On Mon, Oct 24, 2011 at 10:40:47PM +1100, Matt wrote: > Hi, > > I've been trying to setup some recently-acquired hardware and > encountered what I perceive to be possible bugs in the xHCI code. > > Specifically, I've got a Shuttle SH51R4, which utilises an ASMedia > ASM1042 chipset (seemingly xHCI compliant) ASMedia is supposed to be complaint to the xHCI spec, but I've never had one of their host controllers to test with. It's entirely possible they have some quirks or vendor-specific responses to xHCI commands. > and I'm trying to connect > a HotWay h82-su3s2 external drive chassis. Inside the chassis, I > have connected four Hitachi Deskstar 7k3000 drives. For > confirmation purposes, I have successfully used this setup in > Windows 7 and achieved well over 100MB/s transfer rates on a > per-drive basis. The config also works when plugged into a USB 2 > port in all Linux Kernel's I've tried. > > Without exception, I've experienced catastrophic failures using the > xHCI drivers in varying kernel versions. Under CentOS 6.0's > customised 2.6.32 kernel, connecting the drive chassis hard-locks > the PC and I would assume has caused a kernel panic. Under more > recent kernels, a volume of errors are produced, notably 'WARN: > Stalled endpoint' which then seems to degenerate into a cascade of > resulting errors. The debugging below is on 3.1.0-RC4 however is > equally representative of several other kernel versions tested. > > dmesg: http://pastebin.com/raw.php?i=PdUqCtzw Please enable CONFIG_USB_XHCI_HCD_DEBUGGING and turn off CONFIG_USB_STORAGE_DEBUGGING, and repost your dmesg. There are a couple commands failing, but without the xHCI debugging on I can't see why. Sarah Sharp -- 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
