On Sat, 10 Jul 2010, Bob Copeland wrote: > Greetings! > > The Rio Karma USB storage driver broke for me in recent kernels, while > normal USB storage works fine. I bisected it back to this commit: > > commit 0ede76fcec5415ef82a423a95120286895822e2d > Author: Alan Stern <stern@xxxxxxxxxxxxxxxxxxx> > Date: Fri Mar 5 15:10:17 2010 -0500 > > USB: remove uses of URB_NO_SETUP_DMA_MAP > > ...but this didn't cleanly revert on current head (and I'm not sure > that bisect makes sense). In theory it could cause a problem, but I don't see how. More debugging would help. > The "dma XXX already free" line is new, does > this ring any bells? Sounds like a double free but I didn't see > anything obvious. Karma uses us->iobuf for its command transfers. > > dmesg on broken kernel: > > [ 2283.257349] Initializing USB Mass Storage driver... > [ 2283.257951] usbcore: registered new interface driver usb-storage > [ 2283.257955] USB Mass Storage support registered. > [ 2283.436633] ehci_hcd 0000:00:1d.7: dma_pool_free buffer-128, dma a0bd000 already free > [ 2283.436837] ums-karma: probe of 5-3:1.0 failed with error 1 > [ 2283.437101] usbcore: registered new interface driver ums-karma What does usbmon reveal? Or CONFIG_USB_STORAGE_DEBUG? Alan Stern -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
