On 4/8/25 14:16, Nelson H. F. Beebe wrote:
Thanks for this analysis of the gzip-1.13.56-e549 test failures on
DragonFlyBSD 6.4.0:
If you can concoct a small C program that is likely to be runnable
across a range of systems, then I'll be happy to configurize it, and
then do tests builds on our test farm.
Since I'm guessing about the bug, it'd likely be more efficient to see
what system calls the failing gzip is executing, before writing a small
C program. Please run this:
mkdir d
echo >d/f
chmod a-w d
LC_ALL=C truss -o gzip.truss ./gzip d/f
and then look at the tail of gzip.truss. Here's what I see in gzip.truss
on Solaris 10 with bleeding-edge gzip:
lwp_sigmask(SIG_SETMASK, 0x60005003, 0x00000000) = 0xFFBFFEFF [0x0000FFFF]
openat64(3, "f.gz", O_WRONLY|O_CREAT|O_EXCL, 0600) Err#13 EACCES
[file_dac_write]
lwp_sigmask(SIG_SETMASK, 0x00000000, 0x00000000) = 0xFFBFFEFF [0x0000FFFF]
fstat64(2, 0xFFBFE9A0) = 0
write(2, "\n g z i p : d / f . g".., 33) = 33
_exit(1)
