Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

We were leaking pmd pages when 3_LEVEL_PGTABLES was enabled.  This fixes that.
Signed-off-by: NJeff Dike <jdike@addtoit.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Change syscall-stub's data to include a "expected retval".

Stub now checks syscalls retval and aborts execution of syscall list, if
retval != expected retval.

run_syscall_stub prints the data of the failed syscall, using the data pointer
and retval written by the stub to the beginning of the stack.

one_syscall_stub is removed, to simplify code, because only some instructions
are saved by one_syscall_stub, no host-syscall.

Using the stub with additional data (modify_ldt via stub)
is prepared also.
Signed-off-by: NBodo Stroesser <bstroesser@fujitsu-siemens.com>
Signed-off-by: NJeff Dike <jdike@addtoit.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

This change enables SKAS0/SKAS3 to work with all combinations of /proc/mm and
PTRACE_FAULTINFO being available or not.

Also it changes the initialization of proc_mm and ptrace_faultinfo slightly,
to ease forcing SKAS0 on a patched host.  Forcing UML to run without /proc/mm
or PTRACE_FAULTINFO by cmdline parameter can be implemented with a setup
resetting the related variable.
Signed-off-by: NBodo Stroesser <bstroesser@fujitsu-siemens.com>
Signed-off-by: NJeff Dike <jdike@addtoit.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

This merges two sets of files which had no business being split apart in the
first place.
Signed-off-by: NJeff Dike <jdike@addtoit.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

This adds VM op batching to skas0.  Rather than having a context switch to and
from the userspace stub for each address space change, we write a number of
operations to the stub data page and invoke a different stub which loops over
them and executes them all in one go.

The operations are stored as [ system call number, arg1, arg2, ... ] tuples.

The set is terminated by a system call number of 0.  Single operations, i.e.
page faults, are handled in the old way, since that is slightly more
efficient.

For a kernel build, a minority (~1/4) of the operations are part of a set.
These sets averaged ~100 in length, so for this quarter, the context switching
overhead is greatly reduced.
Signed-off-by: NJeff Dike <jdike@addtoit.com>
Cc: Paolo Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Running UML inside a detached screen delivers SIGWINCH when UML is not
expecting it.  This patch ignores them.
Signed-off-by: NJeff Dike <jdike@addtoit.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Fix a typo in wait_stub_done.
Signed-off-by: NBodo Stroesser <bstroesser@fujitsu-siemens.com>
Signed-off-by: NJeff Dike <jdike@addtoit.com>
Cc: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

By this point, .is_user has already been set, so this assignment is useless.
Signed-off-by: NJeff Dike <jdike@addtoit.com>
Cc: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

turn many #if $undefined_string into #ifdef $undefined_string to fix some
warnings after -Wno-def was added to global CFLAGS
Signed-off-by: NOlaf Hering <olh@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

This patch implements the clone-stub mechanism, which allows skas0 to run
with proc_mm==0, even if the clib in UML uses modify_ldt.

Note: There is a bug in skas3.v7 host patch, that avoids UML-skas from
running properly on a SMP-box.  In full skas3, I never really saw problems,
but in skas0 they showed up.

More commentary by jdike - What this patch does is makes sure that the host
parent of each new host process matches the UML parent of the corresponding
UML process.  This ensures that any changed LDTs are inherited.  This is
done by having clone actually called by the UML process from its stub,
rather than by the kernel.  We have special syscall stubs that are loaded
onto the stub code page because that code must be completely
self-contained.  These stubs are given C interfaces, and used like normal C
functions, but there are subtleties.  Principally, we have to be careful
about stack variables in stub_clone_handler after the clone.  The code is
written so that there aren't any - everything boils down to a fixed
address.  If there were any locals, references to them after the clone
would be wrong because the stack just changed.
Signed-off-by: NBodo Stroesser <bstroesser@fujitsu-siemens.com>
Signed-off-by: NJeff Dike <jdike@addtoit.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

UML has had two modes of operation - an insecure, slow mode (tt mode) in
which the kernel is mapped into every process address space which requires
no host kernel modifications, and a secure, faster mode (skas mode) in
which the UML kernel is in a separate host address space, which requires a
patch to the host kernel.

This patch implements something very close to skas mode for hosts which
don't support skas - I'm calling this skas0.  It provides the security of
the skas host patch, and some of the performance gains.

The two main things that are provided by the skas patch, /proc/mm and
PTRACE_FAULTINFO, are implemented in a way that require no host patch.

For the remote address space changing stuff (mmap, munmap, and mprotect),
we set aside two pages in the process above its stack, one of which
contains a little bit of code which can call mmap et al.

To update the address space, the system call information (system call
number and arguments) are written to the stub page above the code.  The
%esp is set to the beginning of the data, the %eip is set the the start of
the stub, and it repeatedly pops the information into its registers and
makes the system call until it sees a system call number of zero.  This is
to amortize the cost of the context switch across multiple address space
updates.

When the updates are done, it SIGSTOPs itself, and the kernel process
continues what it was doing.

For a PTRACE_FAULTINFO replacement, we set up a SIGSEGV handler in the
child, and let it handle segfaults rather than nullifying them.  The
handler is in the same page as the mmap stub.  The second page is used as
the stack.  The handler reads cr2 and err from the sigcontext, sticks them
at the base of the stack in a faultinfo struct, and SIGSTOPs itself.  The
kernel then reads the faultinfo and handles the fault.

A complication on x86_64 is that this involves resetting the registers to
the segfault values when the process is inside the kill system call.  This
breaks on x86_64 because %rcx will contain %rip because you tell SYSRET
where to return to by putting the value in %rcx.  So, this corrupts $rcx on
return from the segfault.  To work around this, I added an
arch_finish_segv, which on x86 does nothing, but which on x86_64 ptraces
the child back through the sigreturn.  This causes %rcx to be restored by
sigreturn and avoids the corruption.  Ultimately, I think I will replace
this with the trick of having it send itself a blocked signal which will be
unblocked by the sigreturn.  This will allow it to be stopped just after
the sigreturn, and PTRACE_SYSCALLed without all the back-and-forth of
PTRACE_SYSCALLing it through sigreturn.

This runs on a stock host, so theoretically (and hopefully), tt mode isn't
needed any more.  We need to make sure that this is better in every way
than tt mode, though.  I'm concerned about the speed of address space
updates and page fault handling, since they involve extra round-trips to
the child.  We can amortize the round-trip cost for large address space
updates by writing all of the operations to the data page and having the
child execute them all at the same time.  This will help fork and exec, but
not page faults, since they involve only one page.

I can't think of any way to help page faults, except to add something like
PTRACE_FAULTINFO to the host.  There is PTRACE_SIGINFO, but UML doesn't use
siginfo for SIGSEGV (or anything else) because there isn't enough
information in the siginfo struct to handle page faults (the faulting
operation type is missing).  Adding that would make PTRACE_SIGINFO a usable
equivalent to PTRACE_FAULTINFO.

As for the code itself:

- The system call stub is in arch/um/kernel/sys-$(SUBARCH)/stub.S.  It is
  put in its own section of the binary along with stub_segv_handler in
  arch/um/kernel/skas/process.c.  This is manipulated with run_syscall_stub
  in arch/um/kernel/skas/mem_user.c.  syscall_stub will execute any system
  call at all, but it's only used for mmap, munmap, and mprotect.

- The x86_64 stub calls sigreturn by hand rather than allowing the normal
  sigreturn to happen, because the normal sigreturn is a SA_RESTORER in
  UML's address space provided by libc.  Needless to say, this is not
  available in the child's address space.  Also, it does a couple of odd
  pops before that which restore the stack to the state it was in at the
  time the signal handler was called.

- There is a new field in the arch mmu_context, which is now a union.
  This is the pid to be manipulated rather than the /proc/mm file
  descriptor.  Code which deals with this now checks proc_mm to see whether
  it should use the usual skas code or the new code.

- userspace_tramp is now used to create a new host process for every UML
  process, rather than one per UML processor.  It checks proc_mm and
  ptrace_faultinfo to decide whether to map in the pages above its stack.

- start_userspace now makes CLONE_VM conditional on proc_mm since we need
  separate address spaces now.

- switch_mm_skas now just sets userspace_pid[0] to the new pid rather
  than PTRACE_SWITCH_MM.  There is an addition to userspace which updates
  its idea of the pid being manipulated each time around the loop.  This is
  important on exec, when the pid will change underneath userspace().

- The stub page has a pte, but it can't be mapped in using tlb_flush
  because it is part of tlb_flush.  This is why it's required for it to be
  mapped in by userspace_tramp.

Other random things:

- The stub section in uml.lds.S is page aligned.  This page is written
  out to the backing vm file in setup_physmem because it is mapped from
  there into user processes.

- There's some confusion with TASK_SIZE now that there are a couple of
  extra pages that the process can't use.  TASK_SIZE is considered by the
  elf code to be the usable process memory, which is reasonable, so it is
  decreased by two pages.  This confuses the definition of
  USER_PGDS_IN_LAST_PML4, making it too small because of the rounding down
  of the uneven division.  So we round it to the nearest PGDIR_SIZE rather
  than the lower one.

- I added a missing PT_SYSCALL_ARG6_OFFSET macro.

- um_mmu.h was made into a userspace-usable file.

- proc_mm and ptrace_faultinfo are globals which say whether the host
  supports these features.

- There is a bad interaction between the mm.nr_ptes check at the end of
  exit_mmap, stack randomization, and skas0.  exit_mmap will stop freeing
  pages at the PGDIR_SIZE boundary after the last vma.  If the stack isn't
  on the last page table page, the last pte page won't be freed, as it
  should be since the stub ptes are there, and exit_mmap will BUG because
  there is an unfreed page.  To get around this, TASK_SIZE is set to the
  next lowest PGDIR_SIZE boundary and mm->nr_ptes is decremented after the
  calls to init_stub_pte.  This ensures that we know the process stack (and
  all other process mappings) will be below the top page table page, and
  thus we know that mm->nr_ptes will be one too many, and can be
  decremented.

Things that need fixing:

- We may need better assurrences that the stub code is PIC.

- The stub pte is set up in init_new_context_skas.

- alloc_pgdir is probably the right place.
Signed-off-by: NJeff Dike <jdike@addtoit.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

There is absolutely no reason to flush the kernel's VM area during a
tlb_flush_mm.

This results in a noticable performance increase in the kernel build
benchmark.
Signed-off-by: NJeff Dike <jdike@addtoit.com>
Cc: Paolo Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

user_time_init_skas and user_time_init_tt were essentially the same.  So, this
merges them, deleting the mode-specific functions and declarations.
Signed-off-by: NJeff Dike <jdike@addtoit.com>
Cc: Paolo Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

In skas mode, the call to uml_idle_timer permanently shut off the virtual
timer, resulting in no timer ticks to anything but the idle thread.  This is
likely the cause of the soft lockups that are seen sporadically in recent
UMLs.
Signed-off-by: NJeff Dike <jdike@addtoit.com>
Cc: Paolo Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Fix the do_fork calling convention: normal arch pass the regs and the new sp
value to do_fork instead of NULL.

Currently the arch-independent code ignores these values, while the UML code
(actually it's copy_thread) gets the right values by itself.

With this patch, things are fixed up.
Signed-off-by: NPaolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: NJeff Dike <jdike@addtoit.com>
Cc: Paolo Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

It turns out that we need to check for pending signals when a newly forked
process is run for the first time.  With strace -f, strace needs to know about
the forked process before it gets going.  If it doesn't, then it ptraces some
bogus values into its registers, and the process segfaults.  So, I added calls
to interrupt_end, which does that, plus checks for reschedules.  There
shouldn't be any of those, but x86 does the same thing, so I'm copying that
behavior to be safe.
Signed-off-by: NJeff Dike <jdike@addtoit.com>
Cc: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

This fixes some x86_64 bugs -

- maybe_map returns -1 on error instead of 0, which is interpreted as
  physical address 0

- removed an include of ipc.h, which isn't needed

- fixed the calculation of signal frame location

- the signal delivery code is now immune to the stack expansion check

- added a missing include
Signed-off-by: NJeff Dike <jdike@addtoit.com>
Cc: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

tt-mode closes switch_pipes in exit_thread_tt and kills processes in
switch_to_tt, if the exit_state is EXIT_DEAD or EXIT_ZOMBIE.

In very rare cases the exiting process can be scheduled out after having set
exit_state and closed switch_pipes (from release_task it calls proc_pid_flush,
which might sleep).  If this process is to be restarted, UML failes in
switch_to_tt with:

   write of switch_pipe failed, err = 9

We fix this by closing switch_pipes not in exit_thread_tt, but later in
release_thread_tt.  Additionally, we set switch_pipe[0] = 0 after closing.
switch_to_tt must not kill "from" process depending on its exit_state, but
must kill it after release_thread was processed only, so it examines
switch_pipe[0] for its decision.
Signed-off-by: NBodo Stroesser <bstroesser@fujitsu-siemens.com>
Signed-off-by: NJeff Dike <jdike@addtoit.com>
Cc: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

So, there I was, looking at my own code, wondering what the magic setjmp
return values did.  This patch turns the constants that are used to make
requests of the initial thread into meaningful symbols.
Signed-off-by: NJeff Dike <jdike@addtoit.com>
Cc: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Remove the __deprecated from verify_area_skas and verify_area_tt.  Since
verify_area is itself marked __deprecated, and it is the only caller of
these, then they don't need to be marked.  Marking them only makes the
build noisier.
Signed-off-by: NJeff Dike <jdike@addtoit.com>
Cc: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

This makes SIGWINCH work again, and fixes a couple of SIGWINCH-associated
crashes.  First, the sigio thread disables SIGWINCH because all hell breaks
loose if it ever gets one and tries to call the signal handling code.  Second,
there was a problem with deferencing tty structs after they were freed.  The
SIGWINCH support for a tty wasn't being turned off or freed after the tty went
away.
Signed-off-by: NJeff Dike <jdike@addtoit.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

This patch removes the arch-specific fault/trap-infos from thread and
skas-regs.

It adds a new struct faultinfo, that is arch-specific defined in
sysdep/faultinfo.h.

The structure is inserted in thread.arch and thread.regs.skas and
thread.regs.tt

Now, segv and other trap-handlers can copy the contents from regs.X.faultinfo
to thread.arch.faultinfo with one simple assignment.

Also, the number of macros necessary is reduced to

FAULT_ADDRESS(struct faultinfo)
    extracts the faulting address from faultinfo

FAULT_WRITE(struct faultinfo)
    extracts the "is_write" flag

SEGV_IS_FIXABLE(struct faultinfo)
    is true for the fixable segvs, i.e. (TRAP == 14)
    on i386

UPT_FAULTINFO(regs)
    result is (struct faultinfo *) to the faultinfo
    in regs->skas.faultinfo

GET_FAULTINFO_FROM_SC(struct faultinfo, struct sigcontext *)
    copies the relevant parts of the sigcontext to
    struct faultinfo.

On SIGSEGV, call user_signal() instead of handle_segv(), if the architecture
provides the information needed in PTRACE_FAULTINFO, or if PTRACE_FAULTINFO is
missing, because segv-stub will provide the info.

The benefit of the change is, that in case of a non-fixable SIGSEGV, we can
give user processes a SIGSEGV, instead of possibly looping on pagefault
handling.

Since handle_segv() sikked arch_fixup() implicitly by passing ip==0 to segv(),
I changed segv() to call arch_fixup() only, if !is_user.
Signed-off-by: NBodo Stroesser <bstroesser@fujitsu-siemens.com>
Signed-off-by: NJeff Dike <jdike@addtoit.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

mk_ptregs converted.  Nothing new here, it's the same situation as with
mk_user_constants.
Signed-off-by: NAl Viro <viro@parcelfarce.linux.theplanet.co.uk>
Signed-off-by: NJeff Dike <jdike@addtoit.com>
Cc: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

There were still a few comments left refering to verify_area, and two
functions, verify_area_skas & verify_area_tt that just wrap corresponding
access_ok_skas & access_ok_tt functions, just like verify_area does for
access_ok - deprecate those.

There was also a few places that still used verify_area in commented-out
code, fix those up to use access_ok.

After applying this one there should not be anything left but finally
removing verify_area completely, which will happen after a kernel release
or two.
Signed-off-by: NJesper Juhl <juhl-lkml@dif.dk>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

GCC 2.95 uses __va_copy instead of va_copy.  Handle it inside compiler.h
instead of in a casual file, and avoid the risk that this breaks with a newer
compiler (which it could do).
Signed-off-by: NPaolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Initial git repository build. I'm not bothering with the full history,
even though we have it. We can create a separate "historical" git
archive of that later if we want to, and in the meantime it's about
3.2GB when imported into git - space that would just make the early
git days unnecessarily complicated, when we don't have a lot of good
infrastructure for it.

Let it rip!