diff options
Diffstat (limited to 'winsup/cygwin/dcrt0.cc')
| -rw-r--r-- | winsup/cygwin/dcrt0.cc | 42 |
1 files changed, 7 insertions, 35 deletions
diff --git a/winsup/cygwin/dcrt0.cc b/winsup/cygwin/dcrt0.cc index 38b4b67f3..727163e08 100644 --- a/winsup/cygwin/dcrt0.cc +++ b/winsup/cygwin/dcrt0.cc @@ -425,9 +425,7 @@ check_sanity_and_sync (per_process *p) /* struct without updating SIZEOF_PER_PROCESS [it makes them think twice */ /* about changing it]. */ if (sizeof (per_process) != SIZEOF_PER_PROCESS) - { - api_fatal ("per_process sanity check failed"); - } + api_fatal ("per_process sanity check failed"); /* Make sure that the app and the dll are in sync. */ @@ -590,48 +588,17 @@ child_info * get_cygwin_startup_info () { STARTUPINFO si; - DWORD zeros[sizeof (child_proc_info->zero) - / sizeof (child_proc_info->zero[0])] = {0}; GetStartupInfo (&si); child_info *res = (child_info *) si.lpReserved2; - /* It appears that when running under WOW64 on Vista 64, the first DWORD - value in the datastructure lpReserved2 is pointing to (zero[0] in - Cygwin), has to reflect the size of that datastructure as used in the - Microsoft C runtime (a count value, counting the number of elements in - two subsequent arrays, BYTE[count and HANDLE[count]), even though the C - runtime isn't used. Otherwise, if zero[0] is 0 or too small, the - datastructure gets overwritten. - - This seems to be a bug in Vista's WOW64, which apparently copies the - lpReserved2 datastructure not using the cbReserved2 size information, - but using the information given in the first DWORD within lpReserved2 - instead. 32 bit Windows and former WOW64 don't care if zero[0] is 0 - or a sensible non-0 count value. However, it's not clear if a non-0 - count doesn't result in trying to evaluate the content, so we do this - really only for Vista 64 for now. - - exec/spawn as well as fork write an appropriate value into zero[0] now, - depending on the wincap.needs_count_in_si_lpres2 flag. The value is - sizeof (child_info_*) / 5 which results in a count which covers the - full datastructure, plus not more than 4 extra bytes. This is ok as - long as the child_info structure is cosily stored within a bigger - datastructure. */ - if (wincap.needs_count_in_si_lpres2 ()) - zeros[0] = si.cbReserved2 / 5; - if (si.cbReserved2 < EXEC_MAGIC_SIZE || !res - || memcmp (res->zero, zeros, sizeof (res->zero)) != 0) + || res->intro != PROC_MAGIC_GENERIC || res->magic != CHILD_INFO_MAGIC) res = NULL; else { if ((res->intro & OPROC_MAGIC_MASK) == OPROC_MAGIC_GENERIC) multiple_cygwin_problem ("proc intro", res->intro, 0); - else if (res->intro == PROC_MAGIC_GENERIC - && res->magic != CHILD_INFO_MAGIC) - multiple_cygwin_problem ("proc magic", res->magic, - CHILD_INFO_MAGIC); else if (res->cygheap != (void *) &_cygheap_start) multiple_cygwin_problem ("cygheap base", (DWORD) res->cygheap, (DWORD) &_cygheap_start); @@ -718,6 +685,11 @@ child_info_spawn::handle_spawn () envc = moreinfo->envc; if (!dynamically_loaded) cygheap->fdtab.fixup_after_exec (); + if (__stdin >= 0) + cygheap->fdtab.move_fd (__stdin, 0); + if (__stdout >= 0) + cygheap->fdtab.move_fd (__stdout, 1); + ready (true); /* Need to do this after debug_fixup_after_fork_exec or DEBUGGING handling of |