[ILUG] C++ woes...

[Brendan Kehoe]

. I wonder if some kind soul out there would take the time to test on
> a RH6 box (with egcs and libstdc++) whether the Wavelet compression
> package available at
> http://www.cs.dartmouth.edu/~gdavis/wavelet/wavelet.html will compile on
> your machine

I've been able to build it successfully on a 6.0 system using the same
versions of the tools as you cited in an earlier message:

egcs-1.1.2-12
egcs-c++-1.1.2-12
libstdc++-2.9.0-12

I was able to verify the behavior of the programs:

% ./raw2pgm  /usr/doc/mgetty-1.1.14/samples/new_fax.pbm 37 37  blah.pgm
Error: Read < 1369 chars when loading file /usr/doc/mgetty-1.1.14/samples/new_fax.pbm

raw2pgm: global.cc:83: volatile void error(char * ...): Assertion `0' failed.
Aborted

... (Yup, 37^2 is bigger than 1315, the size of the pbm file.)

% ./raw2pgm  /usr/doc/mgetty-1.1.14/samples/new_fax.pbm 36 36  blah.pgm
% ./encode blah.pgm fudge.out 10
File blah.pgm is of type PGM, is 35 x 35 with max gray level 255
Reading 35 x 35 image blah.pgm, writing fudge.out
Compression ratio 10:1
Target rate = 58 bytes
Set  Precision     Rate                Distortion
 0:      3           13.66    3.42        838.87    209.72
 1:      2            6.17    3.08        157.85     78.93
 2:      2            5.17    2.58        297.98    148.99
 3:      1            1.58    1.58         59.49     59.49
 4:      2           18.79    3.13       1028.43    171.41
 5:      1           10.71    1.79        841.96    140.33
 6:      1            7.49    1.87       1783.28    445.82
 7:      1           27.42    1.37       5502.92    275.15
 8:      2           54.55    2.73       6211.62    310.58
 9:      0            0.00    0.00      10861.64    678.85
10:      0            0.00    0.00      97282.70   1201.02
11:      1           95.67    1.18      75919.30    937.28
12:      0            0.00    0.00     127268.88   1571.22
13:      0            0.00    0.00     255557.38    835.15
14:      1          222.10    0.73     282147.31    922.05
15:      0            0.00    0.00     219202.14    758.48

total rate = 57.9146
total dist = 1.08496e+06
total coeffs = 1225
RMS error = 29.7604
PSNR (transform domain) = 18.658
%

Are you building it with the default use of -O6 for its optimization?  What
happens if you build it with no optimization at all?  (Mine did build with
those, but you never know...)

What does ldd give you for the encode executable?  Mine reports:

     libstdc++-libc6.1-1.so.2 => /usr/lib/libstdc++-libc6.1-1.so.2 (0x40019000)
     libm.so.6 => /lib/libm.so.6 (0x4005b000)
     libc.so.6 => /lib/libc.so.6 (0x40077000)
     /lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x40000000)

> Thanks a lot, if you get the time, and never mind if you don't. By the
> way, are RedHat 5.* and RedHat 6.* not executable incompatible, or is it
> just me again? If so, why? And why was gcc dropped from redhat in 6.0? 

I don't know, but egcs has now taken the role of the FSF's official gcc, thus
the 2.95.* releases.

Hope this helps,
B

-- 
Brendan Kehoe

Web page: http://www.zen.org/~brendan/