You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Set to 1 if an integrity or consistency check triggered by the LFSR timer or via CHECK_TRIGGER is pending.
18
ro
0x0
DAI_IDLE
Set to 1 if the DAI is idle and ready to accept commands.
17
ro
0x0
BUS_INTEG_ERROR
This bit is set to 1 if a fatal bus integrity fault is detected. This error triggers a fatal_bus_integ_error alert.
16
ro
0x0
KEY_DERIV_FSM_ERROR
Set to 1 if the key derivation FSM has reached an invalid state. This raises an fatal_check_error alert and is an unrecoverable error condition.
15
ro
0x0
SCRAMBLING_FSM_ERROR
Set to 1 if the scrambling datapath FSM has reached an invalid state. This raises an fatal_check_error alert and is an unrecoverable error condition.
14
ro
0x0
LFSR_FSM_ERROR
Set to 1 if the LFSR timer FSM has reached an invalid state. This raises an fatal_check_error alert and is an unrecoverable error condition.
13
ro
0x0
TIMEOUT_ERROR
Set to 1 if an integrity or consistency check times out. This raises an fatal_check_error alert and is an unrecoverable error condition.
12
ro
0x0
LCI_ERROR
Set to 1 if an error occurred in the LCI. If set to 1, SW should check the ERR_CODE register at the corresponding index.
11
ro
0x0
DAI_ERROR
Set to 1 if an error occurred in the DAI. If set to 1, SW should check the ERR_CODE register at the corresponding index.
10
ro
0x0
LIFE_CYCLE_ERROR
Set to 1 if an error occurred in this partition. If set to 1, SW should check the ERR_CODE register at the corresponding index.
9
ro
0x0
SECRET2_ERROR
Set to 1 if an error occurred in this partition. If set to 1, SW should check the ERR_CODE register at the corresponding index.
8
ro
0x0
SECRET1_ERROR
Set to 1 if an error occurred in this partition. If set to 1, SW should check the ERR_CODE register at the corresponding index.
7
ro
0x0
SECRET0_ERROR
Set to 1 if an error occurred in this partition. If set to 1, SW should check the ERR_CODE register at the corresponding index.
6
ro
0x0
HW_CFG1_ERROR
Set to 1 if an error occurred in this partition. If set to 1, SW should check the ERR_CODE register at the corresponding index.
5
ro
0x0
HW_CFG0_ERROR
Set to 1 if an error occurred in this partition. If set to 1, SW should check the ERR_CODE register at the corresponding index.
4
ro
0x0
ROT_CREATOR_AUTH_STATE_ERROR
Set to 1 if an error occurred in this partition. If set to 1, SW should check the ERR_CODE register at the corresponding index.
3
ro
0x0
ROT_CREATOR_AUTH_CODESIGN_ERROR
Set to 1 if an error occurred in this partition. If set to 1, SW should check the ERR_CODE register at the corresponding index.
2
ro
0x0
OWNER_SW_CFG_ERROR
Set to 1 if an error occurred in this partition. If set to 1, SW should check the ERR_CODE register at the corresponding index.
1
ro
0x0
CREATOR_SW_CFG_ERROR
Set to 1 if an error occurred in this partition. If set to 1, SW should check the ERR_CODE register at the corresponding index.
0
ro
0x0
VENDOR_TEST_ERROR
Set to 1 if an error occurred in this partition. If set to 1, SW should check the ERR_CODE register at the corresponding index.
ERR_CODE
This register holds information about error conditions that occurred in the agents
interacting with the OTP macro via the internal bus. The error codes should be checked
if the partitions, DAI or LCI flag an error in the STATUS register, or when an
INTR_STATE.otp_error has been triggered. Note that all errors trigger an otp_error
interrupt, and in addition some errors may trigger either an fatal_macro_error or an
fatal_check_error alert.
Returned if the OTP macro command was invalid or did not complete successfully due to a macro malfunction. This error should never occur during normal operation and is not recoverable. This error triggers an fatal_macro_error alert.
0x2
MACRO_ECC_CORR_ERROR
A correctable ECC error has occured during an OTP read operation. The corresponding controller automatically recovers from this error when issuing a new command.
0x3
MACRO_ECC_UNCORR_ERROR
An uncorrectable ECC error has occurred during an OTP read operation. This error should never occur during normal operation and is not recoverable. If this error is present this may be a sign that the device is malfunctioning. This error triggers an fatal_macro_error alert.
0x4
MACRO_WRITE_BLANK_ERROR
This error is returned if a programming operation attempted to clear a bit that has previously been programmed to 1. The corresponding controller automatically recovers from this error when issuing a new command. Note however that the affected OTP word may be left in an inconsistent state if this error occurs. This can cause several issues when the word is accessed again (either as part of a regular read operation, as part of the readout at boot, or as part of a background check). It is important that SW ensures that each word is only written once, since this can render the device useless.
0x5
ACCESS_ERROR
This error indicates that a locked memory region has been accessed. The corresponding controller automatically recovers from this error when issuing a new command.
0x6
CHECK_FAIL_ERROR
An ECC, integrity or consistency mismatch has been detected in the buffer registers. This error should never occur during normal operation and is not recoverable. This error triggers an fatal_check_error alert.
0x7
FSM_STATE_ERROR
The FSM of the corresponding controller has reached an invalid state, or the FSM has been moved into a terminal error state due to an escalation action via lc_escalate_en_i. This error should never occur during normal operation and is not recoverable. If this error is present, this is a sign that the device has fallen victim to an invasive attack. This error triggers an fatal_check_error alert.
DIRECT_ACCESS_REGWEN
Register write enable for all direct access interface registers.
This bit controls whether the DAI registers can be written.
Write 0 to it in order to clear the bit.
Note that the hardware also modulates this bit and sets it to 0 temporarily
during an OTP operation such that the corresponding address and data registers
cannot be modified while an operation is pending. The DAI_IDLE status bit
will also be set to 0 in such a case.
This is the address for the OTP word to be read or written through
the direct access interface. Note that the address is aligned to the access size
internally, hence bits 1:0 are ignored for 32bit accesses, and bits 2:0 are ignored
for 64bit accesses.
For the digest calculation command, set this register to the partition base offset.
DIRECT_ACCESS_WDATA
Write data for direct accesses.
Hardware automatically determines the access granule (32bit or 64bit) based on which
partition is being written to.
Writing 1 to this bit triggers a consistency check. SW should monitor STATUS.CHECK_PENDING
and wait until the check has been completed. If there are any errors, those will be flagged
in the STATUS and ERR_CODE registers, and via interrupts and alerts.
CHECK_TRIGGER . INTEGRITY
Writing 1 to this bit triggers an integrity check. SW should monitor STATUS.CHECK_PENDING
and wait until the check has been completed. If there are any errors, those will be flagged
in the STATUS and ERR_CODE registers, and via the interrupts and alerts.
Timeout value in cycles for the for the integrity and consistency checks. If an integrity or consistency
check does not complete within the timeout window, an error will be flagged in the STATUS register,
an otp_error interrupt will be raised, and an fatal_check_error alert will be sent out. The timeout should
be set to a large value to stay on the safe side. The maximum check time can be upper bounded by the
number of cycles it takes to readout, scramble and digest the entire OTP array. Since this amounts to
roughly 25k cycles, it is recommended to set this value to at least 100'000 cycles in order to stay on the
safe side. A value of zero disables the timeout mechanism (default).
INTEGRITY_CHECK_PERIOD
This value specifies the maximum period that can be generated pseudo-randomly.
Only applies to the HW_CFG* and SECRET* partitions once they are locked.
The pseudo-random period is generated using a 40bit LFSR internally, and this register defines
the bit mask to be applied to the LFSR output in order to limit its range. The value of this
register is left shifted by 8bits and the lower bits are set to 8'hFF in order to form the 40bit mask.
A recommended value is 0x3_FFFF, corresponding to a maximum period of ~2.8s at 24MHz.
A value of zero disables the timer (default). Note that a one-off check can always be triggered via
CHECK_TRIGGER.INTEGRITY.
CONSISTENCY_CHECK_PERIOD
This value specifies the maximum period that can be generated pseudo-randomly.
This applies to the LIFE_CYCLE partition and the HW_CFG* and SECRET* partitions once they are locked.
The pseudo-random period is generated using a 40bit LFSR internally, and this register defines
the bit mask to be applied to the LFSR output in order to limit its range. The value of this
register is left shifted by 8bits and the lower bits are set to 8'hFF in order to form the 40bit mask.
A recommended value is 0x3FF_FFFF, corresponding to a maximum period of ~716s at 24MHz.
A value of zero disables the timer (default). Note that a one-off check can always be triggered via
CHECK_TRIGGER.CONSISTENCY.
When cleared to 0, read access to the ROT_CREATOR_AUTH_STATE partition is locked. Write 0 to clear this bit.
VENDOR_TEST_DIGEST
Integrity digest for the VENDOR_TEST partition.
The integrity digest is 0 by default. Software must write this
digest value via the direct access interface in order to lock the partition.
After a reset, write access to the VENDOR_TEST partition is locked and
the digest becomes visible in this CSR.
Integrity digest for the CREATOR_SW_CFG partition.
The integrity digest is 0 by default. Software must write this
digest value via the direct access interface in order to lock the partition.
After a reset, write access to the CREATOR_SW_CFG partition is locked and
the digest becomes visible in this CSR.
Integrity digest for the OWNER_SW_CFG partition.
The integrity digest is 0 by default. Software must write this
digest value via the direct access interface in order to lock the partition.
After a reset, write access to the OWNER_SW_CFG partition is locked and
the digest becomes visible in this CSR.
Integrity digest for the ROT_CREATOR_AUTH_CODESIGN partition.
The integrity digest is 0 by default. Software must write this
digest value via the direct access interface in order to lock the partition.
After a reset, write access to the ROT_CREATOR_AUTH_CODESIGN partition is locked and
the digest becomes visible in this CSR.
Integrity digest for the ROT_CREATOR_AUTH_STATE partition.
The integrity digest is 0 by default. Software must write this
digest value via the direct access interface in order to lock the partition.
After a reset, write access to the ROT_CREATOR_AUTH_STATE partition is locked and
the digest becomes visible in this CSR.
Integrity digest for the HW_CFG0 partition.
The integrity digest is 0 by default. The digest calculation can be triggered via the DIRECT_ACCESS_CMD.
After a reset, the digest then becomes visible in this CSR, and the corresponding partition becomes write-locked.
Integrity digest for the HW_CFG1 partition.
The integrity digest is 0 by default. The digest calculation can be triggered via the DIRECT_ACCESS_CMD.
After a reset, the digest then becomes visible in this CSR, and the corresponding partition becomes write-locked.
Integrity digest for the SECRET0 partition.
The integrity digest is 0 by default. The digest calculation can be triggered via the DIRECT_ACCESS_CMD.
After a reset, the digest then becomes visible in this CSR, and the corresponding partition becomes write-locked.
Integrity digest for the SECRET1 partition.
The integrity digest is 0 by default. The digest calculation can be triggered via the DIRECT_ACCESS_CMD.
After a reset, the digest then becomes visible in this CSR, and the corresponding partition becomes write-locked.
Integrity digest for the SECRET2 partition.
The integrity digest is 0 by default. The digest calculation can be triggered via the DIRECT_ACCESS_CMD.
After a reset, the digest then becomes visible in this CSR, and the corresponding partition becomes write-locked.
Any read to this window directly maps to the corresponding offset in the creator and owner software
config partitions, and triggers an OTP readout of the bytes requested. Note that the transaction
will block until OTP readout has completed.