[ I tried sending this last night from my Adaptec email address and have yet to see it on the list. Sorry if this is dup for any of you. ] For the past few months, Adaptec Inc, has been working to enhance MD. The goals of this project are: o Allow fully pluggable meta-data modules o Add support for Adaptec ASR (aka HostRAID) and DDF (Disk Data Format) meta-data types. Both of these formats are understood natively by certain vendor BIOSes meaning that arrays can be booted from transparently. o Improve the ability of MD to auto-configure arrays. o Support multi-level arrays transparently yet allow proper event notification across levels when the topology is known to MD. o Create a more generic "work item" framework which is used to support array initialization, rebuild, and verify operations as well as miscellaneous tasks that a meta-data or RAID personality may need to perform from a thread context (e.g. spare activation where meta-data records may need to be sequenced carefully). o Modify the MD ioctl interface to allow the creation of management utilities that are meta-data format agnostic. A snapshot of this work is now available here: http://people.freebsd.org/~gibbs/linux/SRC/emd-0.7.0-tar.gz This snapshot includes support for RAID0, RAID1, and the Adaptec ASR and DDF meta-data formats. Additional RAID personalities and support for the Super90 and Super 1 meta-data formats will be added in the coming weeks, the end goal being to provide a superset of the functionality in the current MD. A patch to fs/partitions/check.c is also required for this release to function correctly: http://people.freebsd.org/~gibbs/linux/SRC/md_announce_whole_device.diff As the file name implies, this patch exposes not only partitions on devices, but all "base" block devices to MD. This is required to support meta-data formats like ASR and DDF that typically operate on the whole device. Nothing in the implementation prevents any meta-data format from being used on a partition, but BIOS boot support is only available in the non-partitioned mode. Since the current MD notification scheme does not allow MD to receive notifications unless it is statically compiled into the kernel, we would like to work with the community to develop a more generic notification scheme to which modules, such as MD, can dynamically register. Until that occurs, these EMD snapshots will require at least md.c to be a static component of the kernel. For those wanting to test out this snapshot with an Adaptec HostRAID U320 SCSI controller, you will need to update your kernel to use version 2.0.8 of the aic79xx driver. This driver defaults to attaching to 790X controllers operating in HostRAID mode in addition to those in direct SCSI mode. This feature can be disabled using a module or kernel command option. Driver source and BK send patches for this driver can be found here: http://people.freebsd.org/~gibbs/linux/SRC/aic79xx-linux-2.6-20040316-tar.gz http://people.freebsd.org/~gibbs/linux/SRC/aic79xx-linux-2.6-20040316.bksend.gz Architectural Notes =================== The major areas of change in "EMD" can be categorized into: 1) "Object Oriented" Data structure changes These changes are the basis for allowing RAID personalities to transparently operate on "disks" or "arrays" as member objects. While it has always been possible to create multi-level arrays in MD using block layer stacking, our approach allows MD to also stack internally. Once a given RAID or meta-data personality is converted to the new structures, this "feature" comes at no cost. The benefit to stacking internally, which requires a meta-data format that supports this, is that array state can propagate up and down the topology without the loss of information inherent in using the block layer to traverse levels of an array. 2) Opcode based interfaces. Rather than add additional method vectors to either the RAID personality or meta-data personality objects, the new code uses only a few methods that are parameterized. This has allowed us to create a fairly rich interface between the core and the personalities without overly bloating personality "classes". 3) WorkItems Workitems provide a generic framework for queuing work to a thread context. Workitems include a "control" method as well as a "handler" method. This separation allows, for example, a RAID personality to use the generic sync handler while trapping the "open", "close", and "free" of any sync workitems. Since both handlers can be tailored to the individual workitem that is queued, this removes the need to overload one or more interfaces in the personalities. It also means that any code in MD can make use of this framework - it is not tied to particular objects or modules in the system. 4) "Syncable Volume" Support All of the transaction accounting necessary to support redundant arrays has been abstracted out into a few inline functions. With the inclusion of a "sync support" structure in a RAID personality's private data structure area and the use of these functions, the generic sync framework is fully available. The sync algorithm is also now more like that in 2.4.X - with some updates to improve performance. Two contiguous sync ranges are employed so that sync I/O can be pending while the lock range is extended and new sync I/O is stalled waiting for normal I/O writes that might conflict with the new range complete. The syncer updates its stats more frequently than in the past so that it can more quickly react to changes in the normal I/O load. Syncer backoff is also disabled anytime there is pending I/O blocked on the syncer's locked region. RAID personalities have full control over the size of the sync windows used so that they can be optimized based on RAID layout policy. 5) IOCTL Interface "EMD" now performs all of its configuration via an "mdctl" character device. Since one of our goals is to remove any knowledge of meta-data type in the user control programs, initial meta-data stamping and configuration validation occurs in the kernel. In general, the meta-data modules already need this validation code in order to support auto-configuration, so adding this capability adds little to the overall size of EMD. It does, however, require a few additional ioctls to support things like querying the maximum "coerced" size of a disk targeted for a new array, or enumerating the names of installed meta-data modules, etc. This area of EMD is still in very active development and we expect to provide a drop of an "emdadm" utility later this week. 6) Meta-data and Topology State To support pluggable meta-data modules which may have diverse policies, all embedded knowledge of the MD SuperBlock formats has been removed. In general, the meta-data modules "bid" on incoming devices that they can manage. The high bidder is then asked to configure the disk into a reasonable topology that can be managed by a RAID personality and the MD core. The bidding process allows a more "native" meta-data module to outbid a module that can handle the same format in "compatibility" mode. It also allows the user to load a meta-data module update during install scenarios even if an older module is compiled statically into the kernel. Once the topology is created, all information needed for normal operation is available to the MD core and/or RAID personalities via direct variable access (at times protected by locks or atomic ops of course). Array or member state changes occur via calling into the meta-data personality associated with that object. The meta-data personality is then responsible for changing the state visible to the rest of the code and notifying interested parties. This async design means that a RAID module noticing an I/O failure on one member and posting that event to one meta-data module, may cause a chain of notifications all the way to the top-level array object owned by another RAID/meta-data personality. The entire topology is reference counted such that objects will only disappear from the topology once they have transitioned to the FAILED state and all I/O (each I/O holds a reference) ceases. 7) Correction of RAID0 Transform The RAID0 transform's "merge function" assumes that the incoming bio's starting sector is the same as what will be presented to its make_request function. In the case of a partitioned MD device, the starting sector is shifted by the partition offset for the target offset. Unfortunately, the merge functions are not notified of the partition transform, so RAID0 would often reject requests that span "chunk" boundaries once shifted. The fix employed here is to determine if a partition transform will occur and take this into account in the merge function. Adaptec is currently validating EMD through formal testing while continuing the build-out of new features. Our hope is to gather feedback from the Linux community and adjust our approach to satisfy the community's requirements. We look forward to your comments, suggestions, and review of this project. -- Justin - To unsubscribe from this list: send the line "unsubscribe linux-raid" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
