According to Hans-Peter Nilsson: > I've committed the following patch to the main CVS trunk .. > It lets htdig finish successfully on the sun-sparc-solaris-2.5 > machine I debugged on with the testcases I got (although I > stopped Frank's testcase after some 2100 pages), and to my > knowledge it was the only sparc-related bug that is in the way > for a successful upgrade from 3.1.0b4 to 3.1.0. > > The bug was that I missed that some machines do not allow an > unaligned access for "unsigned short". (Temporary insanity; I'm > not [normally] clueless about that issue... Egg in my face.) I guess a bit of that egg is on my face too. I have access to an old Sun sparc system, running Red Hat 4.2. I should have tried a pre-release version on it, but I held off for the final release. This old sparc is so slow that I hate compiling on it any more than I need to, so I try to minimise the number of htdig releases I build and test on it to the major releases for which I build RPMs. In this case, it would have helped to discover the problem sooner. Oddly enough, when I built the 3.1.0 source on this sparc yesterday, without the alignment bug fix patch, the code seemed to run fine, but htsearch didn't find anything (I got the nomatch.html page). There were no bus errors or segfaults, no core dumps, no complaints. Even if I had tested it out on the sparc earlier, I wouldn't have known where to begin looking for the cause of the searches failing. The alignment patch seems to have fixed the problem, though. Thanks for the fix, Hans-Peter! -- Gilles R. Detillieux E-mail: <[EMAIL PROTECTED]> Spinal Cord Research Centre WWW: http://www.scrc.umanitoba.ca/~grdetil Dept. Physiology, U. of Manitoba Phone: (204)789-3766 Winnipeg, MB R3E 3J7 (Canada) Fax: (204)789-3930 ------------------------------------ To unsubscribe from the htdig3-dev mailing list, send a message to [EMAIL PROTECTED] containing the single word "unsubscribe" in the SUBJECT of the message.
