[sane-devel] 39ceeae6 breaks md5 auth

[Olaf Meeuwissen]

Hi James,

Thanks for the report.

James Ring writes:

> Confirmed that with the offending patch, md5.c produces incorrect
> digests for known input/output pairs. We should roll it back.
>
> Also I couldn't reproduce (with gcc 7.2.0) the compiler warning that
> the original change was supposed to fix.

Hmm, neither can I.  In neither the debian-8-mini nor debian-9-mini CI
environments.  But you can still see it in the CI logs for the last
pipeline that ran before the offending commit.

  https://gitlab.com/sane-project/backends/pipelines/4150861

Only the fedora-24-clang log doesn't have it.

> On Tue, Jan 2, 2018 at 9:57 AM, James Ring <sjr at jdns.org> wrote:
>> [...snip...]
>>
>> Reverting that commit restores the functionality. I haven't figured
>> out what the problem is from a cursory inspection of the code, I'll
>> continue staring at it.

I was about to revert the commit but looking at it now I'm wondering
what I was thinking when I committed that :-(  Changing the pointer
type to something of a different size *obviously* screws up the array
indexing!

I've cooked up a fix for that (based on e895ee55).  Could you give the
attached patch a try?

Hope this helps,
--
Olaf Meeuwissen, LPIC-2            FSF Associate Member since 2004-01-27
 GnuPG key: F84A2DD9/B3C0 2F47 EA19 64F4 9F13  F43E B8A4 A88A F84A 2DD9
 Support Free Software                        https://my.fsf.org/donate
 Join the Free Software Foundation              https://my.fsf.org/join
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Fix-array-indexing.patch
Type: text/x-diff
Size: 1478 bytes
Desc: not available
URL: <http://lists.alioth.debian.org/pipermail/sane-devel/attachments/20180103/85d52a97/attachment.patch>