I'm confused the gianfar driver is for the MPC8540 FEC. I believe the initial question was regarding the FCC (on CPM / MPC8560).
As to why your having poor performance on the MPC8540 FEC, that seems odd -- possibly some issue with the Intel PHY. I do not believe we are seeing anything similar on the ADS. - kumar On Mar 11, 2004, at 11:14 AM, Matthew S. McClintock wrote: > > On Thu, 2004-03-11 at 10:16, Dan Malek wrote: > >> Do you keep up to date with the linuxppc trees? Some of us have been >> pushing lots of CPM updates lately. We recently updated all of the >> common code for 82xx and 85xx. If this doesn't work there must be >> some I/O or board configuration that is amiss. > > I'm running the latest linuxppc-2.4 branch. I'm not using the > MPC8540ADS > board, the board I am using has an Intel LXT971A PHY attached to the > FCC. I've added support for that physical layer to > drivers/gianfar_phys.c. I'm guessing something is wrong with that code > that is making it go so slow, but I have been unable to pinpoint what > might be wrong. > > I've noticed that the interrupts for the FCC come in at a very very > slow > pace, that is the only thing I have really noticed that is odd. I've > also tried changing the priority of the interrupt via the IIVPR > register > for the FCC with no results. I can't think of anything else I can > investigate, unless you have any ideas? > > Thanks, > Matthew > > -- > Matthew S. McClintock <mattsm at arlut.utexas.edu> > > ** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/
