Luk Claes <[email protected]> writes:
> I am the s390 experimental buildd admin. An approximation of who is, can
> be found on experimental.buildd.net. The new wanna build maintainers are
> thinking of providing something more suited and more accurate after
> lenny...
Ah, excellent, thank you!
>> If you have a moment, could you try running:
>>
>> ls -l /dev/full
>
> # ls -l /dev/full
> crw-rw-rw- 1 root root 1, 5 Sep 15 2006 /dev/full
Hm, well, it's not that then....
>> tests/util/tokens-t
>
> 2# tests/util/tokens-t
> 10
> ok 1
> ok 2
> ok 3
> ok 4
> ok 5
> ok 6
> ok 7
> ok 8
> ok 9
> not ok 10
> wanted: -2
> seen: 0
However, the program successfully wrote to /dev/full. Which seems wrong.
Maybe there's something I'm missing about s390 or about the test suite
code.
Unfortunately, it looks like raptor.debian.org has no chroots configured
for schroot and no compilers installed in the main OS, so I can't easily
run a simplified test on s390. I was trying to run:
#include <errno.h>
#include <fcntl.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
int
main(void)
{
int fd, status;
fd = open("/dev/full", O_RDWR);
if (fd < 0) {
fprintf(stderr, "failed to open /dev/full: %s\n", strerror(errno));
exit(1);
}
status = write(fd, "foo", 3);
printf("status = %d (%s)\n", status, strerror(errno));
return 0;
}
just as a sanity check. On i386, it returns:
windlord:~> ./foo
status = -1 (No space left on device)
This particular test case hasn't changed in several releases and remctl
previously built on s390, so something must have changed....
--
Russ Allbery ([email protected]) <http://www.eyrie.org/~eagle/>
--
To UNSUBSCRIBE, email to [email protected]
with a subject of "unsubscribe". Trouble? Contact [email protected]
