<larsc> mwalle: do you have an crti.o in your lm32 toolchain?
<larsc> hm... now i have a toolchain and a rootfs, but nothing happens after "VFS: Mounted root (ext2 filesystem) readonly on device 1:0."
<larsc> mwalle: what is your make version?
<larsc> mwalle: i removed lm32-softemu/config-* and now it works....
<larsc> qemu still crashes though
<larsc> mwalle: as soon as irqs are enabled
<larsc> hm... ok that was due to a half applied patch
<larsc> but it still crashes when entering usermode
<larsc> IN: PC=0
<larsc> ie=00000003 (IE=1 EIE=1 BIE=0) im=0000000b ip=00000010
<larsc> it looks like an interrupt within a systemcall
<larsc> hm the syscall handler should increase the return address before calling eret, yet this seems not to happen in qemu
<larsc> hm... eret is called from _restore_irq_frame_and_return and not from _restore_and_return_exception which explains why ea is not incremented
<larsc> hm, or maybe not
<vr1> salut
<lekernel> salut
<vr1> est-ce que milkymist est une bonne plateforme pour traiter du son ?
<lekernel> ça peut se faire oui... quel genre de traitement?
<vr1> de la reco
<vr1> (analyse statistique)
<vr1> g un proto qui fct bien en C/Linux sur un x86_64
<vr1> j'aimerais l'embarquer, je me demande si le milkymist tient la charge
<lekernel> avec un accélérateur fpga, il n'y a sûrement pas de problème. mais il faudra le développer.
<vr1> c-a-d le code en C pur tiendra pas la charge ?
<lekernel> aucune idée
<vr1> le CPU est equivalent à un ARM7 ou ARM9 ?
<lekernel> plutôt arm7
<vr1> ok je vois
<vr1> la possibilité de faire un bout de code FPGA pour accélerer est intéressante
<vr1> si on developpe un soft sur la plateforme, quelle en est la license ?
<vr1> je veux dire: est-on obligé d'opensourcer ?
<lekernel> le design FPGA est largement sous GPL, pour le soft C il n'y a pas de restriction de licence
<vr1> ok
<lekernel> donc si tu modifies le design FPGA et que tu le redistribues, en général oui
<vr1> ok
<lekernel> ceci dit, une licence ARM pour faire la même chose coûterait dans les 100k¬, donc choisis :)
<vr1> est-il possible de dériver le HW pour en lever les composants non vidéo ? (dans le cas d'une production)
<vr1> pardon
<vr1> je me suis mal exprimé
<vr1> est-il possible d'enlever certains composants du design HW ?
<vr1> par exemple enlever les composants vidéo
<vr1> pour faire une BOM moins cher dans le cas d'une prod
<vr1> et est-ce autorisé ?
<vr1> par ex plateforme de son uniquement
<lekernel> oui, c'est parfaitement autorisé
<vr1> quelle est la BOM que l'on peut espérer pour une telle plateforme ? juste une idée de prix approximative
<lekernel> y compris la commercialisation de ton design, tant que les licences GPL et CC-BY-SA sont respectées
<vr1> ok
<lekernel> le PCB/schéma est en CC-BY-SA ...ceci dit, le design FPGA (en GPL) est assez portable également
<lekernel> il tourne aussi sur plusieurs cartes xilinx/altera
<vr1> salut les gars
<wpwrak> lekernel: i hope i haven't just added a 2 month delay to MM1 progress ;-)
<lekernel> wpwrak: ?
<wpwrak> lekernel: see the qi-hw mailing list ;-)
<wpwrak> lekernel: well, or maybe don't :)
<lekernel> the robot?
<wpwrak> yeah
<wpwrak> looks cool, doesn't it ? and it also looks kinda doable ...
<lekernel> heard of boston dynamics?
<lekernel> (big dog etc.)
<wpwrak> yup
<wpwrak> kinda scary that device. looks a bit too "alive"
<wpwrak> what i like about the rotopod is its symmetry - all the six axes are identical. so you don't need to design each axis separately, like in traditional CNC machines.
<lekernel> !karma test
<CIA-8> milkymist: Sebastien Bourdeauducq newmac * r99be230 / (10 files in 3 dirs): minimac2: TX working in simulation - http://bit.ly/fcr6dh
<larsc> mwalle: something is wrong with the scall code generation. http://pastebin.com/3KVHpa7t there is only an scall instruction at 40017c58
<larsc> finally:
<larsc> VFS: Mounted root (ext2 filesystem) readonly on device 1:0.
<larsc> HELLO WORLD
<larsc> Kernel panic - not syncing: Attempted to kill init!
<mwalle> larsc:
<mwalle> is scall broken?
<larsc> mwalle: i reverted d53066a5578945748dd7d8099c96de13f9a69211 and now it works again
<mwalle> larsc: could you please paste the code arount 40017c58?
<larsc> well, there is one scall plus a few mv
<mwalle> do you know what happens?
<larsc> not really
<larsc> for some reasons scall exceptions are also created for the lines following the real scall
<mwalle> mh so why doesn't this happen with test_scall.S?!
<larsc> test_scall.S?
<mwalle> qemu/tests/lm32/
<mwalle> make check
<larsc> hm
<mwalle> larsc: could you upload your kernel and initrd?
<mwalle> mh i think i dont need it
<mwalle> happens for me too
<mwalle> mh ok, works again ;)
<mwalle> larsc: i just uploaded the fix
<larsc> nice
<larsc> :)
<mwalle> narf, commit contains opengl 'fix'..
<mwalle> larsc: i guess you already pulled it :)
<larsc> fell free to rebase
<mwalle> done
<TS-Labs> hello!
<TS-Labs> i have a question about Navre AVR clone
<mwalle> hi
<TS-Labs> as it is said in description, the classic avr instruction set is supported. but when reading the sources I see that the MUL's aren't implemented
<mwalle> i guess it isnt needed by us. so it may not be implemented
<mwalle> esp. since multipliers needs lots of ressourecs
<TS-Labs> yeah, ok )
<mwalle> or its done in software
<TS-Labs> in fact, 8bits by 8 multiplier ate ~110 LE's on Acex I Altera
<mwalle> pipelined?
<TS-Labs> no, just * in verilog )
<TS-Labs> combinatorial logics
<TS-Labs> 4 or 5 summators of different sizes
<mwalle> so its likely that this will be the critical path :)
<TS-Labs> yeah )
<TS-Labs> i used 40 mhz, worked ok, but i didn't measure its latency
<lekernel> multiply isn't part of the 'classic' instruction set
<lekernel> (and btw modern synthesizers can pipeline '*' in verilog)
<larsc> mwalle: I finally managed to get OpenWrt to build bootable images :)
<wolfspraul> larsc: !
<wolfspraul> :-)