bug-cvs
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [bug #18530] Sparc SIGBUS in md5 routines


From: Mark D. Baushke
Subject: Re: [bug #18530] Sparc SIGBUS in md5 routines
Date: Thu, 21 Dec 2006 09:39:43 -0800

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Ingolf Steinbach <Ingolf.Steinbach@jena-optronik.de> writes:

> Mark D. Baushke schrieb:
> > RCS file: /cvsroot/cvs/ccvs/src/cvs.h,v
> [...]
> > +typedef union {
> > +    uint32_t int_checksum[4];
> > +    unsigned char char_checksum[16];
> > +} checksum_t;
> 
> <nit_pick>
> Wouldn't it be more consistent to use
>     uint8_t byte_checksum[16];
> here rather than relying on char being 8-bit?
> </nit_pick>
> 
> (But I understand that platforms with non-8-bit chars might not
> be supported anyway.)

Hmmm... you may be right, but the holiday is getting in the way of my
doing much with CVS sources over the next few weeks. 

The patch I put into the source base was intended to have as minimal an
impact as possible in order to fix the problem you reported.

CVS is using 'unsigned char' in most places rather than uint8_t at
present, but I suppose that could be revisited where it makes sense such
as in this case.

        -- Mark
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (FreeBSD)

iD8DBQFFisbfCg7APGsDnFERApEJAJ9c6LvoukHG4Q0TQZjLMd8XgnfH6QCfaSfS
yoVMKfSBP5VMpCePs8k1Glw=
=mzib
-----END PGP SIGNATURE-----




reply via email to

[Prev in Thread] Current Thread [Next in Thread]