From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-kernel-owner@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
	id S933687AbbDILxy (ORCPT <rfc822;w@1wt.eu>);
	Thu, 9 Apr 2015 07:53:54 -0400
Received: from cantor2.suse.de ([195.135.220.15]:41642 "EHLO mx2.suse.de"
	rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
	id S1755557AbbDILxw (ORCPT <rfc822;linux-kernel@vger.kernel.org>);
	Thu, 9 Apr 2015 07:53:52 -0400
Date: Thu, 9 Apr 2015 13:53:47 +0200
From: Jan Kara <jack@suse.cz>
To: Fabian Frederick <fabf@skynet.be>
Cc: linux-kernel@vger.kernel.org, Jan Kara <jack@suse.cz>
Subject: Re: [PATCH V2 4/9 linux-next] udf: improve error management in
 udf_CS0toUTF8()
Message-ID: <20150409115347.GJ18044@quack.suse.cz>
References: <1428521039-18491-1-git-send-email-fabf@skynet.be>
 <1428521039-18491-5-git-send-email-fabf@skynet.be>
 <20150409083607.GD18044@quack.suse.cz>
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="OBd5C1Lgu00Gd/Tn"
Content-Disposition: inline
In-Reply-To: <20150409083607.GD18044@quack.suse.cz>
User-Agent: Mutt/1.5.21 (2010-09-15)
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org


--OBd5C1Lgu00Gd/Tn
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline

On Thu 09-04-15 10:36:07, Jan Kara wrote:
> On Wed 08-04-15 21:23:54, Fabian Frederick wrote:
> > udf_CS0toUTF8() now returns -EINVAL on error.
> > udf_load_pvoldesc() and udf_get_filename() do the same.
> > 
> > Suggested-by: Jan Kara <jack@suse.cz>
> > Signed-off-by: Fabian Frederick <fabf@skynet.be>
>   After some thought it's probably better API that udf_CS0toUTF8() simply
> returns 0 when 0-lenght string is provided as an input (there's nothing
> invalid in that from the point of view of that conversion functions). We
> cannot actually ever call udf_CS0toUTF8() with such string currently but I
> think it's better for future. So I've changed this and merged the patch.
  Just for reference, attached is the resulting patch.

								Honza

> > ---
> >  fs/udf/super.c   | 23 ++++++++++++++---------
> >  fs/udf/unicode.c |  9 +++++----
> >  2 files changed, 19 insertions(+), 13 deletions(-)
> > 
> > diff --git a/fs/udf/super.c b/fs/udf/super.c
> > index 6299f34..c6a8f5f 100644
> > --- a/fs/udf/super.c
> > +++ b/fs/udf/super.c
> > @@ -927,17 +927,22 @@ static int udf_load_pvoldesc(struct super_block *sb, sector_t block)
> >  #endif
> >  	}
> >  
> > -	if (!udf_build_ustr(instr, pvoldesc->volIdent, 32))
> > -		if (udf_CS0toUTF8(outstr, instr)) {
> > -			strncpy(UDF_SB(sb)->s_volume_ident, outstr->u_name,
> > -				outstr->u_len > 31 ? 31 : outstr->u_len);
> > -			udf_debug("volIdent[] = '%s'\n",
> > -				  UDF_SB(sb)->s_volume_ident);
> > -		}
> > +	if (!udf_build_ustr(instr, pvoldesc->volIdent, 32)) {
> > +		ret = udf_CS0toUTF8(outstr, instr);
> > +		if (ret < 0)
> > +			goto out_bh;
> > +
> > +		strncpy(UDF_SB(sb)->s_volume_ident, outstr->u_name,
> > +			outstr->u_len > 31 ? 31 : outstr->u_len);
> > +		udf_debug("volIdent[] = '%s'\n", UDF_SB(sb)->s_volume_ident);
> > +	}
> >  
> >  	if (!udf_build_ustr(instr, pvoldesc->volSetIdent, 128))
> > -		if (udf_CS0toUTF8(outstr, instr))
> > -			udf_debug("volSetIdent[] = '%s'\n", outstr->u_name);
> > +		ret = udf_CS0toUTF8(outstr, instr);
> > +		if (ret < 0)
> > +			goto out_bh;
> > +
> > +		udf_debug("volSetIdent[] = '%s'\n", outstr->u_name);
> >  
> >  	ret = 0;
> >  out_bh:
> > diff --git a/fs/udf/unicode.c b/fs/udf/unicode.c
> > index 41c3bef..9008a36 100644
> > --- a/fs/udf/unicode.c
> > +++ b/fs/udf/unicode.c
> > @@ -89,7 +89,7 @@ static void udf_build_ustr_exact(struct ustr *dest, dstring *ptr, int exactsize)
> >   * 				both of type "struct ustr *"
> >   *
> >   * POST-CONDITIONS
> > - *	<return>		Zero on success.
> > + *	<return>		>= 0 on success.
> >   *
> >   * HISTORY
> >   *	November 12, 1997 - Andrew E. Mileski
> > @@ -104,7 +104,7 @@ int udf_CS0toUTF8(struct ustr *utf_o, const struct ustr *ocu_i)
> >  	ocu_len = ocu_i->u_len;
> >  	if (ocu_len == 0) {
> >  		memset(utf_o, 0, sizeof(struct ustr));
> > -		return 0;
> > +		return -EINVAL;
> >  	}
> >  
> >  	cmp_id = ocu_i->u_cmpID;
> > @@ -112,7 +112,7 @@ int udf_CS0toUTF8(struct ustr *utf_o, const struct ustr *ocu_i)
> >  		memset(utf_o, 0, sizeof(struct ustr));
> >  		pr_err("unknown compression code (%d) stri=%s\n",
> >  		       cmp_id, ocu_i->u_name);
> > -		return 0;
> > +		return -EINVAL;
> >  	}
> >  
> >  	ocu = ocu_i->u_name;
> > @@ -350,7 +350,8 @@ int udf_get_filename(struct super_block *sb, uint8_t *sname, int slen,
> >  
> >  	udf_build_ustr_exact(unifilename, sname, slen);
> >  	if (UDF_QUERY_FLAG(sb, UDF_FLAG_UTF8)) {
> > -		if (!udf_CS0toUTF8(filename, unifilename)) {
> > +		ret = udf_CS0toUTF8(filename, unifilename);
> > +		if (ret < 0) {
> >  			udf_debug("Failed in udf_get_filename: sname = %s\n",
> >  				  sname);
> >  			goto out2;
> > -- 
> > 1.9.1
> > 
> -- 
> Jan Kara <jack@suse.cz>
> SUSE Labs, CR
-- 
Jan Kara <jack@suse.cz>
SUSE Labs, CR

--OBd5C1Lgu00Gd/Tn
Content-Type: text/x-patch; charset=us-ascii
Content-Disposition: attachment; filename="0004-udf-improve-error-management-in-udf_CS0toUTF8.patch"

>>From 2fd89b7d8600488b00eb2d826750f3800fd5fd9b Mon Sep 17 00:00:00 2001
From: Fabian Frederick <fabf@skynet.be>
Date: Wed, 8 Apr 2015 21:23:54 +0200
Subject: [PATCH 4/8] udf: improve error management in udf_CS0toUTF8()

udf_CS0toUTF8() now returns -EINVAL on error.
udf_load_pvoldesc() and udf_get_filename() do the same.

Suggested-by: Jan Kara <jack@suse.cz>
Signed-off-by: Fabian Frederick <fabf@skynet.be>
Signed-off-by: Jan Kara <jack@suse.cz>
---
 fs/udf/super.c   | 23 ++++++++++++++---------
 fs/udf/unicode.c |  7 ++++---
 2 files changed, 18 insertions(+), 12 deletions(-)

diff --git a/fs/udf/super.c b/fs/udf/super.c
index 6299f341967b..c6a8f5f79443 100644
--- a/fs/udf/super.c
+++ b/fs/udf/super.c
@@ -927,17 +927,22 @@ static int udf_load_pvoldesc(struct super_block *sb, sector_t block)
 #endif
 	}
 
-	if (!udf_build_ustr(instr, pvoldesc->volIdent, 32))
-		if (udf_CS0toUTF8(outstr, instr)) {
-			strncpy(UDF_SB(sb)->s_volume_ident, outstr->u_name,
-				outstr->u_len > 31 ? 31 : outstr->u_len);
-			udf_debug("volIdent[] = '%s'\n",
-				  UDF_SB(sb)->s_volume_ident);
-		}
+	if (!udf_build_ustr(instr, pvoldesc->volIdent, 32)) {
+		ret = udf_CS0toUTF8(outstr, instr);
+		if (ret < 0)
+			goto out_bh;
+
+		strncpy(UDF_SB(sb)->s_volume_ident, outstr->u_name,
+			outstr->u_len > 31 ? 31 : outstr->u_len);
+		udf_debug("volIdent[] = '%s'\n", UDF_SB(sb)->s_volume_ident);
+	}
 
 	if (!udf_build_ustr(instr, pvoldesc->volSetIdent, 128))
-		if (udf_CS0toUTF8(outstr, instr))
-			udf_debug("volSetIdent[] = '%s'\n", outstr->u_name);
+		ret = udf_CS0toUTF8(outstr, instr);
+		if (ret < 0)
+			goto out_bh;
+
+		udf_debug("volSetIdent[] = '%s'\n", outstr->u_name);
 
 	ret = 0;
 out_bh:
diff --git a/fs/udf/unicode.c b/fs/udf/unicode.c
index 41c3bef1d226..35cc9477b066 100644
--- a/fs/udf/unicode.c
+++ b/fs/udf/unicode.c
@@ -89,7 +89,7 @@ static void udf_build_ustr_exact(struct ustr *dest, dstring *ptr, int exactsize)
  * 				both of type "struct ustr *"
  *
  * POST-CONDITIONS
- *	<return>		Zero on success.
+ *	<return>		>= 0 on success.
  *
  * HISTORY
  *	November 12, 1997 - Andrew E. Mileski
@@ -112,7 +112,7 @@ int udf_CS0toUTF8(struct ustr *utf_o, const struct ustr *ocu_i)
 		memset(utf_o, 0, sizeof(struct ustr));
 		pr_err("unknown compression code (%d) stri=%s\n",
 		       cmp_id, ocu_i->u_name);
-		return 0;
+		return -EINVAL;
 	}
 
 	ocu = ocu_i->u_name;
@@ -350,7 +350,8 @@ int udf_get_filename(struct super_block *sb, uint8_t *sname, int slen,
 
 	udf_build_ustr_exact(unifilename, sname, slen);
 	if (UDF_QUERY_FLAG(sb, UDF_FLAG_UTF8)) {
-		if (!udf_CS0toUTF8(filename, unifilename)) {
+		ret = udf_CS0toUTF8(filename, unifilename);
+		if (ret < 0) {
 			udf_debug("Failed in udf_get_filename: sname = %s\n",
 				  sname);
 			goto out2;
-- 
2.1.4


--OBd5C1Lgu00Gd/Tn--