On Thu, Aug 27, 2015 at 12:59:31AM +0200, Paul Kocialkowski wrote: > Le mercredi 26 août 2015 à 09:23 -0400, Tom Rini a écrit : > > On Wed, Aug 26, 2015 at 12:51:07PM +0200, Paul Kocialkowski wrote: > > > > [snip] > > > On all OMAP platforms, USB peripheral loading is done via bulk USB and I > > > suspect it is the same on am33xx platforms. > > > > Actually no! :( > > > > On OMAP3/4/5 (so dra7xx/am57xx) it's the same bulk USB. On > > AM33xx/AM43xx it's (in ROM) USB RNDIS. That's why on am33xx/am43xx we > > continue on with USB RNDIS (which is what SPL_USBETH_SUPPORT implements, > > not the other USB "ethernet" class). > > Interesting. So with this in mind, would it be acceptable to assume that > BOOT_DEVICE_USB and BOOT_DEVICE_USBETH are always two distinct boot > devices and thus not have to care about whether they're equal or not and > just assume they're not, like any other BOOT_DEVICE? > > This would greatly simplify the logic and make everything more readable. > > I can submit a new patch set in that direction tomorrow.
Yes, lets go with that. -- Tom
signature.asc
Description: Digital signature
_______________________________________________ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
