[ace-users] Help ! Porting ACE Proactor Aio application on Linux

[h w]

I found it.

Thanks

On Fri, Mar 4, 2016 at 1:08 PM, Steve Huston <shuston at riverace.com> wrote:

> I recommend using Google to look for it. It’s been at least 10 years since
> I saw it.
>
>
>
> *From:* h w [mailto:hw77377 at gmail.com]
> *Sent:* Friday, March 04, 2016 1:00 PM
> *To:* Steve Huston <shuston at riverace.com>
> *Cc:* ace-users at list.isis.vanderbilt.edu
> *Subject:* Re: [ace-users] Help ! Porting ACE Proactor Aio application on
> Linux
>
>
>
> Thank you very much for your help.
>
>
>
> If someone did to emulate aio using reactive mechanisms, is it possible
> to find this code/doc as an example/reference?
>
>
>
> Thanks,
>
>
>
>
>
> This email has been sent from a virus-free computer protected by Avast.
> www.avast.com
> <https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>
>
>
>
> On Wed, Mar 2, 2016 at 5:30 PM, Steve Huston <shuston at riverace.com> wrote:
>
> Thank you for the PROBLEM-REPORT-FORM.
>
>
>
> I looked for a bit and found no new developments in Linux kernel aio
> socket support, so the situation has apparently not changed.
>
>
>
> Your choices are:
>
> 1. Redesign your application to alternate reads and writes; some protocols
> won’t support this well, but if you can it’s the easiest way.
>
> 2. Change to using Reactor (particularly, ACE_Dev_Poll_Reactor) on Linux,
> at least.
>
> 3. Write something to emulate aio using reactive mechanisms. There was an
> effort that did this in ACE quite a few years ago – maybe you can get your
> hands on that and see if it could work for you.
>
>
>
> -Steve
>
>
>
> *From:* ace-users [mailto:ace-users-bounces at list.isis.vanderbilt.edu] *On
> Behalf Of *h w
> *Sent:* Tuesday, March 01, 2016 9:45 AM
> *To:* ace-users at list.isis.vanderbilt.edu
> *Subject:* [ace-users] Help ! Porting ACE Proactor Aio application on
> Linux
>
>
>
> ACE VERSION: 6.3.0
>
>     HOST MACHINE and OPERATING SYSTEM:
>          Centos 6.7   and Linux kernel 2.6.32-573.12.1.el6.i686
>
>
>
>
>     THE $ACE_ROOT/ace/config.h
>
>            #define ACE_LACKS_GETIPNODEBYADDR
>            #define ACE_LACKS_GETIPNODEBYNAME
>
>            #define ACE_HAS_AIO_CALLS
>            #include "ace/config-linux.h"
>
>
>
>
>     THE $ACE_ROOT/include/makeinclude/platform_macros.GNU
>
>             include $(ACE_ROOT)/include/makeinclude/platform_linux.GNU
>
>
>
>
>     CONTENTS OF $ACE_ROOT/bin/MakeProjectCreator/config/default.features
>             no such file
>
>     AREA/CLASS/EXAMPLE AFFECTED:  no
>
>
>     DOES THE PROBLEM AFFECT:
>         COMPILATION?      no
>         LINKING?     no
>             On Unix systems, did you run make realclean first?   Yes, I
> did.
>
>
>
>     SYNOPSIS:
>         Using ACE Proactor AIO framework on network sockets to exchange
> messages between two or more endpoints on  linux OS.  It looked the message
> read or write blocked or in order.
>
>
>     DESCRIPTION:
>
>          I am porting an application into Linux which utilizes ACE
> proactor AIO framework for a network communication ( read/ write socket ).
> It works fine on Windows OS. However, on Linux, The application looked like
> exchange messages in order,  for example, When I tried to send a message to
> the peer, I found that ACE_POSIX_AIOCB_Proactor did invoke aio_write() but
> the message did not get sent out util the peer sent a message out, and then
> got that previous message in. More worse, most time, when the connector
> estalished a connection and tried to send the first message, the system got
> locked up because a read operation probably be put before write
> operation for listening any incoming messages.
>
>
>
>          I searched the issue and found the similar issue report in the
> link but it was several years age:
> https://stevehuston.wordpress.com/2008/11/25/when-is-it-ok-to-use-ace-proactor-on-linux/
>
>
>
> Is it  still the limitation on Linux?  Is there any way to get the
> ACE Proactor AIO work for network communication ( exchange message via
> socket with full duplex) now?  Any help would be greatly appreciated
>
>
>
>
> Thanks!
>
>
>
> This email has been sent from a virus-free computer protected by Avast.
> www.avast.com
> <https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>
>
>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://list.isis.vanderbilt.edu/pipermail/ace-users/attachments/20160307/9d2b9cea/attachment.html>