Ole, RobberA is robbing a bank and posts her accomplice, RobberB, outside the door to keep watch. Independently, RobberC arrives on the scene to rob the bank as well, noticing that the bank is <<being>> robbed, she takes off and vows to give up her life of crime.
Compare: RobberA&B finish their hard days work and leave, when RobberC arrives, .2s later, to Rob a bank that has no money, three dead clerks, a screaming child and cops on their way. RobberC gets hanged for murder. Compare: RobberA&B finish their hard days work and leave, when RobberC arrives, 1s later, and notices the bank security guard coming out the door in pursuit of RobberA&B. RobberC robs a different bank. The case I was differentiating was between 0 && !0 where actionable information is held by the mutex or semaphore itself, and no other variables are required. Obviously in the difference between .2 and 1 is not contained within the semaphore but in some other time variable. With that said, I have no intention of robbing a bank, at least in the foreseeable future, and the question from my side at least was pure pedantry. I withdraw the question ;) Thanks again for your great work, and--just as importantly--careful oversight of this forum. Bradley On Sun, Jan 4, 2015 at 6:26 AM, Ole Tange <o...@tange.dk> wrote: > On Sat, Jan 3, 2015 at 10:09 PM, Bradley <b...@customweather.com> wrote: > >> On Jan 3, 27 Heisei, at 11:21 AM, Ole Tange <o...@tange.dk> wrote: > >>> On Fri, Jan 2, 2015 at 8:53 PM, Bradley Asztalos < > b...@customweather.com> wrote: > >>>> On Fri, Jan 2, 2015 at 7:35 AM, Ole Tange <o...@tange.dk> wrote: > >>>> On Fri, Jan 2, 2015 at 10:51 AM, Bradley Asztalos > >>>> <b...@customweather.com> wrote: > : > >>> Does --semaphoretimeout 0 mean never give up? > >> > >> Yes: Just as if the option had not been given. > >> > >>> If so then -0 could mean give up immediately. > >> > >> -0 = 0, so no. > >> > >> It is implemented in the git version. > : > > So did you provide for a way to give up immediately if sem cannot be > acquired immediately? > > As mentioned, no. > > If you feel that is very important, please provide a situation where > waiting 1 sec for the semaphore before giving up is much worse than > waiting 0.2 sec (which is more or less the time it takes to run sem). > I could not find a situation in which that extra second made a huge > difference. On the contrary I can easily find situations where giving > up immediately is less useful than trying for just a second. > > /Ole >
