From: Andrew Morton <akpm@google.com>
To: Mimi Zohar <zohar@linux.vnet.ibm.com>
Cc: linux-security-module@vger.kernel.org,
Andy Shevchenko <andy.shevchenko@gmail.com>,
Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>,
David Safford <safford@watson.ibm.com>,
"Nicholas A. Bellinger" <nab@linux-iscsi.org>,
target-devel@vger.kernel.org, linux-kernel@vger.kernel.org,
Andrew Morton <akpm@linux-foundation.org>
Subject: Re: [RFC][PATCH 1/5] lib: add unpack_hex_byte()
Date: Mon, 19 Sep 2011 14:19:11 -0700 [thread overview]
Message-ID: <20110919141911.e13569ab.akpm@google.com> (raw)
In-Reply-To: <1316177430-13167-1-git-send-email-zohar@linux.vnet.ibm.com>
On Fri, 16 Sep 2011 08:50:26 -0400
Mimi Zohar <zohar@linux.vnet.ibm.com> wrote:
> Since converting 2 ascii hex digits into a byte with error checks
> is commonly used, we can replace multiple hex_to_bin() calls with
> a single call to unpack_hex_byte().
>
> Changelog:
> - Error checking added based on Tetsuo Handa's patch.
> - Moved the hex2bin code here, making it into a static inline function.
> (Andy Shevchenko's request.)
Why, on earth? Not for performance reasons - the code is already quite
inefficient.
> Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
> ---
> include/linux/kernel.h | 21 +++++++++++++++++++++
> 1 files changed, 21 insertions(+), 0 deletions(-)
>
> diff --git a/include/linux/kernel.h b/include/linux/kernel.h
> index 46ac9a5..d8ea13b 100644
> --- a/include/linux/kernel.h
> +++ b/include/linux/kernel.h
> @@ -385,6 +385,27 @@ extern int hex_to_bin(char ch);
> extern void hex2bin(u8 *dst, const char *src, size_t count);
>
> /*
> + * unpack_hex_byte - convert 2 asii hex digits into a byte
> + * @byte: binary result
> + * @buf: ascii hexadecimal byte string
> + */
> +static inline bool unpack_hex_byte(u8 *byte, const char *buf)
> +{
> + int hi, lo;
> +
> + hi = hex_to_bin(buf[0]);
> + if (hi < 0)
> + return false;
> +
> + lo = hex_to_bin(buf[1]);
> + if (lo < 0)
> + return false;
> +
> + *byte = (hi << 4) | lo;
> + return true;
> +}
Putting the return value into *byte is inefficient too - it forces the
compiler to place the return value into addressible storage (ie: into a
stack temporary). Maybe the compiler can avoid doing that if the code
is inlined, but the code shouldn't be inlined :(
The return value is undocumented.
Why not copy the hex2bin return value semantics rather than creating a
new scheme? Returning "bool false" in response to an error is very
unconventional and unexpected. Kernel code returns a negative value on
error.
Wouldn't it be better to fix hex2bin() so that it returns -1 on error?
Then the above function becomes a one-liner:
return hex2bin(dst, src, 2);
Finally, the name is poor. It starts with "unpack_", so it belongs to
the "unpack" subsystem. There's no such thing. Something like
hex_byte_to_bin() would be better.
next prev parent reply other threads:[~2011-09-19 21:19 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-09-16 12:50 [RFC][PATCH 1/5] lib: add unpack_hex_byte() Mimi Zohar
2011-09-16 12:50 ` [RFC][PATCH 2/5] lib: add error checking to hex2bin Mimi Zohar
2011-09-16 13:13 ` Andy Shevchenko
2011-09-19 21:20 ` Andrew Morton
2011-09-16 12:50 ` [RFC][PATCH 3/5] trusted-keys: check hex2bin result Mimi Zohar
2011-09-16 12:50 ` [RFC][PATCH 4/5] encrypted-keys: " Mimi Zohar
2011-09-16 12:50 ` [RFC][PATCH 5/5] target: " Mimi Zohar
2011-09-16 14:07 ` Tetsuo Handa
2011-09-16 20:21 ` Nicholas A. Bellinger
2011-09-19 11:19 ` Mimi Zohar
2011-09-16 13:11 ` [RFC][PATCH 1/5] lib: add unpack_hex_byte() Andy Shevchenko
2011-09-19 21:19 ` Andrew Morton [this message]
2011-09-19 22:35 ` Mimi Zohar
2011-09-19 22:38 ` Andrew Morton
2011-09-20 6:04 ` Andy Shevchenko
2011-09-20 6:26 ` Andrew Morton
2011-09-23 10:47 ` [PATCH 1/2] kernel.h: rename pack_hex_byte to hex_pack_byte Andy Shevchenko
2011-09-23 10:47 ` [PATCH 2/2] wireless: at76c50x: use native hex_pack_byte() method Andy Shevchenko
2011-09-27 11:51 ` [PATCHv2] " Andy Shevchenko
2011-09-27 12:01 ` [PATCHv2.1] " Andy Shevchenko
2011-09-23 20:18 ` [PATCH 1/2] kernel.h: rename pack_hex_byte to hex_pack_byte Andrew Morton
2011-09-27 10:51 ` Andy Shevchenko
2011-09-27 17:32 ` Andrew Morton
2011-10-10 14:33 ` [PATCHv2 1/5] lib: rename pack_hex_byte to hex_byte_pack Andy Shevchenko
2011-10-10 14:33 ` [PATCHv2 2/5] kgdb: follow " Andy Shevchenko
2011-10-10 14:55 ` Jesper Nilsson
2011-10-10 14:33 ` [PATCHv2 3/5] security: " Andy Shevchenko
2011-10-10 14:33 ` [PATCHv2 4/5] fat: " Andy Shevchenko
2011-10-10 20:02 ` OGAWA Hirofumi
2011-10-10 14:33 ` [PATCHv2 5/5] wireless: at76c50x: " Andy Shevchenko
2011-10-11 13:55 ` John W. Linville
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=20110919141911.e13569ab.akpm@google.com \
--to=akpm@google.com \
--cc=akpm@linux-foundation.org \
--cc=andy.shevchenko@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-security-module@vger.kernel.org \
--cc=nab@linux-iscsi.org \
--cc=penguin-kernel@i-love.sakura.ne.jp \
--cc=safford@watson.ibm.com \
--cc=target-devel@vger.kernel.org \
--cc=zohar@linux.vnet.ibm.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 NNTP newsgroup(s).