[Bug target/64507] New: SH inlined builtin strncmp doesn't return 0 for 0 length

[Bug target/64507] New: SH inlined builtin strncmp doesn't return 0 for 0 length

[chrbr at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64507

            Bug ID: 64507
           Summary: SH inlined builtin strncmp doesn't return 0 for 0
                    length
           Product: gcc
           Version: unknown
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: target
          Assignee: unassigned at gcc dot gnu.org
          Reporter: chrbr at gcc dot gnu.org

Created attachment 34386
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=34386&action=edit
test case

Thought the check was emitted from the builtin generic prologue like for the
constant lengths. which was not :-(

[Bug target/64507] SH inlined builtin strncmp doesn't return 0 for 0 length

[chrbr at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64507

--- Comment #1 from chrbr at gcc dot gnu.org ---
Author: chrbr
Date: Tue Jan  6 11:59:09 2015
New Revision: 219257

URL: https://gcc.gnu.org/viewcvs?rev=219257&root=gcc&view=rev
Log:
PR target/64507
* config/sh/sh-mem.cc (sh_expand_cmpnstr): Check 0 length.


Added:
    trunk/gcc/testsuite/gcc.target/sh/pr64507.c
Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/config/sh/sh-mem.cc
    trunk/gcc/testsuite/ChangeLog

[Bug target/64507] SH inlined builtin strncmp doesn't return 0 for 0 length

[chrbr at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64507

--- Comment #2 from chrbr at gcc dot gnu.org ---
Author: chrbr
Date: Tue Jan  6 12:22:51 2015
New Revision: 219258

URL: https://gcc.gnu.org/viewcvs?rev=219258&root=gcc&view=rev
Log:
PR target/64507
* config/sh/sh-mem.cc (sh_expand_cmpnstr): Check 0 length.


Added:
    branches/gcc-4_9-branch/gcc/testsuite/gcc.target/sh/pr64507.c
Modified:
    branches/gcc-4_9-branch/gcc/ChangeLog
    branches/gcc-4_9-branch/gcc/config/sh/sh-mem.cc
    branches/gcc-4_9-branch/gcc/testsuite/ChangeLog

[Bug target/64507] SH inlined builtin strncmp doesn't return 0 for 0 length

[chrbr at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64507

--- Comment #4 from chrbr at gcc dot gnu.org ---
Actually, the code could be slightly better by factorizing the return sequences

        ;; check length
        mov    #0,r0
        ;; goto .l2
        ;; else init r1,r0
        ;; goto .l1
.l2:
        rtx
    nop
.l1:
    rts    
    sub    r1,r0

into
       ;; check length
       mov r1,r0
       ;; goto l1
       ;; else init r1,r0
.l1:        
       rts
       sub   r1,r0

however this creates an unitilized mov (which we don´t care) but the backend
keeps forcing an useless mov #0,r1
>From gcc-bugs-return-472368-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Wed Jan 07 09:37:41 2015
Return-Path: <gcc-bugs-return-472368-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 31750 invoked by alias); 7 Jan 2015 09:37:41 -0000
Mailing-List: contact gcc-bugs-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
List-Id: <gcc-bugs.gcc.gnu.org>
List-Archive: <http://gcc.gnu.org/ml/gcc-bugs/>
List-Post: <mailto:gcc-bugs@gcc.gnu.org>
List-Help: <mailto:gcc-bugs-help@gcc.gnu.org>
Sender: gcc-bugs-owner@gcc.gnu.org
Delivered-To: mailing list gcc-bugs@gcc.gnu.org
Received: (qmail 31710 invoked by uid 48); 7 Jan 2015 09:37:37 -0000
From: "maurits.de.jong at ericsson dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug c/64509] _Generic throws error in unselected generic association
Date: Wed, 07 Jan 2015 09:37:00 -0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: c
X-Bugzilla-Version: 4.9.2
X-Bugzilla-Keywords:
X-Bugzilla-Severity: normal
X-Bugzilla-Who: maurits.de.jong at ericsson dot com
X-Bugzilla-Status: UNCONFIRMED
X-Bugzilla-Priority: P3
X-Bugzilla-Assigned-To: unassigned at gcc dot gnu.org
X-Bugzilla-Target-Milestone: ---
X-Bugzilla-Flags:
X-Bugzilla-Changed-Fields:
Message-ID: <bug-64509-4-ugDkHeLB4W@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-64509-4@http.gcc.gnu.org/bugzilla/>
References: <bug-64509-4@http.gcc.gnu.org/bugzilla/>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 7bit
X-Bugzilla-URL: http://gcc.gnu.org/bugzilla/
Auto-Submitted: auto-generated
MIME-Version: 1.0
X-SW-Source: 2015-01/txt/msg00362.txt.bz2
Content-length: 645

https://gcc.gnu.org/bugzilla/show_bug.cgi?idd509

--- Comment #4 from Martien de Jong <maurits.de.jong at ericsson dot com> ---
I understand. What a shallow and ugly feature it is then. I think it can only
usefully be employed using a preprocessor macro, yet it is part of the
expression syntax. It is meant to resolve type issues, yet all variants
should be type-sane. You can indeed implement something like tgmath.h with it,
but only if the types can be implictly converted.

BTW, I notice that

./gcc/ginclude/tgmath.h

uses

__builtin_classify_type
__builtin_types_compatible_p
__builtin_choose_expr

Will it move to using _Generic() ?