[RFC V2 0/3] arm64/hw_breakpoint: Enable FEAT_Debugv8p9

[RFC V2 0/3] arm64/hw_breakpoint: Enable FEAT_Debugv8p9

[Anshuman Khandual]

This series enables FEAT_Debugv8p9 thus extending breakpoint and watchpoint
support upto 64. This has been lightly tested and still work is in progress
but would like to get some early feedback on the approach.

Possible impact of context switches while tracing kernel addresses needs to
be evaluated regarding MDSELR_EL1 access. This series is based on v6.10-rc4
after applying the MDSELR_EL1 KVM series.

https://lore.kernel.org/all/20240620065807.151540-1-anshuman.khandual@arm.com/

Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Marc Zyngier <maz@kernel.org>
Cc: Oliver Upton <oliver.upton@linux.dev>
Cc: James Morse <james.morse@arm.com>
Cc: Suzuki K Poulose <suzuki.poulose@arm.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Will Deacon <will@kernel.org>
Cc: Mark Brown <broonie@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: linux-arm-kernel@lists.infradead.org
Cc: kvmarm@lists.linux.dev
Cc: linux-kernel@vger.kernel.org

Changes in RFC V2:

- This series has been split from RFC V1 dealing only with arm64 breakpoints
- Restored back DBG_MDSCR_MASK definition (unrelated change)
- Added preempt_disable()/enable() blocks between selecting banks and registers

Changes in RFC:

https://lore.kernel.org/all/20240405080008.1225223-1-anshuman.khandual@arm.com/

Anshuman Khandual (3):
  arm64/cpufeature: Add field details for ID_AA64DFR1_EL1 register
  arm64/boot: Enable EL2 requirements for FEAT_Debugv8p9
  arm64/hw_breakpoint: Enable FEAT_Debugv8p9

 Documentation/arch/arm64/booting.rst    | 19 ++++++++++
 arch/arm64/include/asm/debug-monitors.h |  1 +
 arch/arm64/include/asm/el2_setup.h      | 27 +++++++++++++
 arch/arm64/include/asm/hw_breakpoint.h  | 50 ++++++++++++++++++++-----
 arch/arm64/include/asm/kvm_arm.h        |  1 +
 arch/arm64/kernel/cpufeature.c          | 21 ++++++++---
 arch/arm64/kernel/debug-monitors.c      | 16 ++++++--
 arch/arm64/kernel/hw_breakpoint.c       | 33 ++++++++++++++++
 8 files changed, 149 insertions(+), 19 deletions(-)

-- 
2.25.1

[RFC V2 1/3] arm64/cpufeature: Add field details for ID_AA64DFR1_EL1 register

[Anshuman Khandual]

This adds required field details for ID_AA64DFR1_EL1, and also drops dummy
ftr_raz[] array which is now redundant. These register fields will be used
to enable increased breakpoint and watchpoint registers via FEAT_Debugv8p9
later.

Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Will Deacon <will@kernel.org>
cc: Mark Brown <broonie@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Marc Zyngier <maz@kernel.org>
Cc: linux-arm-kernel@lists.infradead.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Anshuman Khandual <anshuman.khandual@arm.com>
---
 arch/arm64/kernel/cpufeature.c | 21 ++++++++++++++++-----
 1 file changed, 16 insertions(+), 5 deletions(-)

diff --git a/arch/arm64/kernel/cpufeature.c b/arch/arm64/kernel/cpufeature.c
index 48e7029f1054..12f0a5181bf2 100644
--- a/arch/arm64/kernel/cpufeature.c
+++ b/arch/arm64/kernel/cpufeature.c
@@ -527,6 +527,21 @@ static const struct arm64_ftr_bits ftr_id_aa64dfr0[] = {
 	ARM64_FTR_END,
 };
 
+static const struct arm64_ftr_bits ftr_id_aa64dfr1[] = {
+	ARM64_FTR_BITS(FTR_HIDDEN, FTR_STRICT, FTR_LOWER_SAFE, ID_AA64DFR1_EL1_ABL_CMPs_SHIFT, 8, 0),
+	ARM64_FTR_BITS(FTR_HIDDEN, FTR_STRICT, FTR_LOWER_SAFE, ID_AA64DFR1_EL1_DPFZS_SHIFT, 4, 0),
+	ARM64_FTR_BITS(FTR_HIDDEN, FTR_STRICT, FTR_LOWER_SAFE, ID_AA64DFR1_EL1_EBEP_SHIFT, 4, 0),
+	ARM64_FTR_BITS(FTR_HIDDEN, FTR_STRICT, FTR_LOWER_SAFE, ID_AA64DFR1_EL1_ITE_SHIFT, 4, 0),
+	ARM64_FTR_BITS(FTR_HIDDEN, FTR_STRICT, FTR_LOWER_SAFE, ID_AA64DFR1_EL1_ABLE_SHIFT, 4, 0),
+	ARM64_FTR_BITS(FTR_HIDDEN, FTR_STRICT, FTR_LOWER_SAFE, ID_AA64DFR1_EL1_PMICNTR_SHIFT, 4, 0),
+	ARM64_FTR_BITS(FTR_HIDDEN, FTR_STRICT, FTR_LOWER_SAFE, ID_AA64DFR1_EL1_SPMU_SHIFT, 4, 0),
+	ARM64_FTR_BITS(FTR_HIDDEN, FTR_STRICT, FTR_LOWER_SAFE, ID_AA64DFR1_EL1_CTX_CMPs_SHIFT, 8, 0),
+	ARM64_FTR_BITS(FTR_HIDDEN, FTR_STRICT, FTR_LOWER_SAFE, ID_AA64DFR1_EL1_WRPs_SHIFT, 8, 0),
+	ARM64_FTR_BITS(FTR_HIDDEN, FTR_STRICT, FTR_LOWER_SAFE, ID_AA64DFR1_EL1_BRPs_SHIFT, 8, 0),
+	ARM64_FTR_BITS(FTR_HIDDEN, FTR_STRICT, FTR_LOWER_SAFE, ID_AA64DFR1_EL1_SYSPMUID_SHIFT, 8, 0),
+	ARM64_FTR_END,
+};
+
 static const struct arm64_ftr_bits ftr_mvfr0[] = {
 	ARM64_FTR_BITS(FTR_HIDDEN, FTR_STRICT, FTR_LOWER_SAFE, MVFR0_EL1_FPRound_SHIFT, 4, 0),
 	ARM64_FTR_BITS(FTR_HIDDEN, FTR_STRICT, FTR_LOWER_SAFE, MVFR0_EL1_FPShVec_SHIFT, 4, 0),
@@ -705,10 +720,6 @@ static const struct arm64_ftr_bits ftr_single32[] = {
 	ARM64_FTR_END,
 };
 
-static const struct arm64_ftr_bits ftr_raz[] = {
-	ARM64_FTR_END,
-};
-
 #define __ARM64_FTR_REG_OVERRIDE(id_str, id, table, ovr) {	\
 		.sys_id = id,					\
 		.reg = 	&(struct arm64_ftr_reg){		\
@@ -781,7 +792,7 @@ static const struct __ftr_reg_entry {
 
 	/* Op1 = 0, CRn = 0, CRm = 5 */
 	ARM64_FTR_REG(SYS_ID_AA64DFR0_EL1, ftr_id_aa64dfr0),
-	ARM64_FTR_REG(SYS_ID_AA64DFR1_EL1, ftr_raz),
+	ARM64_FTR_REG(SYS_ID_AA64DFR1_EL1, ftr_id_aa64dfr1),
 
 	/* Op1 = 0, CRn = 0, CRm = 6 */
 	ARM64_FTR_REG(SYS_ID_AA64ISAR0_EL1, ftr_id_aa64isar0),
-- 
2.25.1

Re: [RFC V2 1/3] arm64/cpufeature: Add field details for ID_AA64DFR1_EL1 register

[Will Deacon]

On Thu, Jun 20, 2024 at 02:56:05PM +0530, Anshuman Khandual wrote:
> This adds required field details for ID_AA64DFR1_EL1, and also drops dummy
> ftr_raz[] array which is now redundant. These register fields will be used
> to enable increased breakpoint and watchpoint registers via FEAT_Debugv8p9
> later.
> 
> Cc: Catalin Marinas <catalin.marinas@arm.com>
> Cc: Will Deacon <will@kernel.org>
> cc: Mark Brown <broonie@kernel.org>
> Cc: Mark Rutland <mark.rutland@arm.com>
> Cc: Marc Zyngier <maz@kernel.org>
> Cc: linux-arm-kernel@lists.infradead.org
> Cc: linux-kernel@vger.kernel.org
> Signed-off-by: Anshuman Khandual <anshuman.khandual@arm.com>
> ---
>  arch/arm64/kernel/cpufeature.c | 21 ++++++++++++++++-----
>  1 file changed, 16 insertions(+), 5 deletions(-)
> 
> diff --git a/arch/arm64/kernel/cpufeature.c b/arch/arm64/kernel/cpufeature.c
> index 48e7029f1054..12f0a5181bf2 100644
> --- a/arch/arm64/kernel/cpufeature.c
> +++ b/arch/arm64/kernel/cpufeature.c
> @@ -527,6 +527,21 @@ static const struct arm64_ftr_bits ftr_id_aa64dfr0[] = {
>  	ARM64_FTR_END,
>  };
>  
> +static const struct arm64_ftr_bits ftr_id_aa64dfr1[] = {
> +	ARM64_FTR_BITS(FTR_HIDDEN, FTR_STRICT, FTR_LOWER_SAFE, ID_AA64DFR1_EL1_ABL_CMPs_SHIFT, 8, 0),
> +	ARM64_FTR_BITS(FTR_HIDDEN, FTR_STRICT, FTR_LOWER_SAFE, ID_AA64DFR1_EL1_DPFZS_SHIFT, 4, 0),

I only got this far in the patch, but why is this FTR_STRICT +
FTR_LOWER_SAFE? The behaviour of the cycle counter on an SPE management
event sounds like it would be fine to differ between cores, no?

Please go through all the new fields, bearing in mind that most of the
PMU stuff is per-CPU type rather than global.

Will

Re: [RFC V2 1/3] arm64/cpufeature: Add field details for ID_AA64DFR1_EL1 register

[Anshuman Khandual]

On 8/5/24 19:29, Will Deacon wrote:
> On Thu, Jun 20, 2024 at 02:56:05PM +0530, Anshuman Khandual wrote:
>> This adds required field details for ID_AA64DFR1_EL1, and also drops dummy
>> ftr_raz[] array which is now redundant. These register fields will be used
>> to enable increased breakpoint and watchpoint registers via FEAT_Debugv8p9
>> later.
>>
>> Cc: Catalin Marinas <catalin.marinas@arm.com>
>> Cc: Will Deacon <will@kernel.org>
>> cc: Mark Brown <broonie@kernel.org>
>> Cc: Mark Rutland <mark.rutland@arm.com>
>> Cc: Marc Zyngier <maz@kernel.org>
>> Cc: linux-arm-kernel@lists.infradead.org
>> Cc: linux-kernel@vger.kernel.org
>> Signed-off-by: Anshuman Khandual <anshuman.khandual@arm.com>
>> ---
>>  arch/arm64/kernel/cpufeature.c | 21 ++++++++++++++++-----
>>  1 file changed, 16 insertions(+), 5 deletions(-)
>>
>> diff --git a/arch/arm64/kernel/cpufeature.c b/arch/arm64/kernel/cpufeature.c
>> index 48e7029f1054..12f0a5181bf2 100644
>> --- a/arch/arm64/kernel/cpufeature.c
>> +++ b/arch/arm64/kernel/cpufeature.c
>> @@ -527,6 +527,21 @@ static const struct arm64_ftr_bits ftr_id_aa64dfr0[] = {
>>  	ARM64_FTR_END,
>>  };
>>  
>> +static const struct arm64_ftr_bits ftr_id_aa64dfr1[] = {
>> +	ARM64_FTR_BITS(FTR_HIDDEN, FTR_STRICT, FTR_LOWER_SAFE, ID_AA64DFR1_EL1_ABL_CMPs_SHIFT, 8, 0),
>> +	ARM64_FTR_BITS(FTR_HIDDEN, FTR_STRICT, FTR_LOWER_SAFE, ID_AA64DFR1_EL1_DPFZS_SHIFT, 4, 0),
> 
> I only got this far in the patch, but why is this FTR_STRICT +
> FTR_LOWER_SAFE? The behaviour of the cycle counter on an SPE management
> event sounds like it would be fine to differ between cores, no?

You are right I guess, will this change as FTR_NONSTRICT instead.

> 
> Please go through all the new fields, bearing in mind that most of the
> PMU stuff is per-CPU type rather than global.
Looking at all the register fields in ID_AA64DFR1_EL1 as in the latest
ARM ARM DDI 0487K.a.

Overall there are two different register field categories, the first is
whether an arch feature is enabled or not. This should be all the same
across different cpus on the system, hence FTR_STRICT + FTR_LOWER_SAFE
probably makes sense. The second type is number of HW counter or element
entries per CPU/PMU which can be different across cpus, hence
FTR_NONSTRICT + FTR_LOWER_SAFE might be applicable in such cases.

1. ABL_CMPs
	- Number of breakpoints that support address linking, minus 1

	[FTR_NONSTRICT + FTR_LOWER_SAFE]

2. DPFZS
	- Behavior of the cycle counter when event counting is frozen
	  by a Statistical Profiling management event

	[FTR_NONSTRICT + FTR_LOWER_SAFE]

3. EBEP
	- Exception-based event profiling

	[FTR_STRICT + FTR_LOWER_SAFE]

4. ITE
	- Instrumentation Trace Extension

	[FTR_STRICT + FTR_LOWER_SAFE]

5. ABLE
	- Address Breakpoint Linking Extension

	[FTR_STRICT + FTR_LOWER_SAFE]

6. PMICNTR
	- PMU fixed-function instruction counter
	- Could this vary between cpus ?

	[FTR_NONSTRICT + FTR_LOWER_SAFE]

7. SPMU
	- System PMU extension

	[FTR_STRICT + FTR_LOWER_SAFE]

8. CTX_CMPs
	- Context-aware breakpoints

	[FTR_NONSTRICT + FTR_LOWER_SAFE]

9. WRPs
	- Watchpoints

	[FTR_NONSTRICT + FTR_LOWER_SAFE]

10. BRPs
	- Breakpoints

	[FTR_NONSTRICT + FTR_LOWER_SAFE]

11. SYSPMUID
	- System PMU ID

	[FTR_STRICT + FTR_LOWER_SAFE]

Although please note that existing breakpoint/watchpoint numbers are represented
as FTR_STRICT + FTR_LOWER_SAFE. Hence just wondering if these extended watchpoint
or breakpoint numbers should represented any different.

static const struct arm64_ftr_bits ftr_id_aa64dfr0[] = {
..............
ARM64_FTR_BITS(FTR_HIDDEN, FTR_STRICT, FTR_LOWER_SAFE, ID_AA64DFR0_EL1_CTX_CMPs_SHIFT, 4, 0),
ARM64_FTR_BITS(FTR_HIDDEN, FTR_STRICT, FTR_LOWER_SAFE, ID_AA64DFR0_EL1_WRPs_SHIFT, 4, 0),
ARM64_FTR_BITS(FTR_HIDDEN, FTR_STRICT, FTR_LOWER_SAFE, ID_AA64DFR0_EL1_BRPs_SHIFT, 4, 0),
..............
};

Please suggest if any of the above needs change. Thank you.

- Anshuman

[RFC V2 2/3] arm64/boot: Enable EL2 requirements for FEAT_Debugv8p9

[Anshuman Khandual]

Fine grained trap control for MDSELR_EL1 register needs to be configured in
HDFGRTR2_EL2, and HDFGWTR2_EL2 registers when kernel enters at EL1, but EL2
is also present. This adds a new helper __init_el2_fgt2() initializing this
new FEAT_FGT2 based fine grained registers.

MDCR_EL2.EBWE needs to be enabled for additional (beyond 16) breakpoint and
watchpoint exceptions when kernel enters at EL1, but EL2 is also present.
This updates __init_el2_debug() as required for FEAT_Debugv8p9.

While here, also update booting.rst with MDCR_EL3 and SCR_EL3 requirements.

Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Will Deacon <will@kernel.org>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Marc Zyngier <maz@kernel.org>
Cc: Oliver Upton <oliver.upton@linux.dev>
Cc: linux-arm-kernel@lists.infradead.org
Cc: linux-doc@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Cc: kvmarm@lists.linux.dev
Signed-off-by: Anshuman Khandual <anshuman.khandual@arm.com>
---
 Documentation/arch/arm64/booting.rst | 19 +++++++++++++++++++
 arch/arm64/include/asm/el2_setup.h   | 27 +++++++++++++++++++++++++++
 arch/arm64/include/asm/kvm_arm.h     |  1 +
 3 files changed, 47 insertions(+)

diff --git a/Documentation/arch/arm64/booting.rst b/Documentation/arch/arm64/booting.rst
index b57776a68f15..e69d972018cf 100644
--- a/Documentation/arch/arm64/booting.rst
+++ b/Documentation/arch/arm64/booting.rst
@@ -285,6 +285,12 @@ Before jumping into the kernel, the following conditions must be met:
 
     - SCR_EL3.FGTEn (bit 27) must be initialised to 0b1.
 
+  For CPUs with the Fine Grained Traps (FEAT_FGT2) extension present:
+
+  - If EL3 is present and the kernel is entered at EL2:
+
+    - SCR_EL3.FGTEn2 (bit 59) must be initialised to 0b1.
+
   For CPUs with support for HCRX_EL2 (FEAT_HCX) present:
 
   - If EL3 is present and the kernel is entered at EL2:
@@ -319,6 +325,19 @@ Before jumping into the kernel, the following conditions must be met:
     - ZCR_EL2.LEN must be initialised to the same value for all CPUs the
       kernel will execute on.
 
+  For CPUs with FEAT_Debugv8p9 extension present:
+
+  - If the kernel is entered at EL1 and EL2 is present:
+
+    - HDFGRTR2_EL2.nMDSELR_EL1 (bit 5) must be initialized to 0b1
+    - HDFGWTR2_EL2.nMDSELR_EL1 (bit 5) must be initialized to 0b1
+    - MDCR_EL2.EBWE (bit 43) must be initialized to 0b1
+
+  - If EL3 is present:
+
+    - MDCR_EL3.TDA (bit 9) must be initialized to 0b0
+    - MDCR_EL3.EBWE (bit 43) must be initialized to 0b1
+
   For CPUs with the Scalable Matrix Extension (FEAT_SME):
 
   - If EL3 is present:
diff --git a/arch/arm64/include/asm/el2_setup.h b/arch/arm64/include/asm/el2_setup.h
index fd87c4b8f984..0cb26a9f34c3 100644
--- a/arch/arm64/include/asm/el2_setup.h
+++ b/arch/arm64/include/asm/el2_setup.h
@@ -97,6 +97,14 @@
 						// to own it.
 
 .Lskip_trace_\@:
+	mrs	x1, id_aa64dfr0_el1
+	ubfx	x1, x1, #ID_AA64DFR0_EL1_DebugVer_SHIFT, #4
+	cmp	x1, #ID_AA64DFR0_EL1_DebugVer_V8P9
+	b.lt	.Lskip_dbg_v8p9_\@
+
+	mov	x0, #MDCR_EL2_EBWE
+	orr	x2, x2, x0
+.Lskip_dbg_v8p9_\@:
 	msr	mdcr_el2, x2			// Configure debug traps
 .endm
 
@@ -204,6 +212,24 @@
 .Lskip_fgt_\@:
 .endm
 
+.macro __init_el2_fgt2
+	mrs	x1, id_aa64mmfr0_el1
+	ubfx	x1, x1, #ID_AA64MMFR0_EL1_FGT_SHIFT, #4
+	cmp	x1, #ID_AA64MMFR0_EL1_FGT_FGT2
+	b.lt	.Lskip_fgt2_\@
+
+	mrs	x1, id_aa64dfr0_el1
+	ubfx	x1, x1, #ID_AA64DFR0_EL1_DebugVer_SHIFT, #4
+	cmp	x1, #ID_AA64DFR0_EL1_DebugVer_V8P9
+	b.lt	.Lskip_dbg_v8p9_\@
+
+	mov_q   x0, HDFGWTR2_EL2_nMDSELR_EL1
+	msr_s	SYS_HDFGWTR2_EL2, x0
+	msr_s	SYS_HDFGRTR2_EL2, x0
+.Lskip_dbg_v8p9_\@:
+.Lskip_fgt2_\@:
+.endm
+
 .macro __init_el2_nvhe_prepare_eret
 	mov	x0, #INIT_PSTATE_EL1
 	msr	spsr_el2, x0
@@ -229,6 +255,7 @@
 	__init_el2_nvhe_idregs
 	__init_el2_cptr
 	__init_el2_fgt
+	__init_el2_fgt2
 .endm
 
 #ifndef __KVM_NVHE_HYPERVISOR__
diff --git a/arch/arm64/include/asm/kvm_arm.h b/arch/arm64/include/asm/kvm_arm.h
index b3fb368bcadb..8621a278ce25 100644
--- a/arch/arm64/include/asm/kvm_arm.h
+++ b/arch/arm64/include/asm/kvm_arm.h
@@ -312,6 +312,7 @@
 				 GENMASK(15, 0))
 
 /* Hyp Debug Configuration Register bits */
+#define MDCR_EL2_EBWE		(UL(1) << 43)
 #define MDCR_EL2_E2TB_MASK	(UL(0x3))
 #define MDCR_EL2_E2TB_SHIFT	(UL(24))
 #define MDCR_EL2_HPMFZS		(UL(1) << 36)
-- 
2.25.1

[RFC V2 3/3] arm64/hw_breakpoint: Enable FEAT_Debugv8p9

[Anshuman Khandual]

Currently there can be maximum 16 breakpoints, and 16 watchpoints available
on a given platform - as detected from ID_AA64DFR0_EL1.[BRPs|WRPs] register
fields. But these breakpoint, and watchpoints can be extended further up to
64 via a new arch feature FEAT_Debugv8p9.

This first enables banked access for the breakpoint and watchpoint register
set via MDSELR_EL1, extended exceptions via MDSCR_EL1.EMBWE and determining
available breakpoints and watchpoints in the platform from ID_AA64DFR1_EL1,
when FEAT_Debugv8p9 is enabled.

Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Will Deacon <will@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: linux-arm-kernel@lists.infradead.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Anshuman Khandual <anshuman.khandual@arm.com>
---
 arch/arm64/include/asm/debug-monitors.h |  1 +
 arch/arm64/include/asm/hw_breakpoint.h  | 50 ++++++++++++++++++++-----
 arch/arm64/kernel/debug-monitors.c      | 16 ++++++--
 arch/arm64/kernel/hw_breakpoint.c       | 33 ++++++++++++++++
 4 files changed, 86 insertions(+), 14 deletions(-)

diff --git a/arch/arm64/include/asm/debug-monitors.h b/arch/arm64/include/asm/debug-monitors.h
index 13d437bcbf58..a14097673ae0 100644
--- a/arch/arm64/include/asm/debug-monitors.h
+++ b/arch/arm64/include/asm/debug-monitors.h
@@ -20,6 +20,7 @@
 #define DBG_MDSCR_KDE		(1 << 13)
 #define DBG_MDSCR_MDE		(1 << 15)
 #define DBG_MDSCR_MASK		~(DBG_MDSCR_KDE | DBG_MDSCR_MDE)
+#define DBG_MDSCR_EMBWE		(1UL << 32)
 
 #define	DBG_ESR_EVT(x)		(((x) >> 27) & 0x7)
 
diff --git a/arch/arm64/include/asm/hw_breakpoint.h b/arch/arm64/include/asm/hw_breakpoint.h
index bd81cf17744a..362c4d4a64ac 100644
--- a/arch/arm64/include/asm/hw_breakpoint.h
+++ b/arch/arm64/include/asm/hw_breakpoint.h
@@ -79,8 +79,8 @@ static inline void decode_ctrl_reg(u32 reg,
  * Limits.
  * Changing these will require modifications to the register accessors.
  */
-#define ARM_MAX_BRP		16
-#define ARM_MAX_WRP		16
+#define ARM_MAX_BRP		64
+#define ARM_MAX_WRP		64
 
 /* Virtual debug register bases. */
 #define AARCH64_DBG_REG_BVR	0
@@ -94,13 +94,25 @@ static inline void decode_ctrl_reg(u32 reg,
 #define AARCH64_DBG_REG_NAME_WVR	wvr
 #define AARCH64_DBG_REG_NAME_WCR	wcr
 
+static inline bool is_debug_v8p9_enabled(void)
+{
+	u64 dfr0 = read_sanitised_ftr_reg(SYS_ID_AA64DFR0_EL1);
+	int dver = cpuid_feature_extract_unsigned_field(dfr0, ID_AA64DFR0_EL1_DebugVer_SHIFT);
+
+	return dver == ID_AA64DFR0_EL1_DebugVer_V8P9;
+}
+
 /* Accessor macros for the debug registers. */
 #define AARCH64_DBG_READ(N, REG, VAL) do {\
 	VAL = read_sysreg(dbg##REG##N##_el1);\
+	if (is_debug_v8p9_enabled())	\
+		preempt_enable();	\
 } while (0)
 
 #define AARCH64_DBG_WRITE(N, REG, VAL) do {\
 	write_sysreg(VAL, dbg##REG##N##_el1);\
+	if (is_debug_v8p9_enabled())	\
+		preempt_enable();	\
 } while (0)
 
 struct task_struct;
@@ -138,19 +150,37 @@ static inline void ptrace_hw_copy_thread(struct task_struct *task)
 /* Determine number of BRP registers available. */
 static inline int get_num_brps(void)
 {
-	u64 dfr0 = read_sanitised_ftr_reg(SYS_ID_AA64DFR0_EL1);
-	return 1 +
-		cpuid_feature_extract_unsigned_field(dfr0,
-						ID_AA64DFR0_EL1_BRPs_SHIFT);
+	u64 dfr0, dfr1;
+	int dver, brps;
+
+	dfr0 = read_sanitised_ftr_reg(SYS_ID_AA64DFR0_EL1);
+	dver = cpuid_feature_extract_unsigned_field(dfr0, ID_AA64DFR0_EL1_DebugVer_SHIFT);
+	if (dver == ID_AA64DFR0_EL1_DebugVer_V8P9) {
+		dfr1 = read_sanitised_ftr_reg(SYS_ID_AA64DFR1_EL1);
+		brps = cpuid_feature_extract_unsigned_field_width(dfr1,
+								  ID_AA64DFR1_EL1_BRPs_SHIFT, 8);
+	} else {
+		brps = cpuid_feature_extract_unsigned_field(dfr0, ID_AA64DFR0_EL1_BRPs_SHIFT);
+	}
+	return 1 + brps;
 }
 
 /* Determine number of WRP registers available. */
 static inline int get_num_wrps(void)
 {
-	u64 dfr0 = read_sanitised_ftr_reg(SYS_ID_AA64DFR0_EL1);
-	return 1 +
-		cpuid_feature_extract_unsigned_field(dfr0,
-						ID_AA64DFR0_EL1_WRPs_SHIFT);
+	u64 dfr0, dfr1;
+	int dver, wrps;
+
+	dfr0 = read_sanitised_ftr_reg(SYS_ID_AA64DFR0_EL1);
+	dver = cpuid_feature_extract_unsigned_field(dfr0, ID_AA64DFR0_EL1_DebugVer_SHIFT);
+	if (dver == ID_AA64DFR0_EL1_DebugVer_V8P9) {
+		dfr1 = read_sanitised_ftr_reg(SYS_ID_AA64DFR1_EL1);
+		wrps = cpuid_feature_extract_unsigned_field_width(dfr1,
+								  ID_AA64DFR1_EL1_WRPs_SHIFT, 8);
+	} else {
+		wrps = cpuid_feature_extract_unsigned_field(dfr0, ID_AA64DFR0_EL1_WRPs_SHIFT);
+	}
+	return 1 + wrps;
 }
 
 #ifdef CONFIG_CPU_PM
diff --git a/arch/arm64/kernel/debug-monitors.c b/arch/arm64/kernel/debug-monitors.c
index 64f2ecbdfe5c..3299d1e8dc61 100644
--- a/arch/arm64/kernel/debug-monitors.c
+++ b/arch/arm64/kernel/debug-monitors.c
@@ -23,6 +23,7 @@
 #include <asm/debug-monitors.h>
 #include <asm/system_misc.h>
 #include <asm/traps.h>
+#include <asm/hw_breakpoint.h>
 
 /* Determine debug architecture. */
 u8 debug_monitors_arch(void)
@@ -34,7 +35,7 @@ u8 debug_monitors_arch(void)
 /*
  * MDSCR access routines.
  */
-static void mdscr_write(u32 mdscr)
+static void mdscr_write(u64 mdscr)
 {
 	unsigned long flags;
 	flags = local_daif_save();
@@ -43,7 +44,7 @@ static void mdscr_write(u32 mdscr)
 }
 NOKPROBE_SYMBOL(mdscr_write);
 
-static u32 mdscr_read(void)
+static u64 mdscr_read(void)
 {
 	return read_sysreg(mdscr_el1);
 }
@@ -76,10 +77,11 @@ early_param("nodebugmon", early_debug_disable);
  */
 static DEFINE_PER_CPU(int, mde_ref_count);
 static DEFINE_PER_CPU(int, kde_ref_count);
+static DEFINE_PER_CPU(int, embwe_ref_count);
 
 void enable_debug_monitors(enum dbg_active_el el)
 {
-	u32 mdscr, enable = 0;
+	u64 mdscr, enable = 0;
 
 	WARN_ON(preemptible());
 
@@ -90,6 +92,9 @@ void enable_debug_monitors(enum dbg_active_el el)
 	    this_cpu_inc_return(kde_ref_count) == 1)
 		enable |= DBG_MDSCR_KDE;
 
+	if (is_debug_v8p9_enabled() && this_cpu_inc_return(embwe_ref_count) == 1)
+		enable |= DBG_MDSCR_EMBWE;
+
 	if (enable && debug_enabled) {
 		mdscr = mdscr_read();
 		mdscr |= enable;
@@ -100,7 +105,7 @@ NOKPROBE_SYMBOL(enable_debug_monitors);
 
 void disable_debug_monitors(enum dbg_active_el el)
 {
-	u32 mdscr, disable = 0;
+	u64 mdscr, disable = 0;
 
 	WARN_ON(preemptible());
 
@@ -111,6 +116,9 @@ void disable_debug_monitors(enum dbg_active_el el)
 	    this_cpu_dec_return(kde_ref_count) == 0)
 		disable &= ~DBG_MDSCR_KDE;
 
+	if (is_debug_v8p9_enabled() && this_cpu_dec_return(embwe_ref_count) == 0)
+		disable &= ~DBG_MDSCR_EMBWE;
+
 	if (disable) {
 		mdscr = mdscr_read();
 		mdscr &= disable;
diff --git a/arch/arm64/kernel/hw_breakpoint.c b/arch/arm64/kernel/hw_breakpoint.c
index 722ac45f9f7b..30156d732284 100644
--- a/arch/arm64/kernel/hw_breakpoint.c
+++ b/arch/arm64/kernel/hw_breakpoint.c
@@ -103,10 +103,40 @@ int hw_breakpoint_slots(int type)
 	WRITE_WB_REG_CASE(OFF, 14, REG, VAL);	\
 	WRITE_WB_REG_CASE(OFF, 15, REG, VAL)
 
+static int set_bank_index(int n)
+{
+	int mdsel_bank;
+	int bank = n / 16, index = n % 16;
+
+	switch (bank) {
+	case 0:
+		mdsel_bank = MDSELR_EL1_BANK_BANK_0;
+		break;
+	case 1:
+		mdsel_bank = MDSELR_EL1_BANK_BANK_1;
+		break;
+	case 2:
+		mdsel_bank = MDSELR_EL1_BANK_BANK_2;
+		break;
+	case 3:
+		mdsel_bank = MDSELR_EL1_BANK_BANK_3;
+		break;
+	default:
+		pr_warn("Unknown register bank %d\n", bank);
+	}
+	preempt_disable();
+	write_sysreg_s(mdsel_bank << MDSELR_EL1_BANK_SHIFT, SYS_MDSELR_EL1);
+	isb();
+	return index;
+}
+
 static u64 read_wb_reg(int reg, int n)
 {
 	u64 val = 0;
 
+	if (is_debug_v8p9_enabled())
+		n = set_bank_index(n);
+
 	switch (reg + n) {
 	GEN_READ_WB_REG_CASES(AARCH64_DBG_REG_BVR, AARCH64_DBG_REG_NAME_BVR, val);
 	GEN_READ_WB_REG_CASES(AARCH64_DBG_REG_BCR, AARCH64_DBG_REG_NAME_BCR, val);
@@ -122,6 +152,9 @@ NOKPROBE_SYMBOL(read_wb_reg);
 
 static void write_wb_reg(int reg, int n, u64 val)
 {
+	if (is_debug_v8p9_enabled())
+		n = set_bank_index(n);
+
 	switch (reg + n) {
 	GEN_WRITE_WB_REG_CASES(AARCH64_DBG_REG_BVR, AARCH64_DBG_REG_NAME_BVR, val);
 	GEN_WRITE_WB_REG_CASES(AARCH64_DBG_REG_BCR, AARCH64_DBG_REG_NAME_BCR, val);
-- 
2.25.1