[Bug 51719] New: _pclose() does not cause child stdin to report EOF

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=51719

Bug ID: 51719
           Summary: _pclose() does not cause child stdin to report EOF
           Product: Wine
           Version: unspecified
          Hardware: x86-64
                OS: Linux
            Status: UNCONFIRMED
          Severity: normal
          Priority: P2
         Component: msvcrt
          Assignee: [e-mail]
          Reporter: [e-mail]
      Distribution: ---

Created attachment 70602
  --> https://bugs.winehq.org/attachment.cgi?id=70602
regression test showing the bug

Hi,

Normally when the write end of a pipe is closed, the read end should see that
upon further reads. However wine's _popen()/_pclose() do not work that way - a
child's read from stdin blocks after the parent's _pclose().

I've attached a patch to the tests to show the problem. The tests hang, but
should succeed.

I reproduced this on git commit 8b9f1e12841298591387e2b7590191610a37986e

Regards,

Andres

PS: I encountered this because I was trying to fix a bug in the windows port of
postgres without having to use windows...

[Bug 51719] _pclose() does not cause child stdin to report EOF

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=51719

Piotr Caban  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |[e-mail]
      Fixed by SHA1|                            |e38f3761f890ccc0b32ac16c633
                   |                            |cdf7582511133
         Resolution|---                         |FIXED
             Status|UNCONFIRMED                 |RESOLVED

--- Comment #1 from Piotr Caban  ---
I've sent a patch that fixes attached test. Marking as fixed.

[Bug 51719] _pclose() does not cause child stdin to report EOF

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=51719

--- Comment #2 from Andres Freund  ---
> I've sent a patch that fixes attached test. Marking as fixed.

Awesome. I can confirm that this doesn't just fix the testcase, but also allows
a cross-built postgres to at least get as far as creating the data directory
(i.e. initdb.exe succeeds).

[Bug 51719] _pclose() does not cause child stdin to report EOF

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=51719

Alexandre Julliard  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |CLOSED

--- Comment #3 from Alexandre Julliard  ---
Closing bugs fixed in 6.21.