This is the mail archive of the
mailing list for the GCC project.
Re: PATCH: Eliminate uses of "sleep" in V3 testsuite
On Wed, Jan 05, 2005 at 12:53:14PM +0100, Paolo Carlini wrote:
> Mark Mitchell wrote:
> >So, Zack avoided the problem that the HP-UX kernel refuses to open a
> >FIFO with O_WRONLY | O_TRUNC, but now we have a race condition. I
> >think that the best solution is to remove the child process. Thoughts?
> Ok, now I see what you mean, and I agree with your analysis (but you are
> using the technical terms "child" and "parent" rather unorthodoxly wrt
> your book on Linux Programming ;) : what really count here, is that,
> when an "ate" open fails, we close the fail and return error (i.e.,
> null). The other process is not doing anything useful.
But if there's a fifo with no reader, could open(name, O_WRONLY) block,
so we don't even get to the seek?
As Mark said, using a pipe would avoid these peculiarities of fifos, and
the child would be redundant.
"I have had my results for a long time, but I do not yet know how I am
to arrive at them."
- Karl Friedrich Gauss