| kyak | xiangfu: http://qt-apps.org/content/show.php/kinyin?content=95926 | 05:33 |
|---|---|---|
| kyak | xiangfu: might be interesting for you :) | 05:34 |
| viric | kristianpaul: trabajas en Loongson2f ahora? :) | 06:16 |
| viric | xiangfu: did you commit that 'nerase' change? | 06:17 |
| viric | xiangfu: (I'm lazy to erase the flash... Do you know how should I backup my ubifs for a later nprog ? | 06:17 |
| viric | ) | 06:17 |
| viric | xiangfu: ni hao ma? | 06:21 |
| xiangfu | viric: mount the root to another folder. then use mkfs.ubifs to create an ubi file base on this folder. :) | 06:28 |
| viric | hmmm is it 'the way'? :) | 06:28 |
| xiangfu | viric: that is come out first in my mind. :) | 06:29 |
| viric | ah ok | 06:29 |
| xiangfu | viric: I am not commit that change. | 06:30 |
| viric | xiangfu: why don't you commit it? | 06:31 |
| viric | xiangfu: don't you think it is better than now? | 06:31 |
| xiangfu | viric: I am not well test that change. | 06:31 |
| xiangfu | viric: recently I always working on the openwrt image. | 06:31 |
| xiangfu | viric: and u-boot, a little slow on nerase. :( | 06:32 |
| viric | ok | 06:32 |
| kristianpaul | viric: hace un tiempo tengo uno | 07:09 |
| kristianpaul | viric: y ciertas cosas no ejecutan bien, por falta del SIMD :S | 07:09 |
| kristianpaul | viric: tambien queria comparar el simd del loonsoon con el del xbusrt a ver que puedo esperar de ambos | 07:12 |
| viric | kristianpaul: yo tengo un par de minipc | 07:13 |
| viric | kristianpaul: qué no "ejecuta bien"? | 07:14 |
| kristianpaul | viric: gps-sdr.com | 07:14 |
| viric | no carga | 07:16 |
| kristianpaul | no :( | 07:17 |
| kristianpaul | me hace falta, no se que paso no se a quien contactar para preguntas | 07:17 |
| kristianpaul | almenos tengo el codigo | 07:17 |
| kristianpaul | hmm espera | 07:17 |
| kristianpaul | viric: http://www.ctae.org/sdr/ (mirror del codigo) | 07:18 |
| viric | hey the USRP gnu radio is super expensive :) | 07:21 |
| kristianpaul | yes , but i plan get one any time next year i hope | 07:22 |
| kristianpaul | is relly usefull for debugging | 07:22 |
| kristianpaul | rellay* | 07:22 |
| viric | it looks very ogod | 07:23 |
| viric | good | 07:23 |
| viric | a radio laboratory should have one | 07:23 |
| kristianpaul | viric: less espensive that buying separate radio stuff | 07:24 |
| viric | I agre | 07:24 |
| viric | e | 07:24 |
| kristianpaul | also you add it a ITX board and you're done | 07:24 |
| kristianpaul | i saw openBTS shop and they sell it with on, is good idea | 07:24 |
| viric | can it do CDMA? | 07:24 |
| kristianpaul | i dont know | 07:25 |
| viric | kristianpaul: how you test the speed on the minipc for the gps code' | 07:26 |
| kristianpaul | correlation time, but first i need pass the simd-like self test :) | 07:27 |
| viric | kristianpaul: cómo lo pruebas? | 07:30 |
| viric | can't you test it without simd? | 07:31 |
| kristianpaul | hehe , no i think , simd is heart for acquisition step | 07:32 |
| viric | does the loongson2f have simd instructions for floating point? | 07:36 |
| viric | or your calculations don't involve floating point? | 07:36 |
| kristianpaul | i dont know yet, i need read manual first | 07:37 |
| kristianpaul | float point dont know either, is just FFT from data-like -1,0,1 | 07:37 |
| viric | for what I read... | 07:40 |
| viric | it only has integer SIMD instructions | 07:40 |
| viric | so, they can work as fixed point. | 07:40 |
| kristianpaul | damn mocp dint fit on nanonote screen :/ | 08:25 |
| wpwrak_ | kristianpaul: (simd) you mean the correlation ? | 08:27 |
| kristianpaul | nope | 08:28 |
| kristianpaul | MMX like instructions | 08:28 |
| wpwrak_ | kristianpaul: yes, but for what kind of operation ? | 08:28 |
| kristianpaul | wpwrak_: fft | 08:29 |
| wpwrak_ | ah yes, fft should be happier with that | 08:29 |
| viric | are you using fftw3? | 08:45 |
| kristianpaul | i will for first test, but i want to try re-use gps-sdr code later | 08:47 |
| qi-bot | [commit] Cristian Paul Peñaranda Rojas: Sync signal to mcu is taken from SiGE as result removed non needed code and fixed identation http://qi-hw.com/p/ben-gps-sdr/f7203fa | 09:07 |
| qi-bot | [commit] Cristian Paul Peñaranda Rojas: Updated simulation, reset trigger removed http://qi-hw.com/p/ben-gps-sdr/f0ae342 | 09:07 |
| qi-bot | [commit] Cristian Paul Peñaranda Rojas: Pin asigment is completed http://qi-hw.com/p/ben-gps-sdr/c379297 | 09:07 |
| qi-bot | [commit] Cristian Paul Peñaranda Rojas: Added README with basic description and spected behavior http://qi-hw.com/p/ben-gps-sdr/31ad2b4 | 09:12 |
| qi-bot | [commit] Cristian Paul Peñaranda Rojas: Updated makefile to be more friendly and removed non required features http://qi-hw.com/p/ben-gps-sdr/4e57bf8 | 09:21 |
| kristianpaul | he | 15:03 |
| kristianpaul | something really weird happen after load bistream to fpga, screen just got crazy :/ and i lost ssh conection | 15:03 |
| lekernel | on sie? | 15:04 |
| kristianpaul | yes | 15:04 |
| kristianpaul | hmm is my bitstream | 15:11 |
| kristianpaul | what's wrong with port A :/ | 15:12 |
| kristianpaul | In theory i just send a 4bits data plus a clock signal at 2.048Mhz | 15:16 |
| Madridman | hi | 16:51 |
| Madridman | please anybody could recommend the best book to learn about gpu's? basics | 16:52 |
| Madridman | I would like to learn the typical GPU's architectures, theorical things | 17:06 |
| Madridman | not interested in programming GL and such | 17:07 |
| Madridman | everybody's sleeping? | 17:07 |
| viric | are you more interested in the power supply and clock of the GPU? | 17:08 |
| lekernel | Madridman: if you're interested in how GPUs are made, have a look at my thesis (on milkymist), and specifically the chapter on the texture mapping unit (tmu) | 17:11 |
| lekernel | the tmu implements several fundamental features of GPUs | 17:11 |
| lekernel | but if you have no prior experience with opengl, I'd recommend playing around a bit with it for starters | 17:11 |
| lekernel | it will really help you understand what is going on | 17:11 |
| Madridman | lekernel I would like to check your thesis, where is milymist? | 17:12 |
| Madridman | not interested in opengl, just electronic architecture theorics | 17:12 |
| lekernel | milkymist.org | 17:12 |
| Madridman | i do not pretend to program | 17:12 |
| lekernel | well, opengl understanding will really come handy | 17:12 |
| lekernel | and opengl is already pretty low-level itself | 17:13 |
| Madridman | sure but before it could be good to know how the memory etc. are placed | 17:13 |
| Madridman | all the buses and rest of components | 17:14 |
| viric | lekernel: is the milkymist source available? | 17:14 |
| lekernel | yes, of course | 17:14 |
| viric | (to me, I mean, for example :) | 17:14 |
| Madridman | I know plastation and xbox have nothing to do one each other | 17:14 |
| viric | lekernel: is it in vhdl? | 17:14 |
| lekernel | no, verilog | 17:14 |
| viric | ouch. Bad luck I had. :) | 17:14 |
| lekernel | verilog is little more than a less verbose vhdl, so if you already know VHDL well enough you'll have little trouble with verilog | 17:15 |
| Madridman | lekernel sorry, I accessed the web page http://milkymist.org/ but I don't know where is the theorical part of your thesis | 17:15 |
| viric | I never tried verilog. | 17:15 |
| Madridman | there is a part about architecture? | 17:15 |
| Madridman | (teoric= | 17:15 |
| Madridman | ) | 17:15 |
| lekernel | is it so hard to find that link? http://www.milkymist.org/thesis/thesis.pdf | 17:16 |
| viric | I'll steal it too | 17:17 |
| Madridman | lekernel that fpga can compete with a commercial architecture? | 17:18 |
| Madridman | I mean is slower? | 17:18 |
| lekernel | having seen a demo of ARM Mali on a FPGA, I think it does :) | 17:18 |
| lekernel | at least performance-wise | 17:18 |
| lekernel | but it lacks some features (that I didn't need for my application) to make a complete GPU that supports all OpenGL application | 17:19 |
| viric | lekernel: do you use any floating point there? | 17:19 |
| Madridman | a fpga can be as fast as a commercial video card as ati or nvidia? | 17:20 |
| lekernel | the texturing unit is fixed point, but the pfpu (roughly, a vertex shader) uses 32-bit ieee 754 | 17:20 |
| lekernel | floating point | 17:20 |
| lekernel | Madridman: generally no | 17:20 |
| lekernel | at least for GPU operations | 17:21 |
| Madridman | so fpga's are not as useful as they say | 17:21 |
| Madridman | they has limitations | 17:21 |
| Madridman | (it) | 17:21 |
| lekernel | everything has limitations | 17:22 |
| lekernel | the fpga is however a lot more flexible than a gpu | 17:22 |
| Madridman | ok, do you know a book or web or resource which speaks about fpga's so I could know better them? | 17:22 |
| viric | lekernel: aren't Floating Point operations taking lots of logic in the fpga? | 17:22 |
| Madridman | just general theory | 17:23 |
| lekernel | so if your particular application doesn't map nicely to, say, CUDA code, the FPGA can still deliver there | 17:23 |
| lekernel | viric: no, I'm not using the Opencores FPU ;) | 17:23 |
| viric | lekernel: where did you get the FPU operations from then? | 17:24 |
| lekernel | I wrote some code... | 17:25 |
| viric | ah | 17:25 |
| Madridman | lekernel do you know a book or web or resource which speaks about fpga's so I could learn and know better them? | 17:25 |
| lekernel | it's not 100% IEEE 754 compliant btw | 17:25 |
| viric | ahh. | 17:25 |
| lekernel | the number format and a lot of things are the same | 17:26 |
| viric | Madridman: get a manual from an FPGA seller | 17:26 |
| lekernel | but I don't totally respect details of the standard such as the way of rounding numbers | 17:26 |
| viric | Madridman: get the manual for the spartan3 fpga for example. | 17:26 |
| lekernel | or how to deal with numbers like NaN or infinites | 17:26 |
| Madridman | viric but I want to know the general common basics, not a concrete seller | 17:26 |
| lekernel | again - I don't need those for graphics processing | 17:26 |
| viric | clear. | 17:26 |
| Madridman | Eg. which kind of transistors or biestables are used generically, etc | 17:27 |
| viric | Madridman: that thing you call "generic fpga" does not exist. There are only concrete FPGAs :) | 17:27 |
| Madridman | ok, imagine I'm new (what is absolutelly true), I never heard about a thing called FPGA. You can't give a manual of a concrete implementation because I don't know even what technology we are speaking about | 17:28 |
| viric | Madridman: look for the "Spartan 3 User Guide" | 17:28 |
| Madridman | ok | 17:29 |
| Madridman | that looks as a board to do circuits | 17:30 |
| lekernel | Madridman: http://lekernel.net/presentations/FPGA_Workshops/ | 17:30 |
| lekernel | http://lekernel.net/presentations/FPGA_Workshops/091108_tmplab/scenes_draft.pdf will generally explain you how they work internally, but you may want to have some practical experience before | 17:31 |
| viric | Madridman: I call FPGA a chip that can be programmed *by software* to mimic a logic circuit. Software allows you to rearrange whatever is there inside to get as if you had made a circuit concerning whatever FPGA pins you want. | 17:32 |
| Madridman | ok, but later what you get is a physical circuit isn't it? | 17:33 |
| Madridman | i mean, logical gates interconnected | 17:33 |
| viric | yes. | 17:34 |
| Madridman | ok, that pdf's seems to have theorical information too, that's what I need for starters | 17:35 |
| viric | The circuits there come at some limitations, of maximum clock frequency, maximum number of clock signals, maximum number of flip flops... | 17:35 |
| viric | the more you pay, higher the limits | 17:35 |
| Madridman | I understand | 17:36 |
| Madridman | the scenes draft looks nice | 17:37 |
| viric | there are hardware description languages, and most FPGA sellers offer 'compilers' for those common languages to whatever their FPGA needs. | 17:37 |
| Madridman | yes, as vhdl which I hates | 17:37 |
| viric | So you can describe your circuit in a common way, and expect it to work among different FPGA vendors. | 17:37 |
| viric | Madridman: there is also verilog | 17:37 |
| Madridman | for sure is much better | 17:38 |
| viric | no idea. | 17:38 |
| viric | and how you have such an opinion about that? Some minutes ago you did not know anything about FPGA | 17:38 |
| Madridman | sure | 17:38 |
| Madridman | that's because I need theorical resources, to avoid bugging you | 17:39 |
| Madridman | is reallistic to pretend have a fpga's hardware which competes against commercial circuits (performance, fabrication costs)? | 17:39 |
| viric | no | 17:40 |
| Madridman | so it'sr just experimental things oriented? | 17:40 |
| lekernel | by "commercial circuit", you mean full customs ICs? | 17:40 |
| Madridman | well, a microprocessor yes | 17:40 |
| viric | FPGAs can compete only in the sense that they can be 'reprogrammed' | 17:40 |
| lekernel | in a typical design 90% of the FPGA transistors are wasted | 17:40 |
| viric | lekernel: hey, talk for your designs | 17:41 |
| viric | ;) | 17:41 |
| Madridman | my university research for multiexecution architectures specific for some tasks, so there are just experiments | 17:41 |
| lekernel | but if you compare that to the NRE of spinning an ASIC, sometimes it makes economic sense to keep an FPGA in a product where the FPGA design remains static | 17:41 |
| viric | FPGA over ASICs may have advantages on a) reprogrammability and b) cheaper for small productions | 17:42 |
| kristianpaul | http://en.qi-hardware.com/wiki/ASIC_production_notes | 17:53 |
| Madridman | thank you so much, you both provided me a lot of information | 17:59 |
| buggus | hi | 18:25 |
| kristianpaul | Hello | 18:27 |
| qi-bot | [commit] David Kühling: First shot at an Emacs package. Experimental! http://qi-hw.com/p/openwrt-packages/bea9c01 | 18:35 |
| kristianpaul | Emacs?? | 18:40 |
| kristianpaul | wow | 18:40 |
| kristianpaul | :) | 18:40 |
| viric | kristianpaul: using that will require more than thumbs | 18:48 |
| kristianpaul | wpwrak: do you have a signal generator at home? | 19:53 |
| wpwrak | yup | 20:02 |
| wpwrak | a Picotest G5100A | 20:02 |
| kristianpaul | I'm testing fpga bistream for the serial to parallel stuff on SIE | 20:03 |
| kristianpaul | as soon i get loaded the bitstream the LCM got odd interfereance and i lost ssh link | 20:03 |
| kristianpaul | i'm not sure if is mestastabillity or something ii'm forgoting to set in port A | 20:04 |
| wpwrak | hmm. so the FPGA is crashing the CPU ? | 20:04 |
| kristianpaul | seems | 20:04 |
| kristianpaul | i need dig more, i'll do try and error as i lack scopemeter and signal generators at home | 20:05 |
| wpwrak | maybe you're mis-initializing some of the pins ? | 20:05 |
| kristianpaul | i think | 20:05 |
| kristianpaul | checking that righ now | 20:06 |
| wpwrak | do you have any other program for the fpga that works ? | 20:06 |
| kristianpaul | sure | 20:06 |
| kristianpaul | and it worked as usual | 20:06 |
| wpwrak | odd then | 20:07 |
| wpwrak | well, you can try translating the one that works step by step to the one that doesn't :) | 20:08 |
| kristianpaul | hehe | 20:08 |
| kristianpaul | well i could i just wanted do do it before this ;= | 20:08 |
| kristianpaul | is just too elaborated for what i wanted to do | 20:08 |
| wpwrak | what is too elaborated ? the debugging process ? | 20:09 |
| kristianpaul | no | 20:09 |
| wpwrak | or the "good" one ? | 20:09 |
| kristianpaul | yup | 20:09 |
| kristianpaul | well is NOT so much, it just was intended to drive FPGA as a memory | 20:10 |
| kristianpaul | so you have a buffer to write and read, thats the general idea | 20:11 |
| wpwrak | ah. a bit complex indeed. probably more complex than the serial to parallel converter | 20:11 |
| kristianpaul | yup | 20:12 |
| --- Sun Dec 12 2010 | 00:00 | |
Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!