From: Jan Kiszka <jan.kiszka@web.de>
To: Nicholas Rotella <nrotella@usc.edu>, xenomai@xenomai.org
Subject: Re: [Xenomai] Xenomai and usb4rt IRQ issues
Date: Tue, 09 Sep 2014 08:09:05 +0200 [thread overview]
Message-ID: <540E9981.1000903@web.de> (raw)
In-Reply-To: <CAMoEH9VtdHc6usxhhYGVn74ssWC_6CQSPzK0mLSQ4Wmog3KG5A@mail.gmail.com>
On 2014-08-27 04:59, Nicholas Rotella wrote:
> Hi all,
>
> I have some issues getting real-time USB working with my device, an
> inertial sensor which normally uses the non-RT cdc-acm driver. I'm
> running Ubuntu 12.04 (linux kernel version 3.8.13) with Xenomai 2.6.3.
> I know that this is the Xenomai mailing list so I'm not sure if anyone
> is very familiar with the usb4rt project (cloned from
> http://git.kiszka.org/?p=usb4rt.git;a=summary and working off the
> cdc-acm branch) but I thought it was worth a try (anyway the issue may
> be more general).
>
> Short story is this: I am able to compile usb4rt, load the drivers,
> open a real-time USB port via the rt_cdc_acm module and write to the
> device. It responds, but I am never able to read data from it. I'm
> sure it's sending data because IRQs are triggered and nothing else
> shares this interrupt line.
>
> Long story is this: I am able to load the correct modules and bind
> them to a UHCI-compatible USB card (USB 1.1), my device is recognized
> and the rt_cdc_acm driver gets loaded properly, creating rtser0 (as
> confirmed by checking /proc/xenomai/rtdm/rtser0/information). I can
> then use RT serial
> (http://www.xenomai.org/documentation/trunk/html/api/group__rtserial.html)
> commands to interact with the virtual serial device. I can open the
> port and send commands, which I'm sure the device is receiving.
> However, I receive no data when I try to read from the device. I can
> put this device into a "streaming" mode in which it sends packets in
> an uninterrupted fashion and, by watching /proc/xenomai/irq, I can see
> that packets are coming in from the device (the rt_uhci driver's IRQ
> count increases steadily). My suspicion is that there is somehow an
> IRQ mismatch which causes the usb4rt drivers to miss the IRQs issued
> by the device when data is ready for reading, but I am not sure.
>
> Any thoughts or advice on how to proceed would be much appreciated.
> The following is a portion of the script I use to compile and bring up
> the usb4rt drivers (after unbinding the non-RT drivers from devices
> which would otherwise share the same IRQ line that usb4rt uses):
>
> git clone git://git.kiszka.org/usb4rt.git
> git checkout cdc-acm
> cd /usr/src/usb4rt
> ./configure --enable-drv-cdc-acm --disable-bandw-reclam
> --with-xeno-user-dir=/usr/
> make && make install
> echo -n "0000:05:00.0" > /sys/bus/pci/drivers/uhci_hcd/unbind
> echo -n "0000:00:1a.0" > /sys/bus/pci/drivers/ehci-pci/unbind
> sleep 1
> insmod /usr/local/usb4rt/modules/rt_usbcore.ko
> sleep 1
> insmod /usr/local/usb4rt/modules/rt_uhci_hcd.ko irq=16
> insmod /usr/local/usb4rt/modules/rt_cdc_acm.ko vendor=0x199b
> product=0x3065 start_index=0
>
> The following is the portion of my systemlog pertaining to the loading
> of the usb4rt drivers:
>
> [23407.104492] ********** Realtime USB-Core Module 0.0.5 ***********
> [23407.104494] RT-USBCORE: Max 16 Controller
> [23407.104495] RT-USBCORE: Max 128 USB-Devices
> [23407.104496] RT-USBCORE: Common debugging: disabled
> [23407.104497] RT-USBCORE: Queue head debugging: disabled
> [23407.104498] RT-USBCORE: Transfer descriptor debugging: disabled
> [23407.104498] RT-USBCORE: Time debugging: disabled
> [23407.104503] RT-USBCORE: Control-URB @ 0xffff8803b5429200 with 8
> Byte Packet-Size
> [23407.104504] RT-USBCORE: Control-URB @ 0xffff8803b5428e00 with 16
> Byte Packet-Size
> [23407.104505] RT-USBCORE: Control-URB @ 0xffff8803b542a200 with 32
> Byte Packet-Size
> [23407.104506] RT-USBCORE: Control-URB @ 0xffff8803b542b000 with 64
> Byte Packet-Size
> [23407.104507] RT-USBCORE: Initialize Hub-List
> [23407.104507] RT-USBCORE: Initialize Controller-List
> [23407.104508] RT-USBCORE: Loading Completed (1152 Byte allocated)
> [23408.103279] ********* Realtime Driver for Universal Host Controller
> 0.0.5 **********
> [23408.103283] RT-UHC-Driver: Searching for Universal-Host-Controller
> [23408.103284] RT-UHC-Driver: -> Use only HC with IRQ 16
> [23408.103288] USB Universal Host Controller found : Vendor = 0x1106,
> Device = 0x3038, IRQ = 16, IO-Port = 0x00002020 (32 Bytes)
> [23408.103290] RT-UHC-Driver: Request IO-Port @ 0x00002020 (32 Byte)
> for UHC[0] ... [OK]
> [23408.103292] RT-UHC-Driver: Request RTDM IRQ 16 ... [OK]
> [23408.103349] RT-USBCORE: Register Host-Controller Driver
> [23408.103350] RT-USBCORE: Host-Controller added to USB-Controller-List
> [23408.548646] RT-USBCORE: Registering CTRL-URB (8 Byte) @ Host-Controller 0
> [23408.549819] 000:00-000: Setting Address 1
> [23408.549885] 000:00-001: Switch to URB with 64 Byte Packet-Size
> [23408.580736] 000:00-001: Manufacturer : MicroStrain, Inc.
> [23408.581010] 000:00-001: Product : 3DM-GX3-25 Orientation Sensor
> [23408.581205] 000:00-001: Device 1 configured
> [23408.581206] --- DUMP USB-DEVICE @ ffffffffa05773b8 -------------------------
> [23408.581207] Number : 1
> [23408.581208] Vendor : 0x199b
> [23408.581209] Product : 0x3065
> [23408.581210] Root-Hub-Port : 0x00
> [23408.581210] Class : 0x02
> [23408.581211] Subclass : 0x00
> [23408.581212] Protocol : 0x00
> [23408.581213] In use : no
> [23408.581214] Status : 0x00
> [23408.581214] Speed : 0x02
> [23408.581215] Host-Controller @ 0xffff8803f5381880
> [23408.581216]
> [23408.581217] Endpoint ---0 ---1 ---2 ---3 ---4 ---5 ---6
> ---7 ---8 ---9 --10 --11 --12 --13 --14 --15
> [23408.581218] Ctrl-Mask IN X - - - - - -
> - - - - - - - - -
> [23408.581220] Ctrl-Mask OUT X - - - - - -
> - - - - - - - - -
> [23408.581222] Bulk-Mask IN - X - - - - -
> - - - - - - - - -
> [23408.581224] Bulk-Mask OUT - - - X - - -
> - - - - - - - - -
> [23408.581225] Int-Mask IN - - X - - - -
> - - - - - - - - -
> [23408.581227] Int-Mask OUT - - - - - - -
> - - - - - - - - -
> [23408.581229] Iso-Mask IN - - - - - - -
> - - - - - - - - -
> [23408.581230] Iso-Mask OUT - - - - - - -
> - - - - - - - - -
> [23408.581232] Toggle-Mask IN - - - - - - -
> - - - - - - - - -
> [23408.581234] Toggle-Mask OUT - - - - - - -
> - - - - - - - - -
> [23408.581236] Max-Packet IN 0064 0064 0008 0000 0000 0000 0000
> 0000 0000 0000 0000 0000 0000 0000 0000 0000
> [23408.581237] Max-Packet OUT 0064 0000 0000 0064 0000 0000 0000
> 0000 0000 0000 0000 0000 0000 0000 0000 0000
> [23408.581238] --- END USB-DEVICE --------------------------
> [23408.581239] --- LIST USB-DEVICES -------------------------
> [23408.581240] No HCD VENDOR PROD CLS SCLS PROT ->CTRL CTRL->
> ->BULK BULK-> ->INT INT-> ->ISOC ISOC-> STATE SPEED USED
> [23408.581242] 001 000 0x199b 0x3065 0x02 0x00 0x00 0x0001 0x0001
> 0x0002 0x0008 0x0004 0x0000 0x0000 0x0000 00000 00002 -
> [23408.581244] --- END USB-DEVICES --------------------------
> [23408.581258] RT-UHC-Driver: Loading Completed (48 Byte allocated)
> [23408.582245] --- LIST USB-DEVICES -------------------------
> [23408.582248] No HCD VENDOR PROD CLS SCLS PROT ->CTRL CTRL->
> ->BULK BULK-> ->INT INT-> ->ISOC ISOC-> STATE SPEED USED
> [23408.582251] 001 000 0x199b 0x3065 0x02 0x00 0x00 0x0001 0x0001
> 0x0002 0x0008 0x0004 0x0000 0x0000 0x0000 00000 00002 X
> [23408.582254] --- END USB-DEVICES --------------------------
> [23408.582255] rt_cdc_acm[0]: device found, vendor=0x199b, product=0x3065
>
> I can provide more detailed info for debugging if it helps. There is
Does this issue persist?
> also this warning when I compile usb4rt which seems a little
> suspicious:
>
> /usr/src/usb4rt/host/rt_uhci.c: In function ‘rt_irq_handler’:
> /usr/src/usb4rt/host/rt_uhci.c:854:12: warning: ‘p_td’ may be used
> uninitialized in this function [-Wuninitialized]
> /usr/src/usb4rt/host/rt_uhci.c:968:9: note: ‘p_td’ was declared here
This is harmless (but will be silenced soon): p_td is never touched in
case it remained uninitialized.
Jan
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: OpenPGP digital signature
URL: <http://www.xenomai.org/pipermail/xenomai/attachments/20140909/1b3ba27f/attachment.sig>
next prev parent reply other threads:[~2014-09-09 6:09 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-08-27 2:59 [Xenomai] Xenomai and usb4rt IRQ issues Nicholas Rotella
2014-09-09 6:09 ` Jan Kiszka [this message]
2014-09-09 6:28 ` Nicholas Rotella
2014-09-09 6:30 ` Jan Kiszka
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=540E9981.1000903@web.de \
--to=jan.kiszka@web.de \
--cc=nrotella@usc.edu \
--cc=xenomai@xenomai.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.