Changeset 23e1f18 in rtems-testing


Ignore:
Timestamp:
Jul 23, 2009, 8:32:19 PM (10 years ago)
Author:
Joel Sherrill <joel.sherrill@…>
Branches:
4.11, 8895273c193787f84c4585a10f6d6aceb3b25dc4
Children:
25ee588
Parents:
361fda6
Message:

2009-07-23 Joel Sherrill <joel.sherrill@…>

  • Explanations.txt: Eliminate more interrupt critical section cases.
  • gen_report: Note which spintrcritical* tests cannot detect when they hit the case and just run until killed.
Location:
rtems-coverage
Files:
3 edited

Legend:

Unmodified
Added
Removed
  • rtems-coverage/ChangeLog

    r361fda6 r23e1f18  
     12009-07-23      Joel Sherrill <joel.sherrill@OARcorp.com>
     2
     3        * Explanations.txt: Eliminate more interrupt critical section cases.
     4        * gen_report: Note which spintrcritical* tests cannot detect when they
     5        hit the case and just run until killed.
     6
    172009-07-23      Joel Sherrill <joel.sherrill@OARcorp.com>
    28
  • rtems-coverage/Explanations.txt

    r361fda6 r23e1f18  
    8686Simple Test Case
    8787It looks like we don't have a good sporadic server test.  This is
    88 the normal case of changin priority.
    89 +++
    90 
    91 timerreset.c:60
    92 Simple Test Case
    93 new error case
    94 +++
    95 
    96 timerserverfireafter.c:80
    97 Interrupt Synchronization
    98 timer must be inserted by an ISR while we are inserting it
     88the normal case of changing priority.
    9989+++
    10090
     
    169159+++
    170160
    171 timerfireafter.c:76
    172 Interrupt Synchronization
    173 timer must be inserted or state altered by ISR
    174 +++
    175 
    176161threadinitialize.c:91
    177162Simple Test Case
     
    196181Need to call iterate over threads in a test where there is an API configured
    197182which does not have threads.  We may be able to add this to sp54.
    198 +++
    199 
    200 threadqenqueuepriority.c:99
    201 Interrupt Synchronization
    202 This case is where we are iterating to enqueue a thread into a priority
    203 based thread queue but the thread we are looking at gets unblocked when
    204 we flash interrupts.  It is NOT the thread we are unblocking.
    205 Forward Search case.
    206 
    207 NOTE: Do not remove this from the explanations since the test can vary.
    208 +++
    209 
    210 threadqenqueuepriority.c:146
    211 Interrupt Synchronization
    212 This case is where we are iterating to enqueue a thread into a priority
    213 based thread queue but the thread we are looking at gets unblocked when
    214 we flash interrupts.  It is NOT the thread we are unblocking.
    215 Reverse Search case.
    216 
    217 NOTE: Do not remove this from explanations.
    218183+++
    219184
     
    242207This is processing a timeout on a thread (that is the thread executing)
    243208that has also had its request satisfied while it is being enqueued.
    244 +++
    245 
    246 watchdoginsert.c:52
    247 Interrupt Synchronization
    248 This is when a higher priority interrupt inserts a timer while a lower
    249 priority interrupt is inserting one.
    250 +++
    251 
    252 watchdogremove.c:41
    253 Interrupt Synchronization
    254 Remove a watchdog while it is being inserted.
    255 +++
    256 
    257 watchdogremove.c:64
    258 Interrupt Synchronization
    259 Remove a watchdog while it is being removed.
    260209+++
    261210
     
    286235killinfo.c:328
    287236Simple Test Case
    288 This is an error case where we can't allocate another signinfo structure.
    289 +++
    290 
    291 timerinserthelper.c:45
    292 Interrupt Synchronization
    293 We just removed the timer but the watchdog timer is being inserted by
    294 a higher priority interrupt.  We may be able to get this to happen
    295 with Classic API Timers.  But that is unclear.
    296 +++
     237This is an error case where we can't allocate another siginfo structure.
     238+++
  • rtems-coverage/gen_report

    r361fda6 r23e1f18  
    6363        *fatal*) ;;
    6464        *stackchk*) ;;
     65        # These do not know when they hit the case so run until killed.
     66        *spintrcritical06*) ;;
     67        *spintrcritical07*) ;;
    6568        *spintrcritical09*) ;;
     69        *spintrcritical13*) ;;
     70        *spintrcritical14*) ;;
    6671        *)
    6772          grep "END OF" $f >/dev/null
     
    95100${NM} -g -n hello.exe >hello.num
    96101
    97 # Basic validity checking 
     102# Basic validity checking
    98103tlow=`grep "T ${lowsym}$" hello.num | cut -d' ' -f1`
    99104if [ X${tlow} = X ] ; then
     
    119124    -T ${TARGET} -e hello.exe -m merged.cov *.cov; \
    120125  echo "====== 15 Largest Range Sizes ======" ; \
    121   grep ^Size report.txt | cut -d':' -f2 | sort -n | uniq -c | tail -15 ; 
     126  grep ^Size report.txt | cut -d':' -f2 | sort -n | uniq -c | tail -15 ;
    122127  echo "=====================================" ) >summary.txt
    123128
Note: See TracChangeset for help on using the changeset viewer.