This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH] Fix PR rtl-optimization/pr60663
- From: Jakub Jelinek <jakub at redhat dot com>
- To: Zhenqiang Chen <zhenqiang dot chen at linaro dot org>
- Cc: "gcc-patches at gcc dot gnu dot org" <gcc-patches at gcc dot gnu dot org>
- Date: Wed, 26 Mar 2014 08:00:44 +0100
- Subject: Re: [PATCH] Fix PR rtl-optimization/pr60663
- Authentication-results: sourceware.org; auth=none
- References: <CACgzC7BAgdoKmKdNTnqoB+WgihgrAhSc4=FdXVUCRB7Pw6T+6Q at mail dot gmail dot com>
- Reply-to: Jakub Jelinek <jakub at redhat dot com>
On Wed, Mar 26, 2014 at 02:16:16PM +0800, Zhenqiang Chen wrote:
> The patch checks the number of the expected operands in
> ASM_OPERANDS_TEMPLATE with the same logic as it in output_asm_insn to
> make sure the ASM_OPERANDS are legal.
>
> Bootstrap and no make check regression on X86-64 and ARM chromebook.
>
> OK for trunk?
No, this is very wrong. How many operands you refer to and how many times
to each is completely unrelated to the fact that CSE should never modify
asm insns to drop some of the outputs.
You can refer to no operands at all and still it would be invalid to drop
the outputs, or you can have output template like "%0 %0 %0 %0 %0 ... million times %0".
Jakub