On Fri, 16 Jan 1998, Bill Hawes wrote:
> > I tried to compile gcc-2.7.2.3 with 'make bootstrap'; after that, 'make
> > compare' complained about differing results of the two compile runs. The
> > differences are not always in the same files or places.
>
> If you use the cmp utility on two sets of files, what sort of
> differences to you note? Are the file lengths different, or only a few
> bytes? Is it always certain bits that have changed?
The file lengths always are the same. I just now thought of comparing versions
compiled under 2.0.x and 2.1.x and noticed that differences occur always in
places where the file from 2.0.x has zeroes; a typical result looks like this:
(from cmp -i 16 -l -c)
bi-reverse.o:
-rw-r--r-- 1 root root 3240 Jan 16 23:40 bi-reverse.o
2466 0 ^@ 331 M-Y
2467 0 ^@ 30 ^X
2468 0 ^@ 10 ^H
cexp.o:
-rw-r--r-- 1 root root 29448 Jan 16 23:34 cexp.o
7115 0 ^@ 4 ^D
crtbegin.o:
-rw-r--r-- 1 root root 1220 Jan 16 23:55 crtbegin.o
61 0 ^@ 130 X
62 0 ^@ 160 p
63 0 ^@ 30 ^X
64 0 ^@ 10 ^H
276 0 ^@ 10 ^H
1195 0 ^@ 30 ^X
1196 0 ^@ 10 ^H
crtbeginS.o:
-rw-r--r-- 1 root root 1296 Jan 16 23:55 crtbeginS.o
53 0 ^@ 144 d
54 0 ^@ 357 M-o
55 0 ^@ 377 M-^?
56 0 ^@ 277 M-?
crtend.o:
-rw-r--r-- 1 root root 1216 Jan 16 23:55 crtend.o
53 0 ^@ 33 ^[
54 0 ^@ 201 M-^A
55 0 ^@ 176 ~
56 0 ^@ 4 ^D
57 0 ^@ 377 M-^?
58 0 ^@ 377 M-^?
61 0 ^@ 17 ^O
62 0 ^@ 204 M-^D
63 0 ^@ 201 M-^A
64 0 ^@ 2 ^B
146 0 ^@ 203 M-^C
147 0 ^@ 76 >
148 0 ^@ 71 9
276 0 ^@ 372 M-z
crtendS.o:
-rw-r--r-- 1 root root 1416 Jan 16 23:55 crtendS.o
57 0 ^@ 121 Q
explow.o:
-rw-r--r-- 1 root root 38152 Jan 16 23:42 explow.o
4506 0 ^@ 10 ^H
4507 0 ^@ 36 ^^
4508 0 ^@ 10 ^H
genoutput.o:
-rw-r--r-- 1 root root 30416 Jan 16 23:51 genoutput.o
5898 0 ^@ 364 M-t
5899 0 ^@ 30 ^X
5900 0 ^@ 10 ^H
getopt.o:
-rw-r--r-- 1 root root 3768 Jan 16 23:55 getopt.o
3743 0 ^@ 27 ^W
3744 0 ^@ 10 ^H
reorg.o:
-rw-r--r-- 1 root root 8400 Jan 16 23:50 reorg.o
7738 0 ^@ 141 a
7739 0 ^@ 164 t
7740 0 ^@ 151 i
sdbout.o:
-rw-r--r-- 1 root root 2180 Jan 16 23:44 sdbout.o
1519 0 ^@ 65 5
1520 0 ^@ 120 P
2155 0 ^@ 1 ^A
underscore.o:
-rw-r--r-- 1 root root 1884 Jan 16 23:55 underscore.o
1859 0 ^@ 114 L
1860 0 ^@ 66 6
> Is this with the unfsd server?
Universal NFS Server 2.2beta29. Every kernel I got knfsd to run on (2.0.33
and later 2.1.x) shows a tendency to loose the NFS connection and require
a server reboot to recover (with both NFS implementations); with 2.1.x this
happens after only a few minutes. Apart from that knfsd is even slower than
unfsd, perhaps because my server has only 8MB memory, and seems to refuse to
create unix domain sockets.
MfG, Ulrich
---
He had the kind of real deep tan that rich people spend ages trying to achieve
with expensive holidays and bits of tinfoil, when really all you need to do to
obtain one is to work your arse off in the open air every day.
T. Pratchett, "Equal Rites"