On Thu, Aug 29, 2019 at 10:27:16AM -0400, Derrick Stolee wrote: > > I don't think we should need such a call. For one thing, if it were > > necessary, that would mean we're not writing out the packet at all. But > > your whole problem is that we're writing the message twice, one of which > > comes from the packet. > > The problem the flush() was trying to solve was the new "Broken pipe" error, > which I had assumed was due to a communication race. (Looking at the message > more closely now, I see that Szeder was able to repro this broken pipe both > with and without my change. I am still unable to repro the broken pipe.) I think the broken pipe is coming the _other_ way. We do send the packet from the server to the client, but since the client is still writing when the server has hung up, we get a write error instead of seeing the error packet. So any fixes there have to happen on the client side. I am still confused about why the client is writing in this case, per the argument in 014ade7484 (upload-pack: send ERR packet for non-tip objects, 2019-04-13). It would be nice to use GIT_TRACE_PACKET to see what it's trying to write, but I still haven't been able to reproduce the issue. -Peff
