On 17 Sep 2002, Gerhard Kutzelnigg <[EMAIL PROTECTED]> wrote: > The other problem is about handling of interrupts while in system calls and > the EINTR error number. On Linux, everything mostly works fine. On Solaris > there are problems. It looks you just forgot to check for the returncode of > -1 and errno==EINTR in io.c. While not needed for sendfile(), there still are > 2 read() and 2 write() function calls which should check this condition and > simply retry. On Linux it doesn't hurt and on Solaris it doesn't > crash ;-))
Where does it fail with EINTR? Can you please post a logfile demonstrating the problem? I checked tonight and it looks like the only place where a nonrestartable signal can occur is when it's launching the compiler on the server, and at that point it is expected. -- Martin _______________________________________________ distcc mailing list [EMAIL PROTECTED] http://lists.samba.org/cgi-bin/mailman/listinfo/distcc
