Return-Path: MIME-Version: 1.0 In-Reply-To: <7edd28a2-c446-2323-169a-b8630fdb1557@gmail.com> References: <20180514235614.25231-1-inga.stotland@intel.com> <5b6349ba-e0dd-8b56-5654-c1e0e20be4a8@gmail.com> <7edd28a2-c446-2323-169a-b8630fdb1557@gmail.com> From: Luiz Augusto von Dentz Date: Thu, 17 May 2018 19:03:39 +0300 Message-ID: Subject: Re: [PATCH BlueZ] shared/io-ell: Add support for ELL based IO handling To: Denis Kenzior Cc: Marcel Holtmann , "Stotland, Inga" , "linux-bluetooth@vger.kernel.org" Content-Type: text/plain; charset="UTF-8" List-ID: Hi Denis, On Thu, May 17, 2018 at 6:01 PM, Denis Kenzior wrote: > Hi Luiz, > > On 05/17/2018 03:01 AM, Luiz Augusto von Dentz wrote: >> >> Hi, >> On Wed, May 16, 2018 at 5:42 PM Denis Kenzior wrote: >> >>> Hi, >> >> >>>>> + >>>>> + do { >>>>> + ret = writev(fd, iov, iovcnt); >>>>> + } while (ret < 0 && errno == EINTR); >>>> >>>> >>>> explain this one to me. Or maybe Luiz should explain it since he >> >> introduced this. >>>> >>>> >> >>> I'm curious why not use TEMP_FAILURE_RETRY macro? ell already uses that >>> for clarity. >> >> >> Szymon was actually suggesting that we should have something similar in >> ell, e.g. l_io_send, it would be nice if that would handle iovec similarly >> to how is done here but perhaps using the TEMP_FAILURE_RETRY as suggested >> by Denis. >> > > I would rather not do that actually. It is already weird that you guys have > io_send but not io_recv. And since io is not only used with streams / > files, but with datagram sockets as well, we would have to add a bunch of > methods for symmetry. Not to mention that somehow 'send' uses iovecs and > the convention inside ell for this is to suffix the method with a 'v'. Ive thought of adding io_recv, it just that it was not necessary by anyone currently, anyway I guess having _v alternative makes ell play nice if the application is using iovec, or you are saying that with ell the application would have to use the fd directly and handle errors? > Also, just a nitpick, but io_send should really be io_writev. You're just > confusing every hardcore UNIX user here since a vectorized send is > 'sendmsg'. Yep, changing it is not a problem though since it just an internal API. > By the way, correct me if I'm wrong, but I think there is zero need for > TEMP_FAILURE_RETRY inside BlueZ since it uses signalfd and non-blocking IO > anyway. The only reason ell uses this macro is because we can't assume how > signals are used in the enclosing application. Usually, BlueZ tools would use signalfd, including the daemon, we could perhaps remove the handling of EINTR if we could detect if signalfd is in fact being used but I don't know if that is possible, so better be save than sorry. -- Luiz Augusto von Dentz