Commit e5b39b0c authored by simonmar's avatar simonmar
Browse files

[project @ 2005-11-19 11:44:32 by simonmar]

something has gone wrong; I don't have time right now to find out
exactly what, so revert rev. 1.22 in an attempt to fix it.
parent 942818de
......@@ -249,8 +249,6 @@ but uses $$dyncall if necessary to cope, just in case you aren't.
function and these markers is shredded by the mangler.
-------------------------------------------------------------------------- */
#ifndef FB_
#if __GNUC__ < 3
/* The following __DISCARD__() has become necessary with gcc 2.96 on x86.
* It prevents gcc from moving stack manipulation code from the function
* body (aka the Real Code) into the function prologue, ie, from moving it
......@@ -260,17 +258,9 @@ but uses $$dyncall if necessary to cope, just in case you aren't.
* it just doesn't choose to do it at the moment.
* -= chak
*/
#define FB_ __asm__ volatile ("--- BEGIN ---"); __DISCARD__ ();
#else
/* The __DISCARD__() doesn't appear to be necessary with gcc >= 3.2 at
* least, and it does cause some difficulty, preventing gcc from
* optimising around the beginning of the function. In particular,
* gcc leaves some stack assignments in the prologue when the call is
* present. --SDM
*/
#define FB_ __asm__ volatile ("--- BEGIN ---");
#endif
#ifndef FB_
#define FB_ __asm__ volatile ("--- BEGIN ---"); __DISCARD__ ();
#endif
#ifndef FE_
......
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment