Re: toolchain/53355: MKREPRO fallout with MKDEBUG

[christos%zoulas.com@localhost (Christos Zoulas)]

On Jun 10, 10:40am, martin%NetBSD.org@localhost (martin%NetBSD.org@localhost) wrote:
-- Subject: toolchain/53355: MKREPRO fallout with MKDEBUG

| Trying to reproduce a releng auto cluster build of netbsd-8
| locally fails to create the same results. There are binary
| difference in the debug sections. Two random examples:
| 
| --- local/usr/lib/libexpat.so.2.3
| +++ releng/usr/lib/libexpat.so.2.3
| readelf --wide --decompress --hex-dump=.gnu_debuglink {}
| @@ -1,5 +1,5 @@
|   
| Hex dump of section '.gnu_debuglink':
|    0x00000000 6c696265 78706174 2e736f2e 322e332e libexpat.so.2.3.
| -  0x00000010 64656275 67000000 17bc9869          debug......i
| +  0x00000010 64656275 67000000 66cddb6a          debug...f..j
| --- local/bin/pwd
| +++ releng/bin/pwd
|  readelf --wide --decompress --hex-dump=.gnu_debuglink {}
| @@ -1,4 +1,4 @@
|   
|  Hex dump of section '.gnu_debuglink':
| -  0x00000000 7077642e 64656275 67000000 dee95ad5 pwd.debug.....Z.
| +  0x00000000 7077642e 64656275 67000000 bb5469dd pwd.debug....Ti.
| 
| 
| All differences I looked at are exactly like this, and always only a single
| line of differences early at the start of the hexdump.

I believe this is the checksum. So this means that the debug files
must be different. Have you compared the debug sets?

christos