• State Resolved
  • Replies 14 replies
  • Subscribers 102 subscribers
  • Views 210 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

TMS320C6655: Issue with Cache_all(L2WB) Function in Our System

H J
Prodigy 30 points
Part Number: TMS320C6655

Tool/software:

I'm experiencing an issue that only occurs in low-temperature(-40°C) conditions.

During the execution of the Cache_all function, the cacheReg variable remains set to 1, causing the while(*cacheReg) loop to run continuously. Could you provide any insights or solutions for this issue?

We found that even if we use a debugger to draw a function called roadsymbol and run, 
the *cacheReg value is still set to 1, so we cannot exit the while statement. 
We are using a gel file, but we do not know where to modify it. need help

I Search 

===========================================================================

Void Cache_all(volatile UInt32 *cacheReg)
{
UInt mask;

/* disable interrupts */
mask = Hwi_disable();

/* wait for any previous cache operation to complete */
while (*L2WWC != 0) {
/* open a window for interrupts */
Hwi_restore(mask);

/* disable interrupts */
mask = Hwi_disable();
}

/* perform global write back of cache */
*cacheReg = 1;

/* restore interrupts */
Hwi_restore(mask);

/* wait until cache operation completes */
while (*cacheReg) {
;
}
}

  • 0
    Prodigy 30 points

    GEL FILE SETTING INFORM

    Program.sectMap["sharedL2"] = "DDR3";  // <== Is this L2SRAM have to setting 
    Program.sectMap["systemHeap"] = "DDR3";
    Program.sectMap[".sysmem"] = "DDR3";
    Program.sectMap[".args"] = "DDR3";
    Program.sectMap[".cio"] = "DDR3";
    Program.sectMap[".far"] = "DDR3";
    Program.sectMap[".rodata"] = "DDR3";
    Program.sectMap[".neardata"] = "DDR3";
    Program.sectMap[".cppi"] = "DDR3";
    Program.sectMap[".init_array"] = "DDR3";
    Program.sectMap[".qmss"] = "DDR3";
    Program.sectMap[".cinit"] = "DDR3";
    Program.sectMap[".bss"] = "DDR3";
    Program.sectMap[".const"] = "DDR3";
    Program.sectMap[".text"] = "DDR3";
    Program.sectMap[".code"] = "DDR3";
    Program.sectMap[".switch"] = "DDR3";
    Program.sectMap[".data"] = "DDR3";
    Program.sectMap[".fardata"] = "DDR3";
    Program.sectMap[".args"] = "DDR3";
    Program.sectMap[".cio"] = "DDR3";
    Program.sectMap[".vecs"] = "DDR3";
    Program.sectMap["emacComm"] = "DDR3"
    Program.sectMap["platform_lib"] = "DDR3"
    Program.sectMap[".far:taskStackSection"] = "DDR3";
    Program.sectMap[".stack"] = "DDR3";
    Program.sectMap[".far:IMAGEDATA"] = {loadSegment: "DDR3", loadAlign: 8};
    Program.sectMap[".far:NDK_OBJMEM"] = {loadSegment: "DDR3", loadAlign: 8};
    Program.sectMap[".far:NDK_PACKETMEM"] = {loadSegment: "DDR3", loadAlign: 128};

    Program.sectMap["bitddr3"] = "DDR3";
    Program.sectMap["bitiram"] = "L2SRAM";

  • 0 in reply to H J
    TI__Mastermind 23745 points

    H J,

    Let me look at it and get back.

    Regards

    Shankari G

  • 0 in reply to H J
    TI__Mastermind 23745 points

    HJ,

    What is the temperature range up to which it works fine?

    --

    Is this GEL file provided by TI? or done by the customer?

    If you do not have the GEL file provided, I have uploaded here. 3718.evmc6657l.gel

    Please compare it with yours.

    --

    To modify the GEL file, you can follow this link.

    https://software-dl.ti.com/ccs/esd/documents/users_guide/ccs_debug-gel.html

    Regards

    Shankari G

  • 0 in reply to Shankari G
    Prodigy 30 points

    The file name is like this: evmc6657l.gel (ti 제공 파일 입니다.)

    It seems to work at -30°C

    We are using this: TMS320C6655CZH
    I'm still debugging, but it stays in while(cachereg). Why is that?

    Void Cache_all(volatile UInt32 *cacheReg)
    {
        UInt mask;

        /* disable interrupts */
        mask = Hwi_disable();

        /* wait for any previous cache operation to complete */
        while (*L2WWC != 0) {
            /* open a window for interrupts */
            Hwi_restore(mask);

            /* disable interrupts */
            mask = Hwi_disable();
        }

        /* perform global write back of cache */
        *cacheReg = 1;

        /* restore interrupts */
        Hwi_restore(mask);

        /* wait until cache operation completes */
        while (*cacheReg) {
            ;
        }
    }
  • 0 in reply to H J
    TI__Mastermind 23745 points

    HJ,

    If you have C6657 TI EVM, please do the same test and check whether results are same.... ?

    Compare your gel file with the one I updated above.

    Let me know if there is any difference.

    Regards

    Shankari G

  • 0 in reply to Shankari G
    Prodigy 30 points

    I'll check your gel file....

    I changed the register below and confirmed that it works.
    I'm worried. Are you okay? Can you reply?

    /* New value with DYN_ODT disabled and SDRAM_DRIVE = RZQ/7 //0x63222A32; // last config write DRAM init occurs */
    /* DDR_SDCFG = 0x63026AB2; */
    TEMP = 0;
    TEMP |= 0x3 << 29; // SDRAM_TYPE bit field 31:29 (fixed value)
    TEMP |= 0x0 << 27; // IBANK_POS bit field 28:27
    TEMP |= 0x4 << 24; // DDR_TERM bit field 26:24
    TEMP |= 0x2 << 21; // DYN_ODT bit field 22:21
    TEMP |= 0x1 << 18; // SDRAM_DRIVE bit field 19:18
    TEMP |= 0x2 << 16; // CWL bit field 17:16
    TEMP |= 0x1 << 14; // NM bit field 15:14
    TEMP |= 0xB << 10; // CL bit field 13:10
    TEMP |= 0x5 << 7; // ROWSIZE bit field 9:7
    TEMP |= 0x3 << 4; // IBANK bit field 6:4
    TEMP |= 0x0 << 3; // EBANK bit field 3:3
    TEMP |= 0x2; // PAGESIZE bit field 2:0
    DDR_SDCFG = TEMP;

  • 0 in reply to H J
    TI__Mastermind 23745 points

    HJ,

    Have you done the same cache test at (-40°C) on TI C6657 EVM? Does it work or fail?

    --

    According to the datasheet, C6655 should work with the temp below...

    Commercial Temperature: – 0°C to 85°C

    Extended Temperature: – –40°C to 100°C

    Regards

    Shankari G

  • 0 in reply to Shankari G
    Prodigy 30 points 
    Problem verification has been completed.
    It's a DSP problem and the difference is that one is marked YB10-32ZDHW9 and the other is marked YB10-38ZCQT9. Both are of the Extended Temperature (A) type. Is this a problem with the individual products? Otherwise, you may need to check if it is a product LOT issue. Can I know the LOT number information?
     
  • 0 in reply to H J
    TI__Mastermind 23745 points

    HJ,

    Whatever may be the LOT number, as per the C665x datasheet, it should support the extended temperature range -  –40°C to 100°C

    Do you have C6657 Ti EVM with you? 

    Have you done the same cache test at (-40°C) on TI C6657 EVM? Does it work or fail?

    --

    PLEASE ANSWER MY QUESTIONS TO HELP YOU BETTER.

    Regards

    Shankari G

  • 0 in reply to Shankari G
    Prodigy 30 points 
    There seems to be no problem with the EVM and other boards as well, 
    but only the board with the 'YB10-38ZCQT9' mark is experiencing problems.
  • 0 in reply to H J
    TI__Mastermind 23745 points

    HJ,

    OK.

    I have another suggestion.

    1. In the same board, try removing " YB10-38ZCQT9" and replace the device-YB10-32ZDHW9.

    2. In the board with YB10-32ZDHW9, replace YB10-38ZCQT9 and observe the behavior.

    This test will confirm whether it is a board issue or the device issue.

    ---

    Meanwhile, 

    Let me forward to our hardware expert.

    Thanks for your patience.

    Regards

    Shankari G

  • 0 in reply to Shankari G
    TI__Mastermind 49492 points

    Hello HJ,

    Can you summarize ... how many systems have you built/tested that are passing and how many are failing?

    Thanks,

    Kyle

  • 0 in reply to kcastille
    Prodigy 30 points 
    I am currently trying to test by changing tms320c6655. Currently there is only one problem.
  • +1 in reply to H J
    TI__Mastermind 23745 points

    HJ,

    These conversations may not help either of us in solving the problem.

    Insufficient data may not lead both of us anywhere.

    Please answer Kyle query in previous post.

    Regards

    Shankari G