Menu
▾
▴
[Libusb-devel] Communication problems with the new Via MS10000E
|
From: Simon M. <sim...@gm...> - 2004-12-22 16:58:26
|
Hi I have just gotten one of the first Via MS10000E mother boards from the production (that is at least what the salesman says), and needs to download software to the Cypress Semiconductor FX2 development board. We need to do it from Linux kernel 2.4 (the SUSE 9.0 distribution) and we does it using the fxload program. We are able always to download small programs (< 512 bytes) to the FX2 board, but when downloading larger programs the process does one of the following: 1. Downloads the software without problems (< 30% of the time) 2. Downloads the software, but it takes many seconds (the software is < 20 KBytes - works < 30% of the time) 3. Fails witht the error message: "write external: Connection timed out" In /var/log/messages the following will appear (the clock is not set correctly): Dec 22 18:30:09 linux kernel: sda : READ CAPACITY failed. Dec 22 18:30:09 linux kernel: sda : status = 1, message = 00, host = 0, driver = 08 Dec 22 18:30:09 linux kernel: Current sd00:00: sense key Not Ready Dec 22 18:30:09 linux kernel: Additional sense indicates Medium not present Dec 22 18:30:09 linux kernel: sda : block size assumed to be 512 bytes, disk size 1GB. Dec 22 18:30:10 linux kernel: sda: Write Protect is off Dec 22 18:30:10 linux kernel: sda: I/O error: dev 08:00, sector 0 Dec 22 18:30:10 linux kernel: I/O error: dev 08:00, sector 0 Dec 22 18:30:10 linux kernel: I/O error: dev 08:00, sector 2097144 Dec 22 18:30:10 linux kernel: I/O error: dev 08:00, sector 2097144 Dec 22 18:30:10 linux kernel: I/O error: dev 08:00, sector 0 Dec 22 18:30:10 linux kernel: I/O error: dev 08:00, sector 0 Dec 22 18:30:10 linux kernel: unable to read partition table Dec 22 18:30:10 linux kernel: I/O error: dev 08:00, sector 0 Dec 22 18:30:10 linux kernel: sda : READ CAPACITY failed. Dec 22 18:30:10 linux kernel: sda : status = 1, message = 00, host = 0, driver = 08 Dec 22 18:30:10 linux kernel: Current sd00:00: sense key Not Ready Dec 22 18:30:10 linux kernel: Additional sense indicates Medium not present Dec 22 18:30:10 linux kernel: sda : block size assumed to be 512 bytes, disk size 1GB. Dec 22 18:30:10 linux kernel: sda: Write Protect is off Dec 22 18:30:10 linux kernel: sda: I/O error: dev 08:00, sector 0 Dec 22 18:30:10 linux kernel: I/O error: dev 08:00, sector 0 Dec 22 18:30:10 linux kernel: I/O error: dev 08:00, sector 2097144 Dec 22 18:30:10 linux kernel: I/O error: dev 08:00, sector 2097144 Dec 22 18:30:10 linux kernel: I/O error: dev 08:00, sector 0 Dec 22 18:30:10 linux kernel: I/O error: dev 08:00, sector 0 Dec 22 18:30:10 linux kernel: unable to read partition table Dec 22 18:30:10 linux kernel: sda : READ CAPACITY failed. Dec 22 18:30:10 linux kernel: sda : status = 1, message = 00, host = 0, driver = 08 Dec 22 18:30:10 linux kernel: Current sd00:00: sense key Not Ready Dec 22 18:30:10 linux kernel: Additional sense indicates Medium not present Dec 22 18:30:10 linux kernel: sda : block size assumed to be 512 bytes, disk size 1GB. Dec 22 18:30:10 linux kernel: sda: Write Protect is off Dec 22 18:30:10 linux kernel: sda: I/O error: dev 08:00, sector 0 Dec 22 18:30:10 linux kernel: I/O error: dev 08:00, sector 0 Dec 22 18:30:10 linux kernel: I/O error: dev 08:00, sector 2097144 Dec 22 18:30:10 linux kernel: I/O error: dev 08:00, sector 2097144 Dec 22 18:30:10 linux kernel: I/O error: dev 08:00, sector 0 Dec 22 18:30:10 linux kernel: I/O error: dev 08:00, sector 0 Dec 22 18:30:10 linux kernel: unable to read partition table Dec 22 18:30:10 linux kernel: I/O error: dev 08:00, sector 0 Note the errors related to the sda SCSI disc, which is probably the Compact Flash slot on the mother board. When this log was generated no CF card was present, but if we insert a CF card the error messages related to sda disapplears. 4. Fails with the error message: "write external: Protocol error" In /var/log/messages the following is written (this is from 4 different attempts to download the same program): Dec 22 18:22:54 linux kernel: usbdevfs: USBDEVFS_CONTROL failed dev 2 rqt 64 rq 163 len 665 ret -71 Dec 22 18:23:30 linux kernel: usbdevfs: USBDEVFS_CONTROL failed dev 2 rqt 64 rq 163 len 665 ret -71 Dec 22 18:23:31 linux kernel: usbdevfs: USBDEVFS_CONTROL failed dev 2 rqt 64 rq 163 len 70 ret -71 Dec 22 18:23:33 linux kernel: usbdevfs: USBDEVFS_CONTROL failed dev 2 rqt 64 rq 163 len 481 ret -71 Note that "len" (probably the number of bytes downloaded to the device) is not the same every time. The most confusing part is, that the four situations does not appear to be associated with any event or condition (except for downloading of programs < 512 bytes). At one time a program downloads without any error and a few seconds later it will fail to download. In each case we have checked with the output of the lsusb command that we try to download to the FX2 board and not the "USB mass storage device. Output of the lsusb command: Bus 005 Device 001: ID 0000:0000 Bus 004 Device 001: ID 0000:0000 Bus 003 Device 001: ID 0000:0000 Bus 002 Device 001: ID 0000:0000 Bus 001 Device 001: ID 0000:0000 Bus 001 Device 004: ID 0547:1002 Anchor Chips, Inc. Bus 001 Device 003: ID 040d:6207 VIA Technologies, Inc. hwinfo --usb gives the following output (only the FX2 device and the CF slot is in this post): 03: SCSI 103.0: 10600 Disk [Created at usb.113] Unique ID: B3Fu.g96JPYNf3VD Parent ID: B3Fu.d82w8F5aOr5 Hardware Class: disk Model: "VIA-P VT6205-CF 0.90 V" Hotplug: USB Vendor: "VIA-P" Device: "VT6205-CF 0.90 V" Revision: "0.90" Serial ID: "000000000001" USB GUID: 040d62070000000000000001 Driver: "usb-storage" Device File: /dev/sda Speed: 1.5 Mbps Size: 2097151 sectors a 512 bytes Geometry (Logical): CHS 1023/64/32 Drive status: no medium Config Status: cfg=no, avail=yes, need=no, active=unknown Attached to: #2 (Hub) 04: USB 104.0: 0000 Unclassified device [Created at usb.113] Unique ID: B3Fu.CfRvIJKgcAB Parent ID: B3Fu.d82w8F5aOr5 Hardware Class: unknown Model: "Anchor Chips Unclassified device" Hotplug: USB Vendor: usb 0x0547 "Anchor Chips Inc." Device: usb 0x1002 USB GUID: 054710020000000000000000 USB Device status: no driver loaded Speed: 1.5 Mbps Config Status: cfg=no, avail=yes, need=no, active=unknown Attached to: #2 (Hub) When we reset the FX2 board (i.e. unplugs the USB device and plugs it in again) we get the following messages in /var/log/messages: Dec 22 18:53:49 linux kernel: usb.c: USB disconnect on device 00:10.4-2 address 5 Dec 22 18:53:50 linux kernel: hub.c: new USB device 00:10.4-2, assigned address 6 Dec 22 18:53:50 linux kernel: usb.c: config descriptor too short (expected 171, got 128) Dec 22 18:53:50 linux kernel: usb.c: unable to get device 6 configuration (error=-22) Dec 22 18:53:50 linux kernel: hub.c: new USB device 00:10.4-2, assigned address 7 Dec 22 18:53:50 linux kernel: usb.c: USB device 7 (vend/prod 0x4b4/0x8613) is not claimed by any active driver. We have tried to run with the same harddrive (and FX2 board) on a Via PD6000E board and an 500 MHz (USB 1.1) Via board (I don't know the product id), without any problems. The same is true for a Embedded Computing ECM-5716 board, which ran with no problems either. Finally is th following selected output from the lsmod command: Module Size Used by Not tainted usb-storage 65312 0 scsi_mod 97108 5 [sd_mod st sr_mod sg usb-storage] usb-uhci 22224 0 (unused) ehci-hcd 16588 0 (unused) usbcore 58668 1 [usb-storage usb-uhci ehci-hcd] Of cause I have tried attaching the FX2 with an other cable - with the same problem. I am really stuck now - something is different in the new VIA board, maby it's a timing issue? Is there an hornest chance that it is somehing I can correct? -- Merry Christmas, Simon Mikkelsen, B.Sc. -------------------------------------------- Gram & Juhl Aps, cvr 19 86 32 71 Hirsevej 16, DK-6100 Haderslev phone: +45 70 20 21 69 fax: +45 73 53 56 20 http://www.gramjuhl.dk -------------------------------------------- Phone-direct: +45 73 53 56 53 |