[Arm-netbook] [review] SoC proposal
ikgeorgiev at gmail.com
Thu Feb 9 14:08:30 GMT 2012
2012/2/8 lkcl luke <luke.leighton at gmail.com>
> folks, hi,
> i've not mentioned this before, publicly, but some time last year when
> this initiative was still being considered we were finding it so
> ridiculously difficult to find cooperative CPU companies that we gave
> serious consideration to putting a deal together to create a new
> (entirely software-based) SoC that would be entirely FSF-Endorseable
> as it would be entirely software-programmable.
> based around something like the ARC 32-bit RISC core (which was a hell
> of a lot better than ARM's offerings at the time), the idea was to put
> at least 4 of them down in 28nm, where they would quite happily run at
> at least 1.5ghz.
> the problem was that ARC's RISC core design wasn't SMP capable, and
> the team behind it didn't feel comfortable doing SMP cache coherency.
> we _did_ come up with a mad and very simple scheme to do cache
> coherency as a software interrupt (similar to how MMU page-swapping is
> done) and even discussed it on LKML thanks to alan cox and some input
> from another guy from intel, it turns out to have been a workable
> ... but, we didn't pursue it.
> some time last month i went "hang on a minute, maybe now's a better
> time" so i've done a draft of the interfaces (DDR3 RAM will be
> included) the number of pins required is surprisingly low, which will
> get the cost down.
> i'd therefore greatly appreciate some help reviewing the pinouts. the
> company that i've found that has an alternative 32-bit RISC core not
> only has SMP cache coherency already done, but also they have an
> absolutely amazing set of Instruction extensions, including DSP,
> Audio, Video, Base-band (for RF handset processing) and much more.
> the plan is, therefore, to target this CPU at a very very wide range
> of markets, based on it having:
> * 8 CPUs at 1.2ghz or above
> * SMP Cache Coherency
> * 32-bit DDR3 1333mhz RAM (with a 2nd version having 2 DDR3 interfaces)
> * virtually everything software-programmable (with the exception of
> CABAC decode)
> meaning that it will do 3D graphics _and_ 1080p Video entirely in
> software. the interfaces i've selected so far will include:
> * HDMI Out _and_ In
> * 24-pin RGB/TTL
> * 2-channel LVDS
> * 2 PCIe (2-lane each)
> * USB-OTG, USB-2 and USB-3
> * SATA-3
> * NAND controller (8-bit with 4-way CS)
> * 3 SD/MMC interfaces
> * 3 SPI interfaces
> * 3 UARTs, 3 IIC interfaces, CAN-Bus, 2 PS/2, Touchscreen, 3 PWMs
> * 2 MPEG Transport Stream Interfaces
> * Smartcard Interface
> that means that it could be used in at least the following products:
> * Laptops, Netbooks, Tablets, Desktops, PCs, NAS-Boxes
> * PVRs, TVs, Set-Top Boxes, Satellite Decoders
> * Smartphones, Base Stations, GNU/Radio SDR Products
> bearing in mind that this will be out some time in middle of 2013, if
> it's started soon, i could reallly do with some help reviewing the
> interfaces and capabilities, to make sure i've got it right.
> arm-netbook mailing list arm-netbook at lists.phcomp.co.uk
> Send large attachments to arm-netbook at files.phcomp.co.uk
The direction that you are evaluating is similar to that of NVidia.
According to the source (full of rumors and not the most credible one
according to me) a future product of Nvidia with code name Denver (or Tegra
5) is a SoC that contains together CPU and GPU. The CPU side will be "a
software/firmware based ‘code morphing’ CPU like Transmeta". (The latter
had VLIW-like architecture that relies on software compiler). The release
date of Denver is planned in 2013/2014 time frame. :)
If the free/open source software developers have enough space to play with
the processor, proposed by you, they will not let you down. More over if
you give them more freedom than NVidia solution. So I am full of positivism
about your direction.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the arm-netbook