From: Christophe G. <gr...@cg...> - 2005-12-13 12:04:06
|
On Tue, 13 Dec 2005, Keith MARSHALL wrote: > David Lamy-Charrier wrote: >> I just investigated in details the compiler options available for >> MSVC6.0 (cl.exe) and it appears that /Oy (Frame-Pointer Omission) >> is enabled by default in "Release" build: >> It certainly does not help GDB to decode the stack frame .... > > Indeed, it wouldn't. The GCC equivalent is `-fomit-frame-pointer'; > the documentation for that warns that it `makes debugging impossible > on some machines': > http://gcc.gnu.org/onlinedocs/gcc-3.4.4/gcc/Optimize-Options.html#index-fomit_002dframe_002dpointer-371 > >> I will have a try without this option to see if it improves stack >> decoding. > > BTW, `-fomit-frame-pointer' is automatically enabled at all levels > of optimisation in GCC, except `-O0'; it has always struck me as Not exactly. Extract of man page "-O also turns on -fomit-frame-pointer on machines where doing so does not interfere with debugging." Christophe --------------------------------------------------------------------- ,-~~-.___. ._. / | ' \ | |"""""""""| -= GRENIER Christophe =- ( ) 0 | | | \_/-, ,----' | | | ==== !_!--v---v--" http://www.cgsecurity.org / \-'~; |""""""""| / __/~| ._-""|| | Email: gr...@cg... =( _____|_|____||________| --------------------------------------------------------------------- |