| |
| |
| |
|
Page: 1 2 3 4 5 6 7 8
Comments:
<0> no, really.
<0> that looks fine. but since i have no idea what you are doing it might be wrong even so.
<1> dvd
<0> it smells like a dvd, sure. pal perhaps.
<1> indeed
<0> that looks like ffmpeg, so i'd guess dvdauthor is next.
<0> then again there are other potential steps, perhaps you want a menu, or some alternate audio streams or angles.
<1> I did -aspect 4:3 on ffmpeg and it totaly ****ed up, I've hardcoded -s 720x576 this time instead :P
<1> **** that
<1> dvdauthor tgz downloadeddeded
<1> missing a package.. *google*
<1> I bet *something* doesn't compile at some point on this dep chain :P
<0> wouldn't surprise me since you are compiling from the source tarball.
<1> but it's so fun! ;)
<1> LIBXML test program was built but could not be ran this time
<1> eww it's a gnome package
<1> i'm going to be here all day
<1> hooray, it didn't fail any deps
<1> that's a first ;)
<1> maaan this **** is full of warnings
<1> someone should fix this poop
<0> feel free.
<1> hmm video came out a bit choppy
<1> dunno if that's mplayer or teh mpg
<1> twkm: Know how to use dvdauthor?
<0> nope.
<1> everything seems to use an xml file
<0> pile of xml ****, then it works.
<1> hmm I can't get it to work menuless :/
<0> then make a menu.
<1> heh, I can't even get it to work WITH a menu acctualy
<1> No .IFO files to process
<1> aha
<1> got it :)
<1> df -h
<1> oops
<2> fag
<1> SHUT UP :( I'll get my bofriend on to you!
<0> careful with that lithp.
<3> hey, can somebody remind me as to if you can p*** by reference in C? Or is that just C++?
<0> just c++
<0> well, and ada and pl/1 and ...
<3> ok thanks
<3> I wish I had never learned C++
<0> i hear they have an injection that'll reverse it.
<0> of course it gives you ebola, so you have to want to avoid c++ pretty strongly.
<3> hmmm, I'll have to put some serious thought into that
<3> right now, ebola sounds like the better choice
<4> alzheimer definitely reverses it
<3> ah but one has to wait so many years for that sweet release
<3> hmmm ok random wierd question: is it possible to put an array within a function call? IE something like this dosomething({0,1,2}); where that function takes a pointer (to an array) as it's parameter
<1> pretty sure you can't
<1> it's the same thing as int array[] = {0,1,2]; though
<3> I've never seen it done, but I think it'd be a convienent shorthand
<3> except it would never have a name ***igned to it, and the memory for it would be gone as soon as the function returned
<4> seems to me c99 has something for this
<0> it does.
<0> foo((int[]){0,1,2})
<5> Ah. I like Google Calendar :)
<6> hi i have. two threads running and i want to send a message to the other thread asynchronously
<6> pthread_kill only sends a signal but no information is p***ed
<6> how can i p*** information? something similair to sigqueue
<4> oof course you can send messages to the other queue
<4> s/queue/thread/
<6> the other thread could be blocked though
<4> there are million ways
<6> that's the problem
<5> kvasov: You could write your own message p***ing functions.
<6> well i am doing that for other cases where there is no possibility of a long long wait
<6> using queues
<5> One of the easiest methods might be to have thread #1 write to a pipe that thread #2 reads.
<6> well in this case thread #2 is doing a wait
<6> it can't read a pipe
<5> What's it waiting for?
<6> it's waiting for a process which i need to schedule via the asynchronous message
<6> hence it needs to receive the asynchronous message and continue waiting
<6> otherwise there is a deadlock possibility
<5> Ah...so you need to wake it up.
<6> wake up who
<4> ssomething is wrong in your description, kvasov
<6> AndroMish, why is that
<6> ok lets say i have it like this
<0> make a queue.
<6> thread #1 is looping scheduling processes on the system
<5> Do you mean...starting processes?
<6> when a process does a trap, thread #1 receives the notification via wait
<6> no
<6> scheduling them
<5> Interesting.
<6> i use ptrace
<0> ugh.
<6> thread #1 could potentially start an system call handler in an another thread and continue scheduling other processes
<0> ptrace and threads. funny man.
<5> I haven't done unix programming that low level.
<5> So I'm useless.
<6> i have GUI that's why
<6> the ptrace stuff is simple
<6> i just have trouble getting the gui to work nicely with this
<4> what you are saying kvasov, the 2nd thread is waiting for messages from 1st thread
<4> right ?
<6> 1st thread is doing a wait on a process. it is the only thread that can schedule processes
<6> if no processes are scheduled and it's waiting, it's starvation
<0> i suggest a queue.
<6> hence, it needs to receive a message and schedule the process and continue waiting
<6> the queue would be fine, but if it's hanging in wait, it can't read the queue
<0> a hung thread is hung, you are fooked.
<6> the only way out of it is through pthread_kill. the bad thing is that it only sends a signal, no information (ie. which pid to schedule). i dont want to use global variables
<6> no. if it receives SIGUSR1, it will scheudle a process and continue running
<0> then it isn't hanging, it is merely blocked.
<6> yeah
<0> signals and threads are bad juju though.
<6> true
<6> there seems to be no other way out
<6> i just want to send an integer. i am not mixing any more than that
<0> anyway, provide an idle process, for it to wait upon when it has nothing else to do.
<4> exactly mu thought
<4> use dummy process to alert the thread
<0> then you can suspend the idle process, which should provide a waiting thread with a status update.
<6> yeah that's kind of messy...
<0> threads and signals and ptrace ... you are in the land of messy.
<6> yeah but dummy process is a step above
<0> hell, you may be its king.
<0> so why not try for emperor (with an idle process)?
<6> heh
<6> it can be much worse trust me :-)
<6> there aren't any virtual machines involved
<6> what i dont get it. how can you simulate a SIGCHLD for example
<6> if you send it with kill, you can't p*** any status information
<0> kill is the wrong tool.
<6> why is that
<0> if you want to p*** a value you need sigqueue.
<6> you cant sigqueue with threads
<6> even if you ignore it in all threads except for one, it wont deliver the signal
<0> well, a) yes you can, and b) see what i mean about threads and signals?
<0> what is wrong with your threading library?
<6> nothing
<4> it's not as bad as you thyink
<4> signal only signals to the other thread that there's something in the msg queue
<0> sigqueue delivers an asynchronous signal, which is sent to the process, which is delivered to any thread which is not masking it.
<4> now, the msg queue shall be implemented separately from signal;s
<7> I'm too lazy to bugger around with threads & signals <shudder>
<0> to smart!
<0> err, too
<4> the signal tells the toher thread to check the msg queue, that's it
<6> i just tried doing this and it kept delivering signals to the same thread even though i ignored it in that thread
<4> you don't really need signalue
<7> I just use mutexes and p*** stuff directly
Return to
#c
or
Go to some related
logs:
#politics
#solaris
#beginner
#politics
#beginner
#computers
#hardware
#sql
#computers
#computers
|
|