RM48L952: CRC calculates random numbers even if there is no data change in the data blocks. Just increased memory of one of the data blocks within range.

Hi Zubair,

Is it possible to attach the both the codes (Before increasing size and after increasing size of data blocks), so that i can debug the issue at my end.

Even you can do a private message to me.

--
Thanks & regards,
Jagadish.

Hello Jagdish,
I have already sent you the changes in private, but for larger audience I am posting it here as well. 
Before : APPCALRAM  (RW) : origin=0x0801C010 length=0x00002FF0
After    : APPCALRAM  (RW) : origin=0x0801C010 length=0x00003FF0.

It works, if we make following changes , possible cause maybe overlapping of data causing wrong CRC check

* @brief Virtual Sector Configuration */
const Fee_VirtualSectorConfigType Fee_VirtualSectorConfiguration[TI_FEE_NUMBER_OF_VIRTUAL_SECTORS] =
{

/* Virtual Sector 1 */
{
/* Virtual sector number */ 1U,
/* Bank */ 7U,
/* Start Sector */ (Fapi_FlashSectorType)0U,
/* End Sector */ (Fapi_FlashSectorType)1U                    // changed from 0
},
/* Virtual Sector 2 */
{
/* Virtual sector number */ 2U,
/* Bank */ 7U,
/* Start Sector */ (Fapi_FlashSectorType)2U,                 // changed from 1
/* End Sector */ (Fapi_FlashSectorType)3U                   // chnaged from 1
}
,

};


/* Block Configurations */
/* SourceId : HL_Fee_SourceId_2 */
/* DesignId : HL_FEE_DesignId_3, HL_FEE_DesignId_8 */
/* Requirements : HL_FEE_SR2, HL_FEE_SR3, HL_FEE_SR4, HL_FEE_SR6, HL_FEE_SR85, HL_FEE_SR86
HL_FEE_SR87, HL_FEE_SR88, HL_FEE_SR89, HL_FEE_SR90, HL_FEE_SR91 */
const Fee_BlockConfigType Fee_BlockConfiguration[TI_FEE_NUMBER_OF_BLOCKS] =
{
/* Block 1 */
{
/* Block number */ 1U,
#pragma diag_push
#pragma diag_suppress 70 /* Justification: Cast needed to get section size from the linker */
#pragma CHECK_MISRA("-11.3") /* Justification: Needed to access variables from Addresses.h */
/* Block size */ (uint16_t)((uint32_t)16384U),         // changed from &__FIRMWARE_CAL_SIZE to  16384U                                                                                      
#pragma CHECK_MISRA("11.3")
#pragma diag_pop
/* Block immediate data used */ TRUE,
/* Number of write cycles */ 0x8U,
/* Device Index */ 0x00000000U,
/* Number of DataSets */ 1U,
/* EEP number */ 0U
}

,
/* Block 2 */
{
/* Block number */ 2U,
#pragma diag_push
#pragma diag_suppress 70 /* Justification: Cast needed to get section size from the linker */
#pragma CHECK_MISRA("-11.3") /* Justification: Needed to access variables from Addresses.h */
/* Block size */ (uint16_t)((uint32_t)16384U),        // changed from &__FIRMWARE_CAL_SIZE to 16384U
#pragma CHECK_MISRA("11.3")
#pragma diag_pop
/* Block immediate data used */ TRUE,
/* Number of write cycles */ 0x8U,
/* Device Index */ 0x00000000U,
/* Number of DataSets */ 1U,
/* EEP number */ 0U
}

Hi Zubair,

Zubair Muzaffar said:

I have already sent you the changes in private, but for larger audience I am posting it here as well. 
Before : APPCALRAM  (RW) : origin=0x0801C010 length=0x00002FF0
After    : APPCALRAM  (RW) : origin=0x0801C010 length=0x00003FF0.

On your RAM side you are just increasing the memory of 0x1000 only that is 4KB. So is it really necessary to change the huge changes on FEE side?

I mean earlier you just configured each virtual sector to 16KB of one physical sector but now you increased that to the 32KB by combining two physical sectors into one virtual sector.

So, it is just a 4KB block increasing on RAM side, so why don't you keep same settings on FEE side and do the test.

--
Thanks & regards,
Jagadish.

Hi Jagadish,
Thanks for answering, we already did that, that was the first impression but it failed.

Hi Zubair,

I need your complete project or any other simplest project to debug this issue at my end. Otherwise it would be very difficult to find out root cause from above information.

--
Thanks & regards,
Jagadish.