This is the mail archive of the newlib@sourceware.org mailing list for the newlib project.
| Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
|---|---|---|
| Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |
| Other format: | [Raw text] | |
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On 06/24/2013 11:36 AM, Corinna Vinschen wrote: > > As far as I can see, destructing _sig_func needs synchronization. > If you free it, a _signal_r call could come in and access the > free'd memory. If it already has been set to NULL, _signal_r calls > _init_signal_r and reallocates it. So we either skip this problem > for now, or you have to make sure that nobody accesses _sig_func > during destruction. > Considering that on a second though the problem looks even worse (malloc isn't guaranteed to by signal-safe...), I opted for a simpler solution. This patch has two advantages: - - it fixes the leak about _signal_buf and _misc. - - it leaves a comment about the _sig_func issue being unsolved, which I presume is better than nothing. newlib/ChangeLog 2013-06-24 Terraneo Federico <fede.tft@hotmail.it> * libc/reent/reent.c (_reclaim_reent): Free also _misc and _signal_buf. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQEcBAEBAgAGBQJRyCkdAAoJECkLFtN5Xr9f6+cH/jLEkP/4Z9GWj30CZc4wBkCM dD2Nn3yAFB9sURuR3If1KNbLCMH5AxeOT20BI5CkBbf78leh3p83AVooXtJNzzuH ZViaJpNMUbr+VjqUA92xXibxduFWXPzhQyEbMWvGcUMjE9Ie2YCYPVpOe4A458yB opwujHsuu8mpIbjKhgpjvSLtZaxyLZWQV3ECn52Zn0dxE/trEFkkscktAZzswjbO C+ZIlo47DucjoLYTjdKU1Y0n1y4+0zQ+Hm26Eu4gGuMnLuufopmDU4c8ZwqShsyc umpDzfYAmwRb6dHd7P1gKHRh22YvUceYIyjbwA3GDTTdF6MZV+yiiTSgGCD+UZI= =kYb6 -----END PGP SIGNATURE-----
diff -ruN a/newlib/libc/reent/reent.c b/newlib/libc/reent/reent.c
--- a/newlib/libc/reent/reent.c 2013-05-16 14:29:28.000000000 +0200
+++ b/newlib/libc/reent/reent.c 2013-06-24 12:59:12.599966562 +0200
@@ -87,6 +87,10 @@
_free_r (ptr, ptr->_localtime_buf);
if (ptr->_asctime_buf)
_free_r (ptr, ptr->_asctime_buf);
+ if (ptr->_signal_buf)
+ _free_r (ptr, ptr->_signal_buf);
+ if (ptr->_misc)
+ _free_r (ptr, ptr->_misc);
#endif
#ifndef _REENT_GLOBAL_ATEXIT
@@ -110,6 +114,11 @@
if (ptr->_cvtbuf)
_free_r (ptr, ptr->_cvtbuf);
+ /* We should free _sig_func to avoid a memory leak, but how to
+ do it safely considering that a signal may be delivered immediately
+ after the free?
+ if (ptr->_sig_func)
+ _free_r (ptr, ptr->_sig_func);*/
if (ptr->__sdidinit)
{
Attachment:
reclaim_reent2.patch.sig
Description: Binary data
| Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
|---|---|---|
| Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |