perftools.rb 0.5.4 → 0.5.5

data/README.md

@@ -0,0 +1,177 @@
+# perftools.rb
+
+    google-perftools for ruby code
+    (c) 2010 Aman Gupta (tmm1)
+    http://www.ruby-lang.org/en/LICENSE.txt
+
+## Usage (in a webapp)
+
+  Use [rack-perftools_profiler](https://github.com/bhb/rack-perftools_profiler):
+
+    require 'rack/perftools_profiler'
+    config.middleware.use ::Rack::PerftoolsProfiler, :default_printer => 'gif'
+
+  Simply add `profile=true` to profile a request:
+
+    curl -o 10_requests_to_homepage.gif "http://localhost:3000/homepage?profile=true&times=10"
+
+## Usage (from Ruby)
+
+  Run the profiler with a block:
+
+    require 'perftools'
+    PerfTools::CpuProfiler.start("/tmp/add_numbers_profile") do
+      5_000_000.times{ 1+2+3+4+5 }
+    end
+
+  Start and stop the profiler manually:
+
+    require 'perftools'
+    PerfTools::CpuProfiler.start("/tmp/add_numbers_profile")
+    5_000_000.times{ 1+2+3+4+5 }
+    PerfTools::CpuProfiler.stop
+
+## Usage (externally)
+
+  Profile an existing ruby application without modifying it:
+
+    $ CPUPROFILE=/tmp/my_app_profile \
+      RUBYOPT="-r`gem which perftools | tail -1`" \
+      ruby my_app.rb
+
+## Profiler Modes
+
+The profiler can be run in one of many modes, set via an environment
+variable before the library is loaded:
+
+  * `CPUPROFILE_REALTIME=1`
+
+    Use walltime instead of cputime profiling. This will capture all time spent in a method, even if it does not involve the CPU.
+
+    For example, `sleep()` is not expensive in terms of cputime, but very expensive in walltime. walltime will also show functions spending a lot of time in network i/o.
+
+  * `CPUPROFILE_OBJECTS=1`
+
+    Profile object allocations instead of cpu/wall time. Each sample represents one object created inside that function.
+
+  * `CPUPROFILE_METHODS=1`
+
+    Profile method calls. Each sample represents one method call made inside that function.
+
+The sampling interval of the profiler can be adjusted to collect more
+(for better profile detail) or fewer samples (for lower overhead):
+
+  * `CPUPROFILE_FREQUENCY=500`
+
+    Default sampling interval is 100 times a second. Valid range is 1-4000
+
+## Reporting
+
+    pprof.rb --text /tmp/add_numbers_profile
+
+    pprof.rb --pdf /tmp/add_numbers_profile > /tmp/add_numbers_profile.pdf
+
+    pprof.rb --gif /tmp/add_numbers_profile > /tmp/add_numbers_profile.gif
+
+    pprof.rb --callgrind /tmp/add_numbers_profile > /tmp/add_numbers_profile.grind
+    kcachegrind /tmp/add_numbers_profile.grind
+
+    pprof.rb --gif --focus=Integer /tmp/add_numbers_profile > /tmp/add_numbers_custom.gif
+
+    pprof.rb --text --ignore=Gem /tmp/my_app_profile
+
+
+  For more options, see [pprof documentation](http://google-perftools.googlecode.com/svn/trunk/doc/cpuprofile.html#pprof)
+
+
+### Examples
+
+#### pprof.rb --text
+
+    Total: 1735 samples
+        1487  85.7%  85.7%     1735 100.0% Integer#times
+         248  14.3% 100.0%      248  14.3% Fixnum#+
+
+#### pprof.rb --gif
+
+  * Simple [require 'rubygems'](http://perftools-rb.rubyforge.org/examples/rubygems.gif) profile
+
+  * Comparing redis-rb [with](http://perftools-rb.rubyforge.org/examples/redis-rb.gif) and [without](http://perftools-rb.rubyforge.org/examples/redis-rb-notimeout.gif) SystemTimer based socket timeouts
+
+  * [Sinatra](http://perftools-rb.rubyforge.org/examples/sinatra.gif) vs. [Merb](http://perftools-rb.rubyforge.org/examples/merb.gif) vs. [Rails](http://perftools-rb.rubyforge.org/examples/rails.gif)
+
+  * C-level profile of EventMachine + epoll + Ruby threads [before](http://perftools-rb.rubyforge.org/examples/eventmachine-epoll+nothreads.gif) and [after](http://perftools-rb.rubyforge.org/examples/eventmachine-epoll+threads.gif) a [6 line EM bugfix](http://timetobleed.com/6-line-eventmachine-bugfix-2x-faster-gc-1300-requestssec/)
+
+  * C-level profile of a [ruby/rails vm](http://perftools-rb.rubyforge.org/examples/ruby_interpreter.gif)
+    * 12% time spent in re_match_exec because of excessive calls to rb_str_sub_bang by Date.parse
+
+
+## Installation
+
+  Just install the gem, which will download, patch and compile google-perftools for you:
+
+    sudo gem install perftools.rb
+
+  Or build your own gem:
+
+    git clone git://github.com/tmm1/perftools.rb
+    cd perftools.rb
+    gem build perftools.rb.gemspec
+    gem install perftools.rb
+
+
+  You'll also need graphviz to generate call graphs using dot:
+
+    sudo brew    install graphviz ghostscript # osx
+    sudo apt-get install graphviz ps2pdf      # debian/ubuntu
+
+## Advantages over ruby-prof
+
+* Sampling profiler
+
+  * perftools samples your process using setitimer() so it can be used in production with minimal overhead.
+
+
+## Profiling the Ruby VM and C extensions
+
+  To profile C code, download and build an unpatched perftools (libunwind or ./configure --enable-frame-pointers required on x86_64).
+
+  Download:
+
+    wget http://google-perftools.googlecode.com/files/google-perftools-1.6.tar.gz
+    tar zxvf google-perftools-1.6.tar.gz
+    cd google-perftools-1.6
+
+  Compile:
+
+    ./configure --prefix=/opt
+    make
+    sudo make install
+
+  Profile:
+
+    export LD_PRELOAD=/opt/lib/libprofiler.so                 # for linux
+    export DYLD_INSERT_LIBRARIES=/opt/lib/libprofiler.dylib   # for osx
+    CPUPROFILE=/tmp/ruby_interpreter.profile ruby -e' 5_000_000.times{ "hello world" } '
+
+  Report:
+
+    pprof `which ruby` --text /tmp/ruby_interpreter.profile
+
+
+## TODO
+
+  * Add support for heap profiling to find memory leaks (PerfTools::HeapProfiler)
+  * Allow both C and Ruby profiling
+  * Add setter for the sampling interval
+
+
+## Resources
+
+  * [GoRuCo 2009 Lightning Talk on perftools.rb](http://goruco2009.confreaks.com/30-may-2009-18-35-rejectconf-various-presenters.html) @ 21:52
+
+  * [Ilya Grigorik's introduction to perftools.rb](http://www.igvita.com/2009/06/13/profiling-ruby-with-googles-perftools/)
+
+  * [Google Perftools](http://code.google.com/p/google-perftools/)
+
+  * [Analyzing profiles and interpreting different output formats](http://google-perftools.googlecode.com/svn/trunk/doc/cpuprofile.html#pprof)

data/ext/extconf.rb

@@ -15,11 +15,12 @@ if RUBY_VERSION >= "1.9"
   begin
     require "ruby_core_source"
   rescue LoadError
-    STDERR.puts "\n\n"
-    STDERR.puts "***************************************************************************************"
-    STDERR.puts "******************** PLEASE RUN gem install ruby_core_source FIRST ********************"
-    STDERR.puts "***************************************************************************************"
-    exit(1)
+    require 'rubygems/dependency_installer'
+    installer = Gem::DependencyInstaller.new
+    installer.install 'ruby_core_source'
+    Gem.refresh
+
+    require "ruby_core_source"
   end
 end
 

data/perftools.rb.gemspec

@@ -1,7 +1,7 @@
 spec = Gem::Specification.new do |s|
   s.name = 'perftools.rb'
-  s.version = '0.5.4'
-  s.date = '2010-11-12'
+  s.version = '0.5.5'
+  s.date = '2011-01-23'
   s.rubyforge_project = 'perftools-rb'
   s.summary = 'google-perftools for ruby code'
   s.description = 'A sampling profiler for ruby code based on patches to google-perftools'

metadata

@@ -5,8 +5,8 @@ version: !ruby/object:Gem::Version
   segments: 
   - 0
   - 5
-  - 4
-  version: 0.5.4
+  - 5
+  version: 0.5.5
 platform: ruby
 authors: 
 - Aman Gupta
@@ -14,7 +14,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2010-11-12 00:00:00 -06:00
+date: 2011-01-23 00:00:00 -08:00
 default_executable: 
 dependencies: []
 
@@ -27,13 +27,11 @@ extensions:
 extra_rdoc_files: []
 
 files: 
-- README
+- README.md
 - bin/pprof.rb
 - ext/extconf.rb
 - ext/perftools.c
 - ext/src/google-perftools-1.6.tar.gz
-- objalloc_tests/Makefile
-- objalloc_tests/trap.c
 - patches/perftools-debug.patch
 - patches/perftools-frames.patch
 - patches/perftools-gc.patch

data/README

@@ -1,167 +0,0 @@
-google-perftools for ruby code
-  (c) 2009 Aman Gupta (tmm1)
-
-=== Usage
-
-  Run the profiler with a block:
-
-    require 'perftools'
-    PerfTools::CpuProfiler.start("/tmp/add_numbers_profile") do
-      5_000_000.times{ 1+2+3+4+5 }
-    end
-
-  Start and stop the profiler manually:
-
-    require 'perftools'
-    PerfTools::CpuProfiler.start("/tmp/add_numbers_profile")
-    5_000_000.times{ 1+2+3+4+5 }
-    PerfTools::CpuProfiler.stop
-
-  Profile an existing ruby application without modifying it:
-
-    $ CPUPROFILE=/tmp/my_app_profile RUBYOPT="-r`gem which perftools | tail -1`" ruby my_app.rb
-
-  Change the sampling interval (defaults to 100 times a second, valid range is 1-4000):
-
-    $ CPUPROFILE_FREQUENCY=500 ruby my_app.rb
-
-  Use walltime instead of cputime profiling:
-
-    $ CPUPROFILE_REALTIME=1 ruby my_app.rb
-
-  Profile object allocations instead of cpu/wall time:
-
-    $ CPUPROFILE_OBJECTS=1 ruby my_app.rb
-
-  Profile method calls:
-
-    $ CPUPROFILE_METHODS=1 ruby my_app.rb
-
-
-=== Reporting
-
-  pprof.rb --text /tmp/add_numbers_profile
-
-  pprof.rb --pdf /tmp/add_numbers_profile > /tmp/add_numbers_profile.pdf
-
-  pprof.rb --gif /tmp/add_numbers_profile > /tmp/add_numbers_profile.gif
-
-  pprof.rb --callgrind /tmp/add_numbers_profile > /tmp/add_numbers_profile.grind
-  kcachegrind /tmp/add_numbers_profile.grind
-
-  pprof.rb --gif --focus=Integer /tmp/add_numbers_profile > /tmp/add_numbers_custom.gif
-
-  pprof.rb --text --ignore=Gem /tmp/my_app_profile
-
-
-  For more options, see http://google-perftools.googlecode.com/svn/trunk/doc/cpuprofile.html#pprof
-
-
-=== Examples
-
-  pprof.rb --text
-
-    Total: 1735 samples
-        1487  85.7%  85.7%     1735 100.0% Integer#times
-         248  14.3% 100.0%      248  14.3% Fixnum#+
-
-  pprof.rb --gif
-
-    Simple require 'rubygems' profile
-
-      http://perftools-rb.rubyforge.org/examples/rubygems.gif
-
-    Comparing redis-rb with and without SystemTimer based socket timeouts
-
-      http://perftools-rb.rubyforge.org/examples/redis-rb.gif
-      http://perftools-rb.rubyforge.org/examples/redis-rb-notimeout.gif
-
-    Sinatra vs. Merb vs. Rails
-
-      http://perftools-rb.rubyforge.org/examples/sinatra.gif
-      http://perftools-rb.rubyforge.org/examples/merb.gif
-      http://perftools-rb.rubyforge.org/examples/rails.gif
-
-    C-level profile of EventMachine + epoll + Ruby threads
-      before http://timetobleed.com/6-line-eventmachine-bugfix-2x-faster-gc-1300-requestssec/
-
-      http://perftools-rb.rubyforge.org/examples/eventmachine-epoll+nothreads.gif
-      http://perftools-rb.rubyforge.org/examples/eventmachine-epoll+threads.gif
-
-    C-level profile of the ruby interpreter
-      12% time spent in re_match_exec because of excessive calls to rb_str_sub_bang by Date.parse
-      easily fixed by using the ThirdBase gem
-
-      http://perftools-rb.rubyforge.org/examples/ruby_interpreter.gif
-
-
-=== Installation
-
-  Just install the gem, which will download, patch and compile google-perftools for you:
-
-    sudo gem install perftools.rb
-
-  Or use the dev gem from github:
-
-    gem install -s http://gems.github.com tmm1-perftools.rb
-
-  Or build your own gem:
-
-    git clone git://github.com/tmm1/perftools.rb
-    cd perftools.rb
-    gem build perftools.rb.gemspec
-    gem install perftools.rb
-
-
-  You'll also need graphviz to generate call graphs using dot:
-
-    sudo port install graphviz     # osx
-    sudo apt-get install graphviz  # debian/ubuntu
-
-
-=== Advantages over ruby-prof
-
-  Sampling profiler
-
-    perftools samples your process using setitimer() so it can be used in production with minimal overhead.
-
-
-=== Profiling the Ruby VM and C extensions
-
-  To profile C code, download and build an unpatched perftools (libunwind or ./configure --enable-frame-pointers required on x86_64):
-
-    wget http://google-perftools.googlecode.com/files/google-perftools-1.3.tar.gz
-    tar zxvf google-perftools-1.3.tar.gz
-    cd google-perftools-1.3
-
-    ./configure --prefix=/opt
-    make
-    sudo make install
-
-    export LD_PRELOAD=/opt/lib/libprofiler.so                 # for linux
-    export DYLD_INSERT_LIBRARIES=/opt/lib/libprofiler.dylib   # for osx
-    CPUPROFILE=/tmp/ruby_interpreter.profile ruby -e' 5_000_000.times{ "hello world" } '
-
-    pprof `which ruby` --text /tmp/ruby_interpreter.profile
-
-
-=== TODO
-
-  * Add support for heap profiling to find memory leaks (PerfTools::HeapProfiler)
-  * Allow both C and Ruby profiling
-  * Add setter for the sampling interval
-
-
-=== Resources
-
-  GoRuCo 2009 Lightning Talk on perftools.rb
-    http://goruco2009.confreaks.com/30-may-2009-18-35-rejectconf-various-presenters.html at 21:52
-
-  Ilya Grigorik's introduction to perftools.rb
-    http://www.igvita.com/2009/06/13/profiling-ruby-with-googles-perftools/
-
-  Google Perftools
-    http://code.google.com/p/google-perftools/
-
-  Analyzing profiles and interpreting different output formats
-    http://google-perftools.googlecode.com/svn/trunk/doc/cpuprofile.html#pprof

data/objalloc_tests/Makefile

@@ -1,7 +0,0 @@
-default: trap
-
-trap: trap.c
-	gcc -o trap trap.c -ggdb -O0 -Wall
-
-clean:
-	rm -rf *.o trap *.dSYM

data/objalloc_tests/trap.c

@@ -1,103 +0,0 @@
-#define _GNU_SOURCE
-#define _XOPEN_SOURCE 600
-
-#include <assert.h>
-#include <ucontext.h>
-#include <unistd.h>
-#include <signal.h>
-#include <stdio.h>
-#include <stdlib.h>
-#include <string.h>
-#include <sys/mman.h>
-
-
-void
-func() {
-  printf("hi\n");
-}
-
-
-static inline void *
-page_align(void *addr) {
-  assert(addr != NULL);
-  return (void *)((size_t)addr & ~(0xFFFF));
-}
-
-static void
-copy_instructions(void *dest, void *src, size_t count) {
-  assert(dest != NULL);
-  assert(src != NULL);
-
-  void *aligned_addr = page_align(dest);
-  if (mprotect(aligned_addr, (dest - aligned_addr) + count, PROT_READ|PROT_WRITE|PROT_EXEC) != 0)
-    perror("mprotect");
-  memcpy(dest, src, count);
-}
-
-#define NUM_ORIG_BYTES 2
-struct {
-  void *location;
-  unsigned char value;
-} orig_bytes[NUM_ORIG_BYTES];
-
-static inline void**
-uc_get_ip(ucontext_t *uc) {
-  #if defined(__FreeBSD__)
-    return (void**)&uc->uc_mcontext.mc_rip;
-  #elif defined(__dietlibc__)
-    return (void**)&uc->uc_mcontext.rip;
-  #elif defined(__APPLE__)
-    return (void**)&uc->uc_mcontext->__ss.__rip;
-  #else
-    return (void**)&uc->uc_mcontext.gregs[REG_RIP];
-  #endif
-}
-
-static void
-trap_handler(int signal, siginfo_t *info, void *data) {
-  int i;
-  ucontext_t *uc = (ucontext_t *)data;
-  void **ip = uc_get_ip(uc);
-
-  // printf("signal: %d, addr: %p, ip: %p\n", signal, info->si_addr, *ip);
-
-  for (i=0; i<NUM_ORIG_BYTES; i++) {
-    if (orig_bytes[i].location == *ip-1) {
-      // restore original byte
-      copy_instructions(orig_bytes[i].location, &orig_bytes[i].value, 1);
-
-      // setup next breakpoint
-      copy_instructions(orig_bytes[(i+1)%NUM_ORIG_BYTES].location, "\xCC", 1);
-
-      // first breakpoint is the notification
-      if (i == 0)
-        printf(" ---> YOU'RE CALLING FUNC()\n");
-
-      // reset instruction pointer
-      *ip -= 1;
-
-      break;
-    }
-  }
-}
-
-int
-main() {
-  int i;
-  struct sigaction sig = { .sa_sigaction = trap_handler, .sa_flags = SA_SIGINFO };
-  sigemptyset(&sig.sa_mask);
-  sigaction(SIGTRAP, &sig, NULL);
-
-  for (i=0; i<NUM_ORIG_BYTES; i++) {
-    orig_bytes[i].location = func + i;
-    orig_bytes[i].value    = ((unsigned char*)func)[i];
-    copy_instructions(func + i, "\xCC", 1);
-  }
-
-  printf("func: %p\n", func);
-
-  for (i=0; i<10; i++)
-    func();
-
-  return 0;
-}