mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Alex Pennace <alex@pennace.org>
To: Amit Kucheria <amitk@ittc.ku.edu>
Cc: linux-kernel@vger.kernel.org
Subject: Re: using objdump to debug some n/w code in kernel
Date: Sun, 28 Oct 2001 01:39:39 -0500	[thread overview]
Message-ID: <20011028013939.C26005@buick.pennace.org> (raw)
In-Reply-To: <Pine.LNX.4.33.0110280124540.16994-100000@havoc.ittc.ku.edu>
In-Reply-To: <Pine.LNX.4.33.0110280124540.16994-100000@havoc.ittc.ku.edu>

On Sun, Oct 28, 2001 at 01:25:06AM -0500, Amit Kucheria wrote:
> I am using the 'makelst' script in the scripts directory to generate
> interleaved source code and assembly listing. This script uses 'objdump'
> 
> Now my problem is that the original C code and the .lst files dont seem to
> match. They are listed below.
[...]
> Can anybody throw any light on this 'phenomena' ?
> BTW, the string ': "memory");' seems to be repeating in a lot of
> places in the .lst file. Is there something about objdump that i havent
> read up on.
> 
> Regards,
> Amit
> 
> Original code:
> -------------
> /* .......
>    So host > network > gateway entries.
> */
>     for (i=0; i < MAX_ROUTING_ENTRIES; i++)
>     {
>         /* copy the routing entry into our local variable..just to be safe
>            We will optimize later by doing without this copy */
>         memcpy(&tmp_entry, ptr_route, sizeof(struct routing_entry));
> 
>         switch(tmp_entry.rt_flag)
>         {
>             case 'H':
>                 /* if dest ip is a host entry */
> ----- end original code -----------------
> 
> Code generated using 'makelst:
> -------------------------------
> /* ....
>    So host > network > gateway entries.
> */
>     for (i=0; i < MAX_ROUTING_ENTRIES; i++)
> c01b3206:       31 ed                   xor    %ebp,%ebp
> c01b3208:       8d 5c 24 48             lea    0x48(%esp,1),%ebx
> c01b320c:       8d 54 24 68             lea    0x68(%esp,1),%edx
> c01b3210:       8b 49 5c                mov    0x5c(%ecx),%ecx
> c01b3213:       89 4c 24 18             mov    %ecx,0x18(%esp,1)
> c01b3217:       89 54 24 10             mov    %edx,0x10(%esp,1)
> c01b321b:       8d 4c 24 28             lea    0x28(%esp,1),%ecx
> c01b321f:       89 4c 24 14             mov    %ecx,0x14(%esp,1)
> c01b3223:       90                      nop
>         : "memory");
> {
>         int d0, d1, d2;
>         switch (n % 4) {
>                 case 0: COMMON(""); return to;
[...]

memcpy is a macro. See linux/include/asm-i386/string.h.

      reply	other threads:[~2001-10-28  6:39 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2001-10-28  6:25 Amit Kucheria
2001-10-28  6:39 ` Alex Pennace [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20011028013939.C26005@buick.pennace.org \
    --to=alex@pennace.org \
    --cc=amitk@ittc.ku.edu \
    --cc=linux-kernel@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®