This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [PATCH 00/21] PowerPC: multiarch support for PPC64
- From: Adhemerval Zanella <azanella at linux dot vnet dot ibm dot com>
- To: libc-alpha at sourceware dot org
- Date: Thu, 19 Dec 2013 12:23:03 -0200
- Subject: Re: [PATCH 00/21] PowerPC: multiarch support for PPC64
- Authentication-results: sourceware.org; auth=none
- References: <52A70916 dot 2020703 at linux dot vnet dot ibm dot com> <mvm61ql2apz dot fsf at hawking dot suse dot de>
On 19-12-2013 12:10, Andreas Schwab wrote:
> ../sysdeps/powerpc/powerpc64/power7/memrchr.S: Assembler messages:
> ../sysdeps/powerpc/powerpc64/power7/memrchr.S:25: Error: non-constant expression in ".if" statement
> ../sysdeps/powerpc/powerpc64/power7/memrchr.S:25: Error: non-constant expression in ".if" statement
> ../sysdeps/powerpc/powerpc64/power7/memrchr.S:25: Error: non-constant expression in ".if" statement
> ../sysdeps/powerpc/powerpc64/power7/memrchr.S:25: Error: non-constant expression in ".if" statement
>
> Andreas.
>
Thanks for spotting it and I think you built with --enable-profile, right? I don't think this is
really related to my patch, I tried to build sysdeps/powerpc/powerpc64/power7/memrchr.S with
-DPROF and I saw the same issue. I will investigate this today, but looks like this code is
triggering it:
66 .macro SAVE_ARG NARG
67 .if \NARG
68 SAVE_ARG \NARG-1
69 std 2+\NARG,call_mcount_parm_offset-8+8*(\NARG)(1)
70 .endif
71 .endm
Has the .if command been deprecated or is something not right in this code?