[Bug middle-end/23289] [4.1 Regression] tail call optimization not performed

2005-08-09 Thread rguenth at gcc dot gnu dot org
--- Additional Comments From rguenth at gcc dot gnu dot org 2005-08-09 08:52 --- I think I know how to fix this. But some general questions (wrt the x86 ABI): - can we always tail call if either function returns void and the other returns in a register? - can we _not_ tail call if ei

[Bug middle-end/23289] [4.1 Regression] tail call optimization not performed

2005-08-08 Thread pinskia at gcc dot gnu dot org
--- Additional Comments From pinskia at gcc dot gnu dot org 2005-08-08 19:22 --- This worked with "4.1.0 20050714". I think this was caused by: 2005-07-14 Richard Guenther <[EMAIL PROTECTED]> * Makefile.in (explow.o, reg-stack.o): Depend on target.h. * calls.c (expand

[Bug middle-end/23289] [4.1 Regression] tail call optimization not performed

2005-08-08 Thread pinskia at gcc dot gnu dot org
--- Additional Comments From pinskia at gcc dot gnu dot org 2005-08-08 19:16 --- The tree level is catching it, SendMousePosition (w, event) [tail call]; Reduced testcase: extern int SendMousePosition1 (void); void HandleIgnore(void) { SendMousePosition1(); } Note if we change