Pretty much as I guessed and I've changed mine. ... However, from my point of view, as a user I'd rather take the exception when trying to touch zero as opposed to a wrong answer. I think that most mmu aware systems have disallowed access to virtual address zero for some time. If the code came from the embedded space than maybe.

Richard W.

-----Original Message----- From: Wolfgang Denk [mailto:wd@denx.de] Sent: Friday, July 25, 2003 11:08 AM To: Woodruff, Richard Cc: u-boot-users@lists.sourceforge.net Subject: Re: [U-Boot-Users] Should CRC32 be changed to work against address 0 ?

Dear Richard,

in message FD2AC9A020DDD51194710008C7089B200BEE221C@dlee17.itg.ti.com you wrote:

...

CRC32 currently will return 0 if its sent buf address of zero. Is this

Arghhh...

...

proper? My ROM is at zero, so running a CRC32 over it might be a useful thing, however, CRC 32 always returns 0 for a start

address of

...

zero. Is there some reason this needs to be this way? It doesn't seem right to me at first glance.

The reason is that the CRC32 code was designed to be used with user-land applications in a standard OS, where a NULL buffer pointer is always an indication of some problem.

With U-Boot, this is just a bug. Thanks for pointing it out.

Fixed in my local tree. Will push it to CVS later today.

Best regards,

Wolfgang Denk

-- Software Engineering: Embedded and Realtime Systems, Embedded Linux Phone: (+49)-8142-4596-87 Fax: (+49)-8142-4596-88 Email: wd@denx.de ... Jesus cried with a loud voice: Lazarus, come forth; the bug hath been found and thy program runneth. And he that was dead came forth... -- John 11:43-44 [version 2.0?]