Distribution: Gentoo Hardware Environment: Athlon Problem Description: errors in dmesg from ehci, and USB 2 HDD will not work. Steps to reproduce: Plug HDD into USB 2 port. ehci_hcd 0000:00:10.3: EHCI Host Controller ehci_hcd 0000:00:10.3: irq 10, pci mem d69fec00 ehci_hcd 0000:00:10.3: new USB bus registered, assigned bus number 3 ehci_hcd 0000:00:10.3: USB 2.0 enabled, EHCI 1.00, driver 2004-May-10 hub 3-0:1.0: USB hub found hub 3-0:1.0: 6 ports detected ehci_hcd 0000:00:10.3: fatal error ehci_hcd 0000:00:10.3: HC died; cleaning up process `named' is using obsolete setsockopt SO_BSDCOMPAT usb usb3: string descriptor 0 read error: -19 usb usb3: string descriptor 0 read error: -19 usb usb3: string descriptor 0 read error: -19 usb usb3: string descriptor 0 read error: -19 usb usb3: string descriptor 0 read error: -19 usb usb3: string descriptor 0 read error: -19 usb usb3: string descriptor 0 read error: -19 usb usb3: string descriptor 0 read error: -19 usb usb3: string descriptor 0 read error: -19 usb usb3: string descriptor 0 read error: -19 usb usb3: string descriptor 0 read error: -19 usb usb3: string descriptor 0 read error: -19 The drive will detect/work properly with uhci. I tried this with 2.6.8-rc1, as well as 2.6.7-mm6 and 2.6.7-mm7 and have the exact same problem.
Please post "lspci -vv" output, and the corresponding initialization output with CONFIG_USB_DEBUG defined. Also /var/log/dmesg (dmesg output while system is booting) inluding PCI/ACPI/IRQ setup information. Doesn't look like anything related to the Maxtor; EHCI is just getting a fatal error as it starts up. Which is extremely rare, and almost certainly indicates a non-USB problem.
Created attachment 3342 [details] dmesg output from boot (2.6.7) with CONFIG_USB_DEBUG=Y
Created attachment 3343 [details] output from lspci -vv (2.6.7)
*** Bug 2897 has been marked as a duplicate of this bug. ***
I'm guessing this is a PCI setup problem, possibly related to those VIA quirks (which break all VIA hardware I've got access to, fwiw).
Created attachment 3455 [details] output of lspci -vv on Dell Poweredge/Tekram602/FedoraCore2-2.6.6
Created attachment 3456 [details] output of dmesg on Dell Poweredge/Tekram602/FedoraCore2-2.6.6
I have the same problem using Fedora Core 2, kernel 2.6.6-1.435.2.3, on a dell PowerEdge 2600 system (serverworks chipset), equipped with a Tekram DC-602T USB2.0 PCI-card based on a NEC chipset. loading ehci_hcd goes just fine, but as soon as I connect my Maxtor OneTouch USB/firewire drive to one of the Tekrams ports, I get the error mentioned above. It is just the same when I modprobe ehci_hcd with the maxtor already attached, and it looks like this: [root@blue2 /]# modprobe ehci_hcd [root@blue2 /]# Aug 2 08:38:10 blue2 kernel: ehci_hcd 0000:0b:00.2: EHCI Host Controller Aug 2 08:38:10 blue2 kernel: ehci_hcd 0000:0b:00.2: irq 11, pci mem 42899c00 Aug 2 08:38:10 blue2 kernel: ehci_hcd 0000:0b:00.2: new USB bus registered, assigned bus number 2 Aug 2 08:38:10 blue2 kernel: ehci_hcd 0000:0b:00.2: USB 2.0 enabled, EHCI 0.95, driver 2004-May-10 Aug 2 08:38:10 blue2 kernel: hub 2-0:1.0: USB hub found Aug 2 08:38:10 blue2 kernel: hub 2-0:1.0: 5 ports detected Aug 2 08:38:10 blue2 kernel: ehci_hcd 0000:0b:00.2: fatal error Aug 2 08:38:10 blue2 kernel: ehci_hcd 0000:0b:00.2: HC died; cleaning up uname -a Linux blue2 2.6.6-1.435.2.3 #1 Thu Jul 1 08:25:29 EDT 2004 i686 i686 i386 GNU/Linux I have added attachments with outputs from lspci -vv and dmesg. The HDD runs fine both with uhci on the mainboards own USB1 as well as with ohci on the NEC card, but, of course, only very slow.
I forgot to add that the very same harddisk runs with ehci in USB2-mode on the following 2 systems: Acer Travelmate 800 LCib Laptop (Intel Centrino Chipset) Pentium IV 2,4GHz based on an ASUS Mainboard, also Intel chipset. Tested with different kernel versions from 2.6.5 to 2.6.7, also the very same Fedora Kernel 2.6.6 mentioned above.
Is this problem ever going to be worked on? The problem still exists with 2.6.8.1 I currently have a box stuck on 2.6.6 because of this bug which was introduced in the 2.6.7 code.
Note that bug #3033 has a similar report (along with the original bug report) of "hc died" on startup. FWIW I haven't reproduced this with a Maxtor or a few different controllers. So it's not that widespread, though it's sort of puzzling nonetheless. Does 2.6.10-rc2 still have this failure? Does the newish "usb-handoff" kernel boot parameter change the behavior?
Same problem here with Fedora Core 3 with kernel-2.6.9-1.724_FC3 and kernel-2.6.10-1.727_FC3. The parameter "usb-handoff" did not help. It's a PCMCIA-to-USB2.0 Adapter with NEC chipset and a USB2.0 HDD case with Myson Century CS8818 chipset. -------------------------------------------------------------------------------- PCI: Enabling device 0000:03:00.1 (0000 -> 0002) ACPI: PCI interrupt 0000:03:00.1[B] -> GSI 10 (level, low) -> IRQ 10 ohci_hcd 0000:03:00.1: OHCI Host Controller PCI: Setting latency timer of device 0000:03:00.1 to 64 ohci_hcd 0000:03:00.1: irq 10, pci mem e0a7a000 ohci_hcd 0000:03:00.1: new USB bus registered, assigned bus number 5 hub 5-0:1.0: USB hub found hub 5-0:1.0: 2 ports detected ehci_hcd 0000:03:00.2: fatal error ehci_hcd 0000:03:00.2: HC died; cleaning up ohci_hcd 0000:03:00.1: wakeup usb 5-1: new full speed USB device using address 2 usb 5-1: not running at top speed; connect to a high speed hub scsi1 : SCSI emulation for USB Mass Storage devices Vendor: FUJITSU Model: MHR2040AT Rev: 40BA Type: Direct-Access ANSI SCSI revision: 02 SCSI device sdb: 78140160 512-byte hdwr sectors (40008 MB) sdb: assuming drive cache: write through sdb: sdb1 sdb2 sdb3 Attached scsi disk sdb at scsi1, channel 0, id 0, lun 0 USB Mass Storage device found at 2 --------------------------------------------------------------------------------
Created attachment 4337 [details] output of lspci -vv
Created attachment 4338 [details] output of lsusb -vv
Created attachment 4340 [details] patch working for ALI and Intel I've submitted this patch for 2.6.11 but it should do as well with older kernels too. I'm still not sure what's triggering this chip bug, but for at least some systems it clearly solves this bug.
Thanks, the patch fixes the problem.
*** Bug 4007 has been marked as a duplicate of this bug. ***
I've tested the patch and now I can use usb 2.0 high speed mode om Acer TravelMate 4000. But it doesn't really work with my usb harddisk since my computer loses the connection to it very often. Sometimes I have this problem also with another computer but now I have to reboot because the whole usb subsystem hasn't worked anymore.
Thomas, you should have filed a new bug instead of re-opening this old one. As you said, you don't even have the same symptoms. So whatever you're seeing, it's a different bug.
Sorry, I thought the patch has some problems with my system because without the patch full speed mode works without problems and with the patch my system crashes (kernel 2.6.8.1). But now I've also tested the patch with kernel 2.6.10 and with this version it works without problems. My usb harddisk is even more stable than ever before!