Re: on gdb's sizeof versus gcc's sizeof

Path	csiph.com!aioe.org!.POSTED.IXP4sbumLrdkG2n6/syALg.user.gioia.aioe.org!not-for-mail
From	Louis Valence <lvalence@example.com>
Newsgroups	comp.lang.c
Subject	Re: on gdb's sizeof versus gcc's sizeof
Date	Sat, 28 Sep 2019 16:57:37 -0300
Organization	Aioe.org NNTP Server
Lines	73
Message-ID	<86pnjki47i.fsf@example.com> (permalink)
References	<867e5vkgw2.fsf@example.com> <a03jF.4165$hU2.239@fx20.iad> <lnpnjm4y80.fsf@kst-u.example.com>
NNTP-Posting-Host	IXP4sbumLrdkG2n6/syALg.user.gioia.aioe.org
Mime-Version	1.0
Content-Type	text/plain
X-Complaints-To	abuse@aioe.org
User-Agent	Gnus/5.13 (Gnus v5.13) Emacs/24.3 (windows-nt)
X-Notice	Filtered by postfilter v. 0.9.2
Cancel-Lock	sha1:0rY+mjXBmIfboa2AMSq4rIotgQU=
Xref	csiph.com comp.lang.c:146399

Show key headers only | View raw

Keith Thompson <kst-u@mib.org> writes:

> scott@slp53.sl.home (Scott Lurndal) writes:
>> Louis Valence <lvalence@example.com> writes:
>>>Can you explain this?  I don't get it.
>>>
>>>%cat sizeof-long.c
>>>#include <stdio.h>
>>>
>>>int main() {
>>>  printf("sizeof unsigned long: %ld\n", sizeof (unsigned long));
>>>  printf("sizeof long: %ld\n", sizeof (long));
>>>}
>>>
>>>%./sizeof-long
>>>sizeof unsigned long: 8
>>>sizeof long: 8
>>>
>>>%gdb
>>>GNU gdb (Ubuntu 8.0.1-0ubuntu1) 8.0.1
>>>
>>>[...]
>>>
>>>(gdb) p sizeof (unsigned long)
>>>$1 = 4
>>
>> You may want to use a 64-bit build of GDB instead of the 32-bit:
>>
>> (gdb) print sizeof(unsigned long)
>> $1 = 0x8
>> (gdb) quit
>
> I don't believe that's the issue.
>
> Were you debugging a program when you got that result, or did you invoke
> gdb with no arguments as Louis and I both did?
>
> On my system:
>
>     $ echo 'print sizeof (long)' | gdb
>     GNU gdb (Ubuntu 8.1-0ubuntu3.1) 8.1.0.20180409-git
>     [...]
>     This GDB was configured as "x86_64-linux-gnu".
>     [...]
>     (gdb) $1 = 4
>     (gdb) quit
>     $
>
> As far as I can tell, gdb uses the environment of the program it's
> debugging, falling back to some default environment if it's not given
> a program to debug.  In that default -- at least on my system --
> long is 4 bytes.

Thanks for the investigation, Keith.  I think you're correct here.
Here's some evidence.

--8<---------------cut here---------------start------------->8---
%gdb
GNU gdb (Ubuntu 8.0.1-0ubuntu1) 8.0.1
[..]
(gdb) p sizeof (unsigned long)
$1 = 4
(gdb) set architecture
Requires an argument. Valid arguments are i386, i386:x86-64,
i386:x64-32, i8086, i386:intel, i386:x86-64:intel, i386:x64-32:intel,
i386:nacl, i386:x86-64:nacl, i386:x64-32:nacl, auto.
(gdb) set architecture i386:x86-64
The target architecture is assumed to be i386:x86-64
(gdb) p sizeof (unsigned long)
$2 = 8
(gdb) quit
%
--8<---------------cut here---------------end--------------->8---

Back to comp.lang.c | Previous | Next — Previous in thread | Find similar | Unroll thread

Thread

on gdb's sizeof versus gcc's sizeof Louis Valence <lvalence@example.com> - 2019-09-25 22:03 -0300
  Re: on gdb's sizeof versus gcc's sizeof Keith Thompson <kst-u@mib.org> - 2019-09-25 19:22 -0700
  Re: on gdb's sizeof versus gcc's sizeof scott@slp53.sl.home (Scott Lurndal) - 2019-09-26 13:26 +0000
    Re: on gdb's sizeof versus gcc's sizeof Keith Thompson <kst-u@mib.org> - 2019-09-26 13:08 -0700
      Re: on gdb's sizeof versus gcc's sizeof Louis Valence <lvalence@example.com> - 2019-09-28 16:57 -0300

csiph-web