[PDF] Advanced C Programming - Profiling

View - Download Advanced C Programming - Profiling

Previous PDF

Next PDF

Advanced C Programming

Proling

Sebastian Hack

hack@cs.uni-sb.de

Christoph Weidenbach

weidenbach@mpi-inf.mpg.de

25.11.2008computer science

saar land u niversity1 Today

Proling

Invasive Proling

Non-Invasive Proling

Tools gprof gcov valgrind oprole

Conclusion

2

What is a Proler?

Analyse the runtime behavior of the program

I

Which parts (functions, statements, ...)

of a program take how long? I

How often are functions called?

I

Which functions call which

I

Construct the dynamic call graph

I

Memory consumption

I

Memory accesses

Imemory leaks

ICache performance

3

Invasive Proling

I

Modify the program (code instrumentation)

I

Insert calls to functions that record data

I

Advantages:

I

Very precise

ITheoretically at the instruction level

IPrecise call graph

I

Disadvantages:

I

Potentially very high overhead

IDepends on the instrumentation code that is inserted

ICannot prole already running systems

(long running servers)

ICan only prole application (not complete system)

4

Non-Invasive Proling

I

Statistic sampling of the program

I

Use a xed time interval

or

Ha rdwarep erformancecounters (CPU feature)

to trigger sampling events I

Record instruction pointer at each sampling event

I

Advantages:

I

Small overhead

IHardware assisted

ICan prole the whole system (even the kernel!)

I

Disadvantages:

I not precise+\only" statistical data

ICall Graph possibly not complete

+some functions are never sampled 5

Proles

I

Flat Prole

How much time does the program spend in which function? I

Call Graph

Which function calls which function how often?

I

Annotated Sources

Annotate each source line with number of executions 6 gprof I

Mixture of invasive and statistical proling

Invasive Part

I gccinserts calls to a functionmcountinto prologue of each function I

Compile with-gand-pg

I mcountcan gure out its caller+we can construct the call graph I mcountcounts the number of invocations for each function I

Call tomcountis the only instrumentation

+almost as ecient as normal build I After program is run, there is a le calledgmon.outcontaining proling data I Evaluate contents ofgmon.outwithgprof name-of-program 7 gprof

Statistical Part

I

Kernel samples instruction pointer (IP)

on each timer interrupt (100/s) I Increments a counter in a histogram of address ranges +cannot track the exact location where timer interrupt happened I Provides a frequency distribution over code locations I

Beware of low samplerate

I Short running programs will mostly not provide meaningful data I Accumulation of several prole runs is possible:$ ./test_program $ mv gmon.out gmon.sum $ ./test_program $ gprof -s ./test_program gmon.out gmon.sum 8 gcov I

Analyses coverage of program code

I

Which line was executed how often

I

Helps for nding code that

I can prot from optimizations

Ithat is notcovered b ytest cases

I

Use GCC

ags I -fprofile-arcs: collect info about jumps I-ftest-coverage: collect info about code coverage I

Attention:

Multiple co delines might b emerged to one instruction 100: 12:if (a != b)

100: 13: c = 1;

100: 14:else

100: 15: c = 0;

9 valgrind I

JIT-compiler / translator:

I Construct intermediate representation from x86 assembly code

IAdd instrumentation code

ICompile back to x86

I

Done while program is loaded

I

Is not only a proler!

I

No compiler

ags / recompilation needed (though-g -fno-inlineadvisable to analyse output) I Program runtime can degrade drastically due to instrumentation code and recompilation I can escape to debugger on certain events +very handy when debugging memory leaks I

Disadvantage:

I program might run an order of magnitude slower Iprogram might consume an order of magnitude more memory 10 valgrind Tools memcheck I

Redirects calls tomallocand the like

I

Keeps track of all allocated memory

I Instruments references to warn about \bad" memory accesses I uninitialized

Ialready freed

I

Detects memory leaks

I

Warns about jumps taken upon uninitialized values

cachegrind I

Instruments memory accesses

Iquotesdbs_dbs10.pdfusesText_16

[PDF] advanced c programming ppt

[PDF] advanced c# tutorial

[PDF] advanced c++ tutorial pdf

[PDF] advanced calculator app for android

[PDF] advanced cisco router configuration pdf

[PDF] advanced complex analysis pdf

[PDF] advanced computational methods in science and engineering pdf

[PDF] advanced concepts in java

[PDF] advanced css book

[PDF] advanced css3 tutorial pdf free download

[PDF] advanced dance moves ballet

[PDF] advanced db2 sql queries

[PDF] advanced dos commands pdf

[PDF] advanced english class pdf

[PDF] advanced english expressions list