From: nik@tiuk.ti.com (Nick Ing-Simmons) Subject: Re: Function times still broken? 16 Dec 1996 07:06:55 -0800 Sender: daemon@cygnus.com Approved: cygnus.gnu-win32@cygnus.com Distribution: cygnus Message-ID: <199612161225.MAA09513.cygnus.gnu-win32@phoenix> References: <199612160649.XAA10885@joplin.colorado.edu> Original-To: fabio@joplin.colorado.edu Original-Cc: gnu-win32@cygnus.com In-Reply-To: <199612160649.XAA10885@joplin.colorado.edu> from Fabio Somenzi on Sun, 15 Dec 1996 23: 49:39 -0700 Original-Sender: owner-gnu-win32@cygnus.com Fabio Somenzi writes: >When I run the following program under b17.1 on my Win95 laptop >---------------------------------------------------------------- >#include >#include >#include > >int >main() >{ > clock_t time; > struct tms buffer; > > time = times(&buffer); > > printf("buffer = %lu\n", (unsigned long) buffer.tms_utime); > printf(" = %lu\n", (unsigned long) buffer.tms_stime); > printf(" = %lu\n", (unsigned long) buffer.tms_cutime); > printf(" = %lu\n", (unsigned long) buffer.tms_cstime); > printf("time = %lu\n", (unsigned long) time); > > exit(0); > >} >---------------------------------------------------------------- >I get results like this one: > >buffer = 560564963 > = 551978954 > = 0 > = 0 >time = 47919604 > >The first two numbers should give the user and system CPU time of the >process, but they rather seem random. Successive runs of the program >produce increasinv values of "time" (as expected) but identical values >of "buffer" (suspect). As each "run" is a new "process" the buffer values should always be the same (give or take resolution). That said values look weird. - For help on using this list, send a message to "gnu-win32-request@cygnus.com" with one line of text: "help".