<hallo99>
do I need a mips or a mipsel cross compiler for the nanonote?
<hallo99>
I have read somewhere, that mips hardware can switch beetwen little and big endian, so does mips and mipsel work for the nanonote?
<mth>
the kernel and all the libraries are built for mipsel
<mth>
I'm not even sure the CPU can do big endian
<mth>
some MIPS CPUs can switch, but maybe not this one
<wpwrak>
hmm, so is there no known way to remove the hinge / separate bottom shell and screen, without breaking anything ?
<hallo99>
does gcc generate optimized code for the jz4720, or do I have to use --march=mips1?
<wejp>
hallo99, you can use mip32 instead of mips1
<wejp>
mips32 even
<Ornotermes>
wejp: have you updated gmu for nanonote to use flac?
<wejp>
there is no need to update it. Gmu supports flac out of the box, it just needs the right libFLAC. if it is there Gmu can be compiled with flac support and it will work out of the box
<wejp>
as openwrt now seems to come with a more recent libFLAC i guess in the next firmware image Gmu will be included with FLAC enabled
<Ornotermes>
wejp: didn't you see openwrt has updated flac?
<Ornotermes>
oh you did :)
<wejp>
i just said so
<wejp>
;)
<wejp>
maybe i'll create another binary, but it wouldn't be of any use as soon as there is a new firmware image anyway. not sure if it is worth it
<Ornotermes>
i built it with flac my self but it would be simpler for other if the could just compile it
<Ornotermes>
can't you make have a binary for liblac too?
<wejp>
i could but it should be possible to install libflach through opkg on the nanonote
<wejp>
*libFLAC
<Ornotermes>
sure, but having it by default in next relase would be good it it can't be dine before
<Ornotermes>
if it*
<Ornotermes>
done*
<wejp>
sure, if the next gmu release is ready before the next official firmware image, i'll add flac support there of course
<xdpirate>
Yo, I'm having some trouble getting internet access on jlime on the nanonote, anyone available to help?
<lekernel>
or I can use the ml401 to sample the usb signals too...
<lekernel>
or even maybe an avr with v-usb
<xdpirate>
ftr my previous issue was solved, i forgot to set the dns server
<wpwrak>
lekernel: (use the ml401 itself) that's a nice idea :)
<lekernel>
the problem with all those approaches is they take far more time than just connecting a proven sniffer
<wpwrak>
lekernel: for the kind of work you're doing, you really ought to have a decent scope, though. e.g., the rigol ones have deep memory at quite a reasonable price. that is, if you can live with 100 MHz bandwidth and only 2 channels.
<lekernel>
decent scope don't have a decent price, so if I can live without them... :)
<wpwrak>
lekernel: yup, a good sniffer is better. but if debugging the really low-level stuff, you often end up staring at the scope anyway, because you actually want to see what your signals look like.
<lekernel>
hm, I can probably use a verilog simulator for that
<wpwrak>
lekernel: yeah, in the ~1kUSD/EUR segment, rigol is the best you can get. the significantly nicer stuff usually adds a digit to the price tag :-( but yes, i'm lusting after a 500 MHz / 4 chan / 10 MS scope, too :)
<lekernel>
signals aren't that fast (12Mbps) that weird things happen to them
<wpwrak>
lekernel: you can always screw up the analog side :) besides, you need something (host) to talk to your system, don't you ?
<lekernel>
I'm developing the host
<lekernel>
the analog side is a micrel transceiver with few external components
<lekernel>
hard to screw up
<wpwrak>
lekernel: okay, then the device :) or do you already have one you could run in the simulator ?
<lekernel>
plus it probably works already, it correctly detects the pull up resistors of devices
<lekernel>
no, I'd need to develop a model for it too
<lekernel>
that's why a proven sniffer would make me gain a lot of time
<wpwrak>
that's a lot of moving parts then
<wpwrak>
well, if you can't find a sniffer, you may still want to try the scope approach. it worked well for me for writing my little usb stack on the c8051f326.
<ezdagor>
Anyone know how to get the ks7010 driver working without it writing, what I suspect to be, debug messages to the current console?
<unclouded>
ezdagor: does the source for ks7010 suggest conditions under which it might suppress those messages?
<ezdagor>
Checking..
<ezdagor>
Not that I can tell.
<ezdagor>
However there is a debug unit. I wonder if I left it out of the Makefile, if that would make a difference. There aren't too many references to it in the rest of the source code.
<ezdagor>
I'm not too knowledgable in C, though.
<ezdagor>
#ifdef KS_WLAN_DEBUGÂ Â Â Â Â Â Â Â int i; #define HEX_OFFSET "\Â Â Â Â Â Â Â Â Â Â Â Â +0 +1 +2 +3 +4 +5 +6 +7 +8 +9 +A +B +C +D +E +F"Â Â Â Â Â Â Â Â printk(HEX_OFFSET);Â Â Â Â Â Â Â Â for (i=0; i<length; i++) {Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â if (i % 16 == 0) printk("\n%04X-%04X:", i, i+15);Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â printk(" %02X", *(p+i));Â Â Â Â Â Â Â Â }Â Â Â Â Â Â Â Â printk("\n"); #endif
<kristianpaul>
wolfspraul: how i can get ingenic chips?
<kristianpaul>
like in SAKC
<kristianpaul>
gn8 !
<unclouded>
azdagor: and you're seeing console output like "+0 +1 +2 +3 +4 +5 +6 +7 +8 +9 +A..."?
<ezdagor>
It shows a home directory of whoever compiled it, and then a memory address.
<ezdagor>
Over and over again.
<ezdagor>
I removed the debug unit from the Makefile... it didn't compile.