Re: [2/3] server: Use tcdrain() instead of tcflush() to implement FlushFileBuffers() for a COM port.

[Alexandre Julliard]

Dmitry Timoshkov  writes:
> MSDN for FlushFileBuffers says:
> Flushes the buffers of a specified file and causes all buffered data to be
> written to a file.
>
> Linux man page says:
> tcdrain() waits until all output written to the object referred to by fd has
> been transmitted.

It's a blocking call, you can't do that on the server side.

[2/3] server: Use tcdrain() instead of tcflush() to implement FlushFileBuffers() for a COM port.

[Wolfgang Walter]

Hello Dmitry,

we had this problem, too. I sent a patch to wine-devel some time ago (I think 
2008). It had the same flaw as yours. I sent a different patch after Alexandre 
gave me the same answer. I then posted 2009 an different one. It did not make 
it into wine, though. I don't know why as there was no answer. Therefor I keep 
it since then in our own repository together with other serial fixes (i.e. to 
make most USB serial adapters working better).

Attached the patch (rebased on 1.6). We use it rebased on then the actual wine 
versions sincd 2009 and it works for us.

Am Montag, 26. August 2013, 10:56:29 schrieb Alexandre Julliard:
> Dmitry Timoshkov  writes:
> > MSDN for FlushFileBuffers says:
> > Flushes the buffers of a specified file and causes all buffered data to be
> > written to a file.
> >
> > Linux man page says:
> > tcdrain() waits until all output written to the object referred to by fd
> > has been transmitted.
>
> It's a blocking call, you can't do that on the server side.

Regards,

[2/3] server: Use tcdrain() instead of tcflush() to implement FlushFileBuffers() for a COM port.

[Dmitry Timoshkov]

Alexandre Julliard  wrote:
> It's a blocking call, you can't do that on the server side.

Can then something like in the patch from Wolfgang Walter be done
instead?

[2/3] server: Use tcdrain() instead of tcflush() to implement FlushFileBuffers() for a COM port.

[Alexandre Julliard]

Dmitry Timoshkov  writes:
> Alexandre Julliard  wrote:
> > It's a blocking call, you can't do that on the server side.
>
> Can then something like in the patch from Wolfgang Walter be done
> instead?

Something like that, yes.