C6000 Single Core DSP

Forum Post: RE: c6457 Mcbsp used as SPI doubt

yue zhang2 — Sat, 25 May 2013 00:45:00 GMT

RandyP

my TMS320SEVM6457 was producted by einfochip company, and there was a CD in the box which contant the CSL of the board. in the CSL example i cannot find the exactly SPI code, but there are three example of MCBSP which show as follow picture, and i refer to the single channel code because i only need single channel, but i only use some of the code like

CSL_mcbspHwControl (hMcbsp, CSL_MCBSP_CMD_RESET_CONTROL, &ctrlMask);

and the fuction of CSL_mcbspHwSetupRaw (hMcbsp, &MyConfig) from the very begining is write by myself. the while(1) is also write by myself, all i want to do is keep the spi in the dead loop which it can always transmit the code writen in the CSL_mcbspWrite() and i can get continual behavior. as i said, when i close the GRST, i can get proper behavior on the OSC, but when i set GRST as 1, i get nothing. i wonder if you can check the register is configured right or not in my code?

hope for your reply!

Regards,

zhang

Forum Post: RE: C6748 internal SRAM speed : L1D compared to shared RAM.

RandyP — Fri, 24 May 2013 17:26:00 GMT

[quote user="Processor"]One thing I don't understand; if I don't use the cache why would switching it off make any difference ?[/quote]

L1D cache is used to cache the L2 SRAM space, so that is why your performance went down when you turned off caching and put all the MSMCSRAM buffers into L1D SRAM.

Moving from MSMCSRAM to L2 would likely give some improvement in performance unless the prefetch buffers are helping out a lot. My guess is that moving to L2 would give an improvement. You could try moving as much as you can move easily, and see if there is any improvement.

Another option would be to leave it all in MSMCSRAM and use EDMA to copy portions from MSMCSRAM to L1D SRAM (using L1D as 50-50 cache and SRAM).

Or you can set the L1D cache to 8KB which would give you plenty for your 20KB buffers in MSMCSRAM.

Trying out all these options is the best way to find the best solution. Of course, the EDMA method would require program changes, so that would not be nearly as easy as the other ones.

For all the options where MSMCSRAM is still being used and some L1D cache is enabled, make sure the MAR bits are set. I think that is the default, or maybe required setting.

Regards,
RandyP

Forum Post: RE: Bundle or full License of CCS with OMAP L137 EVM

Rahul Prabhu — Fri, 24 May 2013 16:00:00 GMT

Hi Vitalii,

The CCS version distributed with the OMAPL137 kit today is CCSv3.3. For new development, we recommend using CCSv5. There is a free version of the license for CCSv5 that you can obtain which allows you to use inbuilt emulation on OMAPL137. Please refer to the following wiki for licensing question:

http://processors.wiki.ti.com/index.php/Licensing_-_CCS

Does this help?

Could you also share with us the reason for which you are considering OMAPL137/C6745 over OMAPL138/C6748. If you have not looked at it, I would recommend you to look at OMAPL138/C6748 platform as from software perspective, it has updated software support like low level drivers using Starterware, support for linux kernel 3.3 and support for SYSBIOS drivers.

Regards,

Rahul

Forum Post: RE: C6748 UART REVID1

Shankari G — Fri, 24 May 2013 13:21:00 GMT

Hi Nicolas,

Your findings are right. Thanks for pointing this out. The UART2 REVID at 0x01D0D028 reads the value 0x44141102h.

We will correct it in TRM.

Regards,

Shankari.

-------------------------------------------------------------------------------------------------------

Please click the Verify Answer button on this post if it answers your question.
--------------------------------------------------------------------------------------------------------

Forum Post: RE: Problem with Gel

Sivaraj Kuppuraj — Fri, 24 May 2013 12:20:00 GMT

Hi,

Thanks for your post.

I understand from the above code snippet on the GEL file is as below:

The identifier "gain" is not referenced properly in the code. So, it throws the error message as "identifier not found:gain"

Please make sure whether the identifier "gain" is declared properly in the GEL file, so that, proper memory would be allocated for the identifier "gain".

---------------------------------------------------------------------------------
Please click the Verify Answer button on this post if it answers your question.
---------------------------------------------------------------------------------

Thanks & regards,

Sivaraj K

Forum Post: RE: external event trigger edma fail

RandyP — Fri, 24 May 2013 11:06:00 GMT

Weslly,

Are you using the CSL functions to initialize and operate the EDMA3?

Have you started from a working example for the EDMA?

Most likely, EMRH and SERH are set, and if that is true then no further use of that channel is possible until they are cleared. If I recall correctly, a useful sequence for clearing those is either

1. Repeatedly clear SER/SERH until it returns as 0, then clear EMR/EMRH.

2. Clear SER/SERH four times, then clear EMR/EMRH.

Option 2 is faster because it does not require reads of the SER/SERH registers over the slow Config Bus. I always add code to do option 2 to my initialization sequences.

Regards,
RandyP

Forum Post: RE: C67xx SPI Flash Booting (16bit or 24bit addressing)

Robert Weekly — Thu, 23 May 2013 19:51:00 GMT

Hi Norman,

Thanks for your responses on this. Indeed the processor in question is the C6722, I think this is why I am so confused - as the C674x series definately appears to support the SPI Flash (with 24 bit addressing) but the C6722 doesn't from what I have read to date.

Regards

Rob

Forum Post: RE: C67x DSPLIB FFT Example

Rahul Prabhu — Thu, 23 May 2013 19:32:00 GMT

The calculation results should be the same for both the functions. Let us know if you are still seeing some issues

Regards

Rahul

Forum Post: RE: Maximum Boot image for C6748 using SPI FLASH

Robert Weekly — Thu, 23 May 2013 15:28:00 GMT

Thanks Jason - unfortunately I am using the C6722. I'll keep looking....

Forum Post: RE: Flash Burn LCDK6748 NAND Writer

Ilich Guerrero — Thu, 23 May 2013 15:25:00 GMT

You are a real rock star. Thanks a lot.

Maybe this could be useful in the future. I used the shf_OMAP_L138 method because the method using the project NAND_Writer_DSP didn’t work, well, at least not for me ;-)

Did I say that you are a rock star?

Forum Post: RE: Iam nt getting output

Shankari G — Thu, 23 May 2013 08:13:00 GMT

Hi,

As I have already suggested, I went to TIWikiPages and given the search as "C6713 training" and got the following links.

http://processors.wiki.ti.com/index.php?title=Special%3ASearch&redirs=0&search=C6713+training&fulltext=Search&ns0=1

http://processors.wiki.ti.com/index.php/TMS320C6x1x_DSP_Integration_Workshop

http://software-dl.ti.com/trainingTTO/trainingTTO_public_sw/iw6000/iw6000_workshop_(v3.1a).pdf

http://software-dl.ti.com/trainingTTO/trainingTTO_public_sw/iw6000/iw6000_labs_a nd_solutions_(v3.1a).exe

Regards,

Shankari.

-------------------------------------------------------------------------------------------------------

Please click the Verify Answer button on this post if it answers your question.
--------------------------------------------------------------------------------------------------------

Forum Post: RE: mcbsp, spi and ads1274

Manuel Fernandez — Thu, 23 May 2013 07:47:00 GMT

Hello Mike

I have not designed my own board, I'm working with the dsk6455 kit (it has the TMS320C6455 DSP microprocessor) and the ADS1274EVM. They are connect via the 5-6k interface.

I don't know the Concerto and I unknow if the bus mcbsp is similar to the dsk6455's.

At the begining I worked with FS protocol but I'm working with SPI protocol since a long time.

This is a piece of code where the mcbsp registers are configured.

C64_disableIER(C64_EINT13);

MCBSP_SPCR = 0x00000000;
   MCBSP_RCR = 0x00010080;
   MCBSP_PCR = 0x00000501;
   MCBSP_SRGR = 0x31FF0118;

MCBSP_SPCR = 0x00C02001;

C64_enableIER(C64_EINT13);

You can see that I worked with interrupts, the event that controls mcbsp0 is 40.

That code was working but I don't use it since some years ago.

I hope this can help you.

I'm sorry but my english is little.

Best regards.

Manuel Fernández Ros

Forum Post: RE: dsplib: efficient fft computation of real input

Markus Grunwald — Thu, 23 May 2013 06:40:00 GMT

Since I just have to "undo" the multiplication with the complex As and Bs of the algorithm it's in fact easy: I don't expand the second haf in the forward fft, so I just have to divide by the As and Bs in the inverse fft. That's actually multiplying by their complex conjugate and that is described in the mentioned document - you just have to find it :)

Forum Post: RE: C6455+ NAND flash ??

RandyP — Wed, 22 May 2013 23:24:00 GMT

Shine Lee,

Welcome to the TI E2E forum. I hope you will find many good answers here and in the TI.com documents and in the TI Wiki Pages. Be sure to search those for helpful information and to browse for the questions others may have asked on similar topics.

Some of our more recent DSPs do include direct support features to make it easy to interface to a NAND device. The C6455 does not include those features.

But the EMIFA has standard control signals for reading and writing to an external device, and these could be used to interface to a NAND device. You would have to write the drivers yourself, or search for a TI Third Party who has done this; I am not aware of any TI Third Party companies who have done this, but I do not know all of them.

Refer to the datasheet and application notes from the NAND device vendor to understand how the interface must be designed and operated. Then you can match that requirement to the available address, control, and data lines of the C6455.

It is my understanding that the primary logic we add in some devices is the ability to maintain a checksum of the data received so that errors can be detected and/or corrected. This would have to be done in your software drivers.

There may be generic NAND drivers available on the web. Whether available or not, this would not be a trivial task, but it would also not be a year's worth of effort for you or your team to develop this.

Regards,
RandyP

Forum Post: RE: Process technologies

RandyP — Wed, 22 May 2013 23:03:00 GMT

Nick,

The OMAP-L138 is fabricated in CMOS technology.

I will let someone else address your question on SEL.

What is the term LETth? It did not find anything on an internet search.

Regards,
RandyP

Forum Post: RE: It doesn't display the text i want...

RandyP — Wed, 22 May 2013 22:39:00 GMT

Danny,

If you use printf, or any form like fprintf, in an embedded application, you will likely lose your synchronization to real-time events when those functions execute. This is why we recommend using System_printf instead.

Please refer to the SYS/BIOS documentation for more information on System_printf, or by using the CCSv5 Help->Search tool.

Regards,
RandyP

Forum Post: RE: Converting C5515 Assembly code to C6727B Processorver

Honey Sukesan — Wed, 22 May 2013 13:15:00 GMT

Hi,

Is there any update on the above ported C code?

I am also stuck with a similar problem.

I am trying to run equivalent C function in C5515 DSP itself.

But my equivalent C function is producing results not comparable with the assembly function.

My ultimate aim is to port it to C64x DSP. But before that, i am trying to analyse if my equivalent function

in C is producing the same result with C5515 itself. Please share your thoughts.

Thanks,

Honey S

Forum Post: RE: 6727 Failure to enter boundry scan

Eddy FOK MAN — Wed, 22 May 2013 09:19:00 GMT

Hello,

I have the same problem. May i ask you what you have dont to resolve it.

Thanks,

Eddy

Forum Post: RE: C6748 Subsystem register address!

Norman Wong — Wed, 22 May 2013 05:39:00 GMT

For CSR access, suggest you look at the special features of the compiler. See:

TMS320C6000 Optimizing Compiler v7.4
User's Guide
Literature Number: SPRU187U July 2012
6.5.2 The cregister Keyword
7.5.8 Using Intrinsics for Interrupt Control and Atomic Sections

You could include c6x.h and access CSR as a special extern variableor use the intrinsics _disable_interrupts(), _enable_interrupts() and _restore_interrupts(). Never used any of these myself. I don't know enough of BIOS to say if you are breaking the BIOS way of doing things. No guarantees.

Forum Post: RE: CCS installation error

Robert Applebee — Tue, 21 May 2013 22:19:00 GMT

I just found a thread on this error in the code composer forum. Will try downloading latest installation.