LS1043AQDS QSPI CUSTOM BOARD - INVALID READ

cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

LS1043AQDS QSPI CUSTOM BOARD - INVALID READ

2,343 Views
carlos-m-ribeiro
Contributor I

Hello,

Now, we can booting the u-boot ls1043aqds from qspi used in our custom board but after booting, we can not read in 0x7fb00000 address (our cpld id register):

 

=> md.b 0x7fb00000

7fb00000:"Synchronous Abort" handler, esr 0x96000006

elr: 0000000040165d38 lr : 0000000040165c50 (reloc)

elr: 00000000fbda5d38 lr : 00000000fbda5c50

x0 : 0000000000000009 x1 : 000000000000003a

x2 : 0000000000000020 x3 : 000000007fb00000

x4 : 0000000000000000 x5 : 00000000fbdc2745

x6 : 0000000000000001 x7 : 000000000000000f

x8 : 00000000ffffffd0 x9 : 000000000000000c

x10: 0000000000000010 x11: 0000000000000006

x12: 000000000001869f x13: 0000000000002048

x14: 00000000fbc38a68 x15: 0000000000000008

x16: 0000000000000001 x17: 000000000000010f

x18: 00000000fbc3bdb8 x19: 0000000000000040

x20: 000000007fb00000 x21: 000000007fb00000

x22: 00000000fbdc144f x23: 0000000000000002

x24: 0000000000000010 x25: 0000000000000010

x26: 0000000000000001 x27: 00000000fbc38858

x28: 0000000000000000 x29: 00000000fbc387d0

 

The NXP forum suggested checks if IFC chip-select is configured for the address 0x7fb00000, but how we can do this?

 

The NXP forum also communicated us to provide this output:

 

=> md.l 1530000 40

01530000: 00000401 00000000 00000000 00000000 ................

01530010: 00000000 00000000 00000000 00000000 ................

01530020: 00000000 00000000 00000000 00000000 ................

01530030: 00000000 00000000 00000000 00000000 ................

01530040: 00000000 00000000 00000000 00000000 ................

01530050: 00000000 00000000 00000000 00000000 ................

01530060: 00000000 00000000 00000000 00000000 ................

01530070: 00000000 00000000 00000000 00000000 ................

01530080: 00000000 00000000 00000000 00000000 ................

01530090: 00000000 00000000 00000000 00000000 ................

015300a0: 00000000 00000000 00000000 00000000 ................

015300b0: 00000000 00000000 00000000 00000000 ................

015300c0: 00000000 00000000 00000000 00000000 ................

015300d0: 00000000 00000000 00000000 00000000 ................

015300e0: 00000000 00000000 00000000 00000000 ................

015300f0: 00000000 00000000 00000000 00000000 ................

Should we activate something in defconfig to can read from these addresses?

Can you take a look to our H board file and defconfig that have some changes made by us, we also verified these files before.

 

Best regards

0 Kudos
11 Replies

2,327 Views
Pavel
NXP Employee
NXP Employee

There is no problem for CPLD reading at address 0x7FB00000 on the LS1043ARDB board.

It looks like that CS is not configured for this address. Check your CS configuration for CPLD.

0 Kudos

2,322 Views
carlos-m-ribeiro
Contributor I

But the CS configuration for CPLD is related to u-boot or is an external hardware CPLD configuration?

 

Thank you

0 Kudos

2,318 Views
carlos-m-ribeiro
Contributor I

And we have the ifc mentioned in dts also as:

&ifc {
status = "okay";
#address-cells = <2>;
#size-cells = <1>;
/* NOR, NAND Flashes and FPGA on board */
ranges = <0x0 0x0 0x0 0x60000000 0x08000000
0x1 0x0 0x0 0x7e800000 0x00010000
0x2 0x0 0x0 0x7fb00000 0x00000100>;

nor@0,0 {
compatible = "cfi-flash";
#address-cells = <1>;
#size-cells = <1>;
reg = <0x0 0x0 0x8000000>;
bank-width = <2>;
device-width = <1>;
};

nand@1,0 {
compatible = "fsl,ifc-nand";
#address-cells = <1>;
#size-cells = <1>;
reg = <0x1 0x0 0x10000>;
};

cpld: board-control@2,0 {
compatible = "fsl,ls1043ardb-cpld";
reg = <0x2 0x0 0x0000100>;
};
};

 

What can be wrong?

 

Thank you

0 Kudos

2,342 Views
carlos-m-ribeiro
Contributor I

The files ...

0 Kudos

2,306 Views
Pavel
NXP Employee
NXP Employee

Is there problem under u-boot for CPLD using?

0 Kudos

2,302 Views
carlos-m-ribeiro
Contributor I

Good morning,

With some changes (in files), in our custom board with ls1043aqds QSPI u-boot we have one CPLD configured to 0x7fb00000 but we have successive reads (successive addresses) with the same values:

 
carlosmribeiro_0-1624956174718.jpeg

The expected behavior is (using u-boot in NAND flash):

carlosmribeiro_1-1624956174553.jpeg

What can be wrong?

 

Thank you

0 Kudos

2,274 Views
Pavel
NXP Employee
NXP Employee

Below is log from the LS1043ARDB board. Check reading all CPLD data.

=> md 7fb00000 20

7fb00000: 00020401 00000400 01010100 01000001    ................

7fb00010: fbfbfb00 fbfbfbfb fbfbfbfb fbfbfbfb    ................

7fb00020: 00020401 00000400 01010100 01000001    ................

7fb00030: fbfbfb00 fbfbfbfb fbfbfbfb fbfbfbfb    ................

7fb00040: 00020401 00000400 01010100 01000001    ................

7fb00050: fbfbfb00 fbfbfbfb fbfbfbfb fbfbfbfb    ................

7fb00060: 00020401 00000400 01010100 01000001    ................

7fb00070: fbfbfb00 fbfbfbfb fbfbfbfb fbfbfbfb    ................

=> md.b 7fb00000 80

7fb00000: 01 04 02 00 00 04 00 00 00 01 01 01 01 00 00 01    ................

7fb00010: 00 fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb    ................

7fb00020: 01 04 02 00 00 04 00 00 00 01 01 01 01 00 00 01    ................

7fb00030: 00 fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb    ................

7fb00040: 01 04 02 00 00 04 00 00 00 01 01 01 01 00 00 01    ................

7fb00050: 00 fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb    ................

7fb00060: 01 04 02 00 00 04 00 00 00 01 01 01 01 00 00 01    ................

7fb00070: 00 fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb    ................

0 Kudos

2,252 Views
carlos-m-ribeiro
Contributor I

Hello,

 

If we reading all CPLD data in our custom board, using this board file:

 

=> md 7fb00000 20
7fb00000: 15151515 15151515 15151515 15151515 ................
7fb00010: 15151515 15151515 15151515 15151515 ................
7fb00020: 15151515 15151515 15151515 15151515 ................
7fb00030: 15151515 15151515 15151515 15151515 ................
7fb00040: 15151515 15151515 15151515 15151515 ................
7fb00050: 15151515 15151515 15151515 15151515 ................
7fb00060: 15151515 15151515 15151515 15151515 ................
7fb00070: 15151515 15151515 15151515 15151515 ................
=> md.b 7fb00000 80
7fb00000: 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 ................
7fb00010: 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 ................
7fb00020: 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 ................
7fb00030: 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 ................
7fb00040: 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 ................
7fb00050: 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 ................
7fb00060: 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 ................
7fb00070: 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 ................
=>
 
Best regards
0 Kudos

2,240 Views
Pavel
NXP Employee
NXP Employee

Perhaps there is incorrect answer from CPLD on your board or hardware problem concerning to address lines in your board.

0 Kudos

2,213 Views
carlos-m-ribeiro
Contributor I

Hello,

 

But the CPU address lines connected to the CPLD are the same using U-BOOT for NAND (generated from u-boot-7-2017-LSDK-17.12) where the CPLD read is correct.

 

Thank you

0 Kudos

2,158 Views
Pavel
NXP Employee
NXP Employee

Is CPLD correct on your board?

Is there possibility to check connecting address lines to this CPLD?

0 Kudos