|
hobbesmaster posted:just have a VM for every crazy tool he thinks synthesizing is unimportant for validation, imma guess emulation isn't his forte
|
# ? Aug 5, 2015 16:45 |
|
|
# ? Jun 1, 2024 05:37 |
|
be nice to validation people, unless they just assume the design is right and write a bunch of tests that always make sure the design is what the design is, then smack them!
|
# ? Aug 5, 2015 17:40 |
|
i guess this is as good as time to ask as any. how is chip level validation _supposed_ to work? I get this basic idea: 1) Design reads <IP-SPEC> then implements <IP-RTL>. 2) Validation reads <IP-SPEC"> then tests "IP-RTL" against the spec. how reasonable is it for design to make a mistake, then validation to make the identical mistake that results in the mis-configuration of entire modules? Should validation have a black-box view of the RTL? Should validation have access to any of the definitions internal to the design?
|
# ? Aug 5, 2015 18:09 |
|
pre-si has to be whitebox, post-si has to be blackbox
|
# ? Aug 5, 2015 18:21 |
|
JawnV6 posted:he thinks synthesizing is unimportant for validation, imma guess emulation isn't his forte you'd be correct i don't do emulation
|
# ? Aug 5, 2015 19:14 |
|
Barnyard Protein posted:i guess this is as good as time to ask as any. how is chip level validation _supposed_ to work? you have to have access to actually debug things in pre-si, but a good pre-si validation strategy will include transaction level abstraction and throw a ton of transactions at a design using an implementation of the interfaces written with no reference to the RTL.
|
# ? Aug 5, 2015 19:20 |
|
kwinkles posted:hey thread. question: i have been wanting to get an fpga and start a github with all the poo poo i make it do but i have been too lazy to fpga shop. what is a good one to get that i can do things on for like $1000? where do you even find that out? are there toolchains i can use on my mac? if you want to go brand x, i recommend the avnet microzed for $200 plus an io breakout board for $50. it's a zynq part so it's actually an arm soc with some 7 series fpga fabric on the side, which is a very nice combo. boots linux from a microsd card out of the box, 1gb ram, gigabit enet, dual core cortex-a9 so it's reasonable fast, and loading a bitfile into the fpga fabric is literally "cat yospos.bit >/dev/some/path/i/dont/remember" run the tools in a linux vm. the license voucher you get with xilinx eval boards will require nodelocking based on ethernet MAC, so make one up for a virtual ethernet interface in your vm, nodelock yr license to that, and you will not be forever tied down to one specific computer idk what's out there for brand a BobHoward fucked around with this message at 10:36 on Aug 6, 2015 |
# ? Aug 6, 2015 10:34 |
|
BobHoward posted:if you want to go brand x, i recommend the avnet microzed for $200 plus an io breakout board for $50. it's a zynq part so it's actually an arm soc with some 7 series fpga fabric on the side, which is a very nice combo. boots linux from a microsd card out of the box, 1gb ram, gigabit enet, dual core cortex-a9 so it's reasonable fast, and loading a bitfile into the fpga fabric is literally "cat yospos.bit >/dev/some/path/i/dont/remember" can confirm this is as decent a way to get started as any
|
# ? Aug 6, 2015 17:38 |
|
BobHoward posted:if you want to go brand x, i recommend the avnet microzed for $200 plus an io breakout board for $50. it's a zynq part so it's actually an arm soc with some 7 series fpga fabric on the side, which is a very nice combo. boots linux from a microsd card out of the box, 1gb ram, gigabit enet, dual core cortex-a9 so it's reasonable fast, and loading a bitfile into the fpga fabric is literally "cat yospos.bit >/dev/some/path/i/dont/remember" hmm is that devcfg driver in xilinx mainline? remember what its called?
|
# ? Aug 6, 2015 17:46 |
|
movax posted:hmm is that devcfg driver in xilinx mainline? remember what its called? this help? http://www.xilinx.com/support/answers/46913.html
|
# ? Aug 6, 2015 18:00 |
|
what is the difference between a microcontroller like and arduino and an fpga?
|
# ? Aug 7, 2015 22:51 |
|
Awia posted:what is the difference between a microcontroller like and arduino and an fpga? A microcontroller is a small CPU with some other bits stuck onto the side, and those other bits are fairly inflexible beyond the specific tasks they were designed to perform. For a given task you choose a microcontroller which has all of the peripherals to get a particular job done and then ignore the peripherals you don't care about. An FPGA can be more or less any digital circuit, including a CPU. But it's not as fast as a "real" digital circuit etched into silicon. An FPGA can do most things ("things" typically being "a form of digital I/O that this microcontroller wasn't specifically designed for" or "apply this mathematical formula to this stream of one billion integers") much faster than a microcontroller, to the point where an FPGA is the only practical way of getting that task done. The downside is that FPGAs are a lot more expensive and a lot more difficult to program for. They're often used together, too: for example, an FPGA to do some bulk data processing and a microcontroller to handle control commands sent over USB. The microcontroller will have a hardware USB core that can be configured to autonomously shovel fast streams of data between the FPGA and the host computer's USB without the microcontroller CPU getting involved.
|
# ? Aug 8, 2015 14:51 |
|
Mr Dog posted:A microcontroller is a small CPU with some other bits stuck onto the side, and those other bits are fairly inflexible beyond the specific tasks they were designed to perform. For a given task you choose a microcontroller which has all of the peripherals to get a particular job done and then ignore the peripherals you don't care about. An FPGA can be more or less any digital circuit, including a CPU. But it's not as fast as a "real" digital circuit etched into silicon. don't stop i'm almost there
|
# ? Aug 8, 2015 15:10 |
|
now do a comparison between FPGAs and dedicated DSPs
|
# ? Aug 8, 2015 21:45 |
|
atomicthumbs posted:now do a comparison between FPGAs and dedicated DSPs be sure to include FPGAs with dedicated DSP subsystems too
|
# ? Aug 8, 2015 23:24 |
|
eschaton posted:be sure to include FPGAs with dedicated DSP subsystems too Jesus Christ
|
# ? Aug 9, 2015 00:43 |
|
thank you Mr Dog
|
# ? Aug 9, 2015 00:46 |
|
microprocessors and microcontrollers
|
# ? Aug 9, 2015 02:27 |
|
JawnV6 posted:microprocessors and microcontrollers I started to type out a post to answer this one but god drat is it difficult.
|
# ? Aug 9, 2015 02:32 |
|
i have a hard time with communicating that an explanation is a simplification that won't hold up for more advanced applications of the concept i'm talking about. like for the microprocessor vs microcontroller comparison: the 100k ft view is the differences in level of integration. but there are so many other factors that feed into why and how and what that i get lost in my head about what to write next.
|
# ? Aug 9, 2015 02:37 |
|
A microcontroller is, in general, a low-power, single-chip computer designed to be connected to things that aren't the rest of a computer. A microprocessor is not.
|
# ? Aug 9, 2015 08:24 |
|
JawnV6 posted:microprocessors and microcontrollers processor computers microprocessor computes, on a single chip (!) microcontoller computes and does other stuff, on a single chip like an 8051 is an 8080 or whatever, plus stuff, plus maybe some RAM and ROM and I/O instead of an address bus and data bus
|
# ? Aug 9, 2015 09:47 |
|
ive tried to slice it by how much extra crap you need on board to support a microprocessor, like a uc will boot with a couple wires and nothing so complicated as a PMIC but it's a weird distinction as you get closer to the line
|
# ? Aug 9, 2015 18:21 |
|
internal memory and peripherals? anyone seen a MCU without them?
|
# ? Aug 9, 2015 20:08 |
|
the line is pretty blurred, and it blurs more with older chips. you can get microcontrollers with no internal program memory, and the peripherals consist of a single timer
|
# ? Aug 9, 2015 20:12 |
|
I would draw the line between the philosophy behind the MCU/processors intended use of the internal byte code architecture no one (sane) is making really elaborate operating systems that run on PIC, you... could but it's not suited for it compared to say a Motorola SoC with an arm core
|
# ? Aug 10, 2015 03:42 |
|
Mido posted:I would draw the line between the philosophy behind the MCU/processors intended use of the internal byte code architecture yeah but there are PICs with MIPS cores
|
# ? Aug 10, 2015 03:52 |
|
Barnyard Protein posted:yeah but there are PICs with MIPS cores my language was carefully chosen
|
# ? Aug 10, 2015 03:56 |
|
yeah i get what you mean and i agree, the dividing line is the application, the architecture typically follows from that.
|
# ? Aug 10, 2015 03:59 |
|
my theory, which is mine, is that microcontrollers are distinguished by specialization for running soft- or hard-realtime control code, usually without a full operating system this results in some common characteristics: 1. integrated peripherals, and generally not high speed either, rather whatever's useful for interfacing to random chips (i2c, spi, rs232, gpio, etc) 2. integrated program memory (flash) 3. integrated data memory (small amounts of sram) 4. little or no cache 5. cpu core / integrated peripherals / the bus connecting them together are either the same clock domain or quasi-synchronous. accesses to peripheral registers take a fixed (and very small) number of clock cycles these characteristics make it possible to easily reason about / measure / predict execution timing. if you need to bitbang something or control some physical process or whatever that is exactly what you want this is why arm offers the cortex-m series, they're arm cores specifically optimized for control applications rather than going fast. microcontroller designers could easily use a much faster arm core instead and thanks to moore's law the costs are not prohibitive at all in most cases, but those cores bring with them system architectures designed for higher general purpose performance, which means async crossings, deep queueing, caches, etc. throughput's great, latency and predictability are not
|
# ? Aug 10, 2015 04:56 |
|
i'm finally playing with atmel studio! wow, I really wish arduino's IDE would include some of the handy tools they have in here. like fuckin hate windows though, and my work machine is slower than my personal. I may look into VM'ing or something anyways road plan: 1) keep coding the arduino way with .ino files and use the nice features of atmel studio/visual micro debugging until I can operate in the IDE efficiently 2) start more bitwise functions for I/O and whatever 3) move out of using .ino files and start using .c/.cpp files for the majority of programming 4) start trying out hardware debuggers? 5) thats probably it because IDK if I am going to be able to/be interested in/need to get further into the hardware development
|
# ? Aug 19, 2015 02:49 |
|
Fanged Lawn Wormy posted:like fuckin hate windows though, and my work machine is slower than my personal. I may look into VM'ing or something
|
# ? Aug 19, 2015 03:13 |
|
that's good to know. do you use VMware? I've looked around and it all the options seem to come out about equal but VMware seems to be the pick for what I want
|
# ? Aug 19, 2015 04:09 |
|
BobHoward posted:easily reason about / measure / predict execution timing this whole post is good, thank u. i particularly enjoy this bit to separate out the two i was asked by someone writing a book aimed at makers and a little higher how to describe the difference, and on the spot I couldn't come up with a simple definition that a mechanical engineer could understand.
|
# ? Aug 19, 2015 22:06 |
|
Fanged Lawn Wormy posted:that's good to know. do you use VMware? I've looked around and it all the options seem to come out about equal but VMware seems to be the pick for what I want ive used parallels and it worked fine enough, asking new job to hook me up with that too
|
# ? Aug 19, 2015 22:10 |
|
anyone have experience with RIOT, contiki, or other "iot rtos" solutions? i need a scheduler for a wifi/tcp driver, thinking FreeRTOS will be plenty, but if one of these upstarts is really compelling I'd consider it
|
# ? Aug 21, 2015 21:02 |
|
Don't use TinyOS I mean, not that anybody sane should ever see an OS written in a toy programming language and decide this could be even remotely worth using, what sort of baka would do something like that anyway ahahahahaha *anime sweatdrop* but yeah TinyOS is poo poo, that's what I have to contribute.
|
# ? Aug 21, 2015 21:15 |
|
Mr Dog posted:but yeah TinyOS is poo poo, that's what I have to contribute.
|
# ? Aug 21, 2015 22:58 |
|
i janitored way too many bits today (i'm responsible for the pc software for this daq system, wish i had control over the firmware for the front end, would be way easier than iterating with an engineer in strasbourg who doesn't like to give too many details away)
|
# ? Aug 21, 2015 23:39 |
|
|
# ? Jun 1, 2024 05:37 |
|
so i want to use an fpga for a thing, i think i saw someone mention the mojo v3 board as a good starting point, but i don't know much about this poo poo, anyone who knows more about it willing to slap me upside the head and tell me im an idiot for choosing that?
|
# ? Aug 26, 2015 11:29 |