This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug tree-optimization/33237] [4.3 Regression] Tree memory partitioning is spending 430 seconds of a 490 second compile.
- From: "rguenth at gcc dot gnu dot org" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: 25 Jan 2008 16:40:54 -0000
- Subject: [Bug tree-optimization/33237] [4.3 Regression] Tree memory partitioning is spending 430 seconds of a 490 second compile.
- References: <bug-33237-11700@http.gcc.gnu.org/bugzilla/>
- Reply-to: gcc-bugzilla at gcc dot gnu dot org
------- Comment #8 from rguenth at gcc dot gnu dot org 2008-01-25 16:40 -------
So we come along some big points-to sets (and thus MTAG_ALIASES) when doing
for (i = 1; i < num_ssa_names; i++)
{
...
tag = pi->name_mem_tag;
EXECUTE_IF_SET_IN_BITMAP (MTAG_ALIASES (tag), 0, j, bj)
{
...
such as NMT.62 with 19034 aliases, and that multiple times (for each pointer
that has this NMT! Which in this testcase are a lot).
I think we are also over-counting here - shouldn't we just walk over all
NMTs once?
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33237