Date: Wed, 1 Nov 2000 10:27:36 +0200 (IST) From: Eli Zaretskii X-Sender: eliz AT is To: djgpp AT delorie DOT com Subject: Re: Elapsed time? In-Reply-To: <8tnru8$r4t$1@nnrp1.deja.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII 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 On Wed, 1 Nov 2000 dcasale AT my-deja DOT com wrote: > I think I know where the slowdown is coming from now. I tried > temporarily turning off the file compression and encryption to make > sure my program handled it correctly, with the new buffering code > changes, and the slowdown disappeared entirely. Somehow, file > buffering has caused what used to be about a 10% slowdown to balloon to > a 50% slowdown, but it's coming from inside the compression code. Fascinating! Never heard of code that slowed down the system timer, by just crunching some CPU-intensive computations. > What's annoying is that this code makes _no special interrupt calls_ at > all. It's almost pure C code, in fact, with an interface wrapper class > for the program to call. I suggest to strip the offending code to the bare minimum that still exhibits the problem, and post it here (as ASCII text, not as binary attachment). If you cannot post, or if it's very large, please describe in some pseudo-language the algorithm that the code uses.