From mboxrd@z Thu Jan  1 00:00:00 1970
From: Dibyendu Majumdar <mobile@majumdar.org.uk>
Subject: Re: sparse-llvm array size computation issue
Date: Tue, 28 Mar 2017 22:14:43 +0100
Message-ID: <CACXZuxcozbUcY9Bot65eSXjSxCDTzVbGSqTqJwgRCit24zisWA@mail.gmail.com>
References: <CACXZuxdBRB+mvRJf3VGLK2Po8w9gvSXkjTnt3=UCfoa2Wqm5Ew@mail.gmail.com>
 <CACXZuxd5cgfdE-cPZ4mkLnzy=MNn5B9EJNAYaazV3VV9o1_QKQ@mail.gmail.com>
 <CAExDi1QupTRLhPDh8PmieMQh1zr2nV59JjjQHKd4r9ab4nTu0w@mail.gmail.com> <CACXZuxcnxYvqM8=kDe=AkWuYB=TmU9mwd9gKc9_veZhbD0z1Xw@mail.gmail.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Return-path: <linux-sparse-owner@vger.kernel.org>
Received: from mail-it0-f46.google.com ([209.85.214.46]:36632 "EHLO
        mail-it0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
        with ESMTP id S1752401AbdC1VOp (ORCPT
        <rfc822;linux-sparse@vger.kernel.org>);
        Tue, 28 Mar 2017 17:14:45 -0400
Received: by mail-it0-f46.google.com with SMTP id e75so71016047itd.1
        for <linux-sparse@vger.kernel.org>; Tue, 28 Mar 2017 14:14:44 -0700 (PDT)
In-Reply-To: <CACXZuxcnxYvqM8=kDe=AkWuYB=TmU9mwd9gKc9_veZhbD0z1Xw@mail.gmail.com>
Sender: linux-sparse-owner@vger.kernel.org
List-Id: linux-sparse@vger.kernel.org
To: Luc Van Oostenryck <luc.vanoostenryck@gmail.com>
Cc: Linux-Sparse <linux-sparse@vger.kernel.org>

Hi Luc,

On 28 March 2017 at 22:06, Dibyendu Majumdar <mobile@majumdar.org.uk> wrote:
> On 28 March 2017 at 21:49, Luc Van Oostenryck
> <luc.vanoostenryck@gmail.com> wrote:
>> On Tue, Mar 28, 2017 at 10:41 PM, Dibyendu Majumdar
>> <mobile@majumdar.org.uk> wrote:
>>> On 28 March 2017 at 21:25, Dibyendu Majumdar <mobile@majumdar.org.uk> wrote:
>>>> I am investigating an issue with following test case:
>>>>
>>>> int main(void)
>>>> {
>>>> const char *names[] = {
>>>> "dibyendu",
>>>> "majumdar",
>>>> NULL,
>>>> };
>>>> return 0;
>>>> }
>>>>
>>>> The issue here is that sparse-llvm thinks the array size is zero:
>>>>
>>>> define i32 @main() {
>>>> L0:
>>>>   %names_0000026DE9D2DDA8. = alloca [0 x i8*]
>>>>
>>>> It appears that sparse is correctly calculating the size of the array
>>>> in examine_node_type() in symbol.c, but by the time the symbol gets to
>>>> sparse-llvm the bit_size is somehow changed to -1. I haven't yet
>>>> tracked down where this is happening.
>>>>
>>>
>>> Looks like the computed bit_size is being held on the SYM_NODE but
>>> sparse-llvm looks as the bit_size field in the SYM_ARRAY node. Does
>>> this sound like a problem - i.e. somehow the SYM_ARRAY is not getting
>>> its size set?
>>

I think that there is a bug in examine_node_type() in symbol.c - it
should set the base_type's bit_size perhaps? See the line marked as
FIX below,

/* Unsized array? The size might come from the initializer.. */
if (bit_size < 0 && base_type->type == SYM_ARRAY) {
struct expression *initializer = get_symbol_initializer(sym);
if (initializer) {
struct symbol *node_type = base_type->ctype.base_type;
int count = count_array_initializer(S, node_type, initializer);

if (node_type && node_type->bit_size >= 0)
bit_size = array_element_offset(S->C->target, node_type->bit_size, count);
base_type->bit_size = bit_size;   /*** FIX set base_type->bit_size ***/
}
}

Thanks and Regards
Dibyendu