El 26 de març de 2012 21:30, Robert Millan <[email protected]> ha escrit: > That doesn't mean the 34..64 range is reserved AFAICT. Unless I > missed something, when someone in upstream comes up with a new signal > that has a specific meaning in the kernel, they will use 34 and our > existing binaries would break. > > If we're going to abuse unreserved space, wouldn't it be safer to use > something close to __SIGRTMIN? > > Besides, upstream explicitly reserves SIGTHR (32) for libpthread and > SIGLIBRT (33) for librt (for the same purpose as my patch, in fact). > Wouldn't it be less confusing if we follow upstream on this, at least > for 32 and 33?
Actually, none of this is relevant to this patch now. My initial version added a new signal (35) but it can be done like NPTL instead. >From nptl/pthreadP.h: /* The signal used for asynchronous cancelation. */ #define SIGCANCEL __SIGRTMIN /* Signal needed for the kernel-supported POSIX timer implementation. We can reuse the cancellation signal since we can distinguish cancellation from timer expirations. */ #define SIGTIMER SIGCANCEL (for LinuxThreads, PTHREAD_SIGBASE+1 rather than __SIGRTMIN, it works fine here) -- Robert Millan -- To UNSUBSCRIBE, email to [email protected] with a subject of "unsubscribe". Trouble? Contact [email protected] Archive: http://lists.debian.org/caofdtxpw2uypza_p7tgd2qvjpvcbh2a9yy6eebbro9o9l6p...@mail.gmail.com
