Library order is important

I've written quite extensively about link ordering issues, but I've not discussed the interaction between archive libraries and shared libraries. So let's take a simple program that calls a maths library function:

#include <math.h>

int main()
{
  for (int i=0; i<10000000; i++)
  {
    sin(i);
  }
}

We compile and run it to get the following performance:

bash-3.2$ cc -g -O fp.c -lm
bash-3.2$ timex ./a.out

real           6.06
user           6.04
sys            0.01

Now most people will have heard of the optimised maths library which is added by the flag -xlibmopt. This contains optimised versions of key mathematical functions, in this instance, using the library doubles performance:

bash-3.2$ cc -g -O -xlibmopt fp.c -lm
bash-3.2$ timex ./a.out

real           2.70
user           2.69
sys            0.00

The optimised maths library is provided as an archive library (libmopt.a), and the driver adds it to the link line just before the maths library - this causes the linker to pick the definitions provided by the static library in preference to those provided by libm. We can see the processing by asking the compiler to print out the link line:

bash-3.2$ cc -### -g -O -xlibmopt fp.c -lm
/usr/ccs/bin/ld ... fp.o -lmopt -lm -o a.out...

The flag to the linker is -lmopt, and this is placed before the -lm flag. So what happens when the -lm flag is in the wrong place on the command line:

bash-3.2$ cc -g -O -xlibmopt -lm fp.c
bash-3.2$ timex ./a.out

real           6.02
user           6.01
sys            0.01

If the -lm flag is before the source file (or object file for that matter), we get the slower performance from the system maths library. Why's that? If we look at the link line we can see the following ordering:

/usr/ccs/bin/ld ... -lmopt -lm fp.o -o a.out 

So the optimised maths library is still placed before the system maths library, but the object file is placed afterwards. This would be ok if the optimised maths library were a shared library, but it is not - instead it's an archive library, and archive library processing is different - as described in the linker and library guide:

"The link-editor searches an archive only to resolve undefined or tentative external references that have previously been encountered."

An archive library can only be used resolve symbols that are outstanding at that point in the link processing. When fp.o is placed before the libmopt.a archive library, then the linker has an unresolved symbol defined in fp.o, and it will search the archive library to resolve that symbol. If the archive library is placed before fp.o then there are no unresolved symbols at that point, and so the linker doesn't need to use the archive library. This is why libmopt needs to be placed after the object files on the link line.

On the other hand if the linker has observed any shared libraries, then at any point these are checked for any unresolved symbols. The consequence of this is that once the linker "sees" libm it will resolve any symbols it can to that library, and it will not check the archive library to resolve them. This is why libmopt needs to be placed before libm on the link line.

This leads to the following order for placing files on the link line:

• Object files
• Archive libraries
• Shared libraries

If you use this order, then things will consistently get resolved to the archive libraries rather than to the shared libaries.

It could be worse....

As "guest" pointed out, in my file I/O test I didn't open the file with O_SYNC, so in fact the time was spent in OS code rather than in disk I/O. It's a straightforward change to add O_SYNC to the open() call, but it's also useful to reduce the iteration count - since the cost per write is much higher:

...
#define SIZE 1024

void test_write()
{
  starttime();
  int file = open("./test.dat",O_WRONLY|O_CREAT|O_SYNC,S_IWGRP|S_IWOTH|S_IWUSR);
...

Running this gave the following results:

Time per iteration   0.000065606310 MB/s
Time per iteration   2.709711563906 MB/s
Time per iteration   0.178590114758 MB/s

Yup, disk I/O is way slower than the original I/O calls. However, it's not a very fair comparison since disks get written in large blocks of data and we're deliberately sending a single byte. A fairer result would be to look at the I/O operations per second; which is about 65 - pretty much what I'd expect for this system.

It's also interesting to examine at the profiles for the two cases. When the write() was trapping into the OS the profile indicated that all the time was being spent in system. When the data was being written to disk, the time got attributed to sleep. This gives us an indication how to interpret profiles from apps doing I/O. It's the sleep time that indicates disk activity.

Write and fprintf for file I/O

fprintf() does buffered I/O, where as write() does unbuffered I/O. So once the write() completes, the data is in the file, whereas, for fprintf() it may take a while for the file to get updated to reflect the output. This results in a significant performance difference - the write works at disk speed. The following is a program to test this:

#include <fcntl.h>
#include <unistd.h>
#include <stdio.h>
#include <stdlib.h>
#include <errno.h>
#include <stdio.h>
#include <sys/time.h>
#include <sys/types.h>
#include <sys/stat.h>

static double s_time;

void starttime()
{
  s_time=1.0*gethrtime();
}

void endtime(long its)
{
  double e_time=1.0*gethrtime();
  printf("Time per iteration %5.2f MB/s\n", (1.0*its)/(e_time-s_time*1.0)*1000);
  s_time=1.0*gethrtime();
}

#define SIZE 10*1024*1024

void test_write()
{
  starttime();
  int file = open("./test.dat",O_WRONLY|O_CREAT,S_IWGRP|S_IWOTH|S_IWUSR);
  for (int i=0; i<SIZE; i++)
  {
    write(file,"a",1);
  }
  close(file);
  endtime(SIZE);
}

void test_fprintf()
{
  starttime();
  FILE* file = fopen("./test.dat","w");
  for (int i=0; i<SIZE; i++)
  {
    fprintf(file,"a");
  }
  fclose(file);
  endtime(SIZE);
}

void test_flush()
{
  starttime();
  FILE* file = fopen("./test.dat","w");
  for (int i=0; i<SIZE; i++)
  {
    fprintf(file,"a");
    fflush(file);
  }
  fclose(file);
  endtime(SIZE);
}


int main()
{
  test_write();
  test_fprintf();
  test_flush();
}

Compiling and running I get 0.2MB/s for write() and 6MB/s for fprintf(). A large difference. There's three tests in this example, the third test uses fprintf() and fflush(). This is equivalent to write() both in performance and in functionality. Which leads to the suggestion that fprintf() (and other buffering I/O functions) are the fastest way of writing to files, and that fflush() should be used to enforce synchronisation of the file contents.

Mixing Java and native code

This was a bit of surprise to me. The slides are available from my presentation at JavaOne on mixed language development. What I wasn't expecting was that there would also be a video of the presentation.

Maximising SPARC T4 performance

My presentation from Oracle Open World is available for download.

25 years of SPARC

Looks like an interesting event at the Computer History Museum on 1 November. A panel discussing SPARC at 25: Past, Present and Future. Free sign up.

Current SPARC Architectures

Different generations of SPARC processors implement different architectures. The architecture that the compiler targets is controlled implicitly by the -xtarget flag and explicitly by the -arch flag.

If an application targets a recent architecture, then the compiler gets to play with all the instructions that the new architecture provides. The downside is that the application won't work on older processors that don't have the new instructions. So for developer's there is a trade-off between performance and portability.

The way we have solved this in the compiler is to assume a "generic" architecture, and we've made this the default behaviour of the compiler. The only flag that doesn't make this assumption is -fast which tells the compiler to assume that the build machine is also the deployment machine - so the compiler can use all the instructions that the build machine provides.

The -xtarget=generic flag tells the compiler explicitly to use this generic model. We work hard on making generic code work well across all processors. So in most cases this is a very good choice.

It is also of interest to know what processors support the various architectures. The following Venn diagram attempts to show this:

A textual description is as follows:

• The T1 and T2 processors, in addition to most other SPARC processors that were shipped in the last 10+ years supported V9b, or sparcvis2.
• The SPARC64 processors from Fujitsu, used in the M-series machines, added support for the floating point multiply accumulate instruction in the sparcfmaf architecture.
• Support for this instruction also appeared in the T3 - this is called sparcvis3
• Later SPARC64 processors added the integer multiply accumulate instruction, this architecture is sparcima.
• Finally the T4 includes support for both the integer and floating point multiply accumulate instructions in the sparc4 architecture.

So the conclusion should be:

• Floating point multiply accumulate is supported in both the T-series and M-series machines, so it should be a relatively safe bet to start using it.
• The T4 is a very good machine to deploy to because it supports all the current instruction sets.