New OSDL initiative targets current and next-generation converged handsets
By Bill Weinberg
Open Source Architecture Specialist, OSDL
The global mobile phone market is exploding. In Q2004, the market grew 34 percent as almost 700 million handsets made their way from device OEMs into people’s hands and onto the global voice and data networks (IDC). By 2009, the global installed base will number more than 2.6 billion mobile phones (Gartner). For the IT industry, these numbers are tantalizing orders of magnitude greater than total shipments and installed base for servers, and far greater in volume than the worldwide desktop market. For the Linux software and related hardware segments, the mobile phone market represents both an opportunity to break out and enjoy significant market share in client devices, and to complement the already significant presence of Linux in communications infrastructure.
In 2004 and 2005, Linux has made significant gains as a mobile handset Operating System (OS). Global manufacturers with household brand names like LG, Motorola, NEC, Panasonic, and Samsung today ship two dozen smart phone models based on Linux, complemented by emerging Chinese brands like Datang, e28, Haier, Huawei, and ZTE. Device OEMs, large and small, are choosing Linux as a strategic platform for their smart phones for a mix of technical and economic reasons. On the technical side, they look to Linux for performance, robustness, gold standard TCP/IP networking (such as routing), and flexibility. On the economic front, Linux offers OEMs:
Lower development and deployment costs
More choice of vendors (including roll your own)
A larger open and commercial technology ecosystem
An opportunity to unify the divergent and costly product lines and engineering efforts needed to support multiple product tiers (smart phones, feature phones, and entry level devices), network types (GSM, CDMA, analog, and WiFi), and carrier requirements
For all of these strong technical and economic benefits, Linux phones account today for only about 5 percent of the total market. In the fastest-growing smart phone segment (85 percent/year – Gartner), Linux enjoys a stronger position (25 percent in Q2 2005 – Gartner), far ahead of Windows Mobile, PalmOS, or RIM (but behind SymbianOS).
To bolster this very positive trend, the OSDL is creating a new initiative called Mobile Linux Initiative (MLI) to bring together:
Chipset manufacturers
Linux distribution and platform suppliers
Middleware ISVs
Handset manufacturers
Integrators
Carriers
Operators
On October 17, 2005, in Beijing, the OSDL hosted the launch of this new initiative with a goal of addressing a mix of key technical and economic challenges, from the kernel up, to accelerate Linux adoption on mobile phones and other converged voice and data devices. While the ultimate requirements and development efforts will be driven by the members of this new initiative, the following is intended to give readers insight into MLI goals:
Technical challenges
Existing OSDL members and other industry players have repeatedly asked for Carrier Grade Linux for the cell phone. What they mean by this request is a specification for a unified and reliable Linux-based embedded platform that addresses the needs of device OEMs, without creating non-standard forking versions of the Linux OS. In particular, they would like to see the following (currently) fragmented technologies become mainstream in Linux with standard implementations in the main Linux kernel tree. What follows is a list of probable technology focuses for MLI, with the ultimate gaps identified and crossed in response to specific OSDL member / MLI participant contributions.
Power management
Today, if a portable device manufacturer wants to offer a Linux-based and power-managed device, he or she faces a boggling choice among divergent paradigms. OEMs can look to the desktop where notebook-centric schemes like ACPI and legacy apmd dominate, and indeed occupy most discussions of Linux power management on the kernel mailing list. There also exist power/thermal management schemes for Linux-powered blades. For non-x86/IA-32 hardware, OEMs can turn to ARM’s own power management framework, or work with the various power management schemes present on silicon from more than 200 ARM licenses (such as Intel XScale or IT OMAP). There also exist unique and further divergent energy conservation protocols from MIPS, its licensees, from Freescale for its CPU lines, from IBM for Power Architecture, from Renesas and Hitachi, and so on across the silicon supplier universe. OEMs can also choose schemes like MontaVista’s DPM and other embedded Linux supplier solutions.
While choice is a good thing, too much choice leads to fragmentation. It is thereby very likely that MLI will address the fragmented power management landscape as a Tier 1 requirement as a way to unify the current divergent definitions of this key mobile area.
Radio interface
Companies like Motorola have been building radio sets for nearly a century, and bring this hard won expertise to bear on phone designs, as do other leading players in the mobile marketplace. New entrants, and also new designs from existing suppliers, however, must face an array of daunting design challenges to build a device that meeds the requirements of both carriers and regulatory bodies like the FCC, and do so cost effectively. In today’s crop of Linux-based smart phones, the GPRS interface resides in an encapsulated modem device which can contain an additional CPU core, a DSP, and RF hardware to support wireless communications. Offloading the radio function makes it easier to build a smart phone, but also raises the cost by adding significant components to an already heavy bill of materials. While smart phones offer OEMs sufficient margins to bear this cost, the need for a self-contained modem limits Linux ability to cover the broader market that includes feature phones and entry-level devices.
Some experimental designs today remove the modem and expose the baseband interface to the application OS (as with Nucleus in low-end phones), but doing so exposes Linux to hard real-time requirements that sit at the edge of the response curve of the open source OS (despite huge advances over the last few years in preemptibility). MLI will thereby also likely tackle the best methods for cost-down through exposed baseband design in terms of real-time response, context switch, and availability of quality native call stacks on Linux.
Real time
For both the radio interface, and for other capabilities like multimedia, Linux still needs a nudge in the direction of RTOS-like responsiveness. Moreover, Linux must meet deadlines and switch context with agility in systems whose clocks can scale erratically to conserve battery power, jumping from 200 MHz peak performance down to 40 MHz (or even 0 MHz) and back in response to system policies and peripheral inputs. MLI will also need to quantify the real-time needs of real-world phone designs, and specify how Linux can best serve those requirements.
aSMP / SoC core interconnect
The current generation of ARM-based phone chip sets feature dense silicon crammed with peripherals to support phone functions such as:
Display and keyboard controllers
CODECs
Power management circuitry
WWAN and WiFi interfaces
Flash controllers
These devices can be highly stateful with intricate but hard-to-program shared memory interfaces among them. These channels constitute a troublesome performance bottleneck. MLI will certainly address ways to streamline intercore communications, building bridges among process functions for maximum transparency and performance.
Multimedia
Linux, and indeed other phone Operating Systems, today lack a unified multimedia framework, and exhibit huge challenges for interoperability, API-wise, format-wise, and in terms of a mixture of IP from differing sources. It is in the initial MLI vision to unify the existing multimedia efforts, but to do so in as open and unburdened fashion as possible.
Small footprint
While today’s smartphones ship with as much as 128 MB of flash and 64 MB of RAM, a good phone OS shouldn’t necessarily seek to occupy every last byte of available storage. Every bit used by the OS and middleware is a bit not available to OEMs for their value-added content. And, while Linux can deploy in as small a footprint at 1 MB or less, phone configurations loom much larger (such as Microsoft minimum footprints of 28-30 Mbs). MLI will certainly work with the mainstream kernel community to continue to compress the Linux minimum useful footprint to meet the needs of mobile phones and other embedded devices.
MLI deliverables
When MLI met in October for the first time in Beijing, the new body appointed interim governance and immediately got down to the job of gap analysis. The first likely deliverable will be a preliminary requirements specification, along with use cases and some marketing output. The most important deliverable, as with Carrier Grade Linux, will be the instigation of new open source project work to fill the gaps it identifies, and to bring currently divergent technologies (like those above) into the Linux mainstream.
