Re: tests using FP specifiers in prints
On 09/02/18 13:21, Sebastian Huber wrote: On 09/02/18 08:58, Sebastian Huber wrote: On 08/02/18 16:45, Joel Sherrill wrote: sp19 and tm26 are not in that list. What about them? They should be fixed now. The sp19 fails now on SPARC: *** BEGIN OF TEST SP 19 *** *** TEST VERSION: 5.0.0.dc8fd0152928649ac83632a712d96c0dbeca05ae *** TEST STATE: EXPECTED-PASS *** TEST BUILD: RTEMS_NETWORKING RTEMS_POSIX_API RTEMS_SMP *** TEST TOOLS: 7.3.0 20180125 (RTEMS 5, RSB 703532cb04c6990fb21e97cb7347a16e9df11108, Newlib 3.0.0) FP1 - integer base = (0x6000) FP1 - float base = (NA) FP1 - integer base = (0x6000) FP1 - float base = (NA) TA1 - integer base = (0x1000) TA1 - rtems_clock_get_tod - 09:00:00 12/31/1988 CPU 0: breakpoint 1 hit 0x4000bc98: 94102000 mov 0, %o2 <_Terminate+4> CPU 1: Power down mode CPU 2: Power down mode CPU 3: Power down mode grmon2> bt %pc %sp #0 0x4000bc98 0x40025b68 <_Terminate+0x4> #1 0x4000bce8 0x40025bd0 <_Internal_error+0x8> #2 0x40005cf8 0x40025c30 #3 0x4001f848 0x40025c98 #4 0x400172d8 0x40031ba0 <_vfprintf_r+0x2e8> #5 0x400198d8 0x40031e28 <__sbprintf+0x70> #6 0x40017280 0x40032308 <_vfprintf_r+0x290> #7 0x40014880 0x400324e0 #8 0x40013e44 0x40032540 #9 0x40004688 0x400325a8 #10 0x4000f7a0 0x400326b8 <_Thread_Entry_adaptor_numeric+0x8> #11 0x4000f870 0x40032718 <_Thread_Handler+0x60> #12 0x4000f810 0x40032778 <_Thread_Handler+0> I don't know how to fix this. The Task_1 is created with/without RTEMS_FLOATING_POINT. Changing everything to RTEMS_FLOATING_POINT may change the test objectives. I fixed it like this: https://git.rtems.org/rtems/commit/?id=828049cb47d974fcc816a5bd8868fdaf670201a3 -- Sebastian Huber, embedded brains GmbH Address : Dornierstr. 4, D-82178 Puchheim, Germany Phone : +49 89 189 47 41-16 Fax : +49 89 189 47 41-09 E-Mail : sebastian.hu...@embedded-brains.de PGP : Public key available on request. Diese Nachricht ist keine geschäftliche Mitteilung im Sinne des EHUG. ___ devel mailing list devel@rtems.org http://lists.rtems.org/mailman/listinfo/devel
Re: tests using FP specifiers in prints
On 09/02/18 08:58, Sebastian Huber wrote: On 08/02/18 16:45, Joel Sherrill wrote: sp19 and tm26 are not in that list. What about them? They should be fixed now. The sp19 fails now on SPARC: *** BEGIN OF TEST SP 19 *** *** TEST VERSION: 5.0.0.dc8fd0152928649ac83632a712d96c0dbeca05ae *** TEST STATE: EXPECTED-PASS *** TEST BUILD: RTEMS_NETWORKING RTEMS_POSIX_API RTEMS_SMP *** TEST TOOLS: 7.3.0 20180125 (RTEMS 5, RSB 703532cb04c6990fb21e97cb7347a16e9df11108, Newlib 3.0.0) FP1 - integer base = (0x6000) FP1 - float base = (NA) FP1 - integer base = (0x6000) FP1 - float base = (NA) TA1 - integer base = (0x1000) TA1 - rtems_clock_get_tod - 09:00:00 12/31/1988 CPU 0: breakpoint 1 hit 0x4000bc98: 94102000 mov 0, %o2 <_Terminate+4> CPU 1: Power down mode CPU 2: Power down mode CPU 3: Power down mode grmon2> bt %pc %sp #0 0x4000bc98 0x40025b68 <_Terminate+0x4> #1 0x4000bce8 0x40025bd0 <_Internal_error+0x8> #2 0x40005cf8 0x40025c30 #3 0x4001f848 0x40025c98 #4 0x400172d8 0x40031ba0 <_vfprintf_r+0x2e8> #5 0x400198d8 0x40031e28 <__sbprintf+0x70> #6 0x40017280 0x40032308 <_vfprintf_r+0x290> #7 0x40014880 0x400324e0 #8 0x40013e44 0x40032540 #9 0x40004688 0x400325a8 #10 0x4000f7a0 0x400326b8 <_Thread_Entry_adaptor_numeric+0x8> #11 0x4000f870 0x40032718 <_Thread_Handler+0x60> #12 0x4000f810 0x40032778 <_Thread_Handler+0> I don't know how to fix this. The Task_1 is created with/without RTEMS_FLOATING_POINT. Changing everything to RTEMS_FLOATING_POINT may change the test objectives. -- Sebastian Huber, embedded brains GmbH Address : Dornierstr. 4, D-82178 Puchheim, Germany Phone : +49 89 189 47 41-16 Fax : +49 89 189 47 41-09 E-Mail : sebastian.hu...@embedded-brains.de PGP : Public key available on request. Diese Nachricht ist keine geschäftliche Mitteilung im Sinne des EHUG. ___ devel mailing list devel@rtems.org http://lists.rtems.org/mailman/listinfo/devel
Re: tests using FP specifiers in prints
On 08/02/18 16:45, Joel Sherrill wrote: sp19 and tm26 are not in that list. What about them? They should be fixed now. -- Sebastian Huber, embedded brains GmbH Address : Dornierstr. 4, D-82178 Puchheim, Germany Phone : +49 89 189 47 41-16 Fax : +49 89 189 47 41-09 E-Mail : sebastian.hu...@embedded-brains.de PGP : Public key available on request. Diese Nachricht ist keine geschäftliche Mitteilung im Sinne des EHUG. ___ devel mailing list devel@rtems.org http://lists.rtems.org/mailman/listinfo/devel
Re: tests using FP specifiers in prints
On Thu, Feb 8, 2018 at 9:13 AM, Sebastian Huber < sebastian.hu...@embedded-brains.de> wrote: > On 08/02/18 16:02, Joel Sherrill wrote: > >> Hi >> >> Jennifer noticed that sp19 prints %g instead of numbers now that >> we switched to printk for the tests. I did a quick grep and found these: >> >> testsuites$ grep -rl "%g" . >> ./samples/fileio/init.c >> ./samples/paranoia/paranoia.c >> ./sptests/sp19/first.c >> ./sptests/sp19/fptask.c >> ./sptests/sp19/fptest.h >> ./tmtests/tm26/fptest.h >> >> Are fileio and paranoia using printf()? >> > > They use: > > testsuites/samples/pppd/init.c: rtems_print_printer_fprintf_pu > tc(&rtems_test_printer); > testsuites/samples/loopback/init.c: rtems_print_printer_fprintf_pu > tc(&rtems_test_printer); > testsuites/samples/paranoia/init.c: rtems_print_printer_fprintf_pu > tc(&rtems_test_printer); > testsuites/samples/hello/init.c: rtems_print_printer_fprintf_pu > tc(&rtems_test_printer); > testsuites/samples/capture/init.c: rtems_print_printer_fprintf_pu > tc(&rtems_test_printer); > testsuites/benchmarks/linpack/init.c: rtems_print_printer_fprintf_pu > tc(&rtems_test_printer); > testsuites/benchmarks/whetstone/init.c: rtems_print_printer_fprintf_pu > tc(&rtems_test_printer); > testsuites/benchmarks/dhrystone/init.c: rtems_print_printer_fprintf_pu > tc(&rtems_test_printer); > testsuites/libtests/mathl/init.c: rtems_print_printer_fprintf_pu > tc(&rtems_test_printer); > testsuites/libtests/mathf/init.c: rtems_print_printer_fprintf_pu > tc(&rtems_test_printer); > testsuites/libtests/complex/init.c: rtems_print_printer_fprintf_pu > tc(&rtems_test_printer); > testsuites/libtests/capture01/init.c: rtems_print_printer_fprintf_pu > tc(&rtems_test_printer); > testsuites/libtests/math/init.c: rtems_print_printer_fprintf_pu > tc(&rtems_test_printer); > testsuites/smptests/smpmigration01/init.c: rtems_print_printer_fprintf_pu > tc(&rtems_test_printer); > testsuites/smptests/smpopenmp01/init.c: rtems_print_printer_fprintf_pu > tc(&rtems_test_printer); > testsuites/sptests/spconsole01/init.c: rtems_print_printer_fprintf_pu > tc(&rtems_test_printer); > sp19 and tm26 are not in that list. What about them? > > >> Do we need to eliminate these uses of %g in sptests and tmtests? >> >> --joel >> >> >> >> >> >> >> >> ___ >> devel mailing list >> devel@rtems.org >> http://lists.rtems.org/mailman/listinfo/devel >> > > -- > Sebastian Huber, embedded brains GmbH > > Address : Dornierstr. 4, D-82178 Puchheim, Germany > Phone : +49 89 189 47 41-16 > Fax : +49 89 189 47 41-09 > E-Mail : sebastian.hu...@embedded-brains.de > PGP : Public key available on request. > > Diese Nachricht ist keine geschäftliche Mitteilung im Sinne des EHUG. > > ___ > devel mailing list > devel@rtems.org > http://lists.rtems.org/mailman/listinfo/devel ___ devel mailing list devel@rtems.org http://lists.rtems.org/mailman/listinfo/devel
Re: tests using FP specifiers in prints
On 08/02/18 16:02, Joel Sherrill wrote: Hi Jennifer noticed that sp19 prints %g instead of numbers now that we switched to printk for the tests. I did a quick grep and found these: testsuites$ grep -rl "%g" . ./samples/fileio/init.c ./samples/paranoia/paranoia.c ./sptests/sp19/first.c ./sptests/sp19/fptask.c ./sptests/sp19/fptest.h ./tmtests/tm26/fptest.h Are fileio and paranoia using printf()? They use: testsuites/samples/pppd/init.c: rtems_print_printer_fprintf_putc(&rtems_test_printer); testsuites/samples/loopback/init.c: rtems_print_printer_fprintf_putc(&rtems_test_printer); testsuites/samples/paranoia/init.c: rtems_print_printer_fprintf_putc(&rtems_test_printer); testsuites/samples/hello/init.c: rtems_print_printer_fprintf_putc(&rtems_test_printer); testsuites/samples/capture/init.c: rtems_print_printer_fprintf_putc(&rtems_test_printer); testsuites/benchmarks/linpack/init.c: rtems_print_printer_fprintf_putc(&rtems_test_printer); testsuites/benchmarks/whetstone/init.c: rtems_print_printer_fprintf_putc(&rtems_test_printer); testsuites/benchmarks/dhrystone/init.c: rtems_print_printer_fprintf_putc(&rtems_test_printer); testsuites/libtests/mathl/init.c: rtems_print_printer_fprintf_putc(&rtems_test_printer); testsuites/libtests/mathf/init.c: rtems_print_printer_fprintf_putc(&rtems_test_printer); testsuites/libtests/complex/init.c: rtems_print_printer_fprintf_putc(&rtems_test_printer); testsuites/libtests/capture01/init.c: rtems_print_printer_fprintf_putc(&rtems_test_printer); testsuites/libtests/math/init.c: rtems_print_printer_fprintf_putc(&rtems_test_printer); testsuites/smptests/smpmigration01/init.c: rtems_print_printer_fprintf_putc(&rtems_test_printer); testsuites/smptests/smpopenmp01/init.c: rtems_print_printer_fprintf_putc(&rtems_test_printer); testsuites/sptests/spconsole01/init.c: rtems_print_printer_fprintf_putc(&rtems_test_printer); Do we need to eliminate these uses of %g in sptests and tmtests? --joel ___ devel mailing list devel@rtems.org http://lists.rtems.org/mailman/listinfo/devel -- Sebastian Huber, embedded brains GmbH Address : Dornierstr. 4, D-82178 Puchheim, Germany Phone : +49 89 189 47 41-16 Fax : +49 89 189 47 41-09 E-Mail : sebastian.hu...@embedded-brains.de PGP : Public key available on request. Diese Nachricht ist keine geschäftliche Mitteilung im Sinne des EHUG. ___ devel mailing list devel@rtems.org http://lists.rtems.org/mailman/listinfo/devel
tests using FP specifiers in prints
Hi Jennifer noticed that sp19 prints %g instead of numbers now that we switched to printk for the tests. I did a quick grep and found these: testsuites$ grep -rl "%g" . ./samples/fileio/init.c ./samples/paranoia/paranoia.c ./sptests/sp19/first.c ./sptests/sp19/fptask.c ./sptests/sp19/fptest.h ./tmtests/tm26/fptest.h Are fileio and paranoia using printf()? Do we need to eliminate these uses of %g in sptests and tmtests? --joel ___ devel mailing list devel@rtems.org http://lists.rtems.org/mailman/listinfo/devel