ARM: stm32: Initialize TAMP_SMCR BKP..PROT fields on STM32MP15xx
Subsystems (3)
PATCH v3 1/1 ARM: stm32: Initialize TAMP_SMCR BKP..PROT fields on STM32MP15xx
In case of an OTP-CLOSED STM32MP15xx system, the CPU core 1 cannot be released from endless loop in BootROM only by populating TAMP BKPxR 4 and 5 with magic and branch address and sending SGI0 interrupt from core 0 to core 1 twice. TAMP_SMCR BKP..PROT fields must be initialized as well to release the core 1 from endless loop during the second SGI0 handling on core 1. Initialize TAMP_SMCR to protect the first 32 backup registers, the ones which contain the core 1 magic, branch address and boot information.
This requirement seems to be undocumented, therefore it was necessary to trace and analyze the STM32MP15xx BootROM using OpenOCD and objdump. Ultimately, it turns out that a certain BootROM function reads out the TAMP_SMCR register and tests whether the BKP..PROT fields are non-zero. If they are zero, the BootROM code again waits for SGI0 using WFI, else the execution moves forward until it reaches handoff to the TAMP BKPxR 5 branch address.
This fixes CPU core 1 release using U-Boot PSCI implementation on an OTP-CLOSED system, i.e. system with fuse 0 bit 6 set.
---
Cc: Igor Opaniuk <[email protected]>
Cc: Patrice Chotard <[email protected]>
Cc: Patrick Delaunay <[email protected]>
Cc: Simon Glass <[email protected]
Cc: Simon Glass <[email protected]>
Cc: Tom Rini <[email protected]>
Cc: [email protected]
Cc: [email protected]
---
V2: Fix up the BKPRWD/BKPWD mask typo
V3: - Update the allocation to 0x20/0x20
- Update commit message
- Add RB from Patrick
--- arch/arm/mach-stm32mp/stm32mp1/stm32mp15x.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+)
arch/arm/mach-stm32mp/stm32mp1/stm32mp15x.c +16 -0
@@ -14,6 +14,7 @@#include <asm/arch/sys_proto.h> #include <dm/device.h> #include <dm/uclass.h> +#include <linux/bitfield.h> /* RCC register */ #define RCC_TZCR (STM32_RCC_BASE + 0x00)
@@ -41,6 +42,9 @@#define TZC_REGION_ID_ACCESS0 (STM32_TZC_BASE + 0x114)#define TAMP_CR1 (STM32_TAMP_BASE + 0x00) +#define TAMP_SMCR (STM32_TAMP_BASE + 0x20) +#define TAMP_SMCR_BKPRWDPROT GENMASK(7, 0) +#define TAMP_SMCR_BKPWDPROT GENMASK(23, 16)#define PWR_CR1 (STM32_PWR_BASE + 0x00)#define PWR_MCUCR (STM32_PWR_BASE + 0x14)
@@ -136,6 +140,18 @@ static void security_init(void)*/ writel (0x0 ,TAMP_CR1 ); +/* +* TAMP: Configure non-zero secure protection settings. This is +* checked by BootROM function 35ac on OTP-CLOSED device during +* CPU core 1 release from endless loop. If secure protection +* fields are zero, the core 1 is not released from endless +* loop on second SGI0. +*/ +clrsetbits_le32 (TAMP_SMCR , +TAMP_SMCR_BKPRWDPROT |TAMP_SMCR_BKPWDPROT , +FIELD_PREP (TAMP_SMCR_BKPRWDPROT ,0x20 ) | +FIELD_PREP (TAMP_SMCR_BKPWDPROT ,0x20 )); +/* GPIOZ: deactivate the security */ writel (BIT (0 ),RCC_MP_AHB5ENSETR );writel (0x0 ,GPIOZ_SECCFGR );
Signature · 1 line
2.43.0
Quoted 71 lines · Marek Vasut
In case of an OTP-CLOSED STM32MP15xx system, the CPU core 1 cannot be released from endless loop in BootROM only by populating TAMP BKPxR 4 and 5 with magic and branch address and sending SGI0 interrupt from core 0 to core 1 twice.
TAMP_SMCRBKP..PROT fields must be initialized as well to release the core 1 from endless loop during the second SGI0 handling on core 1. InitializeTAMP_SMCRto protect the first 32 backup registers, the ones which contain the core 1 magic, branch address and boot information.This requirement seems to be undocumented, therefore it was necessary to trace and analyze the STM32MP15xx BootROM using OpenOCD and objdump. Ultimately, it turns out that a certain BootROM function reads out the
TAMP_SMCRregister and tests whether the BKP..PROT fields are non-zero. If they are zero, the BootROM code again waits for SGI0 using WFI, else the execution moves forward until it reaches handoff to the TAMP BKPxR 5 branch address.This fixes CPU core 1 release using U-Boot PSCI implementation on an OTP-CLOSED system, i.e. system with fuse 0 bit 6 set.
Reviewed-by Patrick Delaunay <[email protected]>Signed-off-by Marek Vasut <[email protected]>---
Cc: Igor Opaniuk <[email protected]>
Cc: Patrice Chotard <[email protected]>
Cc: Patrick Delaunay <[email protected]>
Cc: Simon Glass <[email protected]
Cc: Simon Glass <[email protected]>
Cc: Tom Rini <[email protected]>
Cc: [email protected]
Cc: [email protected]
---
V2: Fix up the BKPRWD/BKPWD mask typo
V3: - Update the allocation to 0x20/0x20
- Update commit message
- Add RB from Patrick--- arch/arm/mach-stm32mp/stm32mp1/stm32mp15x.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+)arch/arm/mach-stm32mp/stm32mp1/stm32mp15x.c +9
-0@@ -14,6 +14,7 @@#include <asm/arch/sys_proto.h> #include <dm/device.h> #include <dm/uclass.h> +#include <linux/bitfield.h> /* RCC register */ #define RCC_TZCR (STM32_RCC_BASE + 0x00)@@ -41,6 +42,9 @@#define TZC_REGION_ID_ACCESS0 (STM32_TZC_BASE + 0x114)#define TAMP_CR1 (STM32_TAMP_BASE + 0x00) +#define TAMP_SMCR (STM32_TAMP_BASE + 0x20) +#define TAMP_SMCR_BKPRWDPROT GENMASK(7, 0) +#define TAMP_SMCR_BKPWDPROT GENMASK(23, 16)#define PWR_CR1 (STM32_PWR_BASE + 0x00)#define PWR_MCUCR (STM32_PWR_BASE + 0x14)@@ -136,6 +140,18 @@ static void security_init(void)*/ writel (0x0 ,TAMP_CR1 ); +/* +* TAMP: Configure non-zero secure protection settings. This is +* checked by BootROM function 35ac on OTP-CLOSED device during +* CPU core 1 release from endless loop. If secure protection +* fields are zero, the core 1 is not released from endless
arch/arm/mach-stm32mp/stm32mp1/stm32mp15x.c static void security_init(void) · line 157
Quoted 10 lines
arch/arm/mach-stm32mp/stm32mp1/stm32mp15x.c +7
-0+* loop on second SGI0. +*/ +clrsetbits_le32 (TAMP_SMCR , +TAMP_SMCR_BKPRWDPROT |TAMP_SMCR_BKPWDPROT , +FIELD_PREP (TAMP_SMCR_BKPRWDPROT ,0x20 ) | +FIELD_PREP (TAMP_SMCR_BKPWDPROT ,0x20 )); +/* GPIOZ: deactivate the security */ writel (BIT (0 ),RCC_MP_AHB5ENSETR );writel (0x0 ,GPIOZ_SECCFGR );
Hi Marek
Thanks
Patrice
Quoted 86 lines · Patrice CHOTARD
Quoted 71 lines · Marek Vasut
In case of an OTP-CLOSED STM32MP15xx system, the CPU core 1 cannot be released from endless loop in BootROM only by populating TAMP BKPxR 4 and 5 with magic and branch address and sending SGI0 interrupt from core 0 to core 1 twice.
TAMP_SMCRBKP..PROT fields must be initialized as well to release the core 1 from endless loop during the second SGI0 handling on core 1. InitializeTAMP_SMCRto protect the first 32 backup registers, the ones which contain the core 1 magic, branch address and boot information.This requirement seems to be undocumented, therefore it was necessary to trace and analyze the STM32MP15xx BootROM using OpenOCD and objdump. Ultimately, it turns out that a certain BootROM function reads out the
TAMP_SMCRregister and tests whether the BKP..PROT fields are non-zero. If they are zero, the BootROM code again waits for SGI0 using WFI, else the execution moves forward until it reaches handoff to the TAMP BKPxR 5 branch address.This fixes CPU core 1 release using U-Boot PSCI implementation on an OTP-CLOSED system, i.e. system with fuse 0 bit 6 set.
Reviewed-by Patrick Delaunay <[email protected]>Signed-off-by Marek Vasut <[email protected]>---
Cc: Igor Opaniuk <[email protected]>
Cc: Patrice Chotard <[email protected]>
Cc: Patrick Delaunay <[email protected]>
Cc: Simon Glass <[email protected]
Cc: Simon Glass <[email protected]>
Cc: Tom Rini <[email protected]>
Cc: [email protected]
Cc: [email protected]
---
V2: Fix up the BKPRWD/BKPWD mask typo
V3: - Update the allocation to 0x20/0x20
- Update commit message
- Add RB from Patrick--- arch/arm/mach-stm32mp/stm32mp1/stm32mp15x.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+)arch/arm/mach-stm32mp/stm32mp1/stm32mp15x.c +9
-0@@ -14,6 +14,7 @@#include <asm/arch/sys_proto.h> #include <dm/device.h> #include <dm/uclass.h> +#include <linux/bitfield.h> /* RCC register */ #define RCC_TZCR (STM32_RCC_BASE + 0x00)@@ -41,6 +42,9 @@#define TZC_REGION_ID_ACCESS0 (STM32_TZC_BASE + 0x114)#define TAMP_CR1 (STM32_TAMP_BASE + 0x00) +#define TAMP_SMCR (STM32_TAMP_BASE + 0x20) +#define TAMP_SMCR_BKPRWDPROT GENMASK(7, 0) +#define TAMP_SMCR_BKPWDPROT GENMASK(23, 16)#define PWR_CR1 (STM32_PWR_BASE + 0x00)#define PWR_MCUCR (STM32_PWR_BASE + 0x14)@@ -136,6 +140,18 @@ static void security_init(void)*/ writel (0x0 ,TAMP_CR1 ); +/* +* TAMP: Configure non-zero secure protection settings. This is +* checked by BootROM function 35ac on OTP-CLOSED device during +* CPU core 1 release from endless loop. If secure protection +* fields are zero, the core 1 is not released from endless arch/arm/mach-stm32mp/stm32mp1/stm32mp15x.c static void security_init(void) · line 157
Quoted 10 lines
arch/arm/mach-stm32mp/stm32mp1/stm32mp15x.c +7
-0+* loop on second SGI0. +*/ +clrsetbits_le32 (TAMP_SMCR , +TAMP_SMCR_BKPRWDPROT |TAMP_SMCR_BKPWDPROT , +FIELD_PREP (TAMP_SMCR_BKPRWDPROT ,0x20 ) | +FIELD_PREP (TAMP_SMCR_BKPWDPROT ,0x20 )); +/* GPIOZ: deactivate the security */ writel (BIT (0 ),RCC_MP_AHB5ENSETR );writel (0x0 ,GPIOZ_SECCFGR );Hi Marek
Quoted 1 line
Reviewed-by Patrice Chotard <[email protected]>
Applied on u-boot-stm32/master
Quoted 7 lines
Thanks
Patrice
_______________________________________________
Uboot-stm32 mailing list
[email protected]
https://st-md-mailman.stormreply.com/mailman/listinfo/uboot-stm32