00:03
<
roh >
720p itself means only 'resolution'
00:05
<
roh >
there is some us stuff which uses YUV on analog rca jackets with something similar to 'sync on green'
00:05
<
roh >
so its only 3 rca jacks. analog.
00:05
<
roh >
used mainly in the us and asia. in europe we just used rgb and scart with csync on cvbs pin
00:05
<
roh >
which had the feature that the tv doesnt need to do any conversion. basically the signal directly goes to the amps on the tube-pcb, sync to the deflection circuit
00:06
<
lekernel >
looks great
00:06
<
roh >
yet the limited bandwith on an rca jack never gave proper picture. and in tft/dlp times its just a intermediate format which gets digitized, buffered and scaled anyhow.... on dvi/dlp/hdmi you usually even let the sender do the scaling and send the optimal format to the screen
00:07
<
roh >
the scalers on gpus are usually also much better than the ones in the display controllers (there are still some in there, but they need to be cheap)
00:07
<
lekernel >
i know. we'll also add your beloved hdmi at some point, don't worry :)
00:08
<
roh >
lekernel: nice cable.. now you just need to find out what pins it wires how. this only works with graphics cards which do more than 'normal vga'
00:08
<
lekernel >
but this has a bit less priority than cvbs because 1. I have rarely seen HDMI projectors 2. it needs a little PCB mod to route a HDMI connector to the FPGA
00:10
<
lekernel >
common projector formats are 1. VGA 2. CVBS/S-Video 3. HDMI/DVI, in decreasing order
00:16
<
lekernel >
pins 1/2/3 are RGB, so we could definitely use such a cable to connect to an unmodified M1
00:17
<
lekernel >
assuming we sort out the possible DAC resolution problem
00:17
<
roh >
well.. if the dacs are dc coupled it could work already
00:18
<
lekernel >
you need DC coupling for VGA to work already, no?
00:21
<
lekernel >
also, can we use the cable as a 'natural' oversampling low pass filter? :-)
00:24
<
roh >
lekernel: dunno.
00:24
<
roh >
i am just sure that we need dc coupling to simulate the sync in the dac signal
00:24
<
lekernel >
that should be OK
00:24
<
roh >
then i would buy such a cable and just try how good it works/the picture looks
00:24
<
roh >
my guess is one looses a bit of dynamic with a 8bit dac due to the amount in voltage the sync already 'eats'
00:25
<
lekernel >
you loses 26% of the dynamic range, but 140MHz oversampling would buy you 3 times the resolution
00:26
<
roh >
hm. well.. i guess that calls for rediscussion after trying out
00:44
<
roh >
shouldnt have done that
00:50
<
kristianpaul >
and yes i had troubles with a vga spliter !!!
00:50
<
kristianpaul >
i'll post details soon..
00:51
<
kristianpaul >
troubles = video signal distorcionated/malformed
00:51
<
kristianpaul >
i'll post image soon..
00:51
<
kristianpaul >
still at dorkbot
00:52
<
roh >
is counting hundreds of acryllic spacers
00:58
<
aw_ >
kristianpaul, what's the possible trouble?
01:30
<
kristianpaul >
aw_: the video technician told me about power and frecuency
01:31
<
kristianpaul >
i also cant run mm1 video on a big plasma screen
01:31
<
aw_ >
kristianpaul, related to design of video? or video format.
01:31
<
kristianpaul >
aw_: vga video, dunno
01:31
<
kristianpaul >
just was a tought from the guy
01:31
<
aw_ >
kristianpaul, hmm...got it.
02:19
<
larsc >
lekernel: if i use the 800x600 mode from the rtms fb driver, the image is displaced by 100 pixels or so to the left
03:50
<
kristianpaul >
aw_: sorry
03:52
<
kristianpaul >
i'll drop a mail saturday i hope, when more time i have.. 3 days workshop is energy consuming..
03:52
<
kristianpaul >
btw like 4 guys asked me about milkymist price and avaliabillity
03:53
<
aw_ >
kristianpaul, no sorry. :-)
03:54
<
kristianpaul >
ah i dint tried last msd..
03:54
<
aw_ >
kristianpaul, AFAIK, current only have one M1 available. others are the RC3 ones later. :-)
03:54
<
aw_ >
I'll try latest msd today. :-)
04:28
<
kristianpaul >
okay bed time
04:29
<
aw_ >
kristianpaul, oah...using splitter..good. cu nice night!
04:29
<
xiangfu >
kristianpaul: n8
06:31
<
lekernel >
larsc: press "auto setting" on your screen?
10:11
<
lekernel >
kristianpaul: how do power and frequency of the video signal (?!) relate to driving big plasma screens?
12:26
<
lekernel >
milkymist at tacheles techno party next wednesday :)
13:00
<
kristianpaul >
lekernel: i'm not saying it was a tought/comment from the guy was setting up video
13:01
<
kristianpaul >
wolfspraul: it worked with out the splitter
13:04
<
wolfspraul >
kristianpaul: yes, but do you think 'support' for this splitter is a missing feature/bug?
13:05
<
wolfspraul >
it's something worth remembering maybe, we need an issue tracker :-)
13:05
<
lekernel >
imo it's just a problem coming from the splitter
13:05
<
wolfspraul >
kristianpaul: if it worked, do you have a nice picture that shows it in action? or even video?
13:05
<
lekernel >
the m1 puts out perfectly standard VGA
13:05
<
wolfspraul >
lekernel: fine, but even then we should document it. let's keep an eye on splitters :-)
13:06
<
wolfspraul >
the end users only cares whether it works or not, not who is to blame, or who could or could not fix it.
13:06
<
wolfspraul >
so the best is if we document well
13:06
<
wolfspraul >
then no surprises...
13:06
<
wolfspraul >
to me it's a good finding, I registered 'splitters' as something to watch out for
13:07
<
kristianpaul >
also this was a quick talk i just had 10 minutes to setup and talk..
13:07
<
wolfspraul >
ah ok. that pic looked similar to the bug :-)
13:07
<
kristianpaul >
no much time for take pictures..
13:07
<
kristianpaul >
really?
13:07
<
wolfspraul >
any feedback on milkymist?
13:07
<
wolfspraul >
what was your talk about?
13:07
<
kristianpaul >
yeah
13:07
<
kristianpaul >
milkymist in 7 minutes
13:08
<
kristianpaul >
people dont want to quit the milkydrop just to edit the patch
13:08
<
kristianpaul >
they found it like a lack of real time
13:08
<
kristianpaul >
i mean
13:08
<
kristianpaul >
interrupt perfomance, edit patch and run again was the critic
13:09
<
wolfspraul >
which is a fair point, and I think sebastien thought about a second vga interface many times already
13:09
<
wolfspraul >
we will not introduce it in rc3 though
13:09
<
kristianpaul >
hey, why the last pic looks bad? it is running a patch as usual
13:09
<
kristianpaul >
yes sure, just saying what other commeted
13:09
<
wolfspraul >
maybe I just didn't get it
13:09
<
wolfspraul >
no worries I am not complaining, I was just confused with the splitter bug and that pic
13:10
<
lekernel >
kristianpaul: yes, otoh you can prepare many patches in advance and switch them in no time, e.g. with a midi controller
13:10
<
kristianpaul >
yes i told that lekernel
13:10
<
wolfspraul >
the pic is great. if that is in fact the intended visual effect, then even better!
13:10
<
wolfspraul >
I just didn't know, that's why I ask.
13:10
<
kristianpaul >
but pople want
*real* *time* editing of patches
13:10
<
wolfspraul >
now I know :-)
13:10
<
kristianpaul >
well i got that impresion
13:10
<
lekernel >
kristianpaul: some do, some others don't...
13:11
<
kristianpaul >
also asked for OSC support
13:11
<
kristianpaul >
lekernel: he, sure ;)
13:11
<
lekernel >
OSC is already built in
13:11
<
kristianpaul >
how they use it?
13:11
<
kristianpaul >
please dont RTFM now :-)
13:11
<
lekernel >
well, there's no documentation for it yet :)
13:11
<
lekernel >
so it would be "RTFS" instead :)
13:12
<
lekernel >
but in more friendly terms
13:12
<
lekernel >
you have parameters in the patches iosc1, iosc2, etc.
13:12
<
lekernel >
which are floating point values you can set through the OSC interface over UDP
13:13
<
lekernel >
you also have a MIDI sink (i.e. MIDI transport over OSC) which gets merged with the rest of the MIDI messages coming from the DIN
13:13
<
wolfspraul >
kristianpaul: for that pic 3638, can you give me the exact event name and location?
13:13
<
lekernel >
the possibility to switch patches with an OSC command
13:13
<
wolfspraul >
in case I want to mention it in the news (the pic is nice!)
13:13
<
kristianpaul >
event labsurlab at mamm
13:13
<
wolfspraul >
name of event, location, date
13:14
<
kristianpaul >
was just before dorkbot talk started
13:14
<
lekernel >
and finally, you can print text messages on the screen using another OSC string message
13:14
<
kristianpaul >
mamm = museo de arte moderno de medellin
13:14
<
wolfspraul >
date was April ? 6?
13:14
<
lekernel >
you can try easily with the 'oscsend' utility from liblo
13:14
<
kristianpaul >
yup wolfspraul
13:14
<
wolfspraul >
is there a url for the event?
13:14
<
lekernel >
kristianpaul: do you have pics with the video-in patches?
13:15
<
kristianpaul >
i think mm1 should support the splitter somw way, well or do more test, but all laptops around dint had problem with splitter
13:15
<
kristianpaul >
i think there was just one including mm1
13:15
<
kristianpaul >
lekernel: i know
13:15
<
lekernel >
kristianpaul: what was the problem exactly?
13:16
<
kristianpaul >
vide-in not tested yday
13:16
<
kristianpaul >
image was partially viewved
13:16
<
kristianpaul >
i mean it was kinda zoomed
13:16
<
lekernel >
kristianpaul: please demo the video-in patches. they're really at the next level :)
13:17
<
kristianpaul >
i will, i hope flash mm1 with last firmware today..
13:17
<
lekernel >
kristianpaul: weird
13:17
<
kristianpaul >
lekernel: indeed
13:17
<
wolfspraul >
how long did m1 run at the event in total?
13:17
<
wolfspraul >
just during your talk, or for some longer period?
13:17
<
kristianpaul >
no more than 1 hr
13:17
<
lekernel >
from my knowledge splitters are simple amplifiers... can't really see how that could do that
13:17
<
kristianpaul >
before talk
13:17
<
kristianpaul >
and during
13:17
<
wolfspraul >
ok then I say 'demoed'
13:17
<
lekernel >
could be some murphy-esque interaction between splitter, projector and signal integrity problems
13:18
<
lekernel >
kristianpaul: what flickernoise/soc versions are you using atm?
13:18
<
lekernel >
0.2/RC2 (dec2010 msd) already support the video-in
13:19
<
kristianpaul >
sure it tested video in just dont have the best setup up for a demo
13:20
<
lekernel >
ah, you're having another demo later on?
13:20
<
kristianpaul >
i will do more tests today i hope, and a meeting with a local VJ group saturday
13:20
<
kristianpaul >
i have a big screen for me all day
13:20
<
lekernel >
ah, cool!
13:20
<
kristianpaul >
but the mm1 just never got to work, the screen said no signal detected
13:20
<
lekernel >
you could even invite people to stand in front of the cam :)
13:21
<
lekernel >
what screen is that?
13:21
<
kristianpaul >
but i plug a small screen at works..
13:21
<
kristianpaul >
i'll take specs today
13:21
<
kristianpaul >
noo time yday
13:21
<
lekernel >
first time I hear about screen signal detection problems...
13:22
<
lekernel >
i've plugged mm1 to all sorts of screens and projectors already, without a single issue
13:22
<
lekernel >
is the cable ok?
13:24
<
kristianpaul >
ok sure !, i tested with other laptops, and NO problem
13:24
<
kristianpaul >
i need leve, i'll bnack ramdomly in next horus
13:24
<
kristianpaul >
hours*
13:24
<
kristianpaul >
bye!
17:25
<
CIA-8 >
milkymist: Sebastien Bourdeauducq master * rd8ddaca / (4 files in 4 dirs): Ethernet: workaround for permanent RX FIFO overflow bug -
http://bit.ly/hE7Hdj
17:39
<
mwalle >
larsc: do you still facing problems on real hw?
17:40
<
CIA-8 >
rtems-milkymist: Sebastien Bourdeauducq master * r5456116 / (3 files in 2 dirs): Ethernet: use library CRC -
http://bit.ly/f2BGDr
17:51
<
larsc >
mwalle: nope
17:52
<
larsc >
now i'm facing problmes in qemu... qemu crashes as soon as the kernel has finished loading the initrd image
17:52
<
larsc >
PC is 0 for some reason
17:54
<
mwalle >
on real hw too?
17:54
<
larsc >
mwalle: not that i can see
17:56
<
larsc >
but i'm maybe using an outdated version of qemu
17:57
<
mwalle >
larsc: do you have applied the gcc-fix-lm32-libgcc.patch?
17:57
<
larsc >
i've applied all your patches
17:58
<
mwalle >
is
__PIC__ undefined?
17:58
<
mwalle >
have a look at libgcc/config/lm32/_ashrsi3.S
18:01
<
larsc >
__PIC__ is probably defined then
18:02
<
larsc >
btw. the latest qemu build fails with make[1]:
*** No rule to make target `../ptimer.o', needed by `qemu-system-lm32'.  Stop.
18:03
<
mwalle >
official one?
18:03
<
larsc >
the milkymist branch
18:04
<
mwalle >
i have one fix regarding system calls
18:04
<
mwalle >
that is not in my repo yet
18:06
<
mwalle >
you could try that too
18:08
<
larsc >
hunk#3 doesn't apply, against which branch is the patch?
18:09
<
mwalle >
mh my working one here
18:09
<
mwalle >
i try to merge upstream and milkymist into master
18:11
<
mwalle >
btw, does anybody know how to avoid duplicate commits you get if your patch is merged upstream (and the commitmsg was changed?)
18:11
<
lekernel >
mwalle: got an idea what's happening when the board suddenly freezes and gdb shows it's executing the debug ROM?
18:12
<
lekernel >
some exception? how to know more?
18:12
<
mwalle >
how shows gdb its executing the debug rom?
18:13
<
lekernel >
as soon as it connects, it shows PC is in the debug rom
18:13
<
lekernel >
well, actually past it
18:13
<
lekernel >
(gdb) set remote interrupt-on-connect on
18:13
<
lekernel >
(gdb) target remote /dev/pts/60
18:13
<
lekernel >
Remote debugging using /dev/pts/60
18:13
<
lekernel >
0x1f18e2a4 in ?? ()
18:13
<
lekernel >
looks pretty bad...
18:13
<
mwalle >
mhhh doenst look like a valid PC
18:14
<
lekernel >
ra is messed up too.. 0x1f18e82c
18:16
<
mwalle >
did you debug before?
18:16
<
lekernel >
no, just booted and connected GDB when it froze
18:17
<
mwalle >
is this a new freeze (introduced by the gdb stub?)
18:17
<
lekernel >
in fact, I did that a couple of times already after 'random' freezes, and everytime I got that weird PC
18:17
<
lekernel >
it's an old bug I'm just tracking down now using the new gdb tool :)
18:17
<
mwalle >
x/x 0x1f18e2a4 ?
18:18
<
mwalle >
some valid opcode?
18:18
<
mwalle >
mh there was some fmt modifier to decode that instruction, wasnt it?
18:19
<
mwalle >
x/i 0x1f18e2a4
18:19
<
lekernel >
disassemble
18:20
<
mwalle >
looks like a loop ;)
18:20
<
mwalle >
maybe we should enable BUS_ERRORS
18:21
<
lekernel >
bus error?
18:21
<
mwalle >
but then you would have to signal invalid addresses on the WB
18:21
<
lekernel >
you mean generate a bus error when the cpu tries to access an invalid location?
18:21
<
mwalle >
theres some CFG_BUS_ERRORS or sth like that in the lm32 core
18:22
<
mwalle >
which raises an exception
18:22
<
lekernel >
so it would stop the execution as soon as it's crashing... yeah
18:22
<
lekernel >
we should have an "invalid instruction" exception too
18:22
<
larsc >
hm, interesting: gcc/Makefile:TARGET_LIBGCC2_CFLAGS =
18:22
<
larsc >
gcc/Makefile:echo TARGET_LIBGCC2_CFLAGS = '$(TARGET_LIBGCC2_CFLAGS)' >> tmp-libgcc.mvars
18:22
<
larsc >
yet libgcc.mvars contains TARGET_LIBGCC2_CFLAGS = -fPIC
18:23
<
mwalle >
yep, well you would have to have some special exception handling for data and instruction bus errors, too (eg branch to debug rom
18:25
<
mwalle >
larsc: not for me
18:25
<
mwalle >
TARGET_LIBGCC2_CFLAGS =
18:25
<
larsc >
ah. found it
18:25
<
larsc >
well maybe not
18:25
<
larsc >
but i'm using lm32-*-linux instead of -uclinux
18:26
<
larsc >
and there is a t-linux which sets TARGET_LIBGCC2_CFLAGS to -fPIC
18:35
<
mwalle >
larsc: mh there are some other option wich are set if you have
*uclinux* triplet iirc
18:37
<
mwalle >
lekernel: mh dunno if thats worth the effort
18:38
<
lekernel >
we'll have to fix those bloody intermittent crashes at some point...
18:38
<
lekernel >
fortunately they do not seem to happen when rendering, only when using the gui
18:39
<
lekernel >
and they're quite rare
18:40
<
CIA-8 >
rtems-milkymist: Sebastien Bourdeauducq master * r0c0d5bc / c/src/lib/libbsp/lm32/shared/milkymist_networking/network.c : Ethernet: check received frame CRC -
http://bit.ly/ik4VxK
19:00
<
mwalle >
lekernel: whats behind the ra address?
19:00
<
lekernel >
hm I rebooted
19:00
<
lekernel >
will check next time
19:03
<
mwalle >
if its rom its the content hasnt changed ;)
19:03
<
mwalle >
i would look at my own, but i have a newer rom ;)
19:37
<
CIA-8 >
milkymist: Michael Walle master * r02cdac9 / (cores/monitor/rtl/gdbstub.rom software/gdbstub/gdbstub.c): gdbstub: fix max packet size reporting -
http://bit.ly/gktZAV
19:37
<
lekernel >
mwalle: thanks
20:26
<
lekernel >
this time crash @ 0x1000029c ...
20:27
<
mwalle >
mh this sounds like the gdbstub was already executing when BREAK was received
20:30
<
mwalle >
larsc: lekernel: i updated my qemu repository. please note that there is no milkymist branch anymore.
20:30
<
mwalle >
only master which is synced with upstream
20:32
<
Fallenou >
mwalle: all milkymist stuff is upstream now ?
20:33
<
mwalle >
Fallenou: yes
20:34
<
mwalle >
there are only two missing commits atm, one which stops the vm instead of a cpu_abort and one which fixes debug exception handling
20:37
<
lekernel >
we should also add support for the higher video resolutions, it didn't work last time i tried (segfault)
20:37
<
lekernel >
should be as simple as reading the hres and vres registers and resizing the window (ignoring the other video timing registers)
20:37
<
lekernel >
i'll have a look at that if you don't before me
20:38
<
Fallenou >
mwalle: ok so I will remove the "checkout the milkymist branch" from the wiki :)
20:39
<
Fallenou >
since all is in master now
20:39
<
mwalle >
Fallenou: yeah
20:39
<
mwalle >
lekernel: theres already some resizing code
20:39
<
mwalle >
if you change hres/vres registers
20:40
<
lekernel >
ha? different bug then... didn't investigate yet... but it appeared when I tried to switch to 1025x768
20:41
<
lekernel >
#{[{#~! why are the pc->board transfers 16ko/s... with very low lm32 cpu utilization
20:42
<
mwalle >
lekernel: is there a way to get the flickernoise binary out of the fbi file
20:42
<
lekernel >
dd the header off
20:42
<
mwalle >
how long is it?
20:45
<
mwalle >
i can see any ELF tags in the fbi file
20:48
<
mwalle >
ah because its .ralf
20:55
<
larsc >
mwalle: i still get the same error building qemu
20:56
<
mwalle >
is there a CONFIG_PTIMER in default-configs/lm32-softmmu.mak?
20:59
<
larsc >
i can fix it by prepending the obj file with hw/
20:59
<
larsc >
but then i get the same error for sd.o
20:59
<
mwalle >
do you build in-tree?
21:00
<
mwalle >
could you try a out-of-tree build?
21:02
<
larsc >
`make -C qemu-lm32`?
21:02
<
mwalle >
../qemu/configure --target-list=lm32-softmmu
21:04
<
CIA-8 >
rtems-milkymist: Sebastien Bourdeauducq master * r14cce88 / c/src/lib/libbsp/lm32/shared/milkymist_networking/network.c : Ethernet: copy-free RX -
http://bit.ly/flWS5m
21:05
<
lekernel >
anyone wants my /opt/rtems4.11 with all the flickernoise libs compiled?
21:05
<
larsc >
mwalle: nope, same error
21:06
<
mwalle >
lekernel: yes :)
21:06
<
mwalle >
larsc: grml
21:06
<
mwalle >
whats your configure line?
21:09
<
lekernel >
size 83725495 bytes
21:09
<
larsc >
mwalle: well `../qemu/configure --target-list=lm32-softmmu`
21:11
<
lekernel >
ethernet seems pretty stable now... except that FTP uploads are slow (but downloads aren't)
21:11
<
lekernel >
strange bug..
21:11
<
mwalle >
maybe it'll work for me too ;)
21:12
<
lekernel >
and there are still some rx fifo overflows too :(
21:12
<
lekernel >
even when doubling the on-chip buffering.. if you ping flood with the max packet size, you get a few
21:13
<
lekernel >
(this no longer crashes the board though :)
21:13
<
lekernel >
maybe I could fix that by making the ethernet core DMA to FML
21:13
<
rofl0r >
lekernel, which mtk files specify how stuff is drawn on the screen ? i wanna try to get it to work on directfb
21:13
<
lekernel >
but the problem is the L2 cache should be flushed for retrieving packets then
21:14
<
lekernel >
rofl0r: see fb.c in flickernoise, you just pass MTK a 16bpp framebuffer
21:17
<
lekernel >
for some reason the board sends duplicate TCP acks and messes up with the TCP window
21:18
<
lekernel >
only for receiving data - board->PC is flawless
21:29
<
mwalle >
lekernel: vgafb resizing should be fixed
21:45
<
mwalle >
larsc: any success?
21:48
<
mwalle >
really strange
22:14
<
lekernel >
qemu has a profiler?
22:18
<
lekernel >
mwalle: yup, hires works now
23:04
<
lekernel >
sp            0x8cf628c147808908
23:04
<
lekernel >
ra            0xffff0fbf-61505
23:04
<
lekernel >
PCÂ Â Â Â Â Â Â Â Â Â Â Â 0xffff35dc-51748
23:07
<
kristianpaul >
okay, lets flash finally the MM1 :D
23:35
<
lekernel >
hmm multi-slot RX seems to have fpga bugs
23:45
<
wolfspraul >
lekernel: multi-slot RX?
23:46
<
lekernel >
yes, programming multiple DMA reception buffers into the (softcore) ethernet controller
23:47
<
lekernel >
it seems that when switching quickly to another buffer, some data gets corrupted