[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Out-of-Line memory deallocation: MIG vs. glibc
From: |
Thomas Schwinge |
Subject: |
Re: Out-of-Line memory deallocation: MIG vs. glibc |
Date: |
Tue, 11 Apr 2006 06:45:04 -0400 |
User-agent: |
Mutt/1.5.6+20040907i |
On Mon, Apr 10, 2006 at 06:57:21PM -0400, I wrote:
> tschwinge@clubber:~/tmp/ext2fs/filesystem.0.i$ fakeroot-tcp sh -c `pwd`/echo\
> foo
> err = 0 fdslen = 925
> foo
> Might this actually be a fakeroot-tcp (which is from the Debian fakeroot
> package) problem?
That's indeed the answer to this issue: after building and installing a
newer version of the Debian fakeroot package, everything worked as
expected, again with a `sudo settrans ...' installed file system:
#v+
tschwinge@clubber:~/tmp/ext2fs/filesystem.0.i$ ls -l echo
-rwxr-xr-x 1 tschwinge tschwinge 14072 Apr 10 13:45 echo
tschwinge@clubber:~/tmp/ext2fs/filesystem.0.i$ fakeroot-tcp sh -c `pwd`/echo\
foo
err = 0 fdslen = 3
foo
tschwinge@clubber:~/tmp/ext2fs/filesystem.0.i$ ls -l ping
-rwsrwxr-x 1 root root 24840 Mar 7 10:07 ping
tschwinge@clubber:~/tmp/ext2fs/filesystem.0.i$ fakeroot-tcp sh -c `pwd`/ping >
/dev/null
err = 0 fdslen = 3
#v-
Running it like ...
#v+
tschwinge@clubber:~/tmp/ext2fs/filesystem.0.i$ while fakeroot-tcp sh -c
`pwd`/ping > /dev/null; do date; done
err = 0 fdslen = 3
Tue Apr 11 01:36:23 EDT 2006
err = 0 fdslen = 3
Tue Apr 11 01:36:25 EDT 2006
err = 0 fdslen = 3
[...]
#v-
... didn't do any harm to the system so far (after minutes). And notice
the sheer pace of the command invocations; this is on a otherwise idle
PIII, 1 Ghz, 384 MiB RAM...
I stopped this experiment after a few minutes, because of ...
#v+
tschwinge@clubber:~$ ps -Af | grep faked-tcp | wc -l
247
#v-
Now we have to figure out if the new fakeroot actually only obfuscates
the issues you uncovered, which might very well be, because of the
radically decreased (factor 300) `fdslen'. Is the Hurd supposed to cope
with a `fdslen' of around a thousand; are the valid uses for that?
> > so many [port rights for the fdarray[] argument] that they are
> > automatically transferred out-of-line, which is one of the conditions
> > causing the crash. I wonder whether this is some bug by itself.
>
> Could you please elaborate on that?
Regards,
Thomas