svn commit: trunk/uClibc/libcrypt

Denys Vlasenko vda.linux at googlemail.com
Thu Jun 12 11:02:52 UTC 2008


On Thursday 12 June 2008 10:43, Peter Kjellerstedt wrote:
> > -----Original Message-----
> > From: uclibc-cvs-bounces at uclibc.org [mailto:uclibc-cvs-
> > bounces at uclibc.org] On Behalf Of vda at uclibc.org
> > Sent: den 11 juni 2008 18:12
> > To: uclibc-cvs at uclibc.org
> > Subject: svn commit: trunk/uClibc/libcrypt
> >
> > Author: vda
> > Date: 2008-06-11 09:11:29 -0700 (Wed, 11 Jun 2008)
> > New Revision: 22296
> >
> > Log:
> > libcrypt/md5.c:
> > Make sp, ep and p variables automatic.
> > There is no point in having them static,
> > they never use their values retained from previous call.
> > Replace strlen(__md5__magic) etc by compile-time constants.
> >
> >     text           data     bss     dec     hex filename
> > -   1957              0     132    2089     829 libcrypt/md5.o
> > +   1905              0     120    2025     7e9 libcrypt/md5.o
> >
> >
> >
> > Modified:
> >    trunk/uClibc/libcrypt/md5.c
> >
> >
> > Changeset:
> > Modified: trunk/uClibc/libcrypt/md5.c
> > ===================================================================
> > --- trunk/uClibc/libcrypt/md5.c 2008-06-11 15:43:19 UTC (rev 22295)
> > +++ trunk/uClibc/libcrypt/md5.c 2008-06-11 16:11:29 UTC (rev 22296)
> > @@ -95,8 +95,11 @@
> >  static void __md5_Transform __P((u_int32_t [4], const unsigned char [64]));
> >
> >
> > -static const unsigned char __md5__magic[] = "$1$";     /* This string is magic for this algorithm.  Having
> > -                                                  it this way, we can get better later on */
> > +#define MD5_MAGIC_STR "$1$"
> > +#define MD5_MAGIC_LEN (sizeof(MD5_MAGIC_STR) - 1)
> > +static const unsigned char __md5__magic[] = MD5_MAGIC_STR;
> 
> Shouldn't the comment below be _before_ the string above (even
> before the two defines I would say)? As it now stands at least
> I read it as referring to __md5_itoa64, which I doubt was how
> it was intended...
> 
> > +/* This string is magic for this algorithm.  Having
> > + * it this way, we can get better later on */
> >  static const unsigned char __md5_itoa64[] =            /* 0 ... 63 => ascii - 64 */
> >         "./0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz";
> 
> //Peter

I will remove the comment altogether, it does not seem
to provide any useful information. Knowing that "$1$" is a magic
for md5 is assumed from anyone trying to modify this file.
--
vda



More information about the uClibc mailing list