Multiple ic2 muxes in a device tree

Multiple ic2 muxes in a device tree

[James Baldwin]

[-- Attachment #1.1: Type: text/plain, Size: 4644 bytes --]

I have tried many different combinations based on what I've been able to
find on

the internet over the last week and have reached a dead end.

 

I have the following i2c topology and specify the devices as follows:
Linux 2.6.38.5

only seems to recognize the devices connected directly to the i2c
controller when it boots.

 

I can add the muxes and lm73/m41st85 manually using the echo command
from the console:

"echo pca9547 0x70 > /sys/bus/12c/devices/i2c-2/new_device"

 

This tells me I have three prime candidates for potential issues.

1)  The device tree specification is incorrect

2)  The device tree processing is incorrect (It is 'appears'
correct in the /proc/devicetree hierarchy)

3)  The probing code for i2c devices doesn't currently traverse muxes
properly.

 

As validating the device tree should be the simplest to eliminate (and
is the part I'm

Directly involved with), I'd appreciate any Constructive comments on
solving this problem.

 

             |

i2c ctrl ----|

             |---lm73@0x49

             |

             |---24c02@0x54

             |

             |---pca9547@0x77---|

             |                  |-Chan 0---|---pca9547@0x70

             |                  |          |

             |                  |

             |                  |-Chan 1---|---pca9547@0x71

             |                  |          |

             |                  |

             |                  |-Chan 2---|---pca9547@0x72

             |                  |          |

             |                  |

             |                  |-Chan 3---|---pca9547@0x73

             |                  |          |

             |                  |

             |                  |-Chan 4---|---pca9547@0x74

             |                  |          |

             |                  |

             |                  |-Chan 5---|---pca9547@0x75

             |                  |          |

             |                  |

             |                  |-Chan 6---|---m41st85@0x68

             |                  |

             |                  |-Chan 7---|---lm73@0x48

             |                  |

 

i2c@3100 {

        #address-cells = <1>;

        #size-cells = <0>;

        cell-index = <1>;

        compatible = "fsl-i2c";

        reg = <0x3100 0x100>;

        interrupts = <15 0x8>;

        interrupt-parent = <&ipic>;

        dfsrr;

 

        temp-sensor@49 {

                compatible = "national,lm73";

                reg = <0x49>;

        };

 

        eeprom@54 {

                compatible = "at24,24c02";

                reg = <0x54>;

                pagesize = <4>;

        };

 

        mux@77 {

                #address-cells = <2>;

                #size-cells = <0>;

                compatible = "philips,pca9547";

                reg = <0x77>;

 

                mux@0,70 {

                        #address-cells = <2>;

                        #size-cells = <0>;

                        compatible = "philips,pca9547";

                        reg = <0 0x70>;

                };

 

                mux@0,71 {

                        #address-cells = <2>;

                        #size-cells = <0>;

                        compatible = "philips,pca9547";

                        reg = <1 0x71>;

                };

 

                mux@0,72 {

                        #address-cells = <2>;

                        #size-cells = <0>;

                        compatible = "philips,pca9547";

                        reg = <2 0x72>;

                };

 

                mux@0,73 {

                        #address-cells = <2>;

                        #size-cells = <0>;

                        compatible = "philips,pca9547";

                        reg = <3 0x73>;

                };

 

                mux@0,74 {

                        #address-cells = <2>;

                        #size-cells = <0>;

                        compatible = "philips,pca9547";

                        reg = <4 0x74>;

                };

 

                mux@0,75 {

                        #address-cells = <2>;

                        #size-cells = <0>;

                        compatible = "philips,pca9547";

                        reg = <0 0x75>;

                };

 

                temp-sensor@7,48 {

                        compatible = "national,lm73";

                        reg = <7 0x48 >;

                };

 

                rtc@6,68 {

                        compatible = "stm,m41st85";

                        reg = <6 0x68>;

                };

        };

 

};

 

 

Thanks,

Jim Baldwin




[-- Attachment #1.2: Type: text/html, Size: 31291 bytes --]

[-- Attachment #2: Type: text/plain, Size: 192 bytes --]

_______________________________________________
devicetree-discuss mailing list
devicetree-discuss-uLR06cmDAlY/bJ5BZ2RsiQ@public.gmane.org
https://lists.ozlabs.org/listinfo/devicetree-discuss

RE: Multiple ic2 muxes in a device tree

[Stephen Warren]

James Baldwin wrote at Tuesday, May 10, 2011 1:24 PM:
> I have the following i2c topology and specify the devices as follows:
> i2c@3100 {
>         #address-cells = <1>;
>         #size-cells = <0>;
>         cell-index = <1>;
>         compatible = "fsl-i2c";
>         reg = <0x3100 0x100>;
>         interrupts = <15 0x8>;
>         interrupt-parent = <&ipic>;
>         dfsrr;
> 
>         temp-sensor@49 {
>                 compatible = "national,lm73";
>                 reg = <0x49>;
>         };
> 
>         eeprom@54 {
>                 compatible = "at24,24c02";
>                 reg = <0x54>;
>                 pagesize = <4>;
>         };
> 
>         mux@77 {

Bear in mind I'm very new to devicetree, but in:
http://www.mail-archive.com/linaro-dev-cunTk1MwBs8s++Sfvej+rw@public.gmane.org/msg03910.html
it was suggested to name this node "i2cmux" not "mux".

>                 #address-cells = <2>;
>                 #size-cells = <0>;
>                 compatible = "philips,pca9547";
>                 reg = <0x77>;
> 
>                 mux@0,70 {

... and this node "i2c" not "mux".

The implication was the the node names actually have an effect on the
parsing, and need to match the type of object represented. Does renaming
them following those conventions help?

>                         #address-cells = <2>;
>                         #size-cells = <0>;
>                         compatible = "philips,pca9547";
>                         reg = <0 0x70>;
>                 };
>...

-- 
nvpublic

Re: Multiple ic2 muxes in a device tree

[Mitch Bradley]

I don't know much about how Linux handles i2c hierarchies, but I can speak with some authority about how it the device tree "should" be structured in order to conform to the principles already established for other buses.

Note that *each* pca9547 results in a subtree of the form:

   i2cmux {
      i2c@0 { }
      i2c@1 { }
      i2c@2 { }
      i2c@3 { }
      i2c@4 { }
      i2c@5 { }
      i2c@6 { }
      i2c@7 { }
   }

Each of those 8 "i2c" children represents a complete I2C bus.  When a subordinate pca9547 is connected to one of those buses, the hierarchy expands to:

   i2cmux {
      i2c@0 { 
         i2cmux {
            i2c@0 { }
            i2c@1 { }
            i2c@2 { }
            i2c@3 { }
            i2c@4 { }
            i2c@5 { }
            i2c@6 { }
            i2c@7 { }
         }
      }
      i2c@1 { }
      i2c@2 { }
      i2c@3 { }
      i2c@4 { }
      i2c@5 { }
      i2c@6 { }
      i2c@7 { }
   }

The full device tree representation for the hardware shown below would be:

i2c@3100 {
  #address-cells = <1>;
  #size-cells = <0>;
  cell-index = <1>;
  compatible = "fsl-i2c";
  reg = <0x3100 0x100>;
  interrupts = <15 0x8>;
  interrupt-parent = <&ipic>;
  dfsrr;
  temp-sensor@49 {
    compatible = "national,lm73";
    reg = <0x49>;
  };
  eeprom@54 {
    compatible = "at24,24c02";
    reg = <0x54>;
    pagesize = <4>;
  };
  i2cmux@77 {
    #address-cells = <1>;
    #size-cells = <0>;
    compatible = "philips,pca9547";
    reg = <0x77>;
    i2c@0 {
      #address-cells = <1>;
      #size-cells = <0>;
      compatible = "philips,pca9547-channel";
      reg = <0>;
      i2cmux@70 {
        #address-cells = <1>;
        #size-cells = <0>;
        compatible = "philips,pca9547";
        reg = <0x70>;
        {Potentially 8 Subordinate nodes, each named "i2c", with children}
      }
    };
    i2c@1 {
      #address-cells = <1>;
      #size-cells = <0>;
      compatible = "philips,pca9547-channel";
      reg = <1>;
      i2cmux@71 {
        #address-cells = <1>;
        #size-cells = <0>;
        compatible = "philips,pca9547";
        reg = <0x71>;
         {Potentially 8 Subordinate nodes, each named "i2c", with children}
      };
    };
    i2c@2 {
      #address-cells = <1>;
      #size-cells = <0>;
      compatible = "philips,pca9547-channel";
      reg = <2>;
      i2cmux@72 {
        #address-cells = <1>;
        #size-cells = <0>;
        compatible = "philips,pca9547";
        reg = <0x72>;
        {Potentially 8 Subordinate nodes, each named "i2c", with children}
      };
    i2c@3 {
      #address-cells = <1>;
      #size-cells = <0>;
      compatible = "philips,pca9547-channel";
      reg = <3>;
      i2cmux@73 {
        #address-cells = <1>;
        #size-cells = <0>;
        compatible = "philips,pca9547";
        reg = <0x73>;
        {Potentially 8 Subordinate nodes, each named "i2c", with children}
      };
    i2c@4 {
      #address-cells = <1>;
      #size-cells = <0>;
      compatible = "philips,pca9547-channel";
      reg = <4>;
      i2cmux@74 {
        #address-cells = <1>;
        #size-cells = <0>;
        compatible = "philips,pca9547";
        reg = <0x74>;
        {Potentially 8 Subordinate nodes, each named "i2c", with children}
      };
    };
    i2c@5 {
      #address-cells = <1>;
      #size-cells = <0>;
      compatible = "philips,pca9547-channel";
      reg = <5>;
      i2cmux@75 {
        #address-cells = <1>;
        #size-cells = <0>;
        compatible = "philips,pca9547";
        reg = <0x75>;
        {Potentially 8 Subordinate nodes, each named "i2c", with children}
      };
    i2c@6 {
      #address-cells = <1>;
      #size-cells = <0>;
      compatible = "philips,pca9547-channel";
      reg = <6>;
      rtc@68 {
        compatible = "stm,m41st85";
        reg = <0x68>;
      };
    };
    i2c@7 {
      #address-cells = <1>;
      #size-cells = <0>;
      compatible = "philips,pca9547-channel";
      reg = <7>;
      temp-sensor@48 {
        compatible = "national,lm73";
        reg = <0x48>;
      };
    };
  };
};


I understand the temptation to collapse the hierarchy by combining the mux node with its subordinate buses (using 2 address cells), but in my experience, shortcuts like that have not aged well.  The case where a single-level hierarchy would work would be a simple i2c-to-i2c bridge with only one child bus.  In this case, the existence of multiple independent child buses argues for the two-level hierarchy, satisfying the following requirements:

a) From the standpoint of a parent bus "looking down" at a pca9547, the pca9547 appears to be a single device at a well-defined address.

b) From the standpoint of a child device "looking up" at a pca9547, that child is attached to a specific I2C bus with a standard one-cell I2C address space.

Again, though, I'm unsure that this "correct" hierarchy is compatible with the expectations of Linux drivers.

On 5/10/2011 9:24 AM, James Baldwin wrote:
> I have tried many different combinations based on what I’ve been able to 
> find on
> 
> the internet over the last week and have reached a dead end.
> 
> I have the following i2c topology and specify the devices as follows: 
> Linux 2.6.38.5
> 
> only seems to recognize the devices connected directly to the i2c 
> controller when it boots.
> 
> I can add the muxes and lm73/m41st85 manually using the echo command 
> from the console:
> 
> “echo pca9547 0x70 > /sys/bus/12c/devices/i2c-2/new_device”
> 
> This tells me I have three prime candidates for potential issues.
> 
> 1)The device tree specification is incorrect
> 
> 2)The device tree processing is incorrect (It is ‘appears’
> correct in the /proc/devicetree hierarchy)
> 
> 3)The probing code for i2c devices doesn’t currently traverse muxes 
> properly.
> 
> As validating the device tree should be the simplest to eliminate (and 
> is the part I’m
> 
> Directly involved with), I’d appreciate any Constructive comments on 
> solving this problem.
> 
> |
> 
> i2c ctrl ----|
> 
> |---lm73@0x49
> 
> |
> 
> |---24c02@0x54
> 
> |
> 
> |---pca9547@0x77---|
> 
> | |-Chan 0---|---pca9547@0x70
> 
> | | |
> 
> | |
> 
> | |-Chan 1---|---pca9547@0x71
> 
> | | |
> 
> | |
> 
> | |-Chan 2---|---pca9547@0x72
> 
> | | |
> 
> | |
> 
> | |-Chan 3---|---pca9547@0x73
> 
> | | |
> 
> | |
> 
> | |-Chan 4---|---pca9547@0x74
> 
> | | |
> 
> | |
> 
> | |-Chan 5---|---pca9547@0x75
> 
> | | |
> 
> | |
> 
> | |-Chan 6---|---m41st85@0x68
> 
> | |
> 
> | |-Chan 7---|---lm73@0x48
> 
> | |
> 
> i2c@3100 {
> 
> #address-cells = <1>;
> 
> #size-cells = <0>;
> 
> cell-index = <1>;
> 
> compatible = "fsl-i2c";
> 
> reg = <0x3100 0x100>;
> 
> interrupts = <15 0x8>;
> 
> interrupt-parent = <&ipic>;
> 
> dfsrr;
> 
> temp-sensor@49 {
> 
> compatible = "national,lm73";
> 
> reg = <0x49>;
> 
> };
> 
> eeprom@54 {
> 
> compatible = "at24,24c02";
> 
> reg = <0x54>;
> 
> pagesize = <4>;
> 
> };
> 
> mux@77 {
> 
> #address-cells = <2>;
> 
> #size-cells = <0>;
> 
> compatible = "philips,pca9547";
> 
> reg = <0x77>;
> 
> mux@0,70 {
> 
> #address-cells = <2>;
> 
> #size-cells = <0>;
> 
> compatible = "philips,pca9547";
> 
> reg = <0 0x70>;
> 
> };
> 
> mux@0,71 {
> 
> #address-cells = <2>;
> 
> #size-cells = <0>;
> 
> compatible = "philips,pca9547";
> 
> reg = <1 0x71>;
> 
> };
> 
> mux@0,72 {
> 
> #address-cells = <2>;
> 
> #size-cells = <0>;
> 
> compatible = "philips,pca9547";
> 
> reg = <2 0x72>;
> 
> };
> 
> mux@0,73 {
> 
> #address-cells = <2>;
> 
> #size-cells = <0>;
> 
> compatible = "philips,pca9547";
> 
> reg = <3 0x73>;
> 
> };
> 
> mux@0,74 {
> 
> #address-cells = <2>;
> 
> #size-cells = <0>;
> 
> compatible = "philips,pca9547";
> 
> reg = <4 0x74>;
> 
> };
> 
> mux@0,75 {
> 
> #address-cells = <2>;
> 
> #size-cells = <0>;
> 
> compatible = "philips,pca9547";
> 
> reg = <0 0x75>;
> 
> };
> 
> temp-sensor@7,48 {
> 
> compatible = "national,lm73";
> 
> reg = <7 0x48 >;
> 
> };
> 
> rtc@6,68 {
> 
> compatible = "stm,m41st85";
> 
> reg = <6 0x68>;
> 
> };
> 
> };
> 
> };
> 
> Thanks,
> 
> Jim Baldwin
> 
> 
> 
> *NOTICE:*This email and any files transmitted with it are Enablence 
> confidential and intended solely for the use of the individual or entity 
> to whom they are addressed. If you have received this email in error 
> please notify the sender. This message contains Enablence confidential 
> information and is intended only for the individual named. If you are 
> not the named addressee you should not disseminate, distribute or copy 
> this e-mail. Please notify the sender immediately by e-mail if you have 
> received this e-mail by mistake and delete this e-mail from your system. 
> If you are not the intended recipient you are notified that disclosing, 
> copying, distributing or taking any action in reliance on the contents 
> of this information is strictly prohibited.
> 
> 
> 
> _______________________________________________
> devicetree-discuss mailing list
> devicetree-discuss-uLR06cmDAlY/bJ5BZ2RsiQ@public.gmane.org
> https://lists.ozlabs.org/listinfo/devicetree-discuss

Re: Multiple ic2 muxes in a device tree

[Grant Likely]

On Tue, May 10, 2011 at 11:00 PM, Mitch Bradley <wmb-D5eQfiDGL7eakBO8gow8eQ@public.gmane.org> wrote:
> I don't know much about how Linux handles i2c hierarchies, but I can speak with some authority about how it the device tree "should" be structured in order to conform to the principles already established for other buses.
>
> Note that *each* pca9547 results in a subtree of the form:
[...]
> I understand the temptation to collapse the hierarchy by combining the mux node with its subordinate buses (using 2 address cells), but in my experience, shortcuts like that have not aged well.

... and we should defer to your wisdom in this regard.  :-)

>  The case where a single-level hierarchy would work would be a simple i2c-to-i2c bridge with only one child bus.  In this case, the existence of multiple independent child buses argues for the two-level hierarchy, satisfying the following requirements:
>
> a) From the standpoint of a parent bus "looking down" at a pca9547, the pca9547 appears to be a single device at a well-defined address.
>
> b) From the standpoint of a child device "looking up" at a pca9547, that child is attached to a specific I2C bus with a standard one-cell I2C address space.
>
> Again, though, I'm unsure that this "correct" hierarchy is compatible with the expectations of Linux drivers.

Since Linux doesn't do any of this yet; I think we can make sure Linux
implements something sane.

g.