Gilles Chanteperdrix wrote: > GIT version control wrote: >> Module: xenomai-jki >> Branch: for-upstream >> Commit: 40e5b366397db3dde0ebde861a739dc2cd624b5a >> URL: >> http://git.xenomai.org/?p=xenomai-jki.git;a=commit;h=40e5b366397db3dde0ebde861a739dc2cd624b5a >> >> Author: Jan Kiszka <jan.kis...@siemens.com> >> Date: Fri Feb 12 17:40:09 2010 +0100 >> >> Fix remapping of global sem_heap on fork >> >> As xeno_init_sem_heaps might be called multiple times even without fork, >> we must not remap the global heap here. Register an atfork handler >> instead that handles this case and simple reject multiple heap creation >> requests for the same process. >> >> Signed-off-by: Jan Kiszka <jan.kis...@siemens.com> >> >> --- >> >> src/skins/common/sem_heap.c | 29 ++++++++++++++++++----------- >> 1 files changed, 18 insertions(+), 11 deletions(-) >> >> diff --git a/src/skins/common/sem_heap.c b/src/skins/common/sem_heap.c >> index 89f325a..12b256f 100644 >> --- a/src/skins/common/sem_heap.c >> +++ b/src/skins/common/sem_heap.c >> @@ -1,6 +1,6 @@ >> #include <stdio.h> >> #include <stdlib.h> >> - >> +#include <pthread.h> >> #include <unistd.h> >> #include <sys/types.h> >> #include <sys/stat.h> >> @@ -65,24 +65,31 @@ static void unmap_sem_heap(unsigned long heap_addr, >> unsigned shared) >> munmap((void *) heap_addr, hinfo.size); >> } >> >> +static void remap_on_fork(void) >> +{ >> + unmap_sem_heap(xeno_sem_heap[0], 0); >> + >> + if (xeno_sem_heap[0] == (unsigned long) MAP_FAILED) { >> + perror("Xenomai: mmap(local sem heap)"); >> + exit(EXIT_FAILURE); >> + } > > Looks bogus to me. Where do the remap occur ? > Looks like your test case does not try forking. >
Yep, copy&paste error. Fixed. Jan -- Siemens AG, Corporate Technology, CT T DE IT 1 Corporate Competence Center Embedded Linux _______________________________________________ Xenomai-core mailing list Xenomai-core@gna.org https://mail.gna.org/listinfo/xenomai-core
