README.cma286 2.9 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869
  1. CPU module revisions
  2. --------------------
  3. My cpu module has the model number "CMA286-60-990526-01". My motherboard
  4. has the model number "CMA102-32M-990526-01". These are both fairly old,
  5. and may not reflect current design. In particular, I can see from the
  6. Cogent web site that the CMA286 has been significantly redesigned - it
  7. now has on board RAM (4M), ethernet 10baseT PHY (on SCC2), 2 serial ports
  8. (SMC1 and SMC2), and 48pin DIN for the FEC (if present i.e. MPC860T), and
  9. also the EPROM is 512K.
  10. My CMA286-60 has none of this, and only 128K EPROM. In addition, the CPU
  11. clock is listed as 66MHz, whereas mine is 33.333MHz.
  12. Clocks
  13. ------
  14. Quote from my "CMA286 MPC860/821 User's Manual":
  15. "When setting up the Periodic Interrupt Timer (PIT), be aware that the
  16. CMA286 places the MPC860/821 in PLL X1 Mode. This means that we feed
  17. a 25MHz clock directly into the MPC860/821. This mode sets the divisor
  18. for the PIT to be 512. In addition, the Time Base Register (TMB)
  19. divisor is set to 16."
  20. I interpreted this information to mean that EXTCLK is 25MHz and that at
  21. power on reset, MODCK1=1 and MODCK2=0, which selects EXTCLK as the
  22. source for OSCCLK and PITRTCLK, sets RTDIV to 512 and sets MF (the
  23. multiplication factor) to 1 (I assume this is what they mean by X1
  24. mode above). MF=1 means the cpus internal clock runs at the same
  25. rate as EXTCLK i.e. 25MHz.
  26. Furthermore, since SCCR[TBS] (the Time Base Source selector bit in the
  27. System Clock and Reset Control register) is set in the cpu initialisation
  28. code, the TMBCLK source is forced to be GCLK2 and the TMBCLK prescale is
  29. forced to be 16. This results in TMBCLK=1562500.
  30. One problem - since PITRTCLK source is EXTCLK (25Mhz) and RTDIV is 512,
  31. PITRTCLK will be 48828.125 (huh?). Another quote from the MPC860 Users
  32. Manual:
  33. "When used by the real-time clock (RTC), the PITRTCLK source is first
  34. divided as determined by RTDIV, and then divided in the RTC circuits by
  35. either 8192 or 9600. Therefore, in order for the RTC to count in
  36. seconds, the clock source must satisfy:
  37. (EXTCLK or OSCM) / [(4 or 512) x (8192 or 9600)] = 1
  38. The RTC will operate with other frequencies, but it will not count in
  39. units of seconds."
  40. Therefore, the internal RTC of the MPC860 is not going to count in
  41. seconds, so we must use the motherboard RTC (if we need a RTC).
  42. I presume this means that they do not provide a fixed oscillator for
  43. OSCM. The code in get_gclk_freq() assumes PITRTCLK source is OSCM,
  44. RTDIV is 4, and that OSCM/4 is 8192 (i.e. a ~32KHz oscillator). Since
  45. the CMA286-60 doesn't have this (at least mine doesn't) we can't use
  46. the code in get_gclk_freq().
  47. Finally, it appears that the internal clock in my CMA286-60 is actually
  48. 33.333MHz. Which makes TMBCLK=2083312.5 (another huh?) and
  49. PITRTCLK=65103.515625 (bloody hell!).
  50. If anyone finds anything wrong with the stuff above, I would appreciate
  51. an email about it.
  52. Murray Jensen <Murray.Jensen@csiro.au>
  53. 21-Aug-00