• State
  • Locked Locked
  • Replies 1 reply
  • Subscribers 98 subscribers
  • Views 409 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

using SDcard on 6748, fread is slower than fwrite

cobsonchael
Mastermind 7930 points
Other Parts Discussed in Thread: SYSBIOS

So here is what is going on.

I am using

SYSBIOS 6.35.0.20

XDC 3.24.6.63

BIOSPSP 3.0.1.0

EDMA 2.11.6.1

We are using the standard BIOSPSP drivers to run the SDCard with FatFS.  we have not modified the drivers. the only alteration made was to enable LFN which required a rebuild of SYSBIOS. we are using a custom board with a 6748 DSP and a 4GB microSD card

So we have the SDCard working with FatFS and it has been for a while. we are having a performance issue. for one the data transfer isn't great. we are getting .2325 MB/s with out writes.  this is pretty sad.

the second issue is that reads (fread) is at LEAST 4x slower than writes (fwrite). 

my questions are

1) How do i improve performance on this thing? is the basic PSP drives just crap and a lot of modification needs to be made to it to actually make this thing usable?

2) why on earth are reads slower than writes?  reads should be faster.

i'm sure more information is needed so let me know how i can help you help me.

  • 0
    TI__Mastermind 43955 points

    Hi cobsonchael,

    cobsonchael says said:
    2) why on earth are reads slower than writes?  reads should be faster.

    As per the I/O READ WRITE Performance data measured and projected on BIOSPSP drivers in C6748_BIOSPSP_Datasheet.pdf ( http://processors.wiki.ti.com/index.php/OMAP-L138_Software_Design_Guide#BIOS_Drivers: ) , the READ seems to be performed faster than WRITE.

    In general, my suggestion is to debug the code and analyze where the time consumption is more during one complete cycle of READ operation ( i.e., starting from the application level to the hardware layer of the peripheral. Once we get to know in which layer ( DDA, DDC and CSLR ) of the driver consumes more time during one single READ operation, will get some insights to improve upon the speed. 

     

    Regards,

    Shankari.

     

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

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