mirror/gcc
2
0
Fork 0
mirror of git://gcc.gnu.org/git/gcc.git synced 2025-04-05 13:51:00 +08:00
Code Issues Packages Projects Releases Wiki Activity

flow.c (split_edge): If we have to insert a new jump, make sure to associate it with a basic block.

Browse Source 
* flow.c (split_edge): If we have to insert a new jump, make
	sure to associate it with a basic block.

	* flow.c (commit_one_edge_insertion): A block with one successor
	can end in a JUMP_INSN that is not a simplejump.

From-SVN: r31380
This commit is contained in:
Jeffrey A Law 2000-01-13 05:41:44 +00:00 committed by Jeff Law
parent 499bf451e3
commit 08520ad8d6
2 changed files with 15 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
gcc/ChangeLog
View File

@ -1,3 +1,11 @@
Wed Jan 12 22:34:00 2000 Jeffrey A Law (law@cygnus.com)
* flow.c (split_edge): If we have to insert a new jump, make
sure to associate it with a basic block.
* flow.c (commit_one_edge_insertion): A block with one successor
can end in a JUMP_INSN that is not a simplejump.
2000-01-12 Robert Lipe <robertl@sco.com>
* i386/sco5.h (INIT_SECTION_ASM_OP_COFF): Rename section to "ctor".

12
gcc/flow.c
View File

@ -1387,6 +1387,7 @@ split_edge (edge_in)
pos = emit_jump_insn_after (gen_jump (old_succ->head),
jump_block->end);
jump_block->end = pos;
set_block_for_insn (pos, jump_block);
emit_barrier_after (pos);
/* ... let jump know that label is in use, ... */
@ -1580,13 +1581,14 @@ commit_one_edge_insertion (e)
&& e->src != ENTRY_BLOCK_PTR)
{
bb = e->src;
/* It is possible to have a non-simple jump here. Consider a target
where some forms of unconditional jumps clobber a register. This
happens on the fr30 for example.
We know this block has a single successor, so we can just emit
the queued insns before the jump. */
if (GET_CODE (bb->end) == JUMP_INSN)
{
/* ??? Is it possible to wind up with non-simple jumps? Perhaps
a jump with delay slots already filled? */
if (! simplejump_p (bb->end))
abort ();
before = bb->end;
}
else