Greetings,
* Tom Lane (tgl@sss.pgh.pa.us) wrote:
> I wrote:
> > I wonder if we should drop the idea of returning a positive bytecount
> > after a partial write, and just return the pqsecure_raw_write result,
> > and not reset PqGSSSendConsumed until we write everything presented.
> > In edge cases maybe that would result in some buffer bloat, but it
> > doesn't seem worse than what happens when the very first
> > pqsecure_raw_write returns EINTR.
>
> Here's a patch that fixes it along those lines. I like this better,
> I think, not least because it removes the assumption that "interesting"
> pqsecure_raw_write failures will recur on the next try.
Thanks for looking into this. I agree with this approach and took a
look through the patch and it all makes sense to me.
> Still need to look at syncing the backend with this.
I see you made similar changes to the backend and those also look good.
Thanks again,
Stephen