Juju Migration

From IEEE 1394 FireWire Wiki
Jump to: navigation, search

Migration to the new FireWire driver stack

Watch this page for future updates on migration requirements. (Or better yet, add here what you know about it.)


Contents


Compatibility and stability

In Q4/2010, the new FireWire kernel driver stack (alias Juju) became feature-complete and ready to replace the older ieee1394 stack. Here is a brief comparison to the older stack:

  • Storage devices: The new stack works better.
  • Video devices: The new stack generally works better or as well as the old one.
  • Audio devices: Caveats apply.
  • Controllers: All more or less widespread controllers are known to work. There are known issues with some rare controllers or rarely used software, see below.

Audio devices require kernel 2.6.32 or later, libraw1394 2.0.5 or later, and libffado 2.0.1 or later. Recommended for audio devices are kernel 2.6.36 or later, libraw1394 2.0.7 or later, and libffado 2.1.0 or later.

Some old camcorders require kernel 3.1 or later and libraw1394 2.0.8 or later.

IP over 1394 users should use kernel 2.6.38 or later.

New stack not yet working as the old stack does

  • Controllers:
  • ALi M525x:
    Some cards with these controllers don't work yet. These controllers are comparably rare though. They were available as mid-range priced PCI cards (possibly also CardBus cards) and can be swapped for any cheap FireWire PCI card or CardBus card as a stop-gap solution.
  • Apple UniNorth rev.1:
    There seem to be problems with this controller, found in PowerBook G3 Pismo and G4 Titanium. These controllers are very rare nowadays. It is possible to use a cheap CardBus FireWire controller in all affected machines instead.
  • NVIDIA NForce2:
    May or may not work with the old drivers, do not yet work with the new drivers. These controllers are very rare. It is possible to use a cheap FireWire PCI controller in all affected machines instead. Note that there were reports that NForce2 did not work for some with the old drivers either, so this may not actually be a regression.
  • Pinnacle MovieBoard:
    firewire-ohci hangs in an interrupt loop at shutdown or panics already at startup. These problems did apparently not happen with the older ohci1394 driver, but the extent to which these card worked (or didn't) with ohci1394 is not clear.
  • TI PCILynx TSB12LV21:
    No support for PCILynx or PCILynx2 controllers. Unlike TI OHCI-Lynx, PCILynx do not implement the OHCI-1394 standard and are nowadays very rare, hence it is unlikely that support for them will ever be written for the new driver stack. Note that their support in the old stack is extremely limited as well (no isochronous I/O which means no video, no audio, not even IP over 1394; slow and possibly very unstable SBP-2 support i.e. storage support). For all practical purposes, PCILynx controllers are not actually supported in the old stack either, and there was never any PCILynx owner who cared. (Instead, they use this hardware as a bus analyzer with the stand-alone driver nosy rather than as a controller with the old ieee1394 pcilynx driver.)

Historical note: VIA VT6306 did not work properly for DV capture with gstreamer's dv1394src plugin or with the DV4Linux tool (report). This should be fixed since kernel 3.10. VT6306 are shown by lspci as rev 46; higher chip revisions denote VT6307 or VT6308 which were apparently not affected by this problem.

Installation of old drivers as fallback if the new ones are insufficient

As a migration aid, it is possible until kernel 2.6.36 inclusive to build and install the new and the old drivers together. However, care needs to be taken to keep in control which drivers are loaded. I.e. create proper blacklist entries in /etc/modprobe.conf as explained below to avoid auto-loading of the wrong drivers. Libraw1394 v2 falls back to access the old drivers if the new ones are not loaded/ not bound to hardware, transparently to libraw1394 based libraries and applications.

If you are having trouble with the new drivers, do not hesitate to get in touch via the Mailinglists.


Why the migration?

The code base of the new stack is smaller, cleaner, and closer follows modern Linux kernel programming paradigms. There are already some features in the new stack which the old one lacks, notably bus manager capability and so-called gap count optimization. The latter provides a noticeable speedup of SBP-2 and other asynchronous protocols. Instead of 3 or 5 userspace interfaces in the old stack, there is a single universal interface in the new stack. Furthermore, there are some fundamental bugs and security considerations with the old stack which to a large part motivated the rewrite of the driver stack.

FIXME: Add a Wiki page on FireWire security.


Module auto-loading

How to get auto-loading

The drivers firewire-ohci and firewire-sbp2 contain the module aliases "pci:v*d*sv*sd*bc0Csc00i10*" and "ieee1394:ven*mo*sp0000609Ever00010483*" which should suffice with hotplug scripts and recent coldplug scripts to automatically load these two drivers when respective hardware is detected. The dependence of both drivers on firewire-core is of course recognized by modprobe.

If the kernel was configured without ohci1394 and sbp2 as modules, then firewire-core and firewire-sbp2 also contain the module aliases ohci1394 and sbp2 respectively. That is, "modprobe ohci1394" will load firewire-ohci (and firewire-core) instead of ohci1394 if ohci1394 was excluded from the kernel build.

How to suppress auto-loading

To avoid confusion, it is recommended that either the old or the new driver stack is built, but not both together. It is nevertheless possible to build and install both stacks together. If you chose to do so, you may want to add lines like

# blacklist firewire-ohci
# blacklist firewire-sbp2
# blacklist firewire-net

blacklist ohci1394
blacklist sbp2
blacklist eth1394
blacklist dv1394
blacklist raw1394
blacklist video1394

to /etc/modprobe.d/file_of_your_choice or /etc/modprobe.conf to suppress auto-loading of one of the stacks.

Extremely old modutils which do not support the blacklist keyword can be instructed by configuration entries like "install module /bin/true" to suppress loading of a particular module.

Module blacklisting alone is of course insufficient if the drivers are statically linked into the kernel. But less obvious, it is also insufficient if the set of drivers that you don't want are already loaded by an initrd (initial RAM disk) during system boot. For example, the stock initrd of Ubuntu 10.4 binds ohci1394 to FireWire controllers. A remedy is to rebuild the initrd with a reduced selection of modules.


Character device files, block device files

Basic operation

Out of the box, udevd and udev scripts automatically create and remove

  • /dev/fw* devices exposed by firewire-core (for use by libraries like libraw1394 and libdc1394, provided the libraries are updated),
  • /dev/{sd,sg,sr,st}* devices exposed by SCSI command set drivers (sd_mod, sg, sr_mod, st) with firewire-sbp2 underneath.

Permissions and ownership for /dev/fw*

Without any configuration file, the device files will be only accessible to root. This is fine and intended for SCSI block device files. But it is usually desirable to access the /dev/fw* character device files as non-root user. The example udev rules shown below allow any user in group "video" to access /dev/fw* with programs such as dvgrab, kino, or coriander.

By evaluating information in firewire-core's sysfs files, it is possible to automatically adapt file permissions and ownership of /def/fw* files according to device types:

  • Files pertaining to local nodes (the controllers) and SBP-2 nodes (storage devices, scanners etc.) should get restrictive file permissions and ownership. Actually, the default should be restrictive on most systems.
  • Files pertaining to audio and video devices (AV/C or IIDC compliant nodes) can get liberal permissions so that they are accessible to application programs without root privilege.
  • Old libraw1394 versions and some special-purpose libraw1394 clients (e.g. gscanbus) currently require also access to the local node. Its permissions need to be liberal in those cases too. To the author's knowledge, this does not have any actual security impact on systems on which user access to AV/C and IIDC devices is already allowed, but distributors typically won't want to open this up per default.

The following rules implement this device type dependent control over permissions, using the example "video" group. These rules work with Linux kernel 2.6.31 and later. These rules have been merged into /lib/udev/rules.d/50-udev-default.rules of udev v144.

# /etc/udev/rules.d/example-firewire.rules

# IIDC devices: industrial cameras and some webcams
SUBSYSTEM=="firewire", ATTR{units}=="*0x00a02d:0x00010*", GROUP="video"

# AV/C devices: camcorders, set-top boxes, TV sets, various audio devices, and more
SUBSYSTEM=="firewire", ATTR{units}=="*0x00a02d:0x010001*", GROUP="video"

In udev v161, two more match patterns were added: "*0x00b09d:0x00010*" for IIDC cameras from Point Grey and "*0x00a02d:0x014001*" for AV/C devices with vendor-unique command set.

Unfortunately, FireWire audio devices often adhere to vendor-specific protocols. Furthermore, an "audio" group assignment may be preferred over "video". Therefore, a special ruleset for the FFADO FireWire audio library is provided by recent versions of libffado (60-ffado.rules).

If your application needs access to all nodes, simply use:

SUBSYSTEM=="firewire", GROUP="video"

There are also schemes which are based on access control lists instead of UNIX file permissions. For example, the Fedora Linux distribution currently contains a mechanism to add read and write permission for the locally logged in user to the ACLs of fw* files of some FireWire device types. To this end, recent udev releases have firewire subsystem rules in the file 70-acl.rules file.

Symlinks to SCSI block device files

Reasonably recent udev scripts create symbolic links in /dev/disk/by-id/*, pointing to the block devices of FireWire harddisks. These links are convenient for example to use them in static fstab entries: Their names are always the same because they are based on persistent and unique device identifiers, while the actual device files have arbitrary names that change all the time when disks are plugged in and out.

The names of the by-id links look per default a little bit different with firewire-sbp2 compared to sbp2. To make them look exactly the same, add the following to /etc/modprobe.d/file_of_your_choice or /etc/modprobe.conf:

options sbp2 long_ieee1394_id=y

This option has been added to sbp2 in Linux 2.6.22.

You only need this if you plan to switch between sbp2 and firewire-sbp2 and if you are using the /dev/disk/by-id symlinks.


Hald support

Nothing special is needed; hotplug and hot-removal of firewire-sbp2 driven disks is exposed on desktops just fine.


Initrd

Scripts which generate initrd (an initial RAM disk used during boot) may need to be updated to deal with the new kernel module names firewire-core, firewire-ohci, firewire-sbp2. Scripts within initrd may already work with the new drivers, see #Module auto-loading.


Libraries

libraw1394

At the time of this writing (July 2012), libraw1394 v2.1.0 is the current release and also the recommended stable release for use with the new firewire drivers.

Compatibility with the new drivers is available since libraw1394 v2.0.0, released in July 2008. This version is able to transparently switch between old and new stack, depending on which drivers you have loaded. Bug fixes related to usage with the new kernel drivers followed in further v2.0.x sub-releases.

Either get the latest libraw1394 2.x release, or build libraw1394 from a fresh git checkout:

$ git clone git://git.kernel.org/pub/scm/libs/ieee1394/libraw1394.git
$ cd libraw1394/
$ autoreconf -fi
$ ./configure
$ make
$ sudo make install

The Juju backend of libraw1394 requires the inotify kernel interface. If you have got a very old libc which does not contain support for inotify, you can have a look at the following patches for libraw1394: patch 1, patch 2. They don't apply to recent versions of libraw1394 anymore though.

Libraw1394's Juju backend furthermore expects that the kernel was configured and built with the following options:

CONFIG_INOTIFY_USER=y
CONFIG_EPOLL=y

The inotify option is located in the "File systems" section of the Kconfig dialogs. CONFIG_EPOLL is already enabled in usual kernel configurations and not even visible in the Kconfig dialogs. But if the kernel was configured for embedded systems it may be necessary to switch CONFIG_EPOLL on explicitly; it is then found in the "General setup" section of the kernel configuration.

libdc1394

At the time of this writing (May 2012), libdc1394 v2.2.0 is the current release and also the recommended stable release for use with the new firewire drivers.

There is compatibility with libdc1394 v2, released in January 2008. Some deployed libdc1394 applications still use the older libdc1394 v1 though which has no support for the Juju drivers. Since v2.1.0, libdc1394 transparently supports the old and the new drivers, though with different extent of functionality. Since libdc1394 v2.1.2 from June 2009 and kernel 2.6.30, libdc1394's Juju backend fully supports multiple cameras on the same bus, precise packet timestamps/ framerate measurement, and isochronous resource clean-up even at unclean program termination — i.e. offers same or better functionality with the new drivers compared to the old ones.

You can check libdc1394 v2 out from the project's repository and build it with support for both the new and the old driver stack by the following steps:

$ git clone git://libdc1394.git.sourceforge.net/gitroot/libdc1394/libdc1394
$ cd libdc1394/libdc1394/
$ autoreconf -fi
$ ./configure
$ make
$ sudo make install

(In libdc1394 until v2.0.3 inclusive, a configure switch called "--with-juju-dir" was necessary to support the new driver ABI, e.g. "./configure --with-juju-dir=/usr/src/linux/include". But this also disabled support for the old ieee1394 stack.)

Libdc1394 is able to work with the new drivers even if you have only a libraw1394 v1 installed alongside libdc1394. This is because libdc1394 requires libraw1394 only in order to use the old drivers while it accesses the new drivers without assistance of libraw1394.

No special kernel headers package is required to build libdc1394.

FFmpeg: libavformat

Alas the dv1394 module of FFmpeg's libavformat has not been ported to firewire-core yet. This particularly affects players like MPlayer and Xine. However, you can use the capture utility dvgrab to provide input to players:

$ dvgrab - | xine stdin:/

pwlib: AV/C and IIDC plugins

The AV/C video input module of pwlib a.k.a. ptlib, used in Opal, Ekiga etc. uses the old raw1394_start_iso_rcv API which is no longer implemented in the kernel since 2.6.23.

The IIDC video input module (a.k.a. DC module) of pwlib/ ptlib uses libdc1394 v1 to which the new drivers are not compatible.

Best would be if we had a V4L2 interface to FireWire cameras; then these applications would work out of the box without special 1394 backends and libraries.

Personal tools