rickman wrote:>> > PS: BTW, just out of interest, which stack based processors do you use >> > ? >> >> Have used Novix NC4016, Harros RTX2000 and 2010. More recently I have >> been looking at the Intellasys SeaForth chips but am probably going with >> the Green ArraysGA144for the next projects. > > What sorts of projects do you expect to be using the GA144?One is a logging instrumentation node. The other project, more for fun than anything else, is a small robot with binocular vision sensing. -- ******************************************************************** Paul E. Bennett...............<email://Paul_E.Bennett@topmail.co.uk> Forth based HIDECS Consultancy Mob: +44 (0)7811-639972 Tel: +44 (0)1235-510979 Going Forth Safely ..... EBA. www.electric-boat-association.org.uk.. ********************************************************************
Getting started with ARM processors - recommendations
Started by ●October 25, 2010
Reply by ●November 16, 20102010-11-16
Reply by ●November 16, 20102010-11-16
2010-11-11 22:41, rickman skrev: <snip>>> >>>>> Atmel doesn't make the flavor you need, go pound sand. With ARM you >>>>> can go to another vendor to find just the mix that suits your app >>>>> best. >>>> OK, where do I order samples with 512 kB Flash and floating point unit? >>>> Kinetis wont be available for a year. >> >>> Here is a perfect example of the problems with sole sourced parts. To >>> work with an AVR32 I have to use the Atmel site. There the floating >>> point is listed as something that "will be" offered. So is the site >>> wrong or are you wrong? With the ARM community if I don't like one >>> vendor because they have a poor web site and poor selection guides. >>> In fact, the only selection guide I can find indicates the AVR32 has >>> no peripherals other than Ethernet, USB and... well, that's it! It >>> also shows no parts with FPU... >> >> You can download the AT32UC3C Series Preliminary Datasheet from: >> >> http://www.atmel.com/dyn/resources/prod_documents/doc32117.pdf >> >> Brief Overview at:http://www.atmel.com/dyn/corporate/view_detail.asp?FileName=Atmel_Int... >> >> The part is released to production. >> There are rev C samples and development kits in stock. >> >> The production revision is Rev D. >> The difference between rev C and rev D is increased functionality. >> Rev D includes Ethernet and additional USART and I2C. >> >> >> >>> http://www.atmel.com/dyn/products/param_table.asp?family_id=607&Order... >> >>> This is exactly the sort of stuff I mean when I talk about the >>> advantages of having multiple vendors for your chosen CPU. >> >>>> Truth is that it is so easy to change CPU architecture nowadays, >>>> that you can live with multiple architectures. >>>> If you don't find an ABR32 which will do the job, you >>>> can go to an ARM, but why select a poor chip when you can get a good chip? >>>> That is what consultants do all the time. >> >>> Yes, this is what I would expect to hear from the AVR32 vendor. But >>> why select a dead end chip when I can select any of literally hundreds >>> of variations in an ARM and simply not need to change tools and >>> startup code? But you can ignore what your customers want... >> >>> BTW, when are your CM4 parts due out? Oh, that's right, you haven't >>> gotten your CM0 parts out yet... >> >>> Rick >> >> -- >> Best Regards >> Ulf Samuelsson >> These are my own personal opinions, which may >> or may not be shared by my employer Atmel Nordic AB > > I repeat the question... is the web site wrong or does Atmel consider > this part not yet available? > > RickSince the part was put up on the website 2010-11-10, neither of the alternatives are correct. Best Regards Ulf Samuelsson
Reply by ●November 17, 20102010-11-17
For those interested in practical attacks and practical defenses please see my paper, "Overcoming an untrusted computing base: detecting and removing malicious hardware automatically" from this year's IEEE Security and Privacy. We describe attacks that pass all processor certification tests that can be triggered by opcodes or data. Unfourtanetly the SA article didn't present the latest and greatest in the area. ---Matthew Hicks www.firefalcon.com> On Nov 1, 5:48 pm, "Paul E. Bennett" <Paul_E.Benn...@topmail.co.uk> > wrote: > >> Simon Clubley wrote: >> >>> On 2010-10-31, Paul E. Bennett <Paul_E.Benn...@topmail.co.uk> wrote: >>> >>>> So you think it might just be sensationalism then? >>>> >>> Yes and no. >>> >>> There's clearly a major element of sensationalism, but there's also >>> a real potential issue which has been identified as well. What I >>> don't have a feeling for is where viable reality becomes >>> sensationalism. >>> >> [%X] >> >>> One could argue the hardware based approach allows malicious logic >>> to be more easily hidden, but are there less risky ways of achieving >>> your goals ? >>> >> Software based attacks seem to be doing just fine for the moment. >> >>> If there is a real advantage, is that advantage sufficient for the >>> risks and resources involved in taking a hardware based approach ? >>> >>> I honestly do not know. >>> >> I am not sure anyone does. >> >>> Simon. >>> >>> PS: BTW, just out of interest, which stack based processors do you >>> use ? >>> >> Have used Novix NC4016, Harros RTX2000 and 2010. More recently I have >> been looking at the Intellasys SeaForth chips but am probably going >> with the Green ArraysGA144for the next projects. >> > What sorts of projects do you expect to be using the GA144? > > Rick >
Reply by ●December 21, 20102010-12-21
On 2010-11-17, Matthew Hicks <mdhicks2@uiuc.edu> wrote:> For those interested in practical attacks and practical defenses please see > my paper, "Overcoming an untrusted computing base: detecting and removing > malicious hardware automatically" from this year's IEEE Security and Privacy. > We describe attacks that pass all processor certification tests that can > be triggered by opcodes or data. Unfourtanetly the SA article didn't present > the latest and greatest in the area. >Yes, I know the post I am replying to is a month old, but I have a reading backlog and just got around to reading this paper :-). It's a interesting paper, which covers a detection technique which can be useful in many circumstances. The crucial thing however, is that the technique only works when the overall hardware producer can be trusted as the technique is designed to find a rogue subcontractor who makes unauthorised modifications to a design. It doesn't cover the case when the overall hardware producer is willing to insert a backdoor at the request of a third party (eg: a nation's security organisation). However, it's still a interesting read and explains multiple and viable ways of implementing and triggering backdoors in hardware, one or two of which I had not really considered. Simon. -- Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP Microsoft: Bringing you 1980s technology to a 21st century world
