| stekern | this paper came across my path, thought it could perhaps be of interest to someone here: http://www.ece.ubc.ca/~lemieux/publications/brant-fccm2012.pdf | 05:31 |
|---|---|---|
| wpwrak | quite radical -) | 05:56 |
| larsc | I think I proposed something similar a few months ago. good to know that somebody is actually implementing it | 07:34 |
| wolfspraul | 40 times penalty? | 07:49 |
| wolfspraul | I think an fpga emulation in a high-end icore could compete with that, no? | 07:49 |
| wolfspraul | lm32 would then run at 2mhz, not 80mhz? :-) | 07:50 |
| wolfspraul | don't tell that to werner | 07:50 |
| wolfspraul | well ok, it wouldn't fit anyway | 07:50 |
| wpwrak | hey, still faster than an AVR with factory settings ;-) | 07:50 |
| wpwrak | i just skimmed the paper and didn't see what exactly the 40 times penalty refers to. oddly enough, they seemed to be happy about that factor. so i wonder what exactly it means. | 07:52 |
| wolfspraul | they are happy because it's better than the 100* in other attempts, they say | 07:52 |
| wolfspraul | everything is relative, no? | 07:52 |
| wolfspraul | I'm happy without any reason, now how is that??? | 07:53 |
| wolfspraul | :-) | 07:53 |
| wpwrak | ah, have to put wine on my shopping list, too :) | 07:55 |
| larsc | maybe I'm blind, but I'm unable to find their implementation. All references to "ZUMA FPGA" just point to the paper. | 09:37 |
| larsc | wpwrak: btw. it is not necessarily slower, just smaller | 09:40 |
| larsc | s/wpwrak/wolfspraul/ | 09:40 |
| wolfspraul | yeah who knows :-) I found too many dead ends to still care | 09:46 |
| wolfspraul | a lot of papers and text that is written as if they know *some* stuff and then decided "we have to report our results" | 09:47 |
| wolfspraul | recobus is the most advanced I think, but sadly they also just publish 5-10 page papers for conferences that are dead-ends for learning, in my experience | 09:47 |
| mumptai_ | fpga on fpga? | 09:47 |
| wolfspraul | why not, you never know what becomes possible later, but the #1 thing imho is what lars said - where are their sources and does it actually run? | 09:50 |
| wolfspraul | too much talk... back to coding :-) | 09:50 |
| larsc | mumptai_: yes | 09:51 |
| mumptai_ | if i was developing a new fpga+toolchain, i would probably like to do that, but otherwise? | 09:52 |
| larsc | well with the big vendor fpga you have the problem that you don't really know what's inside the fpga | 09:56 |
| mumptai_ | larsc, and they dont' tell that without an nda, or not at all | 11:09 |
| larsc | exactly | 11:09 |
| larsc | so an open fpga is a way better target for developing tools | 11:10 |
| mumptai_ | but wrapping it again is only a first step | 11:10 |
| lekernel | aha! I got the bios to compile with gcc 4.8/svn and Nick's patches | 11:10 |
| mumptai_ | i would actually like to have a model of the fpga for other reasons, but the vendors seem to dislike that | 11:11 |
| lekernel | ld later prints mysterious/fucked error messages "main.c:(.text+0xc): warning: internal error: dangerous error" though | 11:11 |
| lekernel | but still generates a binary... | 11:12 |
| lekernel | I see gcc copied clang's error/warning reporting :) hehehe | 11:14 |
| larsc | last time I tried to build openwrt with gcc 4.7 it just hogged all memory and cpu, hopefully it works again with those patches | 11:16 |
| larsc | openwrt for lm32 that is | 11:16 |
| lekernel | there's one patch that fixes a bug that resulted in a infinite recursion loop | 11:17 |
| lekernel | is that what you are talking about? | 11:17 |
| larsc | most likely | 11:18 |
| larsc | as i said gcc just started to eat all of my cpu and memory | 11:19 |
| lekernel | now there's still this weird linker problem... | 11:20 |
| wpwrak | (virtual FPGA a "better target") of course, with a factor of 40, that gives you a lot of headroom for making a rather conservative model of the real thing. kinda what wolfgang is doing. | 13:15 |
| wpwrak | a car analogy: since we can't be sure all curves are safe to drive in at 240 km/h, we a) post speed limits of 30 km/h (that could be wolfgang's result) or b) don't allow cars that can go faster than 6 km/h (the 40x margin) | 13:17 |
| wpwrak | while a) would suck, it would still be 5x better than b) ... | 13:17 |
| --- Wed Aug 29 2012 | 00:00 | |
Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!