http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58244
Marc Glisse changed:
What|Removed |Added
Status|UNCONFIRMED |RESOLVED
Resolution|---
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58244
--- Comment #8 from Martin Konôpka ---
Yes, I understand now. Thanks. The lines with the sin() functions were not
evaluated with the local declaration. My apologies for reporting the false bug.
I will try to close the bug if I am allowed.
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58244
--- Comment #7 from Andreas Schwab ---
Now try the same code with -O0.
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58244
--- Comment #6 from Martin Konôpka ---
(I am sorry that I do not understand internals of compilers.)
I first hit the issue in a bigger code having more than 11000 lines. First I
did not understand at all what is going on. I started to isolate the
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58244
--- Comment #5 from Martin Konôpka ---
I do not fully understand the question. In both cases I used the same
optimisation (-O2). My other comment a while ago was lost. I must retype it.
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58244
--- Comment #4 from Andreas Schwab ---
gcc was able to optimize your code to make it 6000 times faster. How is that a
regression?
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58244
--- Comment #3 from Marc Glisse ---
With auxval local, the compiler knows that computing suma, etc is useless and
removes all that code, including the calls to sin. So you would like the
compiler to do the same for a global auxval? That requires t
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58244
--- Comment #2 from Martin Konôpka ---
It it is confirmed, it is a very serious performance issue. In my test the code
with the global declaration executed about 6 times slower than the code
with the local declaration.
For people doing HPC th
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58244
--- Comment #1 from Andrew Pinski ---
Why do you think this is a bug?
