TMS320F28386D: CLA Data ROM Symbols library combinines incompatible files of VCRC library

Hi Henry,

I think this could be due to mismatch of options used in building the symbols library F2838xCPU1_CLADATAROM_Symbols_fpu32.lib. let me check with the owner of the library the options and confirm back. 

Thanks

Aravindhan

Hi Henry,

I tried to combine a project with including VCRC library and the F2838xCPU1_CLADATAROM_Symbols_fpu32.lib and it seems to build fine. Which version of c2000ware you are using, and the compiler version ? are you using the vcrc library as is or trying to build it ? if you can provide a bite more specifics it will be helpful. 

Thanks

Aravindhan

Thanks for replying.

Version used:

1. C2000ware v3.0.3
   1. F2838xCPU(n)_CLADATAROM_Symbols_fpu32.lib from rev0

2. CLA build with Compiler v18.12.5 LTS
3. CPU build with Compiler v18.12.5 LTS

What I have done in my project:

1. Properties on my project for CPU1 and CPU2
   1. In "C2000 Linker -> File Search Path -> include file", add "c28x_vcrc_library_fpu64.lib"
   2. In "C2000 Linker -> File Search Path -> search path", add "../VCU/lib". In this directory, files copied from \C2000Ware_3_03_00_00\libraries\dsp\VCU\c28, exception of "examples" ,"source" and "ccs" folders.
2. CPU project add module like CRC_Calculator.zip

With trial of comparing what's difference between example code and mine, I tried to add "F2838xCPU1_CLADATAROM_Symbols_fpu32.lib" in the "2838x_vcrc_crc_32" example from \C2000Ware_3_03_00_00\libraries\dsp\VCU\c28\examples\crc. It did make it without any #16004 error.

Is that exactly your trial too? Or, what's advice further for checking building error?

Hi Henry,

Yes. I did the same by including the CLA symbols library and built the vcrc_crc_8 example and it built fine. I will check with the compiler team on the build error you are getting for further insight. In you project you want to use CLA and VCRC ? 

Thanks

Aravindhan 

Hi Aravindhan.

Our team used to develop on F28377D, and use CLA and VCU-II for a long time. In my arrogant opinion, if no need of customizing CRC polynomial, coding work in F28386D is the same as usual in F28377D. But now I meet this problem while adding library.

Thanks for reporting to your design team. I'm looking forward to any further advice or instruction about how to solve this.

Henry,

Thanks. Understood what you are trying to do. We are going to try to build the symbols CLA library with vcrc option and see if we can pass it to you to eliminate mismatch of options. Are you urgent need to debug this or you can give couple of days ?

Thanks

Aravindhan 

Hi Ardvindhan.

First, I really appreciate for your fast support. I am required to make it clear before this Friday upon our schedule (kind of urgently), but it still depends on how fast your team find out the main reason why this happens.

I'm still expecting on getting your news further before Friday maybe. Hope this will be done soon!

Henry,

Sure. will keep you deadline request in mind and try to help.

Thanks

Aravindhan

Hi, Aravindhan.

Sorry for delayed response due to vacation of Lunar New Year. I've applied your updated library files and it came out some unexpected messages.

Here's some detailed warnings for references in further:

#10015-D output file "OpenDrive_F2838x_cpu01.out" cannot be loaded and run on a target system
#10015-D output file "OpenDrive_F2838x_cpu02.out" cannot be loaded and run on a target system
#17001-D /home/ubnuser/ti/repos/c2000ware/internal/repos/c2000ware/libraries/dsp/temp/source/vcu2/crc/vcu2_crc_32.asm: relocation type is not supported (type = 'R_0x14' (0), file = "D:/Documents/IssueBranches/fSERVO4096_M3FcsCheckUsingVcrc_4px/Source/lib/VCUDSP/lib/c28x_vcrc_library_fpu64.lib<vcu2_crc_32.obj>", offset = 0x00000079, section = ".text")
#17001-D /home/ubnuser/ti/repos/c2000ware/internal/repos/c2000ware/libraries/dsp/temp/source/vcu2/crc/vcu2_crc_32.asm: relocation type is not supported (type = 'R_0x14' (0), file = "D:/Documents/IssueBranches/fSERVO4096_M3FcsCheckUsingVcrc_4px/Source/lib/VCUDSP/lib/c28x_vcrc_library_fpu64.lib<vcu2_crc_32.obj>", offset = 0x0000007e, section = ".text")
<a href="processors.wiki.ti.com/.../17003"> /home/ubnuser/ti/repos/c2000ware/libraries/math/temp/source/CLAcos.asm: relocation from assembly function "CLAcos" to symbol "CLAsincosTable_Coef0" overflowed; the 25-bit relocated address 0x1001e54 is too large to encode in the 16-bit unsigned field (type = 'R_CLA_ABS16' (16), file = "D:/Documents/IssueBranches/fSERVO4096_M3FcsCheckUsingVcrc_4px/Source/lib/CLAmath/lib/cla2_math_library_datarom_fpu32_eabi.lib<CLAcos.obj>", offset = 0x00000024, section = "Cla1Prog:_CLAcos")
<a href="processors.wiki.ti.com/.../17003"> /home/ubnuser/ti/repos/c2000ware/libraries/math/temp/source/CLAcos.asm: relocation from assembly function "CLAcos" to symbol "CLAsincosTable_Coef0" overflowed; the 25-bit relocated address 0x1001e54 is too large to encode in the 16-bit unsigned field (type = 'R_CLA_ABS16' (16), file = "D:/Documents/IssueBranches/fSERVO4096_M3FcsCheckUsingVcrc_4px/Source/lib/CLAmath/lib/cla2_math_library_datarom_fpu32_eabi.lib<CLAcos.obj>", offset = 0x00000024, section = "Cla1Prog:_CLAcos")
<a href="processors.wiki.ti.com/.../17003"> /home/ubnuser/ti/repos/c2000ware/libraries/math/temp/source/CLAcos.asm: relocation from assembly function "CLAcos" to symbol "CLAsincosTable_Coef1_pos" overflowed; the 25-bit relocated address 0x1001e58 is too large to encode in the 16-bit unsigned field (type = 'R_CLA_ABS16' (16), file = "D:/Documents/IssueBranches/fSERVO4096_M3FcsCheckUsingVcrc_4px/Source/lib/CLAmath/lib/cla2_math_library_datarom_fpu32_eabi.lib<CLAcos.obj>", offset = 0x0000001e, section = "Cla1Prog:_CLAcos")
<a href="processors.wiki.ti.com/.../17003"> /home/ubnuser/ti/repos/c2000ware/libraries/math/temp/source/CLAcos.asm: relocation from assembly function "CLAcos" to symbol "CLAsincosTable_Coef1_pos" overflowed; the 25-bit relocated address 0x1001e58 is too large to encode in the 16-bit unsigned field (type = 'R_CLA_ABS16' (16), file = "D:/Documents/IssueBranches/fSERVO4096_M3FcsCheckUsingVcrc_4px/Source/lib/CLAmath/lib/cla2_math_library_datarom_fpu32_eabi.lib<CLAcos.obj>", offset = 0x0000001e, section = "Cla1Prog:_CLAcos")
<a href="processors.wiki.ti.com/.../17003"> /home/ubnuser/ti/repos/c2000ware/libraries/math/temp/source/CLAcos.asm: relocation from assembly function "CLAcos" to symbol "CLAsincosTable_Coef2" overflowed; the 25-bit relocated address 0x1001e5a is too large to encode in the 16-bit unsigned field (type = 'R_CLA_ABS16' (16), file = "D:/Documents/IssueBranches/fSERVO4096_M3FcsCheckUsingVcrc_4px/Source/lib/CLAmath/lib/cla2_math_library_datarom_fpu32_eabi.lib<CLAcos.obj>", offset = 0x00000018, section = "Cla1Prog:_CLAcos")
<a href="processors.wiki.ti.com/.../17003"> /home/ubnuser/ti/repos/c2000ware/libraries/math/temp/source/CLAcos.asm: relocation from assembly function "CLAcos" to symbol "CLAsincosTable_Coef2" overflowed; the 25-bit relocated address 0x1001e5a is too large to encode in the 16-bit unsigned field (type = 'R_CLA_ABS16' (16), file = "D:/Documents/IssueBranches/fSERVO4096_M3FcsCheckUsingVcrc_4px/Source/lib/CLAmath/lib/cla2_math_library_datarom_fpu32_eabi.lib<CLAcos.obj>", offset = 0x00000018, section = "Cla1Prog:_CLAcos")
<a href="processors.wiki.ti.com/.../17003"> /home/ubnuser/ti/repos/c2000ware/libraries/math/temp/source/CLAcos.asm: relocation from assembly function "CLAcos" to symbol "CLAsincosTable_Coef3_neg" overflowed; the 25-bit relocated address 0x1001e5e is too large to encode in the 16-bit unsigned field (type = 'R_CLA_ABS16' (16), file = "D:/Documents/IssueBranches/fSERVO4096_M3FcsCheckUsingVcrc_4px/Source/lib/CLAmath/lib/cla2_math_library_datarom_fpu32_eabi.lib<CLAcos.obj>", offset = 0x00000012, section = "Cla1Prog:_CLAcos")

Would you please help us solving those warnings? We have used vcu2 in F28377D for a long time and these warnings didn't come out. What I am concerned about is, it seems to be abnormal on some code generating so that we couldn't use this library files.

Thanks for your help in advance.

Henry,

I will ask the CLA expert to take a look at this thread. He will be able to help out.

Thanks

Aravindhan

Henry,

I've re-assigned the thread, but wanted to give a heads up that due to the weather here in Texas many in our team have had intermittent power/access to the web.  Someone will respond as soon as possible, but this may not be until early next week.

Best,

Matthew

Aravindhan K 

The CLA data ROM content can be viewed by the C28x and by the CLA.  Each uses a different memory address.  I suspect the symbol library is using the C28x memory address.  The CLA PC does not have that far of a reach so the linker issues this warning.  The CLA memory address should be used for the symbol library.

Can you check?

Henry,

The other thing you can check is that the CLA prog sections are allocated to memory that the CLA can access.  Since you have this working on 2837x and are using an example from us, I doubt that is the issue.  It is still good to check, though.

Regards

Lori

Hi, Lori.

About the warning I met like

"#17003-D</a> /home/ubnuser/ti/repos/c2000ware/libraries/math/temp/source/CLAcos.asm: relocation from assembly function "CLAcos" to symbol "CLAsincosTable_Coef0" overflowed; the 25-bit relocated address 0x1001e54 is too large to encode in the 16-bit unsigned field (type = 'R_CLA_ABS16' (16), file = "<some library directory>/cla2_math_library_datarom_fpu32_eabi.lib<CLAcos.obj>", offset = 0x00000024, section = "Cla1Prog:_CLAcos") "

I have found an problem probably with mine. Please refer to: https://e2e.ti.com/support/microcontrollers/c2000/f/171/t/942616

What if the updated library offered by Aravindhan on 2021/2/5 is related to this thread? 

If I look into the map file after building, it did fit to the address where messages mentioned:

abs   01001e54  CLAsincosTable_Coef0                                  
abs   01001e56  CLAsincosTable_Coef1                                  
abs   01001e58  CLAsincosTable_Coef1_pos                              
abs   01001e5a  CLAsincosTable_Coef2                                  
abs   01001e5c  CLAsincosTable_Coef3                                  
abs   01001e5e  CLAsincosTable_Coef3_neg                              
abs   01001d4a  CLAsincosTable_Cos0                                   
abs   01001d0a  CLAsincosTable_Sin0                                   
abs   01001e52  CLAsincosTable_TABLE_MASK                             
abs   01001e4e  CLAsincosTable_TABLE_SIZEDivTwoPi                     
abs   01001e50  CLAsincosTable_TwoPiDivTABLE_SIZE                     

Another warning I met like

"#17001-D /home/ubnuser/ti/repos/c2000ware/internal/repos/c2000ware/libraries/dsp/temp/source/vcu2/crc/vcu2_crc_32.asm: relocation type is not supported (type = 'R_0x14' (0), file = "<some library directory>/c28x_vcrc_library_fpu64.lib<vcu2_crc_32.obj>", offset = 0x00000079, section = ".text")"

this is still confusing me.

Although I have found an problem seems not to be my situation. Please refer to: 

I re-check the compiler version is 18.12.5.LTS, which is later to 18.9.0.STS mentioned by spruhs1c.pdf

It seems the later warning message should have another problem to solve, but the former one may tell a problem caused by the updated library offered by Aravindhan just like the thread link(https://e2e.ti.com/support/microcontrollers/c2000/f/171/t/942616).

Please kindly help us solve on those mysteries. Thanks!

Henry,

I will check with the symbols library owner and see if we have to regenrate for CLA address as indicated by Lori. 

Thanks

Aravindhan

Henry,

The symbol library owner is sure that the CLA addresses were used in generating the library Aravindhan K provided. 

1. Can you double check your project and make sure you are linking the new library in?   As you noted we had this issue previously with an old library.
2. Check the build options linker, file search path and see if there is a path to the old library. 
3. I suggest renaming the old library to help us eliminate it from the equation.
4. Also please check your linker .cmd file.  It shouldn't have these symbols mentioned in it.  The symbol library itself takes care of the mapping.

For the compiler version, unless you have a reason for staying on version 18...LTS I would suggest moving to the latest LTS version of the compiler.  This won't solve the issue, but it is a general recommendation as improvements and bug fixes will be in newer versions.

https://www.ti.com/tool/C2000-CGT

Regards

Lori

About this warning

"#17001-D /home/ubnuser/ti/repos/c2000ware/internal/repos/c2000ware/libraries/dsp/temp/source/vcu2/crc/vcu2_crc_32.asm: relocation type is not supported (type = 'R_0x14' (0), file = "<some library directory>/c28x_vcrc_library_fpu64.lib<vcu2_crc_32.obj>", offset = 0x00000079, section = ".text")"

I tried to change the compiler version from 18.12.5LTS to 20.2.4LTS. With using 20.2.4LTS, this warning is missing. So, this warning may tell the new library provided by  Aravindhan K were created by the newer compiler and mine is incompatible to the new library.

Still check on the other warnings(about CLA Data ROM addresses). If I have something new, I'll update on this thread.

Henry,

Good to know it went away with the latest compiler. Will inform the compiler team about this. Can I close the thread now ?

Thanks

Aravindhan

Warning #17003-D about cla2_math_library_datarom_fpu32_eabi.lib still happens.

I have checked those instruction that Lori offered. We don't mapped those data ROM addresses in cmd. file.

In addition, if I remove the part of VCRC such as CRC Obj using, vcrc library...etc and apply the new library offered on 2021/2/5. Those warnings still exist. And the tested program works fine with the rev. 0 library from C2000ware.

So, the new library may still reproduce the same problems just like thread(https://e2e.ti.com/support/microcontrollers/c2000/f/171/t/942616) mentioned. Even the messages of those data ROM addresses from my map. file are identical to the messages the thread posted.

Can you please double check for the new library for me?

Henry,

Okay. Understood. and will investigate further. thanks for additional details. 

Thanks

Aravindhan

Hi Henry,

can you let us know what is the fpu option you are using  ? is it softlib or fpu32 / fpu64 ? The team can help further if you can share a sample project which has this issue..Is this possible ? 

Thanks

Aravindhan

Aravindhan K said:

is it softlib or fpu32 / fpu64 ?

It is fpu64, thanks.

Henry,

will you be able to share your sample project with the issue so we can take a deeper look ?

Thanks

Aravindhan

Hi, Aravindhan

The reproduction of warning #17003-D was confirmed by c2000ware v3.03.00.00 on "cla_cos" project with replacing CLA data ROM library from rev.0 to revision provided on 2021/02/05. 

Warning details:

#10015-D output file "cla_cos.out" cannot be loaded and run on a target system
<a href="http://processors.wiki.ti.com/index.php/Compiler/diagnostic_messages/17003">#17003-D</a> /home/ubnuser/ti/repos/c2000ware/internal/repos/c2000ware/libraries/math/temp/source/CLAcos.asm: relocation from assembly function "CLAcos" to symbol "CLAsincosTable_Coef0" overflowed; the 25-bit relocated address 0x1001e54 is too large to encode in the 16-bit unsigned field (type = 'R_CLA_ABS16' (16), file = "C:/ti/c2000/C2000Ware_3_03_00_00/libraries/math/CLAmath/c28/lib/cla2_math_library_datarom_fpu32_eabi.lib<CLAcos.obj>", offset = 0x00000024, section = "Cla1Prog:_CLAcos")
<a href="http://processors.wiki.ti.com/index.php/Compiler/diagnostic_messages/17003">#17003-D</a> /home/ubnuser/ti/repos/c2000ware/internal/repos/c2000ware/libraries/math/temp/source/CLAcos.asm: relocation from assembly function "CLAcos" to symbol "CLAsincosTable_Coef1_pos" overflowed; the 25-bit relocated address 0x1001e58 is too large to encode in the 16-bit unsigned field (type = 'R_CLA_ABS16' (16), file = "C:/ti/c2000/C2000Ware_3_03_00_00/libraries/math/CLAmath/c28/lib/cla2_math_library_datarom_fpu32_eabi.lib<CLAcos.obj>", offset = 0x0000001e, section = "Cla1Prog:_CLAcos")
<a href="http://processors.wiki.ti.com/index.php/Compiler/diagnostic_messages/17003">#17003-D</a> /home/ubnuser/ti/repos/c2000ware/internal/repos/c2000ware/libraries/math/temp/source/CLAcos.asm: relocation from assembly function "CLAcos" to symbol "CLAsincosTable_Coef2" overflowed; the 25-bit relocated address 0x1001e5a is too large to encode in the 16-bit unsigned field (type = 'R_CLA_ABS16' (16), file = "C:/ti/c2000/C2000Ware_3_03_00_00/libraries/math/CLAmath/c28/lib/cla2_math_library_datarom_fpu32_eabi.lib<CLAcos.obj>", offset = 0x00000018, section = "Cla1Prog:_CLAcos")
<a href="http://processors.wiki.ti.com/index.php/Compiler/diagnostic_messages/17003">#17003-D</a> /home/ubnuser/ti/repos/c2000ware/internal/repos/c2000ware/libraries/math/temp/source/CLAcos.asm: relocation from assembly function "CLAcos" to symbol "CLAsincosTable_Coef3_neg" overflowed; the 25-bit relocated address 0x1001e5e is too large to encode in the 16-bit unsigned field (type = 'R_CLA_ABS16' (16), file = "C:/ti/c2000/C2000Ware_3_03_00_00/libraries/math/CLAmath/c28/lib/cla2_math_library_datarom_fpu32_eabi.lib<CLAcos.obj>", offset = 0x00000012, section = "Cla1Prog:_CLAcos")
<a href="http://processors.wiki.ti.com/index.php/Compiler/diagnostic_messages/17003">#17003-D</a> /home/ubnuser/ti/repos/c2000ware/internal/repos/c2000ware/libraries/math/temp/source/CLAcos.asm: relocation from assembly function "CLAcos" to symbol "CLAsincosTable_TABLE_MASK" overflowed; the 25-bit relocated address 0x1001e52 is too large to encode in the 16-bit unsigned field (type = 'R_CLA_ABS16' (16), file = "C:/ti/c2000/C2000Ware_3_03_00_00/libraries/math/CLAmath/c28/lib/cla2_math_library_datarom_fpu32_eabi.lib<CLAcos.obj>", offset = 0x00000004, section = "Cla1Prog:_CLAcos")
<a href="http://processors.wiki.ti.com/index.php/Compiler/diagnostic_messages/17003">#17003-D</a> /home/ubnuser/ti/repos/c2000ware/internal/repos/c2000ware/libraries/math/temp/source/CLAcos.asm: relocation from assembly function "CLAcos" to symbol "CLAsincosTable_TABLE_SIZEDivTwoPi" overflowed; the 25-bit relocated address 0x1001e4e is too large to encode in the 16-bit unsigned field (type = 'R_CLA_ABS16' (16), file = "C:/ti/c2000/C2000Ware_3_03_00_00/libraries/math/CLAmath/c28/lib/cla2_math_library_datarom_fpu32_eabi.lib<CLAcos.obj>", offset = 0x00000002, section = "Cla1Prog:_CLAcos")
<a href="http://processors.wiki.ti.com/index.php/Compiler/diagnostic_messages/17003">#17003-D</a> /home/ubnuser/ti/repos/c2000ware/internal/repos/c2000ware/libraries/math/temp/source/CLAcos.asm: relocation from assembly function "CLAcos" to symbol "CLAsincosTable_TwoPiDivTABLE_SIZE" overflowed; the 25-bit relocated address 0x1001e50 is too large to encode in the 16-bit unsigned field (type = 'R_CLA_ABS16' (16), file = "C:/ti/c2000/C2000Ware_3_03_00_00/libraries/math/CLAmath/c28/lib/cla2_math_library_datarom_fpu32_eabi.lib<CLAcos.obj>", offset = 0x00000010, section = "Cla1Prog:_CLAcos")
<a href="http://processors.wiki.ti.com/index.php/Compiler/diagnostic_messages/17003">#17003-D</a> relocation from function "Cla1Task2" to symbol "CLAsincosTable_Coef0" overflowed; the 25-bit relocated address 0x1001e54 is too large to encode in the 16-bit unsigned field (type = 'R_CLA_ABS16' (16), file = "./cla_cos.obj", offset = 0x00000074, section = "Cla1Prog:Cla1Task2")
<a href="http://processors.wiki.ti.com/index.php/Compiler/diagnostic_messages/17003">#17003-D</a> relocation from function "Cla1Task2" to symbol "CLAsincosTable_Coef1_pos" overflowed; the 25-bit relocated address 0x1001e58 is too large to encode in the 16-bit unsigned field (type = 'R_CLA_ABS16' (16), file = "./cla_cos.obj", offset = 0x0000006e, section = "Cla1Prog:Cla1Task2")
<a href="http://processors.wiki.ti.com/index.php/Compiler/diagnostic_messages/17003">#17003-D</a> relocation from function "Cla1Task2" to symbol "CLAsincosTable_Coef2" overflowed; the 25-bit relocated address 0x1001e5a is too large to encode in the 16-bit unsigned field (type = 'R_CLA_ABS16' (16), file = "./cla_cos.obj", offset = 0x00000064, section = "Cla1Prog:Cla1Task2")
<a href="http://processors.wiki.ti.com/index.php/Compiler/diagnostic_messages/17003">#17003-D</a> relocation from function "Cla1Task2" to symbol "CLAsincosTable_Coef3_neg" overflowed; the 25-bit relocated address 0x1001e5e is too large to encode in the 16-bit unsigned field (type = 'R_CLA_ABS16' (16), file = "./cla_cos.obj", offset = 0x00000062, section = "Cla1Prog:Cla1Task2")
<a href="http://processors.wiki.ti.com/index.php/Compiler/diagnostic_messages/17003">#17003-D</a> relocation from function "Cla1Task2" to symbol "CLAsincosTable_TABLE_SIZEDivTwoPi" overflowed; the 25-bit relocated address 0x1001e4e is too large to encode in the 16-bit unsigned field (type = 'R_CLA_ABS16' (16), file = "./cla_cos.obj", offset = 0x00000022, section = "Cla1Prog:Cla1Task2")

Please check the detailed project options with attachments : 

ExampleTi_WorkSpace.zip

BTW, example code search libraries by "Include Option" instead of "C2000 Linker". So I just replaced the old "F2838xCPU1_CLADATAROM_Symbols_fpu32.lib" in "\C2000Ware_3_03_00_00\libraries\boot_rom\f2838x\rev0\rom_symbol_libs\cpu1\claDataROM\" with the new one.

In addition, I checked the "CLAsin" example. It did reproduce too. But the "CLAacos" example didn't.

It means CLAacos() function doesn't cause those warnings but CLAsin(), CLAcos() do.

Thanks for help!

Henry,

These details are useful and i have asked BootROM expert to take a look.

Thanks

Aravindhan

Henry,

You have attached the workspace files and i am not able to build the project with that.

You may need to unselect checkbox "copy files to workspace" while importing the project. Reproduce the error in this case and share the files (you can delete other folders to keep the zip file smaller) with us.

I tried updating the lib (F2838xCPU2_CLADATAROM_Symbols_fpu32.lib) with the new one in C2000ware (installed in my PC) and the project builds fine. Seems there is some other setting which is causing the trouble. if you can do the above (marked in bold font) and share the files it will be helpful.

Regards

Baskaran

Hi, Baskaran.

Re-packaging the zip file for you. Please check this out : cla_cos.zip

The reproducing configuration is "f2838x_FLASH".

Furthermore, after the project imported, I just need to set up a few options for reproducing:

With above setting, if using CLA Data ROM rev.0 library, I get errors #16004-D:

#10010 errors encountered during linking; "cla_cos.out" not built
#16004-D file "C:/ti/c2000/C2000Ware_3_03_00_00/libraries/boot_rom/f2838x/rev0/rom_symbol_libs/cpu1/claDataROM/F2838xCPU1_CLADATAROM_Symbols_fpu32.lib<F2838x_CPU1_ROM_out_CLAsincosTable_Coef0_tmp.obj>" has a Tag_VCU attribute value of "1" that is different than one previously seen ("3"); combining incompatible files
#16004-D file "C:/ti/c2000/C2000Ware_3_03_00_00/libraries/boot_rom/f2838x/rev0/rom_symbol_libs/cpu1/claDataROM/F2838xCPU1_CLADATAROM_Symbols_fpu32.lib<F2838x_CPU1_ROM_out_CLAsincosTable_Coef1_pos_tmp.obj>" has a Tag_VCU attribute value of "1" that is different than one previously seen ("3"); combining incompatible files
#16004-D file "C:/ti/c2000/C2000Ware_3_03_00_00/libraries/boot_rom/f2838x/rev0/rom_symbol_libs/cpu1/claDataROM/F2838xCPU1_CLADATAROM_Symbols_fpu32.lib<F2838x_CPU1_ROM_out_CLAsincosTable_Coef2_tmp.obj>" has a Tag_VCU attribute value of "1" that is different than one previously seen ("3"); combining incompatible files
#16004-D file "C:/ti/c2000/C2000Ware_3_03_00_00/libraries/boot_rom/f2838x/rev0/rom_symbol_libs/cpu1/claDataROM/F2838xCPU1_CLADATAROM_Symbols_fpu32.lib<F2838x_CPU1_ROM_out_CLAsincosTable_Coef3_neg_tmp.obj>" has a Tag_VCU attribute value of "1" that is different than one previously seen ("3"); combining incompatible files
#16004-D file "C:/ti/c2000/C2000Ware_3_03_00_00/libraries/boot_rom/f2838x/rev0/rom_symbol_libs/cpu1/claDataROM/F2838xCPU1_CLADATAROM_Symbols_fpu32.lib<F2838x_CPU1_ROM_out_CLAsincosTable_Sin0_tmp.obj>" has a Tag_VCU attribute value of "1" that is different than one previously seen ("3"); combining incompatible files
#16004-D file "C:/ti/c2000/C2000Ware_3_03_00_00/libraries/boot_rom/f2838x/rev0/rom_symbol_libs/cpu1/claDataROM/F2838xCPU1_CLADATAROM_Symbols_fpu32.lib<F2838x_CPU1_ROM_out_CLAsincosTable_TABLE_MASK_tmp.obj>" has a Tag_VCU attribute value of "1" that is different than one previously seen ("3"); combining incompatible files
#16004-D file "C:/ti/c2000/C2000Ware_3_03_00_00/libraries/boot_rom/f2838x/rev0/rom_symbol_libs/cpu1/claDataROM/F2838xCPU1_CLADATAROM_Symbols_fpu32.lib<F2838x_CPU1_ROM_out_CLAsincosTable_TABLE_SIZEDivTwoPi_tmp.obj>" has a Tag_VCU attribute value of "1" that is different than one previously seen ("3"); combining incompatible files
#16004-D file "C:/ti/c2000/C2000Ware_3_03_00_00/libraries/boot_rom/f2838x/rev0/rom_symbol_libs/cpu1/claDataROM/F2838xCPU1_CLADATAROM_Symbols_fpu32.lib<F2838x_CPU1_ROM_out_CLAsincosTable_tmp.obj>" has a Tag_VCU attribute value of "1" that is different than one previously seen ("3"); combining incompatible files
#16004-D file "C:/ti/c2000/C2000Ware_3_03_00_00/libraries/boot_rom/f2838x/rev0/rom_symbol_libs/cpu1/claDataROM/F2838xCPU1_CLADATAROM_Symbols_fpu32.lib<F2838x_CPU1_ROM_out_CLAsincosTable_TwoPiDivTABLE_SIZE_tmp.obj>" has a Tag_VCU attribute value of "1" that is different than one previously seen ("3"); combining incompatible files
gmake: *** [all] Error 2
gmake[1]: *** [cla_cos.out] Error 1

If I changed the library provided on 2021/02/05 F2838xCPU1_CLADATAROM_Symbols_fpu32.lib, I get warnings like:

#10015-D output file "cla_cos.out" cannot be loaded and run on a target system
<a href="http://processors.wiki.ti.com/index.php/Compiler/diagnostic_messages/17003">#17003-D</a>  /home/ubnuser/ti/repos/c2000ware/internal/repos/c2000ware/libraries/math/temp/source/CLAcos.asm: relocation from assembly function "CLAcos" to symbol "CLAsincosTable_Coef0" overflowed; the 25-bit relocated address 0x1001e54 is too large to encode in the 16-bit unsigned field (type = 'R_CLA_ABS16' (16), file = "C:/ti/c2000/C2000Ware_3_03_00_00/libraries/math/CLAmath/c28/lib/cla2_math_library_datarom_fpu32_eabi.lib<CLAcos.obj>", offset = 0x00000024, section = "Cla1Prog:_CLAcos")
<a href="http://processors.wiki.ti.com/index.php/Compiler/diagnostic_messages/17003">#17003-D</a>  /home/ubnuser/ti/repos/c2000ware/internal/repos/c2000ware/libraries/math/temp/source/CLAcos.asm: relocation from assembly function "CLAcos" to symbol "CLAsincosTable_Coef1_pos" overflowed; the 25-bit relocated address 0x1001e58 is too large to encode in the 16-bit unsigned field (type = 'R_CLA_ABS16' (16), file = "C:/ti/c2000/C2000Ware_3_03_00_00/libraries/math/CLAmath/c28/lib/cla2_math_library_datarom_fpu32_eabi.lib<CLAcos.obj>", offset = 0x0000001e, section = "Cla1Prog:_CLAcos")
<a href="http://processors.wiki.ti.com/index.php/Compiler/diagnostic_messages/17003">#17003-D</a>  /home/ubnuser/ti/repos/c2000ware/internal/repos/c2000ware/libraries/math/temp/source/CLAcos.asm: relocation from assembly function "CLAcos" to symbol "CLAsincosTable_Coef2" overflowed; the 25-bit relocated address 0x1001e5a is too large to encode in the 16-bit unsigned field (type = 'R_CLA_ABS16' (16), file = "C:/ti/c2000/C2000Ware_3_03_00_00/libraries/math/CLAmath/c28/lib/cla2_math_library_datarom_fpu32_eabi.lib<CLAcos.obj>", offset = 0x00000018, section = "Cla1Prog:_CLAcos")
<a href="http://processors.wiki.ti.com/index.php/Compiler/diagnostic_messages/17003">#17003-D</a>  /home/ubnuser/ti/repos/c2000ware/internal/repos/c2000ware/libraries/math/temp/source/CLAcos.asm: relocation from assembly function "CLAcos" to symbol "CLAsincosTable_Coef3_neg" overflowed; the 25-bit relocated address 0x1001e5e is too large to encode in the 16-bit unsigned field (type = 'R_CLA_ABS16' (16), file = "C:/ti/c2000/C2000Ware_3_03_00_00/libraries/math/CLAmath/c28/lib/cla2_math_library_datarom_fpu32_eabi.lib<CLAcos.obj>", offset = 0x00000012, section = "Cla1Prog:_CLAcos")
<a href="http://processors.wiki.ti.com/index.php/Compiler/diagnostic_messages/17003">#17003-D</a>  /home/ubnuser/ti/repos/c2000ware/internal/repos/c2000ware/libraries/math/temp/source/CLAcos.asm: relocation from assembly function "CLAcos" to symbol "CLAsincosTable_TABLE_MASK" overflowed; the 25-bit relocated address 0x1001e52 is too large to encode in the 16-bit unsigned field (type = 'R_CLA_ABS16' (16), file = "C:/ti/c2000/C2000Ware_3_03_00_00/libraries/math/CLAmath/c28/lib/cla2_math_library_datarom_fpu32_eabi.lib<CLAcos.obj>", offset = 0x00000004, section = "Cla1Prog:_CLAcos")
<a href="http://processors.wiki.ti.com/index.php/Compiler/diagnostic_messages/17003">#17003-D</a>  /home/ubnuser/ti/repos/c2000ware/internal/repos/c2000ware/libraries/math/temp/source/CLAcos.asm: relocation from assembly function "CLAcos" to symbol "CLAsincosTable_TABLE_SIZEDivTwoPi" overflowed; the 25-bit relocated address 0x1001e4e is too large to encode in the 16-bit unsigned field (type = 'R_CLA_ABS16' (16), file = "C:/ti/c2000/C2000Ware_3_03_00_00/libraries/math/CLAmath/c28/lib/cla2_math_library_datarom_fpu32_eabi.lib<CLAcos.obj>", offset = 0x00000002, section = "Cla1Prog:_CLAcos")
<a href="http://processors.wiki.ti.com/index.php/Compiler/diagnostic_messages/17003">#17003-D</a>  /home/ubnuser/ti/repos/c2000ware/internal/repos/c2000ware/libraries/math/temp/source/CLAcos.asm: relocation from assembly function "CLAcos" to symbol "CLAsincosTable_TwoPiDivTABLE_SIZE" overflowed; the 25-bit relocated address 0x1001e50 is too large to encode in the 16-bit unsigned field (type = 'R_CLA_ABS16' (16), file = "C:/ti/c2000/C2000Ware_3_03_00_00/libraries/math/CLAmath/c28/lib/cla2_math_library_datarom_fpu32_eabi.lib<CLAcos.obj>", offset = 0x00000010, section = "Cla1Prog:_CLAcos")
<a href="http://processors.wiki.ti.com/index.php/Compiler/diagnostic_messages/17003">#17003-D</a>  relocation from function "Cla1Task2" to symbol "CLAsincosTable_Coef0" overflowed; the 25-bit relocated address 0x1001e54 is too large to encode in the 16-bit unsigned field (type = 'R_CLA_ABS16' (16), file = "./cla_cos.obj", offset = 0x00000074, section = "Cla1Prog:Cla1Task2")
<a href="http://processors.wiki.ti.com/index.php/Compiler/diagnostic_messages/17003">#17003-D</a>  relocation from function "Cla1Task2" to symbol "CLAsincosTable_Coef1_pos" overflowed; the 25-bit relocated address 0x1001e58 is too large to encode in the 16-bit unsigned field (type = 'R_CLA_ABS16' (16), file = "./cla_cos.obj", offset = 0x0000006e, section = "Cla1Prog:Cla1Task2")
<a href="http://processors.wiki.ti.com/index.php/Compiler/diagnostic_messages/17003">#17003-D</a>  relocation from function "Cla1Task2" to symbol "CLAsincosTable_Coef2" overflowed; the 25-bit relocated address 0x1001e5a is too large to encode in the 16-bit unsigned field (type = 'R_CLA_ABS16' (16), file = "./cla_cos.obj", offset = 0x00000064, section = "Cla1Prog:Cla1Task2")
<a href="http://processors.wiki.ti.com/index.php/Compiler/diagnostic_messages/17003">#17003-D</a>  relocation from function "Cla1Task2" to symbol "CLAsincosTable_Coef3_neg" overflowed; the 25-bit relocated address 0x1001e5e is too large to encode in the 16-bit unsigned field (type = 'R_CLA_ABS16' (16), file = "./cla_cos.obj", offset = 0x00000062, section = "Cla1Prog:Cla1Task2")
<a href="http://processors.wiki.ti.com/index.php/Compiler/diagnostic_messages/17003">#17003-D</a>  relocation from function "Cla1Task2" to symbol "CLAsincosTable_TABLE_SIZEDivTwoPi" overflowed; the 25-bit relocated address 0x1001e4e is too large to encode in the 16-bit unsigned field (type = 'R_CLA_ABS16' (16), file = "./cla_cos.obj", offset = 0x00000022, section = "Cla1Prog:Cla1Task2")

Henry,

Baskar was able to reproduce the issue and he is looking into what is going on. We will get back to you as soon as we get a clue. Most likely we will spend more dedicated time on Friday and get back to you. 

Thanks

Aravindhan

Hi, Aravindhan.

I received your message about CPU1 and CPU2 library update. After a simple building test, it works fine. Really appreciate that.

May I ask what's wrong with the old libraries? Did this update on the next c200ware?

Hnery,

Thanks for confirming. I suspect the issue was with the compiler options being used. I will confirm with the engineer who rebuilt this and share with you. 

Thanks

Aravindhan

Thanks for answers!