Duck wrote: > The USB blocks in the 8349 and mx31 are basically the same, > with some minor exceptions. The 8349 has one host controller > with two ports plus a dual-role (OTG) controller. The mx31 > has two host controllers (one port each), plus an OTG controller. > The 8349 has some additional snooping registers not needed on > mx31.
Ok. I had noticed the difference in the types of block. So we can assume that the inner of the USB host controller are the same on 8349 and mx31 (if you don't take the snooping registers into account). > > > Since MPC834x USB module seems to be already quite well advanced, I would > > be happy to try to port the work which was done for i.MX31 in order to > > help i.MX31 mainline support, or at least help me to have a functionnal > > USB support on our i.MX31 hardware. > > This is exactly what I'm working on. Good to hear. I don't know how things are going at Freescale for mx31 mainline support. I have seen a few patches going on the arm-linux-kernel mailing list for the mx31 support, but we don't have that much information. This is quite frustrating because we are also doing some tests and developping software for our device, so it would be nice to see how things are evolving and what peripherals get a better support than Freescale first release to avoid developping things twice. For instance we need a better implementation of the CSPI driver but now I'm hesitating to invest myself into it because someone at Freescale may be doing it ... > > The limitation exists in the MIC2536 power switches used on > the mx31ads board, which provide only 150mA per port. This is > reflected in the 'power_budget' fields of the platform_data > structs found in arch/arm/mach-mx3/usb.c, and passed to the > generic ehci code (as Alan points out) in drivers/usb/host/ehci-arc.c: > hcd->power_budget = config->power_budget; > Ok. We are going to use power switches which provide more current per port (500 mA for the LM3526). Alan Stern wrote: > > c->desc.bMaxPower is what the hub makes available for each port and > > udev->bus_mA is what the device requests. > > It's the reverse: (c->desc.bMaxPower * 2) is what the device requests and > udev->bus_mA is what the hub makes available. Ok. So the hub code rejects a device needing 150 mA (which resluts in 300) althoug the MIC2536 is able to provide 150 mA. For me it rejects a configuration that should be accepted, doestn't it ? Thank you for all the information. Regards Valentin Longchamp ------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV _______________________________________________ linux-usb-devel@lists.sourceforge.net To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
