Re: [fpc-devel] profiling under windows

2009-11-23 Thread Florian Klaempfl
Jonas Maebe schrieb: > The only thing that changed in r14137 was adding a prefetch statement > to tgnuassembler.writetree (on i386 you have to compile with > -Cppentium4 or higher for the prefetch statement to do anything > though). Actually pentium3 ;) ___

Re: [fpc-devel] profiling under windows

2009-11-22 Thread Jonas Maebe
On 22 Nov 2009, at 15:31, Paul Ishenin wrote: > Jonas Maebe wrote: >> On 20 Nov 2009, at 18:08, Paul Ishenin wrote: >> >> Did you benchmark whether this changed code is indeed faster and whether it >> makes any difference at all for the compiler? Changing working code because >> it /might/ bec

Re: [fpc-devel] profiling under windows

2009-11-22 Thread Paul Ishenin
Jonas Maebe wrote: On 20 Nov 2009, at 18:08, Paul Ishenin wrote: I have attached compiler gprof output of compiler itself. Using output I was able to find and improve(?) 3 functions I found in output. Did you benchmark whether this changed code is indeed faster and whether it makes

Re: [fpc-devel] profiling under windows

2009-11-21 Thread Jonas Maebe
On 20 Nov 2009, at 18:08, Paul Ishenin wrote: > I have attached compiler gprof output of compiler itself. Using output I was > able to find and improve(?) 3 functions I found in output. Did you benchmark whether this changed code is indeed faster and whether it makes any difference at all for

Re: [fpc-devel] profiling under windows

2009-11-20 Thread Jonas Maebe
On 20 Nov 2009, at 19:03, Sergei Gorelkin wrote: > I did, but using Linux+valgrind rather than cygwin+gprof. IMHO valgrind (in > its callgrind flavour) outputs more useful profile information. > Some time ago I was able to optimize away about 20% of executed CPU > instructions in the compiler,

Re[2]: [fpc-devel] profiling under windows

2009-11-20 Thread JoshyFun
Hello Paul, Friday, November 20, 2009, 6:08:28 PM, you wrote: PI> What I can't understand why the biggest %time has the lowest seconds. E.g: PI> % cumulative self self total PI> time seconds secondscalls Ts/call Ts/call name PI> 5.11 0.36

Re: [fpc-devel] profiling under windows

2009-11-20 Thread Sergei Gorelkin
Paul Ishenin пишет: What I can't understand why the biggest %time has the lowest seconds. E.g: The second column is a cumulative time, i.e. sum of the time for the given function and everything above it. As result I don't understand which functions must be reviewed at first. Have anybody

Re[2]: [fpc-devel] profiling under windows

2009-11-20 Thread JoshyFun
Hello Vincent, Friday, November 20, 2009, 8:30:49 AM, you wrote: >> pp.pas(224,1) Error: Import library not found for c >> pp.pas(224,1) Error: Import library not found for gmon >> pp.pas(224,1) Error: Import library not found for cygwin >> pp.pas(224,1) Error: Import library not found for user32

Re: [fpc-devel] profiling under windows

2009-11-19 Thread Vincent Snijders
Paul Ishenin schreef: Hello, FPC developers' list. I tried to build the compiler with gprof profile code. In the compiler dir I executed: make clean all OPT="-gl -pg" As result of linking I have the next errors: Linking .\pp.exe pp.pas(224,1) Error: Import library not found for c pp.pas(224

[fpc-devel] profiling under windows

2009-11-19 Thread Paul Ishenin
Hello, FPC developers' list. I tried to build the compiler with gprof profile code. In the compiler dir I executed: make clean all OPT="-gl -pg" As result of linking I have the next errors: Linking .\pp.exe pp.pas(224,1) Error: Import library not found for c pp.pas(224,1) Error: Import libra