Skip to content

K
Kernel

openeuler / Kernel
接近 2 年 前同步成功

通知 8
Star 0
Fork 0
K
Kernel
切换分支/标签
  1. 02 3月, 2022 3 次提交
  3. 20 1月, 2022 2 次提交
    • H
      fs/binfmt_elf: use PT_LOAD p_align values for static PIE · 9630f0d6
      由 H.J. Lu 提交于 
      Extend commit ce81bb25 ("fs/binfmt_elf: use PT_LOAD p_align values
for suitable start address") which fixed PIE binaries built with
-Wl,-z,max-page-size=0x200000, to cover static PIE binaries.  This
fixes:

    https://bugzilla.kernel.org/show_bug.cgi?id=215275

Tested by verifying static PIE binaries with -Wl,-z,max-page-size=0x200000 loading.

Link: https://lkml.kernel.org/r/20211209174052.370537-1-hjl.tools@gmail.comSigned-off-by: NH.J. Lu <hjl.tools@gmail.com>
Cc: Chris Kennelly <ckennelly@google.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Song Liu <songliubraving@fb.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Sandeep Patil <sspatil@google.com>
Cc: Fangrui Song <maskray@google.com>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Shuah Khan <shuah@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      9630f0d6
    • Y
      fs/binfmt_elf: replace open-coded string copy with get_task_comm · 95af469c
      由 Yafang Shao 提交于 
      It is better to use get_task_comm() instead of the open coded string
copy as we do in other places.

struct elf_prpsinfo is used to dump the task information in userspace
coredump or kernel vmcore.  Below is the verification of vmcore,

  crash> ps
     PID    PPID  CPU       TASK        ST  %MEM     VSZ    RSS  COMM
        0      0   0  ffffffff9d21a940  RU   0.0       0      0  [swapper/0]
  >     0      0   1  ffffa09e40f85e80  RU   0.0       0      0  [swapper/1]
  >     0      0   2  ffffa09e40f81f80  RU   0.0       0      0  [swapper/2]
  >     0      0   3  ffffa09e40f83f00  RU   0.0       0      0  [swapper/3]
  >     0      0   4  ffffa09e40f80000  RU   0.0       0      0  [swapper/4]
  >     0      0   5  ffffa09e40f89f80  RU   0.0       0      0  [swapper/5]
        0      0   6  ffffa09e40f8bf00  RU   0.0       0      0  [swapper/6]
  >     0      0   7  ffffa09e40f88000  RU   0.0       0      0  [swapper/7]
  >     0      0   8  ffffa09e40f8de80  RU   0.0       0      0  [swapper/8]
  >     0      0   9  ffffa09e40f95e80  RU   0.0       0      0  [swapper/9]
  >     0      0  10  ffffa09e40f91f80  RU   0.0       0      0  [swapper/10]
  >     0      0  11  ffffa09e40f93f00  RU   0.0       0      0  [swapper/11]
  >     0      0  12  ffffa09e40f90000  RU   0.0       0      0  [swapper/12]
  >     0      0  13  ffffa09e40f9bf00  RU   0.0       0      0  [swapper/13]
  >     0      0  14  ffffa09e40f98000  RU   0.0       0      0  [swapper/14]
  >     0      0  15  ffffa09e40f9de80  RU   0.0       0      0  [swapper/15]

It works well as expected.

Some comments are added to explain why we use the hard-coded 16.

Link: https://lkml.kernel.org/r/20211120112738.45980-5-laoar.shao@gmail.comSuggested-by: NKees Cook <keescook@chromium.org>
Signed-off-by: NYafang Shao <laoar.shao@gmail.com>
Reviewed-by: NDavid Hildenbrand <david@redhat.com>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Arnaldo Carvalho de Melo <arnaldo.melo@gmail.com>
Cc: Andrii Nakryiko <andrii.nakryiko@gmail.com>
Cc: Michal Miroslaw <mirq-linux@rere.qmqm.pl>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: David Hildenbrand <david@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Kees Cook <keescook@chromium.org>
Cc: Petr Mladek <pmladek@suse.com>
Cc: Alexei Starovoitov <alexei.starovoitov@gmail.com>
Cc: Andrii Nakryiko <andrii@kernel.org>
Cc: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      95af469c
  5. 10 11月, 2021 2 次提交
    • A
      ELF: simplify STACK_ALLOC macro · a43e5e3a
      由 Alexey Dobriyan 提交于 
      "A -= B; A" is equivalent to "A -= B".

Link: https://lkml.kernel.org/r/YVmcP256fRMqCwgK@localhost.localdomainSigned-off-by: NAlexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      a43e5e3a
    • K
      binfmt_elf: reintroduce using MAP_FIXED_NOREPLACE · 5f501d55
      由 Kees Cook 提交于 
      Commit b212921b ("elf: don't use MAP_FIXED_NOREPLACE for elf
executable mappings") reverted back to using MAP_FIXED to map ELF LOAD
segments because it was found that the segments in some binaries overlap
and can cause MAP_FIXED_NOREPLACE to fail.

The original intent of MAP_FIXED_NOREPLACE in the ELF loader was to
prevent the silent clobbering of an existing mapping (e.g.  stack) by
the ELF image, which could lead to exploitable conditions.  Quoting
commit 4ed28639 ("fs, elf: drop MAP_FIXED usage from elf_map"),
which originally introduced the use of MAP_FIXED_NOREPLACE in the
loader:

    Both load_elf_interp and load_elf_binary rely on elf_map to map
    segments [to a specific] address and they use MAP_FIXED to enforce
    that. This is however [a] dangerous thing prone to silent data
    corruption which can be even exploitable.
    ...
    Let's take CVE-2017-1000253 as an example ... we could end up mapping
    [the executable] over the existing stack ... The [stack layout] issue
    has been fixed since then ... So we should be safe and any [similar]
    attack should be impractical. On the other hand this is just too
    subtle [an] assumption ... it can break quite easily and [be] hard to
    spot.
    ...
    Address this [weakness] by changing MAP_FIXED to the newly added
    MAP_FIXED_NOREPLACE. This will mean that mmap will fail if there is
    an existing mapping clashing with the requested one [instead of
    silently] clobbering it.

Then processing ET_DYN binaries the loader already calculates a total
size for the image when the first segment is mapped, maps the entire
image, and then unmaps the remainder before the remaining segments are
then individually mapped.

To avoid the earlier problems (legitimate overlapping LOAD segments
specified in the ELF), apply the same logic to ET_EXEC binaries as well.

For both ET_EXEC and ET_DYN+INTERP use MAP_FIXED_NOREPLACE for the
initial total size mapping and then use MAP_FIXED to build the final
(possibly legitimately overlapping) mappings.  For ET_DYN w/out INTERP,
continue to map at a system-selected address in the mmap region.

Link: https://lkml.kernel.org/r/20210916215947.3993776-1-keescook@chromium.org
Link: https://lore.kernel.org/lkml/1595869887-23307-2-git-send-email-anthony.yznaga@oracle.comCo-developed-by: NAnthony Yznaga <anthony.yznaga@oracle.com>
Signed-off-by: NAnthony Yznaga <anthony.yznaga@oracle.com>
Signed-off-by: NKees Cook <keescook@chromium.org>
Cc: Russell King <linux@armlinux.org.uk>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Eric Biederman <ebiederm@xmission.com>
Cc: Chen Jingwen <chenjingwen6@huawei.com>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Andrei Vagin <avagin@openvz.org>
Cc: Khalid Aziz <khalid.aziz@oracle.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      5f501d55
  7. 09 10月, 2021 1 次提交
    • E
      coredump: Limit coredumps to a single thread group · 0258b5fd
      由 Eric W. Biederman 提交于 
      Today when a signal is delivered with a handler of SIG_DFL whose
default behavior is to generate a core dump not only that process but
every process that shares the mm is killed.

In the case of vfork this looks like a real world problem.  Consider
the following well defined sequence.

	if (vfork() == 0) {
		execve(...);
		_exit(EXIT_FAILURE);
	}

If a signal that generates a core dump is received after vfork but
before the execve changes the mm the process that called vfork will
also be killed (as the mm is shared).

Similarly if the execve fails after the point of no return the kernel
delivers SIGSEGV which will kill both the exec'ing process and because
the mm is shared the process that called vfork as well.

As far as I can tell this behavior is a violation of people's
reasonable expectations, POSIX, and is unnecessarily fragile when the
system is low on memory.

Solve this by making a userspace visible change to only kill a single
process/thread group.  This is possible because Jann Horn recently
modified[1] the coredump code so that the mm can safely be modified
while the coredump is happening.  With LinuxThreads long gone I don't
expect anyone to have a notice this behavior change in practice.

To accomplish this move the core_state pointer from mm_struct to
signal_struct, which allows different thread groups to coredump
simultatenously.

In zap_threads remove the work to kill anything except for the current
thread group.

v2: Remove core_state from the VM_BUG_ON_MM print to fix
    compile failure when CONFIG_DEBUG_VM is enabled.
Reported-by: NStephen Rothwell <sfr@canb.auug.org.au>

[1] a07279c9 ("binfmt_elf, binfmt_elf_fdpic: use a VMA list snapshot")
Fixes: d89f3847def4 ("[PATCH] thread-aware coredumps, 2.5.43-C3")
History-tree: git://git.kernel.org/pub/scm/linux/kernel/git/tglx/history.git
Link: https://lkml.kernel.org/r/87y27mvnke.fsf@disp2133
Link: https://lkml.kernel.org/r/20211007144701.67592574@canb.auug.org.auReviewed-by: NKees Cook <keescook@chromium.org>
Signed-off-by: N"Eric W. Biederman" <ebiederm@xmission.com>
      0258b5fd
  9. 04 10月, 2021 1 次提交
    • C
      elf: don't use MAP_FIXED_NOREPLACE for elf interpreter mappings · 9b2f72cc
      由 Chen Jingwen 提交于 
      In commit b212921b ("elf: don't use MAP_FIXED_NOREPLACE for elf
executable mappings") we still leave MAP_FIXED_NOREPLACE in place for
load_elf_interp.

Unfortunately, this will cause kernel to fail to start with:

    1 (init): Uhuuh, elf segment at 00003ffff7ffd000 requested but the memory is mapped already
    Failed to execute /init (error -17)

The reason is that the elf interpreter (ld.so) has overlapping segments.

  readelf -l ld-2.31.so
  Program Headers:
    Type           Offset             VirtAddr           PhysAddr
                   FileSiz            MemSiz              Flags  Align
    LOAD           0x0000000000000000 0x0000000000000000 0x0000000000000000
                   0x000000000002c94c 0x000000000002c94c  R E    0x10000
    LOAD           0x000000000002dae0 0x000000000003dae0 0x000000000003dae0
                   0x00000000000021e8 0x0000000000002320  RW     0x10000
    LOAD           0x000000000002fe00 0x000000000003fe00 0x000000000003fe00
                   0x00000000000011ac 0x0000000000001328  RW     0x10000

The reason for this problem is the same as described in commit
ad55eac7 ("elf: enforce MAP_FIXED on overlaying elf segments").

Not only executable binaries, elf interpreters (e.g. ld.so) can have
overlapping elf segments, so we better drop MAP_FIXED_NOREPLACE and go
back to MAP_FIXED in load_elf_interp.

Fixes: 4ed28639 ("fs, elf: drop MAP_FIXED usage from elf_map")
Cc: <stable@vger.kernel.org> # v4.19
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Michal Hocko <mhocko@suse.com>
Signed-off-by: NChen Jingwen <chenjingwen6@huawei.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      9b2f72cc
  11. 04 9月, 2021 2 次提交
    • D
      binfmt: remove in-tree usage of MAP_DENYWRITE · 4589ff7c
      由 David Hildenbrand 提交于 
      At exec time when we mmap the new executable via MAP_DENYWRITE we have it
opened via do_open_execat() and already deny_write_access()'ed the file
successfully. Once exec completes, we allow_write_acces(); however,
we set mm->exe_file in begin_new_exec() via set_mm_exe_file() and
also deny_write_access() as long as mm->exe_file remains set. We'll
effectively deny write access to our executable via mm->exe_file
until mm->exe_file is changed -- when the process is removed, on new
exec, or via sys_prctl(PR_SET_MM_MAP/EXE_FILE).

Let's remove all usage of MAP_DENYWRITE, it's no longer necessary for
mm->exe_file.

In case of an elf interpreter, we'll now only deny write access to the file
during exec. This is somewhat okay, because the interpreter behaves
(and sometime is) a shared library; all shared libraries, especially the
ones loaded directly in user space like via dlopen() won't ever be mapped
via MAP_DENYWRITE, because we ignore that from user space completely;
these shared libraries can always be modified while mapped and executed.
Let's only special-case the main executable, denying write access while
being executed by a process. This can be considered a minor user space
visible change.

While this is a cleanup, it also fixes part of a problem reported with
VM_DENYWRITE on overlayfs, as VM_DENYWRITE is effectively unused with
this patch and will be removed next:
  "Overlayfs did not honor positive i_writecount on realfile for
   VM_DENYWRITE mappings." [1]

[1] https://lore.kernel.org/r/YNHXzBgzRrZu1MrD@miu.piliscsaba.redhat.com/Reported-by: NChengguang Xu <cgxu519@mykernel.net>
Acked-by: N"Eric W. Biederman" <ebiederm@xmission.com>
Acked-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NDavid Hildenbrand <david@redhat.com>
      4589ff7c
    • D
      binfmt: don't use MAP_DENYWRITE when loading shared libraries via uselib() · 42be8b42
      由 David Hildenbrand 提交于 
      uselib() is the legacy systemcall for loading shared libraries.
Nowadays, applications use dlopen() to load shared libraries, completely
implemented in user space via mmap().

For example, glibc uses MAP_COPY to mmap shared libraries. While this
maps to MAP_PRIVATE | MAP_DENYWRITE on Linux, Linux ignores any
MAP_DENYWRITE specification from user space in mmap.

With this change, all remaining in-tree users of MAP_DENYWRITE use it
to map an executable. We will be able to open shared libraries loaded
via uselib() writable, just as we already can via dlopen() from user
space.

This is one step into the direction of removing MAP_DENYWRITE from the
kernel. This can be considered a minor user space visible change.
Acked-by: N"Eric W. Biederman" <ebiederm@xmission.com>
Acked-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NDavid Hildenbrand <david@redhat.com>
      42be8b42
  13. 30 6月, 2021 1 次提交
  15. 18 6月, 2021 1 次提交
  17. 08 3月, 2021 1 次提交
    • A
      coredump: don't bother with do_truncate() · d0f1088b
      由 Al Viro 提交于 
      have dump_skip() just remember how much needs to be skipped,
leave actual seeks/writing zeroes to the next dump_emit()
or the end of coredump output, whichever comes first.
And instead of playing with do_truncate() in the end, just
write one NUL at the end of the last gap (if any).
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
      d0f1088b
  19. 16 2月, 2021 1 次提交
  21. 06 1月, 2021 1 次提交
    • A
      elf_prstatus: collect the common part (everything before pr_reg) into a struct · f2485a2d
      由 Al Viro 提交于 
      Preparations to doing i386 compat elf_prstatus sanely - rather than duplicating
the beginning of compat_elf_prstatus, take these fields into a separate
structure (compat_elf_prstatus_common), so that it could be reused.  Due to
the incestous relationship between binfmt_elf.c and compat_binfmt_elf.c we
need the same shape change done to native struct elf_prstatus, gathering the
fields prior to pr_reg into a new structure (struct elf_prstatus_common).

Fortunately, offset of pr_reg is always a multiple of 16 with no padding
right before it, so it's possible to turn all the stuff prior to it into
a single member without disturbing the layout.

[build fix from Geert Uytterhoeven folded in]
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
      f2485a2d
  23. 05 1月, 2021 1 次提交
    • A
      binfmt_elf: partially sanitize PRSTATUS_SIZE and SET_PR_FPVALID · 8a00dd00
      由 Al Viro 提交于 
      On 64bit architectures that support 32bit processes there are
two possible layouts for NT_PRSTATUS note in ELF coredumps.
For one thing, several fields are 64bit for native processes
and 32bit for compat ones (pr_sigpend, etc.).  For another,
the register dump is obviously different - the size and number
of registers are not going to be the same for 32bit and 64bit
variants of processor.

Usually that's handled by having two structures - elf_prstatus
for native layout and compat_elf_prstatus for 32bit one.
32bit processes are handled by fs/compat_binfmt_elf.c, which
defines a macro called 'elf_prstatus' that expands to compat_elf_prstatus.
Then it includes fs/binfmt_elf.c, which makes all references to
struct elf_prstatus to be textually replaced with struct
compat_elf_prstatus.  Ugly and somewhat brittle, but it works.

However, amd64 is worse - there are _three_ possible layouts.
One for native 64bit processes, another for i386 (32bit) processes
and yet another for x32 (32bit address space with full 64bit
registers).

Both i386 and x32 processes are handled by fs/compat_binfmt_elf.c,
with usual compat_binfmt_elf.c trickery.  However, the layouts
for i386 and x32 are not identical - they have the common beginning,
but the register dump part (pr_reg) is bigger on x32.  Worse, pr_reg
is not the last field - it's followed by int pr_fpvalid, so that
field ends up at different offsets for i386 and x32 layouts.

Fortunately, there's not much code that cares about any of that -
it's all encapsulated in fill_thread_core_info().  Since x32
variant is bigger, we define compat_elf_prstatus to match that
layout.  That way i386 processes have enough space to fit
their layout into.

Moreover, since these layouts are identical prior to pr_reg,
we don't need to distinguish x32 and i386 cases when we are
setting the fields prior to pr_reg.

Filling pr_reg itself is done by calling ->get() method of
appropriate regset, and that method knows what layout (and size)
to use.

We do need to distinguish x32 and i386 cases only for two
things: setting ->pr_fpvalid (offset differs for x32 and
i386) and choosing the right size for our note.

The way it's done is Not Nice, for the lack of more accurate
printable description.  There are two macros (PRSTATUS_SIZE and
SET_PR_FPVALID), that default essentially to sizeof(struct elf_prstatus)
and (S)->pr_fpvalid = 1.  On x86 asm/compat.h provides its own
variants.

Unfortunately, quite a few things go wrong there:
	* PRSTATUS_SIZE doesn't use the normal test for process
being an x32 one; it compares the size reported by regset with
the size of pr_reg.
	* it hardcodes the sizes of x32 and i386 variants (296 and 144
resp.), so if some change in includes leads to asm/compat.h pulled
in by fs/binfmt_elf.c we are in trouble - it will end up using
the size of x32 variant for 64bit processes.
	* it's in the wrong place; asm/compat.h couldn't define
the structure for i386 layout, since it lacks quite a few types
needed for it.  Hardcoded sizes are largely due to that.

The proper fix would be to have an explicitly defined i386 variant
of structure and have PRSTATUS_SIZE/SET_PR_FPVALID check for
TIF_X32 to choose the variant that should be used.  Unfortunately,
that requires some manipulations of headers; we'll do that later
in the series, but for now let's go with the minimal variant -
rename PRSTATUS_SIZE in asm/compat.h to COMPAT_PRSTATUS_SIZE,
have fs/compat_binfmt_elf.c define PRSTATUS_SIZE to COMPAT_PRSTATUS_SIZE
and use the normal TIF_X32 check in that macro.  The size of i386 variant
is kept hardcoded for now.  Similar story for SET_PR_FPVALID.
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
      8a00dd00
  25. 11 12月, 2020 1 次提交
    • E
      coredump: Document coredump code exclusively used by cell spufs · c39ab6de
      由 Eric W. Biederman 提交于 
      Oleg Nesterov recently asked[1] why is there an unshare_files in
do_coredump.  After digging through all of the callers of lookup_fd it
turns out that it is
arch/powerpc/platforms/cell/spufs/coredump.c:coredump_next_context
that needs the unshare_files in do_coredump.

Looking at the history[2] this code was also the only piece of coredump code
that required the unshare_files when the unshare_files was added.

Looking at that code it turns out that cell is also the only
architecture that implements elf_coredump_extra_notes_size and
elf_coredump_extra_notes_write.

I looked at the gdb repo[3] support for cell has been removed[4] in binutils
2.34.  Geoff Levand reports he is still getting questions on how to
run modern kernels on the PS3, from people using 3rd party firmware so
this code is not dead.  According to Wikipedia the last PS3 shipped in
Japan sometime in 2017.  So it will probably be a little while before
everyone's hardware dies.

Add some comments briefly documenting the coredump code that exists
only to support cell spufs to make it easier to understand the
coredump code.  Eventually the hardware will be dead, or their won't
be userspace tools, or the coredump code will be refactored and it
will be too difficult to update a dead architecture and these comments
make it easy to tell where to pull to remove cell spufs support.

[1] https://lkml.kernel.org/r/20201123175052.GA20279@redhat.com
[2] 179e037f ("do_coredump(): make sure that descriptor table isn't shared")
[3] git://sourceware.org/git/binutils-gdb.git
[4] abf516c6931a ("Remove Cell Broadband Engine debugging support").
Link: https://lkml.kernel.org/r/87h7pdnlzv.fsf_-_@x220.int.ebiederm.orgSigned-off-by: NEric W. Biederman <ebiederm@xmission.com>
      c39ab6de
  27. 30 10月, 2020 1 次提交
  29. 26 10月, 2020 2 次提交
  31. 19 10月, 2020 1 次提交
    • J
      binfmt_elf: take the mmap lock around find_extend_vma() · b2767d97
      由 Jann Horn 提交于 
      create_elf_tables() runs after setup_new_exec(), so other tasks can
already access our new mm and do things like process_madvise() on it.  (At
the time I'm writing this commit, process_madvise() is not in mainline
yet, but has been in akpm's tree for some time.)

While I believe that there are currently no APIs that would actually allow
another process to mess up our VMA tree (process_madvise() is limited to
MADV_COLD and MADV_PAGEOUT, and uring and userfaultfd cannot reach an mm
under which no syscalls have been executed yet), this seems like an
accident waiting to happen.

Let's make sure that we always take the mmap lock around GUP paths as long
as another process might be able to see the mm.

(Yes, this diff looks suspicious because we drop the lock before doing
anything with `vma`, but that's because we actually don't do anything with
it apart from the NULL check.)
Signed-off-by: NJann Horn <jannh@google.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Acked-by: NMichel Lespinasse <walken@google.com>
Cc: "Eric W . Biederman" <ebiederm@xmission.com>
Cc: Jason Gunthorpe <jgg@nvidia.com>
Cc: John Hubbard <jhubbard@nvidia.com>
Cc: Mauro Carvalho Chehab <mchehab@kernel.org>
Cc: Sakari Ailus <sakari.ailus@linux.intel.com>
Link: https://lkml.kernel.org/r/CAG48ez1-PBCdv3y8pn-Ty-b+FmBSLwDuVKFSt8h7wARLy0dF-Q@mail.gmail.comSigned-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      b2767d97
  33. 17 10月, 2020 4 次提交
    • J
      binfmt_elf, binfmt_elf_fdpic: use a VMA list snapshot · a07279c9
      由 Jann Horn 提交于 
      In both binfmt_elf and binfmt_elf_fdpic, use a new helper
dump_vma_snapshot() to take a snapshot of the VMA list (including the gate
VMA, if we have one) while protected by the mmap_lock, and then use that
snapshot instead of walking the VMA list without locking.

An alternative approach would be to keep the mmap_lock held across the
entire core dumping operation; however, keeping the mmap_lock locked while
we may be blocked for an unbounded amount of time (e.g.  because we're
dumping to a FUSE filesystem or so) isn't really optimal; the mmap_lock
blocks things like the ->release handler of userfaultfd, and we don't
really want critical system daemons to grind to a halt just because
someone "gifted" them SCM_RIGHTS to an eternally-locked userfaultfd, or
something like that.

Since both the normal ELF code and the FDPIC ELF code need this
functionality (and if any other binfmt wants to add coredump support in
the future, they'd probably need it, too), implement this with a common
helper in fs/coredump.c.

A downside of this approach is that we now need a bigger amount of kernel
memory per userspace VMA in the normal ELF case, and that we need O(n)
kernel memory in the FDPIC ELF case at all; but 40 bytes per VMA shouldn't
be terribly bad.

There currently is a data race between stack expansion and anything that
reads ->vm_start or ->vm_end under the mmap_lock held in read mode; to
mitigate that for core dumping, take the mmap_lock in write mode when
taking a snapshot of the VMA hierarchy.  (If we only took the mmap_lock in
read mode, we could end up with a corrupted core dump if someone does
get_user_pages_remote() concurrently.  Not really a major problem, but
taking the mmap_lock either way works here, so we might as well avoid the
issue.) (This doesn't do anything about the existing data races with stack
expansion in other mm code.)
Signed-off-by: NJann Horn <jannh@google.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Acked-by: NLinus Torvalds <torvalds@linux-foundation.org>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: "Eric W . Biederman" <ebiederm@xmission.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Hugh Dickins <hughd@google.com>
Link: http://lkml.kernel.org/r/20200827114932.3572699-6-jannh@google.comSigned-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      a07279c9
    • J
      coredump: rework elf/elf_fdpic vma_dump_size() into common helper · 429a22e7
      由 Jann Horn 提交于 
      At the moment, the binfmt_elf and binfmt_elf_fdpic code have slightly
different code to figure out which VMAs should be dumped, and if so,
whether the dump should contain the entire VMA or just its first page.

Eliminate duplicate code by reworking the binfmt_elf version into a
generic core dumping helper in coredump.c.

As part of that, change the heuristic for detecting executable/library
header pages to check whether the inode is executable instead of looking
at the file mode.

This is less problematic in terms of locking because it lets us avoid
get_user() under the mmap_sem.  (And arguably it looks nicer and makes
more sense in generic code.)

Adjust a little bit based on the binfmt_elf_fdpic version: ->anon_vma is
only meaningful under CONFIG_MMU, otherwise we have to assume that the VMA
has been written to.
Suggested-by: NLinus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: NJann Horn <jannh@google.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Acked-by: NLinus Torvalds <torvalds@linux-foundation.org>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: "Eric W . Biederman" <ebiederm@xmission.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Hugh Dickins <hughd@google.com>
Link: http://lkml.kernel.org/r/20200827114932.3572699-5-jannh@google.comSigned-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      429a22e7
    • J
      coredump: refactor page range dumping into common helper · afc63a97
      由 Jann Horn 提交于 
      Both fs/binfmt_elf.c and fs/binfmt_elf_fdpic.c need to dump ranges of
pages into the coredump file.  Extract that logic into a common helper.
Signed-off-by: NJann Horn <jannh@google.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Acked-by: NLinus Torvalds <torvalds@linux-foundation.org>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: "Eric W . Biederman" <ebiederm@xmission.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Hugh Dickins <hughd@google.com>
Link: http://lkml.kernel.org/r/20200827114932.3572699-4-jannh@google.comSigned-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      afc63a97
    • C
      fs/binfmt_elf: use PT_LOAD p_align values for suitable start address · ce81bb25
      由 Chris Kennelly 提交于 
      Patch series "Selecting Load Addresses According to p_align", v3.

The current ELF loading mechancism provides page-aligned mappings.  This
can lead to the program being loaded in a way unsuitable for file-backed,
transparent huge pages when handling PIE executables.

While specifying -z,max-page-size=0x200000 to the linker will generate
suitably aligned segments for huge pages on x86_64, the executable needs
to be loaded at a suitably aligned address as well.  This alignment
requires the binary's cooperation, as distinct segments need to be
appropriately paddded to be eligible for THP.

For binaries built with increased alignment, this limits the number of
bits usable for ASLR, but provides some randomization over using fixed
load addresses/non-PIE binaries.

This patch (of 2):

The current ELF loading mechancism provides page-aligned mappings.  This
can lead to the program being loaded in a way unsuitable for file-backed,
transparent huge pages when handling PIE executables.

For binaries built with increased alignment, this limits the number of
bits usable for ASLR, but provides some randomization over using fixed
load addresses/non-PIE binaries.

Tested by verifying program with -Wl,-z,max-page-size=0x200000 loading.

[akpm@linux-foundation.org: fix max() warning]
[ckennelly@google.com: augment comment]
  Link: https://lkml.kernel.org/r/20200821233848.3904680-2-ckennelly@google.comSigned-off-by: NChris Kennelly <ckennelly@google.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Song Liu <songliubraving@fb.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Hugh Dickens <hughd@google.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Sandeep Patil <sspatil@google.com>
Cc: Fangrui Song <maskray@google.com>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Shuah Khan <shuah@kernel.org>
Link: https://lkml.kernel.org/r/20200820170541.1132271-1-ckennelly@google.com
Link: https://lkml.kernel.org/r/20200820170541.1132271-2-ckennelly@google.comSigned-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      ce81bb25
  35. 28 7月, 2020 2 次提交
    • A
      kill elf_fpxregs_t · 7a896028
      由 Al Viro 提交于 
      all uses are conditional upon ELF_CORE_COPY_XFPREGS, which has not
been defined on any architecture since 2010
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
      7a896028
    • A
      introduction of regset ->get() wrappers, switching ELF coredumps to those · b4e9c954
      由 Al Viro 提交于 
      Two new helpers: given a process and regset, dump into a buffer.
regset_get() takes a buffer and size, regset_get_alloc() takes size
and allocates a buffer.

Return value in both cases is the amount of data actually dumped in
case of success or -E...  on error.

In both cases the size is capped by regset->n * regset->size, so
->get() is called with offset 0 and size no more than what regset
expects.

binfmt_elf.c callers of ->get() are switched to using those; the other
caller (copy_regset_to_user()) will need some preparations to switch.
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
      b4e9c954
  37. 05 6月, 2020 1 次提交
  39. 04 6月, 2020 1 次提交
  41. 29 5月, 2020 1 次提交
  43. 21 5月, 2020 1 次提交
    • E
      exec: Generic execfd support · b8a61c9e
      由 Eric W. Biederman 提交于 
      Most of the support for passing the file descriptor of an executable
to an interpreter already lives in the generic code and in binfmt_elf.
Rework the fields in binfmt_elf that deal with executable file
descriptor passing to make executable file descriptor passing a first
class concept.

Move the fd_install from binfmt_misc into begin_new_exec after the new
creds have been installed.  This means that accessing the file through
/proc/<pid>/fd/N is able to see the creds for the new executable
before allowing access to the new executables files.

Performing the install of the executables file descriptor after
the point of no return also means that nothing special needs to
be done on error.  The exiting of the process will close all
of it's open files.

Move the would_dump from binfmt_misc into begin_new_exec right
after would_dump is called on the bprm->file.  This makes it
obvious this case exists and that no nesting of bprm->file is
currently supported.

In binfmt_misc the movement of fd_install into generic code means
that it's special error exit path is no longer needed.

Link: https://lkml.kernel.org/r/87y2poyd91.fsf_-_@x220.int.ebiederm.orgAcked-by: NLinus Torvalds <torvalds@linux-foundation.org>
Reviewed-by: NKees Cook <keescook@chromium.org>
Signed-off-by: N"Eric W. Biederman" <ebiederm@xmission.com>
      b8a61c9e
  45. 08 5月, 2020 2 次提交
  47. 06 5月, 2020 2 次提交
  49. 08 4月, 2020 4 次提交