"/lib/ld-linux.so.2 /usr/bin/emacs" fails: "Memory exhausted"

"/lib/ld-linux.so.2 /usr/bin/emacs" fails: "Memory exhausted"

[Mark Seaborn]

The problem can be summarised very simply:

$ /lib/ld-linux.so.2 /usr/bin/emacs
Memory exhausted--use M-x save-some-buffers RET
$ emacs --version
GNU Emacs 21.4.1
...

In other words, starting GNU Emacs by invoking the dynamic linker
doesn't work, and fails pretty quickly with an incorrect out-of-memory
error.

This is on i386.

This is the first program I've seen that fails when starting using the
dynamic linker this way.  I would like this to work because Plash
(http://plash.beasts.org) launches programs this way.

The only reference to this problem that I have seen is a note in a
Linux kernel patch:

  [SPARC]: Adjust 32-bit ELF_ET_DYN_BASE.

  We were using 0x08000000 instead of TASK_UNMAPPED_BASE
  so that running something like "/lib/ld-linux.so.2 emacs"
  would work.

  The issue there was that wherever /lib/ld-linux.so.2 gets
  mapped (controlled by ELF_ET_DYN_BASE), that is where the
  BSS start for the process ends up.  Now, emacs allocates
  dynamic memory for LISP objects from the BSS, and needs
  the top 4 bits of the virtual address to be clear so that
  it can encode LISP type and GC marking information there.

  But making this obscure emacs case work breaks lots of other
  stuff.  For example, programs with a reasonably large data
  section fail to load via direct ld.so interpreter execution
  because the data section is large enough to begin overlapping
  with the ELF_ET_DYN_BASE area.

  The /lib/ld-linux.so.2 emacs case does not work on a lot of
  platforms due to this issue, including i386, so it is not
  worth making work on sparc either.  It is indeed useful
  sometimes when debugging a new experimental build of glibc
  for example, but people doing that can hack the value of
  ELF_ET_DYN_BASE in their kernels.  Perhaps at some point
  we will make a sysctl controllable value.

  Signed-off-by: David S. Miller <davem@davemloft.net>

<http://www.kernel.org/pub/linux/kernel/v2.6/snapshots/old/patch-2.6.10-rc3-bk14.log>

I don't fully understand this.  Starting programs through
/lib/ld-linux.so.2 does not appear to change the addresses at which
object files get mapped.

Do you know what the problem is and whether it is simple to fix?

For comparison, XEmacs does not have this problem.

Cheers,
Mark

Re: "/lib/ld-linux.so.2 /usr/bin/emacs" fails: "Memory exhausted"

[Richard M. Stallman]

I don't know anything about what's happening here; I am not sure
what ld-linux.so.2 does in the first place, or whether that ought
to work, or what it ought to do.

I suggest you try with the development Emacs sources and see if the
problem still happens.  If it does, I think it will be up to you to
debug it or find someone who wants to debug it.  If you present an
analysis of what the problem is, maybe Emacs developers can fix it.

Re: "/lib/ld-linux.so.2 /usr/bin/emacs" fails: "Memory exhausted"

[Mark Seaborn]

"Richard M. Stallman" <rms@gnu.org> wrote:

> I don't know anything about what's happening here; I am not sure
> what ld-linux.so.2 does in the first place, or whether that ought
> to work, or what it ought to do.

ld-linux.so.2 is the dynamic linker on GNU/Linux:

  $ /lib/ld-linux.so.2 
  Usage: ld.so [OPTION]... EXECUTABLE-FILE [ARGS-FOR-PROGRAM...]
  You have invoked `ld.so', the helper program for shared library executables.
  This program usually lives in the file `/lib/ld.so', and special directives
  in executable files using ELF shared libraries tell the system's program
  loader to load the helper program from this file.  This helper program loads
  the shared libraries needed by the program executable, prepares the program
  to run, and runs it.  You may invoke this helper program directly from the
  command line to load and run an ELF executable file; this is like executing
  that file itself, but always uses this helper program from the file you
  specified, instead of the helper program file specified in the executable
  file you run.  This is mostly of use for maintainers to test new versions
  of this helper program; chances are you did not intend to run this program.
  
    --list                list all dependencies and how they are resolved
    --verify              verify that given object really is a dynamically linked
                          object we can handle
    --library-path PATH   use given PATH instead of content of the environment
                          variable LD_LIBRARY_PATH
    --inhibit-rpath LIST  ignore RUNPATH and RPATH information in object names
                          in LIST

Plash uses a modified version of the dynamic linker, so it launches
programs by invoking ld-linux.so.2 directly.  It also does this as a
technique for exec()'ing an executable from a chroot() jail without
the executable having to be inside the jail.

Mark

Re: "/lib/ld-linux.so.2 /usr/bin/emacs" fails: "Memory exhausted"

[Romain Francoise]

Mark Seaborn <mseaborn@onetel.com> writes:

> Do you know what the problem is and whether it is simple to fix?

I can reproduce this problem with Emacs 21.4 but not with the CVS
version, so it's been fixed already.  It's hard to tell which change
fixed it, since the Emacs 21 and Emacs 22 codebases have diverged
significantly in the 4 years that have passed since the 21.1 release.

-- 
Romain Francoise <romain@orebokech.com> | The sea! the sea! the open
it's a miracle -- http://orebokech.com/ | sea! The blue, the fresh, the
                                        | ever free! --Bryan W. Procter

Re: "/lib/ld-linux.so.2 /usr/bin/emacs" fails: "Memory exhausted"

[Mark Seaborn]

Romain Francoise <romain@orebokech.com> wrote:

> Mark Seaborn <mseaborn@onetel.com> writes:
> 
> > Do you know what the problem is and whether it is simple to fix?
> 
> I can reproduce this problem with Emacs 21.4 but not with the CVS
> version, so it's been fixed already.  It's hard to tell which change
> fixed it, since the Emacs 21 and Emacs 22 codebases have diverged
> significantly in the 4 years that have passed since the 21.1 release.

Thanks for checking.

It turned out that I was wrong, and the use of address space does
change when you invoke ld-linux.so.2 directly: brk() changes where it
allocates memory from.  So I can fill in the gaps from the description
I posted before.  brk() starts allocating from after the BSS
(zero-initialised) segment of the executable that was invoked by
exec().  For normal executables this is after 0x08000000.
ld-linux.so.2 gets loaded at 0x80000000, so brk() follows from
somewhere after that (regardless of what executable ld-linux.so.2
subsequently loads).

Emacs allocates memory using malloc(), which uses brk(), and so it
gets an address with one of the top 4 bits set, which it can't handle.

I would guess that Emacs' use of the top 4 bits hasn't changed but
rather Emacs 22 uses mmap() to allocate memory rather than malloc().

I suppose Emacs 21 could be fixed to use mmap() rather than malloc(),
though I don't know whether you'd want to make a change like that in
the stable branch.  For running programs under Plash, I could change
libc's behaviour so that malloc() uses mmap() rather than brk(), which
would then work with any version of Emacs.

Mark