[Bug 10648] New: Async WSASend does not send all bytes.

http://bugs.winehq.org/show_bug.cgi?id=10648

Summary: Async WSASend does not send all bytes.
           Product: Wine
           Version: 0.9.5.
          Platform: Other
        OS/Version: Linux
            Status: UNCONFIRMED
          Severity: enhancement
          Priority: P2
         Component: test
        AssignedTo: [e-mail]
        ReportedBy: [e-mail]

On Server sending 1500 bytes with non-blocked WSASend and 2 WSABUFFERS will
transfer only 1452 bytes. On client WSARecv called again will be blocked and
does not receive the the remaining bytes.

The completion port receives event on transfered 1452 bytes, and there is no
more followed events on transfered the last 8 bytes.

OS: Ubuntu 7.10

[Bug 10648] Async WSASend does not send all bytes.

http://bugs.winehq.org/show_bug.cgi?id=10648

Austin English  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
          Component|testcases                   |-unknown

--- Comment #1 from Austin English   2008-04-28 16:29:05 ---
Can you upload a small test case (source and binary) showing this problem?

[Bug 10648] Async WSASend does not send all bytes.

http://bugs.winehq.org/show_bug.cgi?id=10648

--- Comment #2 from Andrew Nguyen   2008-11-02 20:44:29 ---
This bug seems to be abandoned.

[Bug 10648] Async WSASend does not send all bytes.

http://bugs.winehq.org/show_bug.cgi?id=10648

Austin English  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |RESOLVED
         Resolution|                            |ABANDONED

--- Comment #3 from Austin English   2008-11-02 21:47:58 ---
Abandoned.

[Bug 10648] Async WSASend does not send all bytes.

http://bugs.winehq.org/show_bug.cgi?id=10648

Austin English  changed:

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

--- Comment #4 from Austin English   2008-11-02 21:48:04 ---
Closing.

[Bug 10648] Async WSASend does not send all bytes.

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=10648

Jinoh Kang  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |[e-mail]

--- Comment #5 from Jinoh Kang  ---
Requesting to reopen the ticket.

The gRPC library relies on the all-or-none behaviour of nonblocking send() on
Winsock.  All applications relying on gRPC thus fails to transmit RPC packets
properly.

It's possible to patch gRPC instead [1] so that it properly handles partial
writes, but it is not always feasible in proprietary software applications.

Possible directions for the fix include:

- In sock_send(), test if it's possible to actually send the full stream data
before calling send().
- On partial write, queue up the rest of the buffer and report success.

[1] https://github.com/grpc/grpc/pull/28432/files

[Bug 10648] gRPC library fails to send RPC packets correctly (nonblocking send() should not perform partial writes)

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=10648

Zebediah Figura  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|ABANDONED                   |---
     Ever confirmed|0                           |1
                 CC|                            |[e-mail]
             Status|CLOSED                      |REOPENED
          Component|-unknown                    |winsock
            Summary|Async WSASend does not send |gRPC library fails to send
                   |all bytes.                  |RPC packets correctly
                   |                            |(nonblocking send() should
                   |                            |not perform partial writes)

--- Comment #6 from Zebediah Figura  ---
(In reply to Jinoh Kang from comment #5)
> Requesting to reopen the ticket.
>
> The gRPC library relies on the all-or-none behaviour of nonblocking send()
> on Winsock.  All applications relying on gRPC thus fails to transmit RPC
> packets properly.
>
> It's possible to patch gRPC instead [1] so that it properly handles partial
> writes, but it is not always feasible in proprietary software applications.

I've reopened, but fixing applications is preferable if at all possible. It's
not clear that any other applications actually care about short write
behaviour.
> Possible directions for the fix include:
> - In sock_send(), test if it's possible to actually send the full stream
>
> data before calling send().

How?
> - On partial write, queue up the rest of the buffer and report success.

This seems to me like the only option feasible right now, although it's
obviously far from ideal.

Of course, we could also try to introduce an extra socket option or flag into
the kernel.

[Bug 10648] gRPC library fails to send RPC packets correctly (nonblocking send() should not perform partial writes)

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=10648

--- Comment #7 from Jinoh Kang  ---
(In reply to Zebediah Figura from comment #6)
> (In reply to Jinoh Kang from comment #5)
> > Requesting to reopen the ticket.
> >
> > The gRPC library relies on the all-or-none behaviour of nonblocking send()
> > on Winsock.  All applications relying on gRPC thus fails to transmit RPC
> > packets properly.
> >
> > It's possible to patch gRPC instead [1] so that it properly handles partial
> > writes, but it is not always feasible in proprietary software applications.
>
> I've reopened, but fixing applications is preferable if at all possible.
> It's not clear that any other applications actually care about short write
> behaviour.
> > Possible directions for the fix include:
> > - In sock_send(), test if it's possible to actually send the full stream
> >
> > data before calling send().
>
> How?

Heuristics.  Actually it would require *both* remedies in order for the final
solution to be plausible at all.
> > - On partial write, queue up the rest of the buffer and report success.
>
> This seems to me like the only option feasible right now, although it's
> obviously far from ideal.

Combined with heuristics above, I think this won't happen that much frequently
in practice.
> Of course, we could also try to introduce an extra socket option or flag
> into the kernel.

Linux maybe, but Darwin...

[Bug 10648] gRPC library fails to send RPC packets correctly (nonblocking send() should not perform partial writes)

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=10648

--- Comment #8 from Zebediah Figura  ---
(In reply to Jinoh Kang from comment #7)
> (In reply to Zebediah Figura from comment #6)
> > (In reply to Jinoh Kang from comment #5)
> > > - In sock_send(), test if it's possible to actually send the full stream
> > >
> > > data before calling send().
> >
> > How?
>
> Heuristics.  Actually it would require *both* remedies in order for the
> final solution to be plausible at all.

What heuristics do you have in mind, exactly?

[Bug 10648] gRPC library fails to send RPC packets correctly (nonblocking send() should not perform partial writes)

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=10648

--- Comment #9 from Jinoh Kang  ---
A recently posted MR,
https://gitlab.winehq.org/wine/wine/-/merge_requests/6490, should fix this bug.
The MR is for "JOY OF PROGRAMMING - Software Engineering Simulator."

[Bug 10648] gRPC library fails to send RPC packets correctly (nonblocking send() should not perform partial writes)

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=10648

--- Comment #10 from Jinoh Kang  ---
This should be fixed in 535c8a84c1ec4391786d8be6f87d683e20223b12.

[Bug 10648] gRPC library fails to send RPC packets correctly (nonblocking send() should not perform partial writes)

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=10648

Paul Gofman  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
    Regression SHA1|                            |535c8a84c1ec4391786d8be6f87
                   |                            |d683e20223b12
                 CC|                            |[e-mail]
         Resolution|---                         |FIXED
             Status|REOPENED                    |RESOLVED

--- Comment #11 from Paul Gofman  ---
Marking resolved.

[Bug 10648] gRPC library fails to send RPC packets correctly (nonblocking send() should not perform partial writes)

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=10648

Paul Gofman  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
    Regression SHA1|535c8a84c1ec4391786d8be6f87 |
                   |d683e20223b12               |
      Fixed by SHA1|                            |535c8a84c1ec4391786d8be6f87
                   |                            |d683e20223b12

[Bug 10648] gRPC library fails to send RPC packets correctly (nonblocking send() should not perform partial writes)

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=10648

Alexandre Julliard  changed:

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

--- Comment #12 from Alexandre Julliard  ---
Closing bugs fixed in 9.18.