On Wed, 2014-04-23 at 17:16 +0300, Diana Craciun wrote: > On 04/19/2014 12:33 AM, Scott Wood wrote: > > On Fri, 2014-04-18 at 18:11 +0300, Diana Craciun wrote: > >> From: Diana Craciun <Diana.Craciun@xxxxxxxxxxxxx> > >> > >> The CoreNet coherency fabric is a fabric-oriented, conectivity > >> infrastructure that enables the implementation of coherent, multicore > >> systems. The CCF acts as a central interconnect for cores, > >> platform-level caches, memory subsystem, peripheral devices and I/O host > >> bridges in the system. > >> > >> Signed-off-by: Diana Craciun <Diana.Craciun@xxxxxxxxxxxxx> > >> --- > >> .../devicetree/bindings/powerpc/fsl/ccf.txt | 36 ++++++++++++++++++++++ > >> 1 file changed, 36 insertions(+) > >> create mode 100644 Documentation/devicetree/bindings/powerpc/fsl/ccf.txt > >> > >> diff --git a/Documentation/devicetree/bindings/powerpc/fsl/ccf.txt b/Documentation/devicetree/bindings/powerpc/fsl/ccf.txt > >> new file mode 100644 > >> index 0000000..f0b7143 > >> --- /dev/null > >> +++ b/Documentation/devicetree/bindings/powerpc/fsl/ccf.txt > >> @@ -0,0 +1,36 @@ > >> +Freescale CoreNet Coherency Fabric(CCF) Device Tree Binding > >> + > >> +DESCRIPTION > >> + > >> +The CoreNet coherency fabric is a fabric-oriented, connectivity infrastructure > >> +that enables the implementation of coherent, multicore systems. > >> + > >> +Required properties: > >> + > >> +- compatible : <string> > >> + Must include "fsl,corenetX-cf", "fsl,corenet-cf" - CoreNet coherency > >> + fabric version X > > Specify "fsl,corenet1-cf" and "fsl,corenet2-cf" rather than > > "fsl,corenetX-cf" (given there's nothing in a chip manual that you can > > correlate with the value of X), and provide example chips for each. > > OK. > > > Also specify that "fsl,corenet-cf" represents the registers that are > > common between the two versions (not arbitrary "fsl,corenetX-cf" -- if > > there's ever an "fsl,corenet3-cf" it may not be compatible with this), > > and is retained for compatibility reasons. > > > > What do you mean by common? There are the csdids and snoop ids > registers which are common between the two versions but only by name > because the register format is not the same. The only difference I see is that corenet2-cf documents certain bits as being for core clusters, and another bit as being for the PAMU, whereas corenet1-cf opaquely describes all the bits as "port id". This isn't really a change, just a documentation difference plus a difference in how cores and PAMUs map to port ids (this differs within corenet1-cf chips as well, as it's based on the number of cores and the number of pamus). That mapping should be expressed in the device tree binding somehow. -Scott -- To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
