X-Authentication-Warning: delorie.com: mail set sender to djgpp-bounces using -f From: "Gundolf" Newsgroups: comp.os.msdos.djgpp Subject: Re: Watt32 with DJGPP - read performance is extreamly poor Date: 21 Jul 2005 12:16:18 -0700 Organization: http://groups.google.com Lines: 14 Message-ID: <1121973378.642936.268920@g14g2000cwa.googlegroups.com> References: <1121737347 DOT 351585 DOT 124950 AT g43g2000cwa DOT googlegroups DOT com> NNTP-Posting-Host: 213.198.84.198 Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" X-Trace: posting.google.com 1121973384 19397 127.0.0.1 (21 Jul 2005 19:16:24 GMT) X-Complaints-To: groups-abuse AT google DOT com NNTP-Posting-Date: Thu, 21 Jul 2005 19:16:24 +0000 (UTC) In-Reply-To: <1121737347.351585.124950@g43g2000cwa.googlegroups.com> User-Agent: G2/0.2 Complaints-To: groups-abuse AT google DOT com Injection-Info: g14g2000cwa.googlegroups.com; posting-host=213.198.84.198; posting-account=rY35YA0AAAAyeF5kjpy8hQOSdJiziKje To: djgpp AT delorie DOT com DJ-Gateway: from newsgroup comp.os.msdos.djgpp Reply-To: djgpp AT delorie DOT com Errors-To: nobody AT delorie DOT com X-Mailing-List: djgpp AT delorie DOT com X-Unsubscribes-To: listserv AT delorie DOT com Precedence: bulk How (often) is OGSocket::recv(...) being called - from a busy loop? What about that _ip_delay1(m_pSock,sock_delay) ... sounds kind of suspicious - anything with big delays happening in there? Not sure about more recent versions of wattcp, but it used to be that the main tcp worker 'process' tcp_tick() was running on the good old 18.18 Hz / 55 msec dos clock. You could try calling tcp_tick() [not quite sutre that was the exact name] before recv() to be sure all 'raw' received data has been processed and is available to fastread(). Just a bunch of hunches ;) Gundolf