IMX8SXL Tamper pin voltage

[reickhoff]

Problem/Observation: 

Using the IMX8SXL. We are using the "TAMPER" pins as GPIO. The idea is that they are used in a way that all 8 pins are pulled up internally. Then there are certain pins that have a 0 ohm pulldown so that the software can read these pins and understand what Hardware it is running on.

The very odd thing that was not expected, is that when the voltage is measured on these lines, we see 2.5V. Our expectation is that they would just be pulled up to the supply that is fed to AG17, which on our board is 1.8V. We have verified that our 1.8V supply is correct. We also do not have a 2.5V supply on the board.

Question:

Is there something special about how these pins are configured that is making these pulled up to 2.5V and not the 1.8V supply? We only see this on the tamper pins. Is there some internal regulator that is used that is not obvious in the documentation? We need to make sure that this is ok for our HW to be verified. 

Thanks

[JosephAtNXP]

Hi,

Was the voltage measurement performed in both IN and OUT pins?

How is the connection in SNVS_LDO_1P8_CAP made?

How did you configure these pins as General Purpose?

Thank you!

[reickhoff]

Was the voltage measurement performed in both IN and OUT pins?

 - Technically only the TAMPER_OUT pins because those pins are actually used by our software team meaning that they purposely set them as GPIO and enable the pullups. The TAMPER_IN pins are not actively used yet to my knowledge

How is the connection in SNVS_LDO_1P8_CAP made?

I measure 1.785V on the LDO output

How did you configure these pins as General Purpose?

- Waiting for feedback from the software team. Is there something specific you are looking for?

[JosephAtNXP]

Hi,

I see everything in the chip-side alright, I would like to see what the GPIOs are connected to, maybe the other circuitry has another pull up to 3.3V or the line has one.

And yes, I am looking in both hardware and software for meeting a document guidelines, could you verify it? It's an AppNote from a colleague attached.

I'll be waiting for the extra information to analyze it, but it is not desired, is another source or pull up that influences on GPIO.

Thank you for you attention and patience.

[snurb588]

From the software side we do the following in the SCFW code and M4 core application. The SCFW simply allocates the tamper pin resources to the M4 core and the M4 core configures the pins as GPIO on LSIO_GPIO2. We are using the auto-generated code from the MCUExpresso tool as well. Here is one of the pin configurations in our M4 pin_mux.c file generated by MCUExpresso:

err = sc_pad_set_all(ipc, SC_P_SNVS_TAMPER_OUT1, 4U, SC_PAD_CONFIG_NORMAL, SC_PAD_ISO_OFF, 0x20 ,SC_PAD_WAKEUP_OFF);/* IOMUXD_SNVS_TAMPER_DIG_OUT1 register modification value */

In the SCFW we don't do anything custom. We just updated the board_system_config function to allocate the pins to the M4:

/* List of resources */
static const sc_rsrc_t rsrc_list[] =
{
SC_R_SYSTEM,
SC_R_IRQSTR_M4_0,
SC_R_MU_5B,
SC_R_MU_8B,
SC_R_GPT_4,
RM_RANGE(SC_R_CAN_0, SC_R_CAN_1),
SC_R_SECO_MU_4,
// PARSHALL: add DMA3 to M4
RM_RANGE(SC_R_DMA_3_CH0, SC_R_DMA_3_CH15),
SC_R_ADC_0,
SC_R_FSPI_0,
RM_RANGE(SC_R_GPIO_1, SC_R_GPIO_3)
};

/* List of pads */
static const sc_pad_t pad_list[] =
{
RM_RANGE(SC_P_ADC_IN1, SC_P_ADC_IN5),
RM_RANGE(SC_P_FLEXCAN0_RX, SC_P_FLEXCAN1_TX),
RM_RANGE(SC_P_SPI0_SCK, SC_P_SPI0_CS0),
RM_RANGE(SC_P_SPI1_SCK, SC_P_SPI1_SDO),
RM_RANGE(SC_P_QSPI0A_DATA1, SC_P_QSPI0A_SCLK),
RM_RANGE(SC_P_QSPI0B_SCLK, SC_P_QSPI0B_DATA2),
RM_RANGE(SC_P_SNVS_TAMPER_OUT1 , SC_P_SNVS_TAMPER_OUT4),
RM_RANGE(SC_P_SNVS_TAMPER_IN0 , SC_P_SNVS_TAMPER_IN3),
};

I'm trying to use AN13365 example code to configure the pins in the SCFW instead of the M4, but the example provided doesn't give all the info you need (specifically what to set for your ipc handle) and wouldn't even build without me editing a few things, so it hasn't been the most helpful. Any guidance there would be helpful too.

[reickhoff]

they are either floating and nothing is attached (internal pullup makes it a 1) or they are shorted to ground with a 0ohm resistor (makes it a 0). So there is no other circuitry or any other voltage supplies.