acpica/tests/aslts/bin

CONTENTS

   Utilities to manager the tests


Do
   The test suite command dispatcher.
   Allows to install AML tests, run them and compare results of runs.
   The output log of tests and results of processing the test logs are
   located into aslts/tmp/RESULTS automatically generated directory.

   Run Do without parameters to print out its Usage.


   The simplest sequence of actions to deal with the test suite
   is like one described below.


   1. Set up the following variables:

      ASL      - pathname of iASL compiler
      acpiexec - pathname of AcpiExec utility
      ASLTSDIR - pathname of root directory of aslts test suite

   2. Add the <aslts/bin> directory to the PATH variable.

   3. Run:
      > Do
        utility prints out its usage
   4. Run (if the tests in aslts/tmp/aml directory are not built
           or you want to re-compiler them):
      > Do 0 aslts
        utility do compile and install all the AML test cases under aslts/tmp/aml
   5. Run:
      > Do 1
        utility do run all the specified test cases in all the specified modes

           all possible modes now are:
              32-bit norm mode
              64-bit norm mode
              32-bit slack mode
              64-bit slack mode

   6. Run:
      > Do 2
        utility do compare the last multi-result of run with the previous one

        or for comparing the particular multi-results

      > Do 2 <first multi-result directory> <second multi-result directory>
             <first multi-result directory>  - new obtained multi-result which
                                               you want to examine either it has
                                               any regressions relative to the
                                               second multi-result
             <second multi-result directory> - benchmark multi-result


   After the tests completed:
     - see aslts/tmp/RESULTS/date.time.revision/Symmary file for summary
     - compare the obtained multi-result with the previous benchmark
       multi-result for to catch regressions

   To adjust the needed set of test cases and modes of runs
   edit manually the aslts/bin/settings file.





Do 2 output:
============

1. Summary information related to the particular test case:

   echo "|$2|$cnt|$pass|$fail|$skip|$start|$finish|$total|$outstand0|$blck|$memtotal|$max0|$max1|$max2|$max3|$max4|$max5|$out0|$out1|$out2|$out3|$out4|$out5|$outstand1|$LargeRefCount|$memtotalbytes|$exceptionsnum|" >> "$3"

outstand0 -- outstanding allocations of the particular test case
             Example from arithmetic test case log:
                Outstanding: 0x11 allocations after execution
outstand1 -- outstanding allocations of the particular test case
             Example from reference test case log:
                ACPI Error (uttrack-0719): 39(27) Outstanding allocations [20074402]

memtotal  -- total memory in bytes consumed during execution of the particular test case
start     -- time when test case started
finish    -- time when test case finished
total     -- duration of test case execution

2. Summary information related to all test cases been run for some test mode (32-bit normal, etc..):

   echo "|TOTAL|$outstand0|$memtotal|${MODES_TIMES[$2]}|$outstand1|$LargeRefCount|$memtotalbytes|"  >> "$path"

totaloutstand00 -- summary of outstand0 of all test cases
totalmem0       -- summary of memtotal of all test cases
totaltime0      -- all the test cases execution duration
totaloutstand01 -- summary of outstand1 of all test cases

-------------<>.