#milkymist IRC log for Saturday, 2012-05-19

wpwraklekernel: ah, another possibility would be if the mouse has more than 8 buttons. or if the firmware thinks it might.00:05
wpwrak(again something proper report processing will fix)00:05
wpwraklekernel: btw, do you remember any reason for the asymmetric bypassing of U14 and U15 ? all this is from before my time with M1 ...00:06
GitHub163[board-m1] adamwang pushed 1 new commit to master: https://github.com/milkymist/board-m1/commit/d374ac92cf09760d4dc5bce7115e521289f5546702:39
GitHub163[board-m1/master] Dram.sch: indicating bypass capacitors to dram pins - Adam Wang02:39
GitHub46[board-m1] adamwang pushed 1 new commit to master: https://github.com/milkymist/board-m1/commit/133288af51c5624b04a0e8e26fb591bc4ed3c65502:49
GitHub46[board-m1/master] Dram.sch: moved C156 to right of C252. - Adam Wang02:49
lekernelI don't know, the layout people placed them this way, and it works so it's ok12:18
wpwrakokay, so it's what i thought. it's funny how such subtle changes happen :)13:03
mischief6any tips on building a lm32 cross compiler14:55
mischief6found this makefile at the bottom of this page, but all the links to tarballs seem outdated. http://www.das-labor.org/wiki/Soc-lm32/lm32-toolchain_install_the_crosscompiler14:56
mischief6can't find gcc-core-4.5-20091112 anywhere14:56
sh4rm4mischief6, http://lekernel.net/blog/2011/10/soc-milkymist-le-developpement-logiciel-en-pratique/15:01
mischief6thx sh4rm4. will see what i can do15:12
sh4rm4otherwise just dl gcc-core 4.5.0 or 4.5.1 instead15:13
sh4rm4it shouldnt be much different15:13
mischief6i think this article says that gcc 4.5 will not build linux correctly, or that there is bugs with the lm32 arch for this version?15:14
sh4rm4the french article ?15:14
Fallenoutry 4.5.2 it should work well (at least C compiler)15:14
mischief6yes sh4rm4.15:14
FallenouC++ compiler is broken AFAIK15:14
Fallenoulm32 toolchain builds well under Mac OS X and Debian15:15
mischief6(translated) 'Although LatticeMico32 architecture is now included as standard in GCC 4.5, this inclusion has been made the addition of bugs that make it impossible to compile a Linux kernel. '15:15
Fallenoumischief6: what do you want to compile ?15:16
Fallenourtems ? flickernoise ? linux ?15:16
mischief6hopefully :-)15:16
mischief6i am trying to understand what is supported, how to run the platform with virtualization, etc15:17
mischief6linux might be first target15:17
Fallenouright now I don't know the status of the lm32 linux port15:17
mischief6perhaps i could try to port oskit if i have time15:18
FallenouI think for a first try to run something on a Milkymist One board (that's what you have, right ? or qemu ?) you should try the bios / rtems / flickernoise15:18
Fallenouthose are more documented15:18
Fallenouand known to be working15:18
mischief6getting anything running would be cool :P15:18
Fallenouon a M1 ?15:19
mischief6yes, sure15:19
Fallenoutry Milkymist bios :)15:19
mischief6i don't have one, but a few people at noisebridge hackerspace are interested in development and possibly working on compiler support15:20
Fallenouoh !15:20
mischief6what is the state of llvm-lm32?15:20
Fallenouworking on compiler support would be awesome :)15:20
Fallenouyou would have to ask JP Bonn about llvm-lm3215:21
FallenouI think we compile a bit of software with it15:21
Fallenounot sure which part15:22
mischief6ok i will check it out15:23
mischief6time for me to sleep though :-)15:23
Fallenougn8 !15:23
mischief6night15:24
Fallenoudoing a quick grep shows15:24
Fallenou./software/include.mak:#CC_normal := clang -march=mico32 -ccc-host-triple mico32-elf15:24
Fallenouit seems to be commented15:24
FallenouI guess by default we only use gcc, maybe because lm32 clang is a work in progress15:24
Fallenoubut I am quite sure we had some results/success with clang lm32 :)15:25
Action: Fallenou does not remember15:25
lekernel<larsc>for example i really don't want to be bother with inserting buffers at the right positions.20:48
lekernel<larsc>i want to write down the supposed functionallity and give the block a overall timing constraint20:48
lekernel==> migen flow with CAL (or similar) will do more or less that...20:48
lekernelany recommendation for a good embedded OS? ideally, RTEMS but without dirty code, without autocrap, without disturbing bugs (like the infamous queue problem), and with a reasonably up to date TCP/IP stack22:29
GitHub23[milkymist-ng] sbourdeauducq pushed 1 new commit to master: http://git.io/UUcHeg22:31
GitHub23[milkymist-ng/master] Add Ethernet MAC - Sebastien Bourdeauducq22:31
lekernelhttps://rwmj.wordpress.com/2012/04/13/genuine-fragment-from-rhel-spec-file/22:37
lekernelhttp://www.xilinx.com/products/design-tools/vivado/implementation/place-and-route/index.htm23:39
lekernelfirst time they talk about this - https://www.google.com/webhp?q=%22Simulated%20Annealing%22%20site%3Axilinx.com#hl=en&safe=off&output=search&sclient=psy-ab&q=%22Simulated+Annealing%22+site%3Axilinx.com&oq=&aq=&aqi=&aql=&gs_l=&pbx=1&bav=on.2,or.r_gc.r_pw.r_cp.r_qf.,cf.osb23:41
--- Sun May 20 201200:00

Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!