At Fri, 26 Feb 2021 17:30:32 +0900 (JST), Kyotaro Horiguchi <horikyota....@gmail.com> wrote in > At Fri, 26 Feb 2021 16:12:39 +0900 (JST), Kyotaro Horiguchi > <horikyota....@gmail.com> wrote in > > This inhibits logCursor being updated. What is worse, I find that > > logCursor movement is quite dubious. > > Using (inCursor - inStart) as logCursor doesn't work correctly if > tracing state desyncs. Once desync happens inStart can be moved at > the timing that the tracing code doesn't expect. - This requires (as I - mentioned upthread) pqReadData to actively reset logCursor, though. + So pgReadData needs to avtively reset logCursor. If logCursor is + actively reset, we no longer need to use (inCursor - inStart) as + logCursor and it is enough that logCursor follows inCursor. > > logCursor should move when bytes are fed to the tracing functoins even > when theyare the last bytes of a message.
regards. -- Kyotaro Horiguchi NTT Open Source Software Center
