This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

CCSV4.2 R1: Memory save not ok

Hi

Using XDS560 on an OMAP4430 board, with latest CCSV4.2_CR1 I can not save data from the cortex m3 internal memory (local address 0)

It was ok with CCSv4.2 M9.

Selecting the CortexM3_0 core, In the memory window, rigth click -> Memory Save

 

The memory saved is not coherent with the value displayed in the memory windows

 

  • Claudio,

    How exactly the saved data is not coherent? I tested here and I could save memory in several data sizes (Type-size parameter) and swapped/unswapped values (Swap parameter)

    If you notice that each byte is prepended with zeros, then try reducing the Type-size. Using your screenshot as an example, if the output file is 00000083 00000080 00000000 00000000 00000083 and so on, the memory save utility is reading the target memory byte-by-byte but saving it as a 32-bit value.

    If your data seems displaced, try to disable the Swap value. For example, if you enable Swap with Type-size 32-bit the output should be 00008083 01000083 5B6B0083 E1A20083 and so on.

    The attached files were obtained from CCS4.2.0.00018 with different settings. If you are seeing something different please let me know, ok?

    Regards,

    Rafael

    S�yc����l@@�OzC�T�(�p�7�[z�X�_,y�o^�8K��\�k�iۀ�^�n�q6M7Zv�x���B�zHlgiT������us
    |6Ս�&��V�X������"�uC�J��ֻ�)�WmK���X�L�7��u�����uƅ_��qm�/>ܤW������X)�
    �ܙ
    ��I��^L��r`�=Q�VF�!�[7�k��n�~s+a��va� Tu_�@!���KI*�(�\"(����"1

    cy�S���@@l�zO��T�Cp�(�7�X�z[y,_��^o���K8�k�\��i��^�q�nZ7M6x�v���z�B�glHTi����u���s6|
    &���V�X�����u�"�J�C����)ӻ�KmW�X��7��L�u������u�q��_>/�m����W����)X���
    I܎
    ^��r��LQ=�`�FV�7[�!��k�s~�n��a+�avuT !@�_KԈ�(�*I"\���(1"�

    cy�S���@@l�zO��T�Cp�(�7�X�z[y,_��^o���K8�k�\��i��^�q�nZ7M6x�v���z�B�glHTi����u���s6|
    &���V�X�����u�"�J�C����)ӻ�KmW�X��7��L�u������u�q��_>/�m����W����)X���
    I܎
    ^��r��LQ=�`�FV�7[�!��k�s~�n��a+�avuT !@�_KԈ�(�*I"\���(1"�

  • rafael.

     

    No the data are really different. It sounds the cortex M3 internal memory (at 0 from cortexm3 side)  is not seen from debugger

    here is the file saved corresponding to the memory windows you can see in the screen shot above

    1651 1 0 0 40
    0x83800000
    0x83000001
    0x83006B5B
    0x8300A2E1
    0x8300653B
    0x0000000A
    0x00000000
    0x00000000
    0x00000000
    0x00000000
    0x00000000
    0x8300766B
    0x00000000
    0x00000000
    0x00000000
    0x00000000
    0x00000000
    0x00000000
    0x00000000
    0x00000000
    0x00000000
    0x00000000
    0x00000000
    0x00000000
    0x00000000
    0x00000000
    0x00000000
    0x00000000
    0x00000000
    0x00000000
    0x00000000
    0x00000000
    0x00000000
    0x00000000
    0x00000000
    0x00000000
    0x00000000
    0x00000000
    0x00000000
    0x00000000
    0x00000000
    0x00000000
    0x00000000
    0x00000000
    0x00000000
    0x00000000
    0x203A81B6
    0xD6884018
    0x2485A3D5
    0xDBFE4DCC
    0x02E18346
    0xE2CA118A
    0xB79EEC4C
    0xFB1811C9
    0x2C6FE7C0
    0x3DD2F066
    0x4D15AD25
    0xD8154C60
    0x42B58D8C
    0x27A8FD26
    0x806EB426
    0x73157307
    0xE06DE488
    0x0DC31233
    

     

    did you try on a OMAP4430 board?  My CCS version is the 4.2.00017

    [Post modification]

    Note that if I save the data from the cortexa9_side, at address 0x55020000 which is the L3 address for M3 internal memory, I see correct values in the saved data

    rgds

  • Claudio,

    The file you sent is a TI data file format, a text file that contains the data in alphanumeric characters instead of the raw binary data. It is saved this way if you name the output file with the extension .dat.

    I got the exact same file output in the file below, by selecting Format as Hex and Type size as 32-bits.

    1651 1 0 0 40
    0x6379A753
    0x00000801
    0xF9ECDF7F
    0x40406CB6
    0x7A4FB515
    0xC054E643
    0x70DA1F28
    0xA504378C
    0x58BD7A5B
    0x792C5F80
    0xBB5E6FEF
    0xF8D94B38
    0x866BD25C
    0x7F80DB69
    0x8BE95EDA
    0x7115AE6E
    0x5A374D36
    0x78F50F76
    0xF105B595
    0x7AF3429B
    0x67116C48
    0x1F541269
    0xE6C3F1C7
    0x7588E0DF
    0x0F007313
    0x36047C0D
    0x26B68DD5
    0x56E61FC2
    0x81581FA5
    0xF2F7A1B6
    0x75F022EB
    0x4AF04307
    0xD61FBFBF
    0x8329D3BB
    0x964B6D57
    0x9958F9EE
    0x3791E84C
    0xBE75EF83
    0xEEF47FA3
    0x85C675FB
    0x71819F5F
    0x3E2F9D6D
    0x7F03A4DC
    0xECF1AC57
    0xFBE419F7
    0xE6295816
    0x99DCF00D
    0x49DC8E0D
    0x5ED811D0
    0x72C3E44C
    0x513DC860
    0xB04656CD
    0x375B9B21
    0x8DB36BD1
    0x737E926E
    0xB899612B
    0xD061760C
    0x75541A20
    0x2140AA5F
    0x4BD488BA
    0x28B82A49
    0x04225C97
    0xD3D40728
    0x3122DCD1
    

    If you want to save the raw binary data, try to name the file something else other than .dat or .out extensions. I saved the file below using the same settings as above but with the extension .bin:

    cy�S���@@l�zO��T�Cp�(�7�X�z[y,_��^o���K8�k�\��i��^�q�nZ7M6x�v���z�B�glHTi����u���s6|
    &���V�X�����u�"�J�C����)ӻ�KmW�X��7��L�u������u�q��_>/�m����W����)X���
    I܎
    ^��r��LQ=�`�FV�7[�!��k�s~�n��a+�avuT !@�_KԈ�(�*I"\���(1"�

    These files were saved straight from the M3 core 0 of an OMAP4:

    Hope this helps,

    Rafael

  • rafael,

     

    what is odd is that I also select the .bin extension, as you can see from the attached files.

    Waht is the difference between the ccsv4.2_0017 (mine) anthe ccsv4.2.00018 (yours)?

    rgds

  • Claudio,

    Very strange indeed; the two versions are absolutely identical from a functional standpoint.

    Were you able to save the file correctly this time or you are still having trouble?

    The file extension does not matter, as long as it is different than .dat or .out - the only two pre-formatted extensions used by CCS.

    Regards,

    Rafael

     

  • raphael,

     

    No I still have the issue.

    Did you install the Normal 0 false false false EN-US X-NONE X-NONE MicrosoftInternetExplorer4 ti_emupack_setup_5.0.219.0.exe emulation patch?

    maybe this could explain..

     


  • Claudio,

    That could be the reason. I have the "stock" 5.0.158.0.

    I will install the newer emupack and give it a shot.

    Cheers,

    Rafael