This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH] Handle weak symbols
At 07:34 08.05.2001, Mark Mitchell wrote:
>Franz --
>
>I have looked at this problem in more detail, and I am not convinced
>it can actually occur in the current sources. We are going to need a
>test-case to convince me that we need to do something here for the
>branch. It cannot occur in some of the obvious places (like
>delete_null_pointer_checks, which insists on having a use of the MEM
>before eliminating a future check), or loop, where as far as I can
>tell, we only move the possibly-trapping instruction out of the loop
>if we are sure it will be executed anyhow.
I gcc-2.95 it happens in one of the jump_optimize calls during cse2, I
didn't find why it doesn't happen in 3.0 anymore yet.
Franz.