[U-Boot] [PATCH] OMAP3 SPI : Fixed bugs related to SPI transfer

[Ajoy Das]

Hi Peter
  I described the entire patch in small paragraphs
  so it is better to understand
   
  The first paragraph is for posted writes
  (courtesy linux OMAP3 SPI driver code)

  The last three paragraphs actually fixes the
  bug related to the operation of the OMAP3 
  SPI controller in TX_ONLY and RX_ONLY mode.
   
  Only after implementing the those three things
  'together' the TX_ONLY and RX_ONLY transfers works perfectly.
  The problem occurs when the RX transfer is immediately
  followed by the TX transfer.  

  So shall I divide them in four patches  according to the paragraphs
  Or shall I make two patches according to the functionality getting fixed 

  Please suggest

Ajoy



________________________________
 From: Peter Korsgaard <jacmet at sunsite.dk>
To: Ajoy Kumar Das <akdas75 at yahoo.in> 
Cc: u-boot at lists.denx.de; Tom Rini <trini at ti.com>; jacopo mondi <j.mondi at voltaelectronics.com> 
Sent: Sunday, 18 November 2012 5:23 AM
Subject: Re: [U-Boot] [PATCH] OMAP3 SPI : Fixed bugs related to SPI transfer
 
>>>>> "Ajoy" == Ajoy Kumar Das <akdas75 at yahoo.in> writes:

Ajoy> From: ajoy <akdas75 at yahoo.in>
Ajoy> Added posted writes (read after writes) to effect the
Ajoy> change immediately for channel confiuration and channel
Ajoy> enable register

Ajoy> Disable the channel to purge receieve data in TX_ONLY
Ajoy> mode transfer otherwise rx data will get affected by
Ajoy> the next immediate RX_ONLY mode transfer

Ajoy> Wait for the EOT bit to be set after last byte has been
Ajoy> loaded to TX shift register in the the TX_ONLY mode.This
Ajoy> ensures TX data has been completely shifted out

Ajoy> Disable the channel in RX_ONLY mode before reading the
Ajoy> last data from  RXX register to prevent the SPI slave
Ajoy> to transmit next word

So it is 4 separate fixes? Could you please split it up in 4 separate
patches?

-- 
Bye, Peter Korsgaard