Add a method to quickly verify whether safe RET operates properly on
a given system using perf tool.

Also, add a selftest which does the same thing.

Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Link: https://lore.kernel.org/r/20240731160531.28640-1-bp@kernel.org

Pull probes updates from Masami Hiramatsu:
 "Uprobes:

   - x86/shstk: Make return uprobe work with shadow stack

   - Add uretprobe syscall which speeds up the uretprobe 10-30% faster.
     This syscall is automatically used from user-space trampolines
     which are generated by the uretprobe. If this syscall is used by
     normal user program, it will cause SIGILL. Note that this is
     currently only implemented on x86_64.

     (This also has two fixes for adjusting the syscall number to avoid
     conflict with new *attrat syscalls.)

   - uprobes/perf: fix user stack traces in the presence of pending
     uretprobe. This corrects the uretprobe's trampoline address in the
     stacktrace with correct return address

   - selftests/x86: Add a return uprobe with shadow stack test

   - selftests/bpf: Add uretprobe syscall related tests.
      - test case for register integrity check
      - test case with register changing case
      - test case for uretprobe syscall without uprobes (expected to fail)
      - test case for uretprobe with shadow stack

   - selftests/bpf: add test validating uprobe/uretprobe stack traces

   - MAINTAINERS: Add uprobes entry. This does not specify the tree but
     to clarify who maintains and reviews the uprobes

  Kprobes:

   - tracing/kprobes: Test case cleanups.

     Replace redundant WARN_ON_ONCE() + pr_warn() with WARN_ONCE() and
     remove unnecessary code from selftest

   - tracing/kprobes: Add symbol counting check when module loads.

     This checks the uniqueness of the probed symbol on modules. The
     same check has already done for kernel symbols

     (This also has a fix for build error with CONFIG_MODULES=n)

  Cleanup:

   - Add MODULE_DESCRIPTION() macros for fprobe and kprobe examples"

* tag 'probes-v6.11' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
  MAINTAINERS: Add uprobes entry
  selftests/bpf: Change uretprobe syscall number in uprobe_syscall test
  uprobe: Change uretprobe syscall scope and number
  tracing/kprobes: Fix build error when find_module() is not available
  tracing/kprobes: Add symbol counting check when module loads
  selftests/bpf: add test validating uprobe/uretprobe stack traces
  perf,uprobes: fix user stack traces in the presence of pending uretprobes
  tracing/kprobe: Remove cleanup code unrelated to selftest
  tracing/kprobe: Integrate test warnings into WARN_ONCE
  selftests/bpf: Add uretprobe shadow stack test
  selftests/bpf: Add uretprobe syscall call from user space test
  selftests/bpf: Add uretprobe syscall test for regs changes
  selftests/bpf: Add uretprobe syscall test for regs integrity
  selftests/x86: Add return uprobe shadow stack test
  uprobe: Add uretprobe syscall to speed up return probe
  uprobe: Wire up uretprobe system call
  x86/shstk: Make return uprobe work with shadow stack
  samples: kprobes: add missing MODULE_DESCRIPTION() macros
  fprobe: add missing MODULE_DESCRIPTION() macro

These warnings are all of the form, "the format specified a short
(signed or unsigned) int, but the value is a full length int".

Acked-by: Muhammad Usama Anjum <usama.anjum@collabora.com>
Signed-off-by: John Hubbard <jhubbard@nvidia.com>
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>

When building with clang, via:

    make LLVM=1 -C tools/testing/selftests

...quite a few functions are variables are generating "unused" warnings.
Fix the warnings by deleting the unused items.

One item, the "nerrs" variable in vsdo_restorer.c's main(), is unused
but probably wants to be returned from main(), as a non-zero result.
That result is also unused right now, so another option would be to
delete it entirely, but this way, main() also gets fixed. It was missing
a return value.

Acked-by: Muhammad Usama Anjum <usama.anjum@collabora.com>
Signed-off-by: John Hubbard <jhubbard@nvidia.com>
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>

When building with clang, via:

    make LLVM=1 -C tools/testing/selftests

...clang warns that -no-pie is "unused during compilation".

This occurs because clang only wants to see -no-pie during linking.
Here, we don't have a separate linking stage, so a compiler warning is
unavoidable without (wastefully) restructuring the Makefile.

Avoid the warning by simply disabling that warning, for clang builds.

Acked-by: Muhammad Usama Anjum <usama.anjum@collabora.com>
Signed-off-by: John Hubbard <jhubbard@nvidia.com>
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>

When building with clang, via:

    make LLVM=1 -C tools/testing/selftests

...the build fails because clang's inline asm doesn't support all of the
features that are used in the asm() snippet in sysret_rip.c.

Fix this by moving the asm code into the clang_helpers_64.S file, where
it can be built with the assembler's full set of features.

Acked-by: Muhammad Usama Anjum <usama.anjum@collabora.com>
Signed-off-by: John Hubbard <jhubbard@nvidia.com>
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>

When building with clang, via:

    make LLVM=1 -C tools/testing/selftests

Fix this by moving the inline asm to "pure" assembly, in two new files:
clang_helpers_32.S, clang_helpers_64.S.

As a bonus, the pure asm avoids the need for ifdefs, and is now very
simple and easy on the eyes.

Acked-by: Muhammad Usama Anjum <usama.anjum@collabora.com>
Signed-off-by: John Hubbard <jhubbard@nvidia.com>
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>

Use fisttps instead of fisttp to specify correctly that the output
variable is of size short.

test_FISTTP.c:28:3: error: ambiguous instructions require an explicit suffix (could be 'fisttps', or 'fisttpl')
   28 |         "       fisttp  res16""\n"
      |          ^
<inline asm>:3:2: note: instantiated into assembly here
    3 |         fisttp  res16
      |         ^

...followed by three more cases of the same warning for other lines.

[jh: removed a bit of duplication from the warnings report, above, and
     fixed a typo in the title]

Signed-off-by: Muhammad Usama Anjum <usama.anjum@collabora.com>
Signed-off-by: John Hubbard <jhubbard@nvidia.com>
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>

When building with clang, via:

    make LLVM=1 -C tools/testing/selftests

...the following build failure occurs in selftests/x86:

   clang: error: cannot specify -o when generating multiple output files

This happens because, although gcc doesn't complain if you invoke it
like this:

    gcc file1.c header2.h

...clang won't accept that form--it rejects the .h file(s). Also, the
above approach is inaccurate anyway, because file.c includes header2.h
in this case, and the inclusion of header2.h on the invocation is an
artifact of the Makefile's desire to maintain dependencies.

In Makefiles of this type, a better way to do it is to use Makefile
dependencies to trigger the appropriate incremental rebuilds, and
separately use file lists (see EXTRA_FILES in this commit) to track what
to pass to the compiler.

This commit splits those concepts up, by setting up both EXTRA_FILES and
the Makefile dependencies with a single call to the new Makefile
function extra-files.

That fixes the build failure, while still providing the correct
dependencies in all cases.

Acked-by: Muhammad Usama Anjum <usama.anjum@collabora.com>
Signed-off-by: John Hubbard <jhubbard@nvidia.com>
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>

Adding return uprobe test for shadow stack and making sure it's
working properly. Borrowed some of the code from bpf selftests.

Link: https://lore.kernel.org/all/20240611112158.40795-5-jolsa@kernel.org/

Acked-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>