3.3.0: Possible regression in cygwin DLL (Win10); fixed in snapshot

[Takashi Yano]

On Wed, 10 Nov 2021 11:34:25 +0100
Corinna Vinschen wrote:
> On Nov 10 17:30, Takashi Yano wrote:
> > On Sat, 6 Nov 2021 12:42:46 +0100
> > Corinna Vinschen wrote:
> > > On Nov  6 15:10, Takashi Yano wrote:
> > > > Unfortunately, these solutions do not resolve the issue
> > > > which is another issue with C# program:
> > > > https://cygwin.com/pipermail/cygwin/2021-March/247987.html
> > > > This still needs FILE_SYNCHRONOUS_IO_NONALERT flag.
> > > 
> > > If we want to add FILE_SYNCHRONOUS_IO_NONALERT, this would have to be
> > > solved by running NtReadFile/NtWriteFile synchronously in a thread,
> > > started on every invocation of raw_read/raw_write.  raw_read/raw_write
> > > would then call cygwait on the thread object.  To break on signal or
> > > thread cancallation events, it would have to call CancelSynchronousIo.
> > > That's certainly doable.
> > 
> > I tried to implement your idea, however, I noticed that
> > NtQueryObject(ObjectNameInformation) call in
> > get_query_hdl_per_process() is blocked while reading the
> > pipe if FIPE_SYNCHRONOUS_IO_NONALERT is set and pipe is
> > in blocking mode.
> > 
> > So I would like to propose alternative implementation with
> > FILE_SYNCHRONOUS_IO_NONALERT being set. Please have a look
> > at attached patch. With this patch, pipe itself in read side
> > is always set to nonblocking mode and simulate the blocking
> > behaviour in raw_read(). This can eliminate creating thread
> > for reading as well as calling CancelSynchronousIo().
> > 
> > Note that setting FILE_SYNCHRONOUS_IO_NONALERT only for read
> > pipe seems to be enough for C# programs.
> > 
> > What do you think of this implementation?
> 
> Can you push this to the topic/pipe branch for playing?  I updated
> the branch to current master.

Thanks. I have just pushed the experimental patch to topic/pipe.
Please try. If something wrong, please point it out.

-- 
Takashi Yano <takashi.yano@nifty.ne.jp>