signals and pthread (again)

Hi NG,

First of all I want to apologize for reposting, but my original message didn't show up in google groups, so I figured it didn't get much further than my ISP.

Recently I've been experimenting with libposix_aio to do asynchronous IO. I've used a 2.6 kernel patch to be able to get a signal on IO completion. I first block the signal and then wait for it using sigwaitinfo. I don't have much experience with signals, so I'm not sure if I'm allowed to use any signal in the aiocb structure, but I use RTMIN.
If I do all this in a single thread it works fine. The thing is that I want to be able to create a few threads using pthread_create that will all wait using sigwaitinfo.
Because I wasn't sure if the signal would be sent to all threads or just one (as I want) I wrote a test program to see which thread got the signal. To my surprise none of the threads got a signal.

As I said before I'm not to experienced with signals, so I hope somebody can help me with this issue, or point me to a site that explains the posix signals.

Thanks,
Mark
.

Relevant Pages

  • Re: A problem with fork() and managing processes
    ... >I was experimenting with fork(), signals, and other process management ...
    (comp.lang.perl.misc)
  • Help please with MS technique
    ... I'm experimenting with M/S. ... Behringer LDC for the side signals in figure 8 mode ... I made a copy of the Side signal wav, ... from a single mic source that was inverted and added ...
    (rec.audio.pro)
  • pthread and signals
    ... I posted this message to comp.os.linux.devlopment.apps knowing that it's a very helpful group. ... Recently I've been experimenting with libposix_aio to do asynchronous IO. ... The thing is that I want to be able to create a few threads using pthread_create that will all wait using sigwaitinfo. ... As I said before I'm not to experienced with signals, so I hope somebody can help me with this issue, or point me to a site that explains the posix signals. ...
    (comp.programming.threads)
  • EINTR causes sigwaitinfo and pthread_kill to become hosed
    ... The attach code mimics the algorithm used by a major software vendor to ... sigwaitinfo which causes the worker thread to effectively suspend (no ... Worker thread becomes consumed with processing the signals being ...
    (Linux-Kernel)
  • Re: FFT and IFFT
    ... is becasue I have multiple signals and therefore it is ... go to the archives in Google groups and search on ... I haven't been able to introduce the time delay and ... greg-heath fft time-advance ...
    (comp.soft-sys.matlab)