Going back to an old topic regarding the mix of peripherals on Stellaris parts, see: http://www.luminarymicro.com/products/whats_new.html -- Regards, Richard. + http://www.FreeRTOS.org 13 official architecture ports, 1000 downloads per week. + http://www.SafeRTOS.com Certified by T�V as meeting the requirements for safety related systems.
New Cortex-M3 parts - This time CAN and Ethernet
Started by ●September 18, 2007
Reply by ●September 18, 20072007-09-18
On Tue, 18 Sep 2007 12:51:32 GMT, "FreeRTOS.org" <noemail@address.com> wrote in comp.arch.embedded:> Going back to an old topic regarding the mix of peripherals on Stellaris > parts, see: http://www.luminarymicro.com/products/whats_new.htmlHave you looked at the STM32 Cortex M3 parts from ST? Much smaller geometry and much lower power consumption, up to 72 MHz, USB and CAN, but no Ethernet. I was at a one-day seminar they put on last week. One thing I don't like about Luminary is that you can't view or download PDF data sheets without registering. -- Jack Klein Home: http://JK-Technology.Com FAQs for comp.lang.c http://c-faq.com/ comp.lang.c++ http://www.parashift.com/c++-faq-lite/ alt.comp.lang.learn.c-c++ http://www.club.cc.cmu.edu/~ajo/docs/FAQ-acllc.html
Reply by ●September 19, 20072007-09-19
> Have you looked at the STM32 Cortex M3 parts from ST?Yep - http://www.freertos.org/portstm32iar.html A very nice system too. My support so far is limited to IAR. For the Stellaris parts I have a nice Eclipse setup now as well as GCC, Rowley and Keil. These will be extended to cover the STM32 parts too.> Much smaller > geometry and much lower power consumption, up to 72 MHz,I'm not sure that is with zero wait states though.. The Stellaris parts work at 50MHz with zero wait states. It would be good to do a comparison.> USB and CAN, > but no Ethernet. I was at a one-day seminar they put on last week.....but the USB and CAN cannot be used simultaniously. This was one of the discussion points of the original thread to which I refered. The Stellaris parts had CAN and Ethernet, but not on the same device. The STM32 parts have USB and CAN, but cannot be used simultaniously. -- Regards, Richard. + http://www.FreeRTOS.org 13 official architecture ports, 1000 downloads per week. + http://www.SafeRTOS.com Certified by T�V as meeting the requirements for safety related systems.
Reply by ●September 19, 20072007-09-19
> Out of curiosity, have you tried the Codesourcery gcc builds? I gather > that the latest release has Thumb-2 and Stellaris support.Again, yes :o) http://www.freertos.org/portLM3Sxxxx_Eclipse.html> > Do you ever do any comparisons between compilers, or do the commercial > compilers' EULAs prevent you for saying anything about them? >Yes I do comparisons, I used to publish some but it is very difficult to draw any conclusions from them as different people want different things (and the arguments go on for days). What in particular are you interested in? -- Regards, Richard. + http://www.FreeRTOS.org 13 official architecture ports, 1000 downloads per week. + http://www.SafeRTOS.com Certified by T�V as meeting the requirements for safety related systems.
Reply by ●September 19, 20072007-09-19
FreeRTOS.org wrote:>> Have you looked at the STM32 Cortex M3 parts from ST? > > Yep - http://www.freertos.org/portstm32iar.html > > A very nice system too. My support so far is limited to IAR. For the > Stellaris parts I have a nice Eclipse setup now as well as GCC, Rowley and > Keil. These will be extended to cover the STM32 parts too. >Out of curiosity, have you tried the Codesourcery gcc builds? I gather that the latest release has Thumb-2 and Stellaris support. Do you ever do any comparisons between compilers, or do the commercial compilers' EULAs prevent you for saying anything about them? mvh., David
Reply by ●September 19, 20072007-09-19
FreeRTOS.org wrote:>> Out of curiosity, have you tried the Codesourcery gcc builds? I gather >> that the latest release has Thumb-2 and Stellaris support. > > Again, yes :o) http://www.freertos.org/portLM3Sxxxx_Eclipse.html > >> Do you ever do any comparisons between compilers, or do the commercial >> compilers' EULAs prevent you for saying anything about them? >> > > Yes I do comparisons, I used to publish some but it is very difficult to > draw any conclusions from them as different people want different things > (and the arguments go on for days). What in particular are you interested > in? >I'm not looking for anything specific, I'm merely curious as to what sort of a difference the Thumb-2 changes in gcc make to code size and speed. We don't use ARM's at the moment (I use Codesourcery for their ColdFire gcc - and some day, I plan to try out the FreeRTOS MG5234 port), but ARMs are always something to consider for new projects. And of course, these arguments about compilers can be entertaining! mvh., David
Reply by ●September 20, 20072007-09-20
"FreeRTOS.org" <noemail@address.com> skrev i meddelandet news:Ta6Ii.16975$c_1.8267@text.news.blueyonder.co.uk...>> Have you looked at the STM32 Cortex M3 parts from ST? > > Yep - http://www.freertos.org/portstm32iar.html > > A very nice system too. My support so far is limited to IAR. For the > Stellaris parts I have a nice Eclipse setup now as well as GCC, Rowley and > Keil. These will be extended to cover the STM32 parts too. > >> Much smaller >> geometry and much lower power consumption, up to 72 MHz, > > I'm not sure that is with zero wait states though.. The Stellaris parts > work at 50MHz with zero wait states. It would be good to do a comparison. >ST can run up to 24 MHz with zero waitstates. At 72 MHz, they need to run with two waitstates so any load from flash will be slow. The preliminary datasheet does not mentions any memory accelerator, so max bandwidth to flash is in this case (24 MHz * 4 bytes) = 96 MB/s. I really doubt you can support 90 MIPS with 96 MB/s flash bandwidth, so either the datasheet brief is not telling the complete story, or the 90 MIPS is only when executing from SRAM The AVR uC3 series, as a comparision, supports 33 MHz on first access and 66 MHz access on subsuquent accesses for a min throughput of 132 MB/s and a max throughput of 266 MB/s from flash. If you compare power consumption, the STM32F103 will use 22 mA at 36 MHz and the uC3B will use 14 mA. The uC3B does not have the CAN yet, so that has to wait for the uC3C coming next year. The biggest drawback of the ST parts seems to be the memory configuration. Very few ARM7 customers are looking for less than 32 kB of SRAM. I see the typical size starts at 128 kB and many people want to migrate abnove 256 kB. With 6-20 kB of SRAM, and 32-128 kB of flash ST must be focusing more on replacing 8 bitters, than replacing other 32 bitters.>> USB and CAN, >> but no Ethernet. I was at a one-day seminar they put on last week. > > > ....but the USB and CAN cannot be used simultaniously. This was one of > the discussion points of the original thread to which I refered. The > Stellaris parts had CAN and Ethernet, but not on the same device. The > STM32 parts have USB and CAN, but cannot be used simultaniously. > > -- > Regards, > Richard. > > + http://www.FreeRTOS.org > 13 official architecture ports, 1000 downloads per week. > > + http://www.SafeRTOS.com > Certified by T�V as meeting the requirements for safety related systems. >-- Best Regards, Ulf Samuelsson This is intended to be my personal opinion which may, or may not be shared by my employer Atmel Nordic AB
Reply by ●September 21, 20072007-09-21
On Sep 20, 2:41 pm, "Ulf Samuelsson" <u...@a-t-m-e-l.com> wrote:> "FreeRTOS.org" <noem...@address.com> skrev i meddelandetnews:Ta6Ii.16975$c_1.8267@text.news.blueyonder.co.uk... > > >> Have you looked at the STM32 Cortex M3 parts from ST? > > > Yep -http://www.freertos.org/portstm32iar.html > > > A very nice system too. My support so far is limited to IAR. For the > > Stellaris parts I have a nice Eclipse setup now as well as GCC, Rowley and > > Keil. These will be extended to cover the STM32 parts too. > > >> Much smaller > >> geometry and much lower power consumption, up to 72 MHz, > > > I'm not sure that is with zero wait states though.. The Stellaris parts > > work at 50MHz with zero wait states. It would be good to do a comparison. > > ST can run up to 24 MHz with zero waitstates. > At 72 MHz, they need to run with two waitstates so any load from flash > will be slow. The preliminary datasheet does not mentions any > memory accelerator, so max bandwidth to flash is in this case > (24 MHz * 4 bytes) = 96 MB/s. > I really doubt you can support 90 MIPS with 96 MB/s flash bandwidth, > so either the datasheet brief is not telling the complete story, > or the 90 MIPS is only when executing from SRAM > > The AVR uC3 series, as a comparision, supports 33 MHz on first > access and 66 MHz access on subsuquent accesses for a > min throughput of 132 MB/s and a max throughput of 266 MB/s > from flash. > > If you compare power consumption, the STM32F103 will use 22 mA at 36 MHz > and the uC3B will use 14 mA. > > The uC3B does not have the CAN yet, so that has to wait for the uC3C > coming next year. > > The biggest drawback of the ST parts seems to be the memory configuration. > Very few ARM7 customers are looking for less than 32 kB of SRAM. > I see the typical size starts at 128 kB and many people want > to migrate abnove 256 kB. > With 6-20 kB of SRAM, and 32-128 kB of flash ST must be focusing > more on replacing 8 bitters, than replacing other 32 bitters.I would rate it somewhere between AVR8 and AVR32, depending on the function. It's a much improved design than the LMI, with better power and clock options. The 1 MSPS 12 bits A2D is a bonus. Overall, not a bad deal for $1.80@10K for the 36MHz version.
Reply by ●September 21, 20072007-09-21
"Ulf Samuelsson" <ulf@a-t-m-e-l.com> wrote in message news:fcuqh0$1kd$1@aioe.org...> "FreeRTOS.org" <noemail@address.com> skrev i meddelandet > news:Ta6Ii.16975$c_1.8267@text.news.blueyonder.co.uk... >>> Have you looked at the STM32 Cortex M3 parts from ST? >> >> Yep - http://www.freertos.org/portstm32iar.html >> >> A very nice system too. My support so far is limited to IAR. For the >> Stellaris parts I have a nice Eclipse setup now as well as GCC, Rowley >> and Keil. These will be extended to cover the STM32 parts too. >> >>> Much smaller >>> geometry and much lower power consumption, up to 72 MHz, >> >> I'm not sure that is with zero wait states though.. The Stellaris parts >> work at 50MHz with zero wait states. It would be good to do a >> comparison. >> > > ST can run up to 24 MHz with zero waitstates. > At 72 MHz, they need to run with two waitstates so any load from flash > will be slow. The preliminary datasheet does not mentions any > memory accelerator, so max bandwidth to flash is in this case > (24 MHz * 4 bytes) = 96 MB/s. > I really doubt you can support 90 MIPS with 96 MB/s flash bandwidth, > so either the datasheet brief is not telling the complete story, > or the 90 MIPS is only when executing from SRAM > > The AVR uC3 series, as a comparision, supports 33 MHz on first > access and 66 MHz access on subsuquent accesses for a > min throughput of 132 MB/s and a max throughput of 266 MB/s > from flash.Did I mention that FreeRTOS.org also supports the UC3? :o) http://www.freertos.org/portAVR32.html -- Regards, Richard. + http://www.FreeRTOS.org 13 official architecture ports, 1000 downloads per week. + http://www.SafeRTOS.com Certified by T�V as meeting the requirements for safety related systems.
Reply by ●September 21, 20072007-09-21
"Ulf Samuelsson" <ulf@a-t-m-e-l.com> wrote in message news:fcuqh0$1kd$1@aioe.org...> "FreeRTOS.org" <noemail@address.com> skrev i meddelandet news:Ta6Ii.16975$c_1.8267@text.news.blueyonder.co.uk... >>> Have you looked at the STM32 Cortex M3 parts from ST? >> >> Yep - http://www.freertos.org/portstm32iar.html >> >> A very nice system too. My support so far is limited to IAR. For the Stellaris parts I have a nice Eclipse setup >> now as well as GCC, Rowley and Keil. These will be extended to cover the STM32 parts too. >> >>> Much smaller >>> geometry and much lower power consumption, up to 72 MHz, >> >> I'm not sure that is with zero wait states though.. The Stellaris parts work at 50MHz with zero wait states. It >> would be good to do a comparison. >> > > ST can run up to 24 MHz with zero waitstates. > At 72 MHz, they need to run with two waitstates so any load from flash > will be slow. The preliminary datasheet does not mentions any > memory accelerator, so max bandwidth to flash is in this case > (24 MHz * 4 bytes) = 96 MB/s.You mean 192MBytes/s as the flash is 64-bit wide. Ie. similar to the 200MBytes/s of the Luminary parts.> I really doubt you can support 90 MIPS with 96 MB/s flash bandwidth, > so either the datasheet brief is not telling the complete story, > or the 90 MIPS is only when executing from SRAMRunning from flash is typically slower than running from SRAM (Luminary parts being the exception). But this is not due to bandwidth limitations. 192MBytes/s provides more than enough bandwidth for running a 16/32-bit instruction set at 72MHz. An average instruction is ~2.7 bytes and takes ~1.3 cycles, ie. 2.7 * 72 / 1.3 = 150MBytes/s bandwidth required. So bandwidth is not an issue. The real issue with waitstates is that branches become more expensive. Wilco
