This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: GCC 4.8 and -ftree-loop-distribute-patterns.
- From: "Joseph S. Myers" <joseph at codesourcery dot com>
- To: Roland McGrath <roland at hack dot frob dot com>
- Cc: Adhemerval Zanella <azanella at linux dot vnet dot ibm dot com>, <libc-alpha at sourceware dot org>
- Date: Fri, 19 Apr 2013 19:35:54 +0000
- Subject: Re: GCC 4.8 and -ftree-loop-distribute-patterns.
- References: <511D4F82 dot 3080604 at redhat dot com> <515F4565 dot 6030301 at linux dot vnet dot ibm dot com> <20130405223023 dot 0DEB62C096 at topped-with-meat dot com> <51601F98 dot 5000606 at linux dot vnet dot ibm dot com> <20130408203339 dot 88DBF2C09F at topped-with-meat dot com>
On Mon, 8 Apr 2013, Roland McGrath wrote:
> Instead, we should have a global solution to the problem of unwanted PLT
> calls in compiler-chosen implicit calls. This could happen anywhere,
> not just in cases that -ftree-loop-distribute-patterns is newly causing.
> For example, any large struct copying might generate a memcpy call; if
> that were done in some file that doesn't happen to #include <string.h>
> for any reason, then we would already be seeing the problem.
As another example, various architectures allow calls to software
floating-point functions (where those functions are defined in libc rather
than just libgcc) in their localplt.data because of the lack of a
mechanism to redirect those libc-local calls away from the PLT.
--
Joseph S. Myers
joseph@codesourcery.com