public inbox for binutils@sourceware.org
 help / color / mirror / Atom feed
From: "H. J. Lu" <hjl@lucon.org>
To: Nick Clifton <nickc@redhat.com>
Cc: binutils@sources.redhat.com, jbeulich@novell.com
Subject: Re: PATCH: Fix buffer overflow in gas
Date: Tue, 02 May 2006 14:20:00 -0000	[thread overview]
Message-ID: <20060502142031.GA10660@lucon.org> (raw)
In-Reply-To: <20060502133130.GB10201@lucon.org>

On Tue, May 02, 2006 at 06:31:30AM -0700, H. J. Lu wrote:
> On Tue, May 02, 2006 at 10:47:27AM +0100, Nick Clifton wrote:
> > Hi H. J.
> > 
> > >There are some potential buffer overflows in gas. 8byte isn't enough
> > >to hold a negative byte. This patch fixes them. Also we should use
> > >snprintf instead of sprintf.
> > 
> > Did you test this patch ?  if so, please could you say how.
> 
> You can put some none ascii char in assembly code.
> 
> > 
> > >2006-05-01  H.J. Lu  <hongjiu.lu@intel.com>
> > >
> > >	* config/tc-i386.c (output_invalid_buf): Change size to 16.
> > >	* config/tc-tic30.c (output_invalid_buf): Likewise.
> > >
> > >	* config/tc-i386.c (output_invalid): Use snprintf instead of
> > >	sprintf.
> > >	* config/tc-ia64.c (declare_register_set): Likewise.
> > >	(emit_one_bundle): Likewise.
> > >	(check_dependencies): Likewise.
> > >	* config/tc-tic30.c (output_invalid): Likewise.
> > 
> > Assuming that you have tested the targets involved and that there were 
> > no regressions then this patch is approved.
> 
> Done.
> 

Jan suggested we should cast int to unsigned char.

foo.s:20: Error: invalid character (0xd6) in mnemonic

is better than

foo.s:20: Error: invalid character (0xfffffffd6) in mnemonic

I will check it in.

H.J.
---
2006-05-02  H.J. Lu  <hongjiu.lu@intel.com>
	    Jan Beulich  <jbeulich@novell.com>

	* config/tc-i386.c (output_invalid_buf): Change size for
	unsigned char.
	* config/tc-tic30.c (output_invalid_buf): Likewise.

	* config/tc-i386.c (output_invalid): Cast none-ascii char to
	unsigned char.
	* config/tc-tic30.c (output_invalid): Likewise.

--- gas/config/tc-i386.c.buf	2006-05-02 06:50:00.000000000 -0700
+++ gas/config/tc-i386.c	2006-05-02 07:12:03.000000000 -0700
@@ -5251,7 +5251,7 @@ md_atof (type, litP, sizeP)
   return 0;
 }
 \f
-static char output_invalid_buf[16];
+static char output_invalid_buf[sizeof (unsigned char) * 2 + 6];
 
 static char *
 output_invalid (c)
@@ -5262,7 +5262,7 @@ output_invalid (c)
 	      "'%c'", c);
   else
     snprintf (output_invalid_buf, sizeof (output_invalid_buf),
-	      "(0x%x)", (unsigned) c);
+	      "(0x%x)", (unsigned char) c);
   return output_invalid_buf;
 }
 
--- gas/config/tc-tic30.c.buf	2006-05-02 06:50:00.000000000 -0700
+++ gas/config/tc-tic30.c	2006-05-02 07:12:11.000000000 -0700
@@ -273,7 +273,7 @@ struct tic30_insn
 struct tic30_insn insn;
 static int found_parallel_insn;
 
-static char output_invalid_buf[16];
+static char output_invalid_buf[sizeof (unsigned char) * 2 + 6];
 
 static char *
 output_invalid (char c)
@@ -283,7 +283,7 @@ output_invalid (char c)
 	      "'%c'", c);
   else
     snprintf (output_invalid_buf, sizeof (output_invalid_buf), 
-	      "(0x%x)", (unsigned) c);
+	      "(0x%x)", (unsigned char) c);
   return output_invalid_buf;
 }
 
2006-05-02  H.J. Lu  <hongjiu.lu@intel.com>

	* config/tc-i386.c (output_invalid_buf): Change size for
	unsigned char.
	* config/tc-tic30.c (output_invalid_buf): Likewise.

	* config/tc-i386.c (output_invalid): Cast none-ascii char to
	unsigned char.
	* config/tc-tic30.c (output_invalid): Likewise.

--- gas/config/tc-i386.c.buf	2006-05-02 06:50:00.000000000 -0700
+++ gas/config/tc-i386.c	2006-05-02 07:12:03.000000000 -0700
@@ -5251,7 +5251,7 @@ md_atof (type, litP, sizeP)
   return 0;
 }
 \f
-static char output_invalid_buf[16];
+static char output_invalid_buf[sizeof (unsigned char) * 2 + 6];
 
 static char *
 output_invalid (c)
@@ -5262,7 +5262,7 @@ output_invalid (c)
 	      "'%c'", c);
   else
     snprintf (output_invalid_buf, sizeof (output_invalid_buf),
-	      "(0x%x)", (unsigned) c);
+	      "(0x%x)", (unsigned char) c);
   return output_invalid_buf;
 }
 
--- gas/config/tc-tic30.c.buf	2006-05-02 06:50:00.000000000 -0700
+++ gas/config/tc-tic30.c	2006-05-02 07:12:11.000000000 -0700
@@ -273,7 +273,7 @@ struct tic30_insn
 struct tic30_insn insn;
 static int found_parallel_insn;
 
-static char output_invalid_buf[16];
+static char output_invalid_buf[sizeof (unsigned char) * 2 + 6];
 
 static char *
 output_invalid (char c)
@@ -283,7 +283,7 @@ output_invalid (char c)
 	      "'%c'", c);
   else
     snprintf (output_invalid_buf, sizeof (output_invalid_buf), 
-	      "(0x%x)", (unsigned) c);
+	      "(0x%x)", (unsigned char) c);
   return output_invalid_buf;
 }
 

  reply	other threads:[~2006-05-02 14:20 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-05-01 18:40 H. J. Lu
2006-05-02  9:48 ` Nick Clifton
2006-05-02 13:31   ` H. J. Lu
2006-05-02 14:20     ` H. J. Lu [this message]
2006-05-02 14:35       ` Jan Beulich
2006-05-02 14:58         ` H. J. Lu
2006-05-02 16:53           ` Andreas Schwab

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=20060502142031.GA10660@lucon.org \
    --to=hjl@lucon.org \
    --cc=binutils@sources.redhat.com \
    --cc=jbeulich@novell.com \
    --cc=nickc@redhat.com \
    /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;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).