Hi! Here are the registers for the "two line serial":
Reg 0x100 WRITE Bits 6-0 CHIPADDRESS Bit 7 Reserved Bits 15-8 BASE ADDRESS Bits 23-16 DATA1 Bit 24 Start transaction. Bit 25 R/#W Read=1, Write=0 Bit 27-26 Size of transaction. 00=1, 11=4 bytes. Bit 29-28 Port selection 00=Invalid, 01=Frontend, 10-EEPROM, 11=Tuner Bit 30 Base address presence indicator. When 1, no base address is generated. Bit 31 Reserved Reg 0x100 READ Bit 0 R/#W - Internal states. Bits 7-1 CHIPADDRESS Bits 15-8 BASE ADDRESS Bits 23-16 DATA1 Bit 24 Status. When=1, transaction underway. =0 engine is free. Bit 25 R/#W Read=1, Write=0 Bit 27-26 Size of transaction. 00=1, 11=4 bytes. Bit 29-28 Port selection 00=Invalid, 01=Frontend, 10-EEPROM, 11=Tuner Bit 30 ACK error. When =1, no ACK from slave. Bit 31 When set, it indicate that the current transaction is done. Reg 0x104 Bit 7-0 Data2 Bit 15-8 Data3 Bit 23-16 Data4 Bit 24 Explicit STOP. When set, the transaction is done without a trailing STOP. FC3 only. Bit 25 When set, generates a STOP event. FC3 only. Bit 31-26 Reserved. Reg 0x108, 0x10c and 0x110 Have the same function, but affecting ports 1(Frontend), 2(EEPROM) and 3(Tuner), respectively. Bits 5-0 thi Time High for i2c timing. The reference is a pre-scale internal clock based on the main clock. Bits 12-8 tlo Time Low for i2c timing. The reference is a pre-scale internal clock based on the main clock. All other bits, reserved. I hope this is not too confusing. Let me know if you have any questions. Augusto *********** REPLY SEPARATOR *********** On 12/9/2003 at 9:47 AM Niklas Peinecke wrote: >Hi, > >well the one and only time that it occured was scary enough (I had to do >a reset of the system and fschk then)! But maybe I'll make a backup of >my system and then have a few more tries. >What do all the registers in i2c_main_write_for_flex2 mean? I pretend >0x40000000 is some kind of "write success" but what are 0x80000000 and >0x01000000 ? > >Niklas > >Augusto Cardoso wrote: >> Hi! >> >> That's what I was thinking, but I tested in a system with multiple 2.6 >only and it was OK. I only tried couple of times. Because I didn't get a >failure, I started thinking about race conditions or other exotic >problems. It could as well be something very simple that for some reason >don't show up in my test. >> My development system has multiple card types. >> >> Can you reproduce this every time? >> >> Augusto >> >> *********** REPLY SEPARATOR *********** >> >> On 12/9/2003 at 9:22 AM Niklas Peinecke wrote: >> >> >>>Hi Augusto, >>> >>>maybe this is just _because_ you have multiple cards? Maybe the mt312 >>>wants some feedback ;) >>> >>>Niklas >>> >>>Augusto Cardoso wrote: >>> >>>>Hi! >>>> >>>>Thanks for the info. >>>> >>>>Does it happen every time? >>>>I have multiple cards in my system and I always load the mt312 and the >>> >>>stv0299. The mt312 probe all cards and finds the right ones. Then the >>>stv0299 find the other cards. >>> >>>>This sounds like a race condition or some other odd situation. >>>>Since I wasn't able to reproduce yet, I'll take another look at the >code. >>>> >>>>Thanks. >>>> >>>> Augusto >>>> >>>>*********** REPLY SEPARATOR *********** >>>> >>>>On 12/9/2003 at 8:18 AM Niklas Peinecke wrote: >>>> >>>> >>>> >>>>>Augusto Cardoso wrote: >>>>> >>>>> >>>>>>Hi! >>>>>> >>>>>>I haven't been able to reproduce this problem. >>>>>>Can some one that has the problem provide some information? >>>>>>For example, order that the modules are loaded, some system >information, >>>>> >>>>>what version of the driver and kernel, etc. >>>>> >>>>>I experienced this last week: >>>>> >>>>>System: K6/2 400Mhz, Skystar2 rev2.6B with FlexcopII, Suse 8.0 (with >>>>>losts of non-standard extensions), kernel 2.4.23, driver dvb-kernel >from >>>>>CVS >>>>> >>>>>I compile the driver using getlinks && make >>>>>and then insmod.sh >>>>>All the modules are loaded without problems (of course, because most >of >>>>>them are unused) then skystar2 and mt312 are loaded. mt312 tries to >>>>>attach itself to skystar2 and hangs during initialize. Both modules >are >>>>>marked as "used" and can therefor no more be removed. >>>>>Nothing to see on dmesg. >>>>> >>>>>Things work without problems when mt312 is left out. >>>>> >>>>>Niklas >>>>> >>>>> >>>>> >>>>>-- >>>>>Info: >>>>>To unsubscribe send a mail to [EMAIL PROTECTED] with "unsubscribe >>> >>>linux-dvb" as subject. >>> >>>> >>>> >>>> >>>> >>>> >>> >>> >>> >>>-- >>>Info: >>>To unsubscribe send a mail to [EMAIL PROTECTED] with "unsubscribe >linux-dvb" as subject. >> >> >> >> >> >> > > > > >-- >Info: >To unsubscribe send a mail to [EMAIL PROTECTED] with "unsubscribe linux-dvb" as >subject. -- Info: To unsubscribe send a mail to [EMAIL PROTECTED] with "unsubscribe linux-dvb" as subject.