jepler: there are all sorts of things that are sort of like line numbers, and I don't understand what they all are -- but I changed AXIS to use the one that EMC uses for showing which line is executing, which seems to be right
Howdy LAdies and Gentlemen
(so I still don't really understand the bug)
I removed a few +1 things, so it must be a good fix
SWPadnos you alive?
Got nvidia? http://www.us-cert.gov/current/current_activity.html#nvgrhpoc
already fixed in ubuntu a while back
LawrenceG did you get those netgear boxes running yet?
03swpadnos 07HEAD * 10emc2/src/hal/utils/halcmd.c:
Fix minor bug in get_input
Fix bug in replace_vars that would cause errors with blank command lines
(the first word of the previous command, or garbage if the first line(s) are
blank, would get executed)
03swpadnos 07HEAD * 10emc2/src/hal/components/weighted_sum.c: Fix a couple of typos
/ns identify afs893h298@#$
pass changed now.
what is that video dyno something ?
of the cnc machine
man, you guys are worthless!
Thank you, you throwback. :)
[09:36:18] <alex_joni> http://www.linuxcnc.org/irc/irc.freenode.net:6667/emc/2006-10-26.txt
jymmmm: use the force (of google) luke
linux decided to say my video drivers where pooh... so I had to recompile.
must have been an upgrade I did.
upgrading to ubuntu 6.10
lets see what havoc I'll get from that
alex_joni: I really tried... but never remember the name of it
reading a good book, but still couldn't sleep... [usually reading long periods of time puts me to sleep]
but this book has made me think... it's called "mindscan"
[09:39:03] <alex_joni> http://www.google.com/search?hl=en&lr=&safe=off&client=opera&rls=en&q=irc+emc+dyno+video&btnG=Search
like I said, you have to remember the name
you said dyno before
and it's not the proper name anyways
go google dyno cnc
that's what I did ;)
so what's been new with you alex_joni?
* alex_joni prepares for a trip
another trip to where bouts?
maybe someone here can give me a good option.
I've currently got 3 HDs in my computer.
200 gig eIDE (split up in a few partitions), and two 200 SATA in Raid-0 ReiserFS.
I dual boot with winXP...
what's an intelligent way to split things up, where I want data to be sharable between the two of them?
okay... tha'ts what I've done with the windows partition...
and using EXT2 with the linux partition.
you can mount ext3 as ext2
however, I think EXT3 is readable by windows now as well.
and still have journaling
yeah... about to upgrade my ext2 to 3 now... just reading about it
[10:09:52] <A-L-P-H-A> http://www.crazysquirrel.com/computing/debian/ext3.jspx
I wonder can I convert an ReiserFS to ext3, and see if windows can read that partition as well.
that'd be sweet
* alex_joni has nfc
I haven't slept yet
freak'n get dark so quick these days :(
not enough sunlight
* anonimasu nods
A-L-P-H-A: very true
good morning all
is a 4GB drive sufficient to install ubunto/emc ?
we don't need two of me!
two of me would be a bonus!
ChrisSmol: yes, that's enough provided you don't want to run lots of other stuff on it too
it will be dedicated for cnc
although.... i might want to install gcc/etc as well. probably not enough room for all that
ChrisSmol: 4GB is big enough for all the packages needed to develop emc2, but it starts getting tight.
i have a 20gb drive i could use as well that has FC on it. but i'm not really doing much with that atm, i had installed mythtv to play with
IIRC I have an 8gb drive and it's enough for compiling emc and running emc
when I wanted a virtualisation of windows for my CADCAM stuff I had to add a 20gb drive though
oh cool -- I didn't realize the old non-live text install mode was on the regular ubuntu 6.06 install CD. It's called "install in OEM mode" or something like that -- second option at the boot menu
I know it doesn't install on 2gb ;). this is a breazy install a long time ago.
jepler: my coworker installed the 'server' install and I noticed it was the text install.
I thought that you had to download the right .iso image to get the different types of install
I saw server on that menu too, though
if they're all the same, why 3 different CD images?
well. my old 4gb drive seems to be flaky. so i will install on the other drive instead i think
I forget .. Ubuntu is like debian, but with various closed source blobs? right?
jepler: odd isn't it :)
ChrisSmol: I bet that you can get mythtv for linux too
yes, i was running it under fedora
kinda lost interest in it. last thing i need to do is watch more tv ;-)
robin_sz: something like that. in addition to including stuff like the nvidia drivers right on the install media (the binary blobs you mention) they say they've also concentrated on making it suitable for end-users, not geeks.
does mythtv work with sat cards?
dunno, i was using it with an ancient video input card i had laying around
worked, but using a whole pc to record tv shows seems like overkill
is it normal for the installer to sit at 15% (scanning files) for a long time?
mouse cursor still moves. no, wait. now it isn't. hmm. not good.
ChrisSmol: how much memory?
that should work I would think. Have you had it installed on this computer before?
the live cd boots ok
no, not ubuntu, but i have installed fedora
it's dead, jim. rebooting
I know I have had issues 'formating' the drive - where I had to run it twice
256MB is supposed to be enough RAM to install from the live CD
[15:41:19] <cradek> http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl/emcinfo.pl?Hardware_Requirements
i'll check the cd just in case
yeah, i saw that it said 256mb was ok. i'll eventually add another 256mb
ok, cd check passed
i'll try again. if it hangs again, i'll steal another 256mb from another peecee is try again
aha. currently at 24% this time
so far so good :)
and of course i changed nothing
i will need to go to the goodwill store to get another monitor for this system, they usually have nice 15 inch crts for $5 or so
argh. it hung again i think. at 59% this time
just for the heck of it, try the memtest boot option next time (unless you've already done so)
i haven't, but i will
what video card?
also, is this the 4G flaky drive, or the other one?
the cursor moves ok until i try to click on the installer status window, then it freezes. the clock on the upper left corner has also stopped updating
no this is another drive, and now i don't think the 4gb was flaky
well, you may want to try doing the text-mode install
internal video on a dell. and the live cd boots up and runs fine.
getting to a desktop on the liveCD is rather memory-intensive
ok. doing the memory test
i think i can grab another 256mb from another peecee, and try again.
I have also had issues with certain video cards. (during install) jeese - how many times have I installed ubuntu?
it's also a resource thing - the CD has a filesystem that gets partially mirrored into a ramdisk for speed (I think), so it's even tighter. add to that the fact that there's no swap until the system is installed, and you have a recipe for disaster on low memory systems
(which it's sad to say, 256M is these days)
I should take the system down to 128 to see how it runs after the install... Because at 128 the live cd becomes unusable.
as long as you have sufficient swap, and possibly needing some tweaking of startup scripts, 128M should work OK
I was running 6.06 on 128 Gateway laptop. Slower than mole asses in january.
memory test was ok. i've just added another 256mb, and will try again
i was assuming it used /dev/ram for the live cd boot
I've run Xubuntu without issues on 96mb ram (I had to install with a text-based installer, naturally)
so 128 should work with gnome
oh, I guess not
xubuntu was relatively snappy with only 96mb ram
almost done booting...
ok, installing again
I just installed debian testing on this laptop, its much more funky than Sarge
I remember it taking a week to install dedrat when I first got this laptop 5 years ago ... drivers and that, testing installed striaght off, one tweak of x11conf and an ndiswrapper for the wireless card and it was done
ChrisSmol: did you get further?
just finished installing
so 256mb may not always be sufficient
I think if you have an existing linux swap partition, the installer uses it during install
this would explain why there are different results on different systems
a saavy person could partition and add swap manually, after booting the livcd, and before running the install
sigh ... right, must go back to work and grab a windows PC. Not having a doz PC at home sucketh.
and --- must find some random sort of Win95 laptop to run this crappy DOS sofware on
ok, seems to boot fine, and emc is there. yay.
03jmkasunich 07HEAD * 10emc2/src/emc/iotask/ioControl.cc: re-order tool prep command and tool number to eliminate a narrow race condition window
Lerneaen_Hydra_ is now known as Lerneaen_Hydra
sorry about the dicconnect/reconnect all the time :/
Just this once .. here's the log: http://www.linuxcnc.org/irc/irc.freenode.net:6667/emc/2006-11-12.txt
any suggestion for motor mounts? i see someone who sells them on ebay, let me find the auction
these nice metal ones: 230048964600 and these are less expensive, but not sure how they will last: 200044582172
did that guy ever put online that jogwheel blog stuff?
*cant remember the name of the guy*
jogwheel blog stuff?
* Lerneaen_Hydra pokes alex
someone said something about a joghweel yesterday
oh, on the mailing list?
[19:04:46] <skunkworks_wrk> http://www.anderswallin.net/2006/11/jogging-emc2/
going to have a look at it later
Just this once .. here's the log: http://www.linuxcnc.org/irc/irc.freenode.net:6667/emc/2006-11-12.txt
anyone using edgy?
many people are
but not for emc
I've been playing around with edgy a bit, it's quite good imo, though not very different from dapper
it seemed a bit snappier in the brief experiments I did
I found my reason for my error.
gnome-terminal doesn't work with my Nividia drivers. lovely
gnome-terminal works great with my nvidia drivers ...
apparently it is trouble for some people: http://ubuntuforums.org/showthread.php?t=276354&highlight=gnome+terminal
I did that... then went here
[20:21:30] <A-L-P-H-A> http://www.nvnews.net/vbulletin/showthread.php?t=78843
not that bid of a deal, considering I can use xterm
just not as nice.
hmmm - I'm not using Xinerama - I'm using TwinView
SWP with ATI? or NVidia?
twinview is nvidia's thing
did you get your encoder wheel straitened out?
yeah it came out nice.
[20:43:45] <owhite> http://www.nilno.com/100_0349.jpg
well well - looks like microsoft might have finally fixed thier nodbc driver. finally
its for a fossilized video game that uses a mechanical encoder for the trackball.
ah - cool. A rapair then?
todays project is to re-align my optics. >>insert miserable looking frowny face here<<
you made that encoder?
sort of a repair -- he thinks that if I put more slots in it, he'll have better resolution.
how'd you get the little slits in it??
* owhite has a laser that cuts sheet metal.
No fun. you have a right angle mirror in it right?
I want a laser that cuts sheet metal!
laser would be nice :)
owhite's rich boy! :P We poor kids dun wanna play with you anymore
heh. I am unbelievably lucky.
skunk: yeah there's a right angle.
how'd you get stuck with that laser? I'll take it off your hands.
the hard part comes with getting that stupid little beam out a 1mm hole.
alpha: bought it from ebay. $28k retail laser for $6.5k. like I said, I'm way lucky.
alpha: zero hours.
is this just the laser unit?
the guy was a laser dealer that had three. *shrugs shoulders*
and he gave me really good consulting.
* skunkworks_wrk always wanted to make a laser - since 8th grade.
dang dang dang.
making a laser isn't hard... making a powerful laser is.
I guess it's just a $ issue
note the "only $6.5k" comment ;)
something tells me I'm not going to get a lot of traction complaining how hard it is to get working sometimes.
owhite: do you use more than co2? Our laser uses nitrogen, co2 and helium. It acually uses co2 the least.
owhite: no :)
co2 only refers to the gas that's in the chamber.
(all three gasses are in the chamber in ours.)
I use oxygen to come out the front port, with the beam, to heat the point where it hits the steel.
what series is this thing?
ah. mine is sealed. I think the bigger (and older) lasers were flowing gas.
what is it called - 'forced beam'? we have a blower.
havent heard that term before.
might not be right. has been a while.
it has a blower that forces the gass thru the chamber. - gets sent through a heat exchanger also.
sealed CO2 laser.
1kw would be nice. could cut 1/4 steel.
mostly used for plastic and wood. little shim stock.
I'd be happy with 100W
owhite: borked any lenses yet? ;)
not completely. I have chips in mine.
was just cleaning it.
Gawd, I'd be sooooooooooo happy with 100W fiber CO2 laser =)
but I promised myself not to complain on #emc.
what kind of focal legnth are you using?
ok - that is what we use for plastics - for dieboard(plywood) we use a 5 inch focal
* owhite is in the middle of finding if its 3.60 to 3.40 inch right at this moment :-)
that would be fun. could you describe the ventilation system?
well is like a big ass hood or a suckydoohickus near the cutting head?
or do you have nothing, like me. :-)
skunkworks you're not using a blower to exhausted the gasses?
owhite: yes - we have a table that is sealed and evacuated from the bottom - also the hood sucks from the top. Just a huge exauste fan mounted on the roof of the building
(there is also a blower forcing the gasses around the laser tube - but that is a sealed circuit.
skunkworks air assist?
hey! you know what... that photo has enough ridge details on his owhite's photo, that anyone with some knowledge could copy his finger prints and commits misc crimes with it. http://www.nilno.com/100_0349.jpg
leave partial finger prints everywhere.
we use compressed air for the tip
ha! but if you did that you'd get linked to all the crimes I've already committed.
do it A-L-P-H-A, do it!
donate your laser to Canada, or prepare to be invaded.
owhite when you were searching for your laser, did you ever come across smaller ones responably priced?
ever seen that mythbuster episode where they use a photocopy of a fingerprint (FP) to fool a supposed top notch biometric scanner?
jymm: hell yeah. ebay has very good prices actually.
[21:02:11] <A-L-P-H-A> http://search.ebay.com/search/search.dll?from=R40&satitle=co2+laser
Jymmm: google. ebay. froogle. llama.
ah yes, this is the part I forgot about -- how to force that air solenoid open when its not under the control of #emc.
owhite: a button ?
owhite timer relay?
anonimasu: this is the problem of having everything under software control. when I'm using a simple calibration board to drive the laser, its not on the emc system, and I have to kludge something.
dumb. I know. but...
owhite: jumper the relay/whatever
or turn the laser on via emc :)
owhite OH.... solid state relay + OFF|AUTO|MANUAL switch
its...confusing....and not perfect.
well, nothing's perfect..
welll, hm. maybe emc2 would let me....
it's all a compromise between what we want and what we can do..
its one of those -- 'just cause I built this thing does not mean I am under any obligation to remember how I wired it'.
btw, what kind of efficiency can you expect to have with a CO2 laser of that size? 5%? 10%?
I think I've seen discusssion of efficiency on Sam's laser FAQ.
any idea of what it is, very roughly
something like that.
is your laser liquid cooled? it would seem to me that 1kW shouldn't be that hard to air-cool
(depending on the surface/shape/material of course)
owhite: laser faq?
* skunkworks_wrk 's name is sam. Was wondering ;)
* owhite goes with the 'disconnect the 9v battery from the smoke alarm and connect the battery to the relay' strategy.
skunkworks: The you should know about your own faq! Eeeeesh
Samantha... how pretty.
actually my dads name is sam - so I am 'son of sam' - watch out.
hermaphrodite begets hermaphrodite.
I feel better today.
* Jymmm sends skunkworks and skunkworks_wrk all of A-L-P-H-A's information, profile, and habits.
gave up on a girl... and feel kinda relieved.
A-L-P-H-A batting for the other team now huh?
Jymmm: I've always been a lesbian trapped in a mans body.
aren't we all
A-L-P-H-A but if they made you female, you would just be playing with yourself way too much!
dude, we have penises... we already play with that too much.
if I had a girl parts... that means I have 3 parts to play with.
A-L-P-H-A Did you notice have everyone else just shutup?
they all know it's true.
A-L-P-H-A maybe they are just "occupied" at the moment.
need to get my act to fuck'n together. bbl.
* owhite is adjusting the laser which involves tedious trial and error tests of height and pulses sent at a card, the smaller the diameter of hole produced on the card the better.
owhite: Do you have a windows box?
one laptop with windows, and the emc/linux machine runnign the laser.
owhite: You *MIGHT* consider grabbing the GCC laser (print) driver... It has settings that you might be able to use to set the pulse rate/dureation. At least for refernce.
that's good to know. what I do is I have a custom circuit board that handles the duration setting.
then I have that board connected to a foot peddle :-)
owhite: Well, their driver has custom settings for various materials and such... they already did the math, so you can copy-n-paste to at least get you in the "ballpark"
nice to know.
what I like is since my table is made of upside down hack saw blades is to get all sweaty, rest my arm on the blades, hit the foot switch, watch the pulse hit the thermal paper, change the height, repeat.
its called bonding through pain.
if I worked in a commercial shop OSHA would shut this joint down in about 2 minutes.
that long huh?
and now its cutting like there's no tomorrow.
happiness is a well-adjusted laser -- and teenage daughter.
let's not go too far here
anybody know what it'd like to have a laptop that's connected by wireless running the linux box that is connected to a cnc system?
er, what it'd _be_ like...
owhite: wouldn't it be: happiness is a well-adjusted teenage daughter -- and laser.
I think the main issue would be wireless on the machine controller PC
03jmkasunich 07HEAD * 10emc2/src/hal/drivers/hal_evoreg.c: replace some uses of hal types with generic types, where the hal types are not being used for hal objects
I don't know how well madwifi or ndiswrapper play with RT
well I know theres some way to run an X system on a windows machine, but dont know much about the details.
owhite: Would you really want your laser under war driving remote control?
it already is.
you are already using wifi?
well, I guess my wireless system is firewalled.
elson: as in Jon?
yes, it's me.
owhite: Um, you do know that wifi is VERY easy to break the keys, right?
jymmm: fair enough.
owhite: And that if they do break the keys, it is a connection INSIDE the firewall.
(under 90% of setups)
I was hoping to ask John Kasunich a question, but it doesn't look like he's online.
I think jmkasunich is here
perhaps in the devel channel ...
03jmkasunich 07HEAD * 10emc2/src/hal/drivers/hal_m5i20.c: replace some uses of hal types with generic types, where the hal types are not being used for hal objects
* owhite is a happy Universal Stepper Controller customer.
*waves at elson*
OK, I didn't know what the difference is between greyed and foreground usernames.
gray usually means away, but it also means "forgot to turn off the away flag when he came back"
Good, glad you are a **happpy** user!
yeah and I just got my UCS to work on emc2.
thanks very much to #emc-devel.
John, I just discovered that the UPC driver has a problem, in fact that's what forced the recent change. It needs a non-contiguous write for the PWM outputs - the control reg is not adjacent.
are there important things in between the used registers?
it _has_ non-contiguous writes, now
I need to know how to deal with the add_wr_funct() call to set up the bit map with a gap.
ok, I thought about that, but didn't make any changes to avoid rocking the boat
one approach is to pass a bitmap, but that means you have to either hard code the bitmaps, or build them at compile time from the various #defines of register numbers
Umm, no, but it is a big gap. I could just span the gap right now and make sure that's the only problem.
another approach is to split out the declaration of regsiters to be written from the adding of the function
as a quick workaround, you could use multiple add_wr_funct calls, but that's not the "right" way to do it
Can I just call add_wr_funct() twice?
but you'd need a dummy function to call for the second wr_funct
yes and no - that is certainly the easiest way, but you'd want to pass NULL as the function for subsequent calls, and add_wr_funct needs to be modified so it doesn't store the NULL param in the function list
elson: I have a question when you're done with your current thread.
Yeah, that's the rub. Well, maybe I should just span the gap for now, it will only add ~3 us per cycle.
lets do it right
we can split add_wr_funct into add_wr_funct() and set_registers_writable() or something like that
one possibility is to use use the min_addr as a mask if min_addr<0 (or something like that)
oops - if max_addr<min_addr or if max_addr<0, use min_addr as a mask
so present calls don't need to be modified, but new ones can use bitmasks
elson: are there lots of little gaps?
How about a call that does exactly what add_wr_funct() does now, and another funct that does the different parts?
(such that you'd needs lots of calls with short ranges in order to convey the entire pattern)
right now, add_wr_funct() does two things
ie - add_wr_funct_mask(yadda yadda)
1) it adds a function to the list to be called each thread
2) it marks a range of registers to be written from cache to hardware
No, this is really the only place I know of a gap in normal operations. The USC has no gap. Maybe I should have changed the addr of the control reg on the UPC, but I didn't.
SWPadnos : grab me when you're done.
ewww - I don't think so ;)
it makes perfect sense to me to either split that into two functions, or to change the way we pass the list of registers to be written
actually, making the masks isn't a big deal. I'd just change it to use a single mask parameter
Right, changing the way we pass the list could be another way.
there are only ~7 calls to add_wr_funct
maybe just pass the bitmap, and provide a start/end to mask generator.
as steve said, passing it as a mask is trivial to implement (actually easier than the original approach)
oh, good idea
the mask generator might even be a macro
Then, you could call the start/end converter, then add one extra bit.
and you can or multiple masks to gether if you have multiple ranges, or just or in the bit if its not a range
ok, I'll do that right now
Yeah! This makes sense!
I just happen to be working in the driver directory
OK, thanks, John!
owhite, you had a question?
elson: while you are here
there is something else I wanted to talk to you about
elson: are you working on another I/O board. someone on here told me you were.
talk to owhite, I'll start on the driver
but don't go away when you are done
I need a bit more i/o above what the UCS has.
Right now, I am working on a threading update for the USC.
was hoping I could stay with the UCS and use the other board as well.
All USC's have an additional 8-bit output port. We have a way
to write an 8-bit numeric value, as for a DAC. We have partial code to support using that port as 8 separate outputs.
beyond the current connectors?
the P8 connector, next to the DB-25
Yes, there's P8 over by the DB-25 connector. There was no software support until the latest driver.
oh yeah? mine is currently unoccupied.
that would be great.
Yup, you would have to solder in a male header in that location.
I want more outputs.
so it sounds like youre still working on the software to support it, true?
These are just plain 5 V TTL outputs, but could easily be wired to an Opto-22 type SSR mounting panel.
I'm implementing a bunch of stuff that had always been planned.
well if you're on your way that works out fine -- I'll start working on the supporting external circuitry.
Well, what we had on EMC1 was pretty mature, but this extra output wasn't implemented.
no problemo. I'll start rigging the electronics to drive what I had in mind.
emc2 can drive my chiller, ventilation, and laser power supply.
The current driver can export a new hal pin extraDAC, and a scaling parameter for it. So, you can get 256 different combinations of the bits. Partially finished is a different code that will export 8 pins extraDout(0..7).
I thought it was all finished.... I guess I lost track of where we were
The other thing I'm just planning now is a brushless drive to complement my brush PWM servo drive.
wait a sec -- the conversation you're having with jmkasunich is about makiing the extra output port work?
not really, but we're working in the same file
its probably a couple hours work
okay that'd have been too funny.
It says "export code goes here" Doesn't look to complicated. I could probably do it myself. Just haven't gotten around to it.
I think my plan was to make sure the first one (the DAC) was using the proper approach
once we know that works and are happy with it, its pretty much cut/paste to do the rest
The numeric version that treats the 8 extra bits as a single number works now. Another option that treats it as 8 separate bits is not finished.
I think the DAC version is now working fine. I made a small change to abs() the value sent to the dac.
what would be great for me is if some day I can run an m-code that operates on that port -- is that the plan?
that's being planned, but it touches many levels other than the driver itself
Yes, there is a way to add or assign new M-codes to the interpreter, and then have them passed through HAL to the aux. I/O functions.
there are also M-codes for turning on/off aux I/O, which would then be connected in HAL to the appropriate pins
I think this M101, etc. is already possible. no?
elson: that abs() change is what I wanted to talk to you about
I'm not sure what version those are in though
I'd like to see a bit param that that can turn the abs() on or off
that happens sometimes
a server glitched somewhere
Thanks, I'm sure you can do this faster than I can.
I called the function that takes a start and end address and makes a bitmap "block"
does that seem clear?
OK, I guess so.
in that case, its (almost) done
almost, because I didn't change any of the existing blocks - you will get the exact same behavior as befere
I'm gonna commit, and leave it to you to fix the gap
As far as I know, there is only one function (PWM generator) that needs the gap.
As soon as you commit, I will pull it down and test it.
you'll need to change "block(START_ADDR,ENDADDR)" to "block(START_ADDR,NEW_END_ADDR) | block (OTHERSTART,OTHEREND)"
am I making sense?
OK, great, thanks for the work! Yes, makes sense.
03jmkasunich 07HEAD * 10emc2/src/hal/drivers/hal_ppmc.c: change method of marking cached EPP addresses from contiguous blocks to a bitmap to allow arbitrary sets of addresses to be cached
there you go
there are some prints of the bitmaps in there from when I was working on it
once everything is good, those should be removed
4 lines starting at 631
Got some new error messages : Depending conv_bit_u8.9
sh hal/components/mkconv.sh bit u32 // < hal/components/conv.comp.in > hal/components/conv_bit_u32.comp
../bin/comp --document -o ../docs/man/man9/conv_bit_u32.9 hal/components/conv_bit_u32.comp
did you have a current checkout before you got my changes?
because that is stuff that jepler added a couple days ago
has nothing to do with what I changec
are those actually error messages?
I don't see the word error or warning in there\
Yes, did a complete checkout just a couple minutes ago. Yes, I didn't think this had anything to do with hal_ppmc.c
Right, not an error, just a new message, but LOTS of them.
thats because he added a LOT of new components
I'm going to go test it now.
I suspect there's some __constant_bit_mask thing in gcc or the kernel, rather than a loop. if I find it, I'll stick it in there
yeah, I just didn't feel like messing around with doing it in a macro
if it ran in realtime I'd care, but since it runs only during init, I figured keep it simple
yeah - it's just annoying :)
03jmkasunich 07HEAD * 10emc2/src/hal/drivers/m5i20.h: replaced HAL types with generic types when referring to hardware registers on the m5i20 card itself
03jmkasunich 07HEAD * 10emc2/src/hal/drivers/hal_m5i20.c: changed three u16 HAL params to u32
rm -rf /
does anyone know the name of the windows application that runs X on a windows machine?
03jmkasunich 07HEAD * 10emc2/src/hal/drivers/plx9030.h: replaced HAL types with generic types when referring to hardware registers on the m5i20 card itself
ha. I already had in installed. :-)
John, it is basically working, after fixing a type. I am getting very narrow pulse widths though, I'm still investigating that.