64/kernel/cpu_errata.c         |  4 ++++
 arch/arm64/kernel/topology.c           |  9 +++++----
 4 files changed, 33 insertions(+), 4 deletions(-)

diff --git a/Documentation/arm64/silicon-errata.rst b/Documentation/arm64/silicon-errata.rst
index fe7b90ac52d7..74ae9e4f637a 100644
--- a/Documentation/arm64/silicon-errata.rst
+++ b/Documentation/arm64/silicon-errata.rst
@@ -155,6 +155,8 @@ stable kernels.
 +----------------+-----------------+-----------------+-----------------------------+
 | ARM            | Cortex-A725     | #3456106        | ARM64_ERRATUM_3194386       |
 +----------------+-----------------+-----------------+-----------------------------+
+| ARM            | Cortex-A725     | #3821522        | ARM64_ERRATUM_3821522       |
++----------------+-----------------+-----------------+-----------------------------+
 | ARM            | Cortex-X1       | #3324344        | ARM64_ERRATUM_3194386       |
 +----------------+-----------------+-----------------+-----------------------------+
 | ARM            | Cortex-X1       | #4193791        | ARM64_ERRATUM_4118414       |
diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
index 370bde2b2b07..1b8ee28f2383 100644
--- a/arch/arm64/Kconfig
+++ b/arch/arm64/Kconfig
@@ -991,6 +991,28 @@ config ARM64_ERRATUM_2457168
 
 	  If unsure, say Y.
 
+config ARM64_ERRATUM_3821522
+	bool "Cortex-A725: 3821522: workaround for possible CNT_CYCLES increment error due to WFE/WFI"
+	depends on ARM64_AMU_EXTN
+	select ARM64_WORKAROUND_BROKEN_AMU_CONSTCNT
+	default y
+	help
+	  This option adds the workaround for ARM Cortex-A725 erratum 3821522.
+
+	  On affected A725 cores, the CNT_CYCLES event may incur a significant
+	  increment error when entering and subsequently exiting WFx.
+	  As a result, the CNT_CYCLES may diverge from the system counter
+	  frequency at which it is expected to increment.
+	  This renders the AMU counter AMEVCNTR01, that implements CNT_CYCLES,
+	  being unreliable and unsuitable for use.
+
+	  Since there is no hardware workaround, reads of the affected CNT_CYCLES
+	  counter return 0 in the relevant paths. This causes users of the counter
+	  to treat it as unavailable and is functionally equivalent to firmware
+	  disabling the affected counter.
+
+	  If unsure, say Y.
+
 config ARM64_WORKAROUND_SPECULATIVE_UNPRIV_LOAD
 	bool
 
diff --git a/arch/arm64/kernel/cpu_errata.c b/arch/arm64/kernel/cpu_errata.c
index 518e3b6877b5..660b3ff16ebc 100644
--- a/arch/arm64/kernel/cpu_errata.c
+++ b/arch/arm64/kernel/cpu_errata.c
@@ -280,6 +280,10 @@ static const struct midr_range workaround_amu_constcnt_list[] = {
 #ifdef CONFIG_ARM64_ERRATUM_2457168
 	/* Cortex-A510 r0p0-r1p1 */
 	MIDR_RANGE(MIDR_CORTEX_A510, 0, 0, 1, 1),
+#endif
+#ifdef CONFIG_ARM64_ERRATUM_3821522
+	/* Cortex-A725 r0p0 - r0p2 */
+	MIDR_RANGE(MIDR_CORTEX_A725, 0, 0, 0, 2),
 #endif
 	{}
 };
diff --git a/arch/arm64/kernel/topology.c b/arch/arm64/kernel/topology.c
index d61d1afdee15..61df0918585b 100644
--- a/arch/arm64/kernel/topology.c
+++ b/arch/arm64/kernel/topology.c
@@ -268,10 +268,11 @@ static void cpu_read_corecnt(void *val)
 static void cpu_read_constcnt(void *val)
 {
 	/*
-	 * Return 0 if the current CPU is affected by erratum 2457168. A value
-	 * of 0 is also returned if the current CPU does not support AMUs or if
-	 * the counter is disabled. A return value of 0 at counter read is
-	 * properly handled as an error case by the users of the counter.
+	 * Return 0 if the current CPU is affected by a HW erratum.
+	 * A value of 0 is also returned if the current CPU does not
+	 * support AMUs or if the counter is disabled. A return
+	 * value of 0 at counter read is properly handled as an error
+	 * case by the users of the counter.
 	 */
 	*(u64 *)val = this_cpu_has_cap(ARM64_WORKAROUND_BROKEN_AMU_CONSTCNT) ?
 		      0UL : read_constcnt();
-- 
2.43.0

Re: [PATCH 6.1.y] arm64: errata: Add Cortex-A725 erratum 3821522 workaround

[Beata Michalska]

On Tue, Oct 06, 2026 at 10:53:32AM +0200, Beata Michalska wrote:
> This one is also missing/mixing things.
> Do let me know if you want me to make the backport.
> 
Posted as a reply to:
https://lore.kernel.org/stable/2026100502-outtakes-dragonfly-ead9@gregkh/T/#t

---
BR
Beata
> ---
> BR
> Beata
> On Tue, Oct 06, 2026 at 04:31:46AM -0400, Sasha Levin wrote:
> > From: Beata Michalska <beata.michalska@arm.com>
> > 
> > [ Upstream commit 45f7304679875eb000d67e194090ee7abd05c89c ]
> > 
> > Cortex-A725 erratum 3821522 affects the CNT_CYCLES event, which can
> > incur a significant increment error when a CPU enters and subsequently
> > exits WFE or WFI, and may no longer track the system counter
> > frequency.
> > 
> > The AMEVCNTR01_EL0 counter is used as the AMU constant counter for
> > frequency invariance and CPPC FFH feedback counters. Wire the affected
> > Cortex-A725 range into the shared broken AMU constant-counter capability
> > so the affected counter is treated as unavailable by returning zero in
> > the AMU counter paths. This prevents the broken counter from being used
> > as a reference source.
> > 
> > The erratum can also affect PMUv3 users of the CNT_CYCLES event,
> > but this workaround intentionally does not change PMU event handling.
> > Hiding or rejecting the PMU event from the erratum code would change
> > the perf-visible PMU event interface, including raw event selection,
> > and would need a separate PMU-specific approach rather than being
> > folded into the AMU reference-counter workaround.
> > 
> > Cc: stable@vger.kernel.org
> > Signed-off-by: Beata Michalska <beata.michalska@arm.com>
> > Reviewed-by: Vladimir Murzin <vladimir.murzin@arm.com>
> > Signed-off-by: Will Deacon <will@kernel.org>
> > [ sashal: Reused ARM64_WORKAROUND_2457168 in place of the unavailable generic AMU capability. ]
> > Signed-off-by: Sasha Levin <sashal@kernel.org>
> > ---
> >  Documentation/arm64/silicon-errata.rst |  2 ++
> >  arch/arm64/Kconfig                     | 21 +++++++++++++++++++++
> >  arch/arm64/kernel/cpu_errata.c         | 24 +++++++++++++++++++-----
> >  arch/arm64/kernel/topology.c           |  9 +++++----
> >  4 files changed, 47 insertions(+), 9 deletions(-)
> > 
> > diff --git a/Documentation/arm64/silicon-errata.rst b/Documentation/arm64/silicon-errata.rst
> > index fe7b90ac52d7b..74ae9e4f637a7 100644
> > --- a/Documentation/arm64/silicon-errata.rst
> > +++ b/Documentation/arm64/silicon-errata.rst
> > @@ -155,6 +155,8 @@ stable kernels.
> >  +----------------+-----------------+-----------------+-----------------------------+
> >  | ARM            | Cortex-A725     | #3456106        | ARM64_ERRATUM_3194386       |
> >  +----------------+-----------------+-----------------+-----------------------------+
> > +| ARM            | Cortex-A725     | #3821522        | ARM64_ERRATUM_3821522       |
> > ++----------------+-----------------+-----------------+-----------------------------+
> >  | ARM            | Cortex-X1       | #3324344        | ARM64_ERRATUM_3194386       |
> >  +----------------+-----------------+-----------------+-----------------------------+
> >  | ARM            | Cortex-X1       | #4193791        | ARM64_ERRATUM_4118414       |
> > diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
> > index 9cb8867eb1205..314a9ac22be54 100644
> > --- a/arch/arm64/Kconfig
> > +++ b/arch/arm64/Kconfig
> > @@ -983,6 +983,27 @@ config ARM64_ERRATUM_2457168
> >  
> >  	  If unsure, say Y.
> >  
> > +config ARM64_ERRATUM_3821522
> > +	bool "Cortex-A725: 3821522: workaround for possible CNT_CYCLES increment error due to WFE/WFI"
> > +	depends on ARM64_AMU_EXTN
> > +	default y
> > +	help
> > +	  This option adds the workaround for ARM Cortex-A725 erratum 3821522.
> > +
> > +	  On affected A725 cores, the CNT_CYCLES event may incur a significant
> > +	  increment error when entering and subsequently exiting WFx.
> > +	  As a result, the CNT_CYCLES may diverge from the system counter
> > +	  frequency at which it is expected to increment.
> > +	  This renders the AMU counter AMEVCNTR01, that implements CNT_CYCLES,
> > +	  being unreliable and unsuitable for use.
> > +
> > +	  Since there is no hardware workaround, reads of the affected CNT_CYCLES
> > +	  counter return 0 in the relevant paths. This causes users of the counter
> > +	  to treat it as unavailable and is functionally equivalent to firmware
> > +	  disabling the affected counter.
> > +
> > +	  If unsure, say Y.
> > +
> >  config ARM64_WORKAROUND_SPECULATIVE_UNPRIV_LOAD
> >  	bool
> >  
> > diff --git a/arch/arm64/kernel/cpu_errata.c b/arch/arm64/kernel/cpu_errata.c
> > index cff85a371c353..fbc80f64f3eeb 100644
> > --- a/arch/arm64/kernel/cpu_errata.c
> > +++ b/arch/arm64/kernel/cpu_errata.c
> > @@ -275,6 +275,21 @@ static const struct arm64_cpu_capabilities arm64_repeat_tlbi_list[] = {
> >  };
> >  #endif
> >  
> > +#if defined(CONFIG_ARM64_ERRATUM_2457168) || \
> > +	defined(CONFIG_ARM64_ERRATUM_3821522)
> > +static const struct midr_range workaround_amu_constcnt_list[] = {
> > +#ifdef CONFIG_ARM64_ERRATUM_2457168
> > +	/* Cortex-A510 r0p0-r1p1 */
> > +	MIDR_RANGE(MIDR_CORTEX_A510, 0, 0, 1, 1),
> > +#endif
> > +#ifdef CONFIG_ARM64_ERRATUM_3821522
> > +	/* Cortex-A725 r0p0 - r0p2 */
> > +	MIDR_RANGE(MIDR_CORTEX_A725, 0, 0, 0, 2),
> > +#endif
> > +	{}
> > +};
> > +#endif
> > +
> >  #ifdef CONFIG_CAVIUM_ERRATUM_23154
> >  static const struct midr_range cavium_erratum_23154_cpus[] = {
> >  	MIDR_ALL_VERSIONS(MIDR_THUNDERX),
> > @@ -737,14 +752,13 @@ const struct arm64_cpu_capabilities arm64_errata[] = {
> >  		ERRATA_MIDR_REV_RANGE(MIDR_CORTEX_A510, 0, 0, 2)
> >  	},
> >  #endif
> > -#ifdef CONFIG_ARM64_ERRATUM_2457168
> > +#if defined(CONFIG_ARM64_ERRATUM_2457168) || \
> > +	defined(CONFIG_ARM64_ERRATUM_3821522)
> >  	{
> > -		.desc = "ARM erratum 2457168",
> > +		.desc = "Broken AMU AMEVCNTR01 (const counter)",
> >  		.capability = ARM64_WORKAROUND_2457168,
> >  		.type = ARM64_CPUCAP_WEAK_LOCAL_CPU_FEATURE,
> > -
> > -		/* Cortex-A510 r0p0-r1p1 */
> > -		CAP_MIDR_RANGE(MIDR_CORTEX_A510, 0, 0, 1, 1)
> > +		CAP_MIDR_RANGE_LIST(workaround_amu_constcnt_list)
> >  	},
> >  #endif
> >  #ifdef CONFIG_ARM64_ERRATUM_2038923
> > diff --git a/arch/arm64/kernel/topology.c b/arch/arm64/kernel/topology.c
> > index 817d788cd8666..575d0f9a76870 100644
> > --- a/arch/arm64/kernel/topology.c
> > +++ b/arch/arm64/kernel/topology.c
> > @@ -268,10 +268,11 @@ static void cpu_read_corecnt(void *val)
> >  static void cpu_read_constcnt(void *val)
> >  {
> >  	/*
> > -	 * Return 0 if the current CPU is affected by erratum 2457168. A value
> > -	 * of 0 is also returned if the current CPU does not support AMUs or if
> > -	 * the counter is disabled. A return value of 0 at counter read is
> > -	 * properly handled as an error case by the users of the counter.
> > +	 * Return 0 if the current CPU is affected by a HW erratum.
> > +	 * A value of 0 is also returned if the current CPU does not
> > +	 * support AMUs or if the counter is disabled. A return
> > +	 * value of 0 at counter read is properly handled as an error
> > +	 * case by the users of the counter.
> >  	 */
> >  	*(u64 *)val = this_cpu_has_cap(ARM64_WORKAROUND_2457168) ?
> >  		      0UL : read_constcnt();
> > -- 
> > 2.53.0
> > 

Re: [PATCH 6.1.y v2 1/2] arm64: errata: Factor out broken AMU const counter cap

[Sasha Levin]

> Move the workaround from the erratum 2457168-specific cpucap to a generic
> broken AMU constant-counter one.

Queued the series for 6.1, thanks.

-- 
Thanks,
Sasha