Technical Support for SPECaccel

(To check for possible updates to this document, please see http://www.spec.org/accel2023/Docs/ )

Contents

1. Introduction

2. Before contacting SPEC

3. Contact Information

4. Limitations on support

5. Useful information to include when contacting SPEC

6. Benchmark source code change requests

Note: links to SPECaccel documents on this web page assume that you are reading the page from a directory that also contains the other SPECaccel documents. If by some chance you are reading this web page from a location where the links do not work, try accessing the referenced documents at one of the following locations:

• www.spec.org/accel2023/Docs/
• The $SPEC/Docs/ (Unix)
• The Docs/ directory on your SPECaccel distribution media.

1. Introduction

Reminder: SPECaccel is a source code benchmark. Using SPECaccel requires access to a computer system running a supported operating system with the necessary compilers and tools needed for generating executable binaries for the system. It is assumed that the person using SPECaccel is comfortable using compiler commands as well as other commands via a shell (aka "command window", "terminal window", "console", "terminal emulator", etc.)

If you are not familiar with the terms above, it is suggested that you research those topics. One useful reference can be found in the file readme1st.html.

SPEC can provide limited technical support for:

• Installing SPECaccel on a UNIX/Linux system.
• Building the SPECaccel tools.
• Using the SPECaccel tools to run the benchmarks.
• Updates for known issues.

2. Before contacting SPEC

Before contacting SPEC for support, it is suggested that you check the SPECaccel benchmark web site: http://www.spec.org/accel2023

For a list of questions that are frequently asked by users of the benchmark suite. Also, please see the section "Troubleshooting" in config.html

3. Contact Information

SPECaccel technical support can be reached by sending email to: [email protected]

Additional/updated contact information can also be found at SPEC's primary Web Site: www.spec.org

4. Limitations on support

SPEC can provide only limited advice regarding:

• Miscompares. If your system gets an answer that the tools consider to be incorrect, the usual advice is try asking your optimizer to be less aggressive. If you do report a miscompare, and if we recognize it, we can tell you so. But if the miscompare requires digging through the entrails, you'll probably have to use a debugger, and you may have to report a bug to your compiler vendor.

• New platforms. The list of supported platforms is in system-requirements.html. If you install on a platform that does not have tools, you'll find guidance in the document tools-build.html about what to do. If you report problems with the build, we might be able to help you; but usually the problem is something that needs to be taken apart on the system where the build is being attempted - i.e. your system, not a system at SPEC. You will find troubleshooting suggestions in tools-build.html.

In general, SPEC can provide technical assistance with its benchmarks, but SPEC does not have the ability to provide support on other vendor's products. Examples of support that SPEC can not provide include:

• Hardware installation issues.
• Hardware problems.
• Non-SPEC software installation issues.
• Non-SPEC software problems.
• Hardware specific tuning.

If known, SPEC will provide a pointer to the appropriate source of information. Otherwise for these and similar issues, it is suggested that the hardware or software vendor be contacted directly.

5. Useful information to include when contacting SPEC

The following information should be provided to SPEC when requesting help with an issue:

1. The revision of the benchmark suite that you're using - please say "runspec -V"
2. The specific nature of the problem and the symptoms. Please give the exact error message. If the message says "see file x/y/z.err", then please include the contents of z.err.
3. Type of hardware/name of system being used.
4. Name and version of the operating system.
5. Name and version of the compiler.
6. If a compilation issue:
   • list of flags being used
   • check if problems exist at lowest level of optimization.
7. Has this issue been tried or seen on other systems?

Additional details related to the problem are also appreciated.

6. Benchmark source code change requests

Although the SPECaccel suite underwent extensive portability testing prior to release, it is possible that new systems, or changes in language standards, might cause there to be a need for source code modifications. Under the run rules, no publication of results may be done using modified source code, unless SPEC has accepted the change. This section describes how to go about creating a request for approval of a source change.

By including all the relevant information, you improve the chances that SPEC will accept your request. Nevertheless, please note that SPEC is not obliged to accept your request. SPEC will consider a request for a source change in a similar manner to how it considers portability flags, weighing aspects such as performance neutrality, amount of code affected, and impact on the original intent of the program.

Proposals for changes to benchmark source code should include the information mentioned in the previous section, plus:

8. A "context diff", typically via diff -u, or, if that is unavailable, diff -C. Specify the original file FIRST. If the changes are extensive, or if you do not have access to a decent diff utility, an alternative would be to attach the files. You can package them up and compress them by saying something like this: spectar cvf - newsrc/* | specxz > proposed.tar.xz
9. A clear statement of what happens without the proposed change.
10. A clear statement of whether a workaround exists. Are there any conditions under which the benchmark can be made to work without the proposed change?
11. If at all possible, a specific reference to the relevant portion of the compiler language standard.
12. If at all possible, results of testing for both correctness and performance when the same change is applied on some hw/sw platform other than the one where the problem was observed.
13. If the problem has been reported to the compiler vendor, a reference to a specific case number, mail archive message number, web forum address, or compiler developer email address.
14. If the problem has not been reported to the compiler vendor, a brief statement as to why not.
15. If the problem needs to be solved by a certain date, please say so, and why.

Example change request: with [cross references] to above list

I would like to propose a change to benchmark 997.oldfort

Using SPECaccel V2.0 [1], benchmark 997.oldfort fails to compile with "FATAL ERROR: foo.f, line 1814: DO loops must end with CONTINUE or END DO"[2]. The error occurs on the TurboBlaster Mark I system [3] using TurboBlasterUnix V1.0 [4] and TurboBlaster Fortran V1.0 [5]. The compiler supports only two language dialects; neither one will compile the benchmark: "--standard:Fortran2003" and "--standard:Fortran2008". Lower optimization levels have no effect.

The problem occurs only on the TurboBlaster. Other compilers (on other systems) issue a warning about the same line of source code, but not a fatal error.[7].

With the following change to 997.oldfort/src/foo.f, I can compile the benchmark:[8]

   $ diff -u foo.f.orig foo.f
   --- foo.f.orig       Fri Jul 15 14:09:28 2011
   +++ foo.f    Sat Jul 29 10:06:34 2009
   @@ -1811,7 +1811,8 @@
          FFY(I) = ZERO
          DDT(I) = ZERO
          DDQ(I) = ZERO
   -  100 DDE(I) = ZERO
   +      DDE(I) = ZERO
   +  100 CONTINUE
          NYQBTTS = 20
          IF(QPM) NYQBTTS=21
    C
   $ 

Without the change, the compiler complains [9]:

    FATAL ERROR: foo.f, line 1814: DO loops must end  either with CONTINUE or END DO 
      100 DDE(I) = ZERO
    --^

I am unable to workaround the problem on this system, because no other Fortran compiler is available, and TurboBlaster Fortran is quite insistent about this point [10].

Although I do not have a copy of the standard, a Fortran 95 compiler on another system does warn about the same statement and specifically claims that it is nonstandard [11]. On that other system, the above source change was tested, and performance was the same (within run-to-run variation - much less than 1% different). [12]. The problem has been reported to the compiler vendor, at www.turboblasterfortran.com/emailforum/msg00416.html [13], and you can see in the replies attached to that message that the vendor seems quite proud of its assertion that it has a fully compliant Fortran2008 compiler with no support for older variations of the language.

It would be very much appreciated if you could accept this change request within the next month, so that I can publish my PhD dissertation before my parents arrive from overseas for a six-month visit [15].

Copyright © 2014-2023 Standard Performance Evaluation Corporation All Rights Reserved