|
From: <Woe...@on...> - 2003-12-30 11:56:24
|
=2D----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi, when is Valgrind's libpthread.so used? I've two applications linked against libqt-mt.so and so against libpthread.so. If I start the apps with valgrind -v app 1 reports =2E.. =3D=3D8915=3D=3D Reading syms from /usr/lib/valgrind/libpthread.so =2E.. =3D=3D8915=3D=3D REPLACING libc(__errno_location) with libpthread(__errno_l= ocation) =3D=3D8915=3D=3D REPLACING libc(__h_errno_location) with libpthread(__h_err= no_location) =3D=3D8915=3D=3D REPLACING libc(__res_state) with libpthread(__res_state) =3D=3D8915=3D=3D =3D=3D8915=3D=3D TRANSLATE: 0x40E1EFA0 redirected to 0x40CFE4A5 =2E.. and app 2 reports =2E.. =3D=3D9074=3D=3D Reading syms from /lib/libpthread-0.10.so =2E.. and a lot of invalid read/write of size 4 which make the report useless. I'm using Debian's sid (Valgrind 2.1.0) but have the same effect on SuSE 9.0 (Valgrind-20031012). How can I convince Valgrind to use its own libpthread.so? Andr=E9 =2D----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.3 (GNU/Linux) iD8DBQE/8WfoGcp/AjG8ndYRAvwDAJ9m6s+oF1pac8mXF4cgTG/WXAfGdgCbBvlb erT+KonQ133g4Gxa7tH8edE=3D =3Dq1D4 =2D----END PGP SIGNATURE----- |