linux.git/arch/x86/tools, branch v6.9

x86, relocs: Ignore relocations in .notes section

2024-03-01T06:34:42+00:00

When building with CONFIG_XEN_PV=y, .text symbols are emitted into
the .notes section so that Xen can find the "startup_xen" entry point.
This information is used prior to booting the kernel, so relocations
are not useful. In fact, performing relocations against the .notes
section means that the KASLR base is exposed since /sys/kernel/notes
is world-readable.

To avoid leaking the KASLR base without breaking unprivileged tools that
are expecting to read /sys/kernel/notes, skip performing relocations in
the .notes section. The values readable in .notes are then identical to
those found in System.map.

Reported-by: Guixiong Wei 
Closes: https://lore.kernel.org/all/20240218073501.54555-1-guixiongwei@gmail.com/
Fixes: 5ead97c84fa7 ("xen: Core Xen implementation")
Fixes: da1a679cde9b ("Add /sys/kernel/notes")
Reviewed-by: Juergen Gross 
Signed-off-by: Kees Cook

Merge tag 'x86-build-2024-01-08' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

2024-01-09T01:22:02+00:00

Pull x86 build updates from Ingo Molnar:

 - Update the objdump & instruction decoder self-test code for better
   LLVM toolchain compatibility

 - Rework CONFIG_X86_PAE dependencies, for better readability and higher
   robustness.

 - Misc cleanups

* tag 'x86-build-2024-01-08' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/tools: objdump_reformat.awk: Skip bad instructions from llvm-objdump
  x86/Kconfig: Rework CONFIG_X86_PAE dependency
  x86/tools: Remove chkobjdump.awk
  x86/tools: objdump_reformat.awk: Allow for spaces
  x86/tools: objdump_reformat.awk: Ensure regex matches fwait

x86/tools: objdump_reformat.awk: Skip bad instructions from llvm-objdump

2024-01-04T09:04:02+00:00

When running the instruction decoder selftest with LLVM=1 and
CONFIG_PVH=y, there is a series of warnings:

  arch/x86/tools/insn_decoder_test: warning: Found an x86 instruction decoder bug, please report this.
  arch/x86/tools/insn_decoder_test: warning: ffffffff81000050     ea                      
  arch/x86/tools/insn_decoder_test: warning: objdump says 1 bytes, but insn_get_length() says 7
  arch/x86/tools/insn_decoder_test: warning: Decoded and checked 7214721 instructions with 1 failures

GNU objdump outputs "(bad)" instead of "", which is already
handled in the bad_expr regex, so there is no warning.

  $ objdump -d arch/x86/platform/pvh/head.o | grep -E '50:\s+ea'
  50:   ea                      (bad)

  $ llvm-objdump -d arch/x86/platform/pvh/head.o | grep -E '50:\s+ea'
        50: ea                            

Add "" to the bad_expr regex to clear up the warning, allowing
the instruction decoder selftest to fully pass with llvm-objdump.

Signed-off-by: Nathan Chancellor 
Signed-off-by: Borislav Petkov (AMD) 
Link: https://lore.kernel.org/r/20231205-objdump_reformat-awk-handle-llvm-objdump-bad_expr-v1-1-b4a74f39396f@kernel.org

x86/paravirt: Remove no longer needed paravirt patching code

2023-12-10T22:34:37+00:00

Now that paravirt is using the alternatives patching infrastructure,
remove the paravirt patching code.

Signed-off-by: Juergen Gross 
Signed-off-by: Borislav Petkov (AMD) 
Acked-by: Peter Zijlstra (Intel) 
Link: https://lore.kernel.org/r/20231210062138.2417-6-jgross@suse.com

x86/tools: Remove chkobjdump.awk

2023-11-30T08:38:12+00:00

This check is superfluous now that the minimum version of binutils to
build the kernel is 2.25. This also fixes an error seen with
llvm-objdump because it does not support '-v' prior to LLVM 13:

  llvm-objdump: error: unknown argument '-v'

Signed-off-by: Nathan Chancellor 
Signed-off-by: Ingo Molnar 
Tested-by: Kees Cook 
Reviewed-by: Kees Cook 
Link: https://github.com/llvm/llvm-project/commit/dde24a87c55f82d8c7b3bf3eafb10f2b9b2b9a01
Link: https://lore.kernel.org/r/20231129-objdump-reformat-llvm-v3-3-0d855e79314d@kernel.org
Closes: https://github.com/ClangBuiltLinux/linux/issues/1362

x86/tools: objdump_reformat.awk: Allow for spaces

2023-11-30T08:38:10+00:00

GNU objdump and LLVM objdump have differing output formats.
Specifically, GNU objump will format its output as: address:hex,
whereas LLVM objdump displays its output as address:hex.

objdump_reformat.awk incorrectly handles this discrepancy due to
the unexpected space and as a result insn_decoder_test fails, as
its input is garbled.

The instruction line being tokenized now handles a space and colon,
or tab delimiter.

Signed-off-by: Samuel Zeter 
Signed-off-by: Nathan Chancellor 
Signed-off-by: Ingo Molnar 
Tested-by: Nathan Chancellor 
Tested-by: Kees Cook 
Reviewed-by: Kees Cook 
Acked-by: Masami Hiramatsu 
Link: https://lore.kernel.org/r/20231129-objdump-reformat-llvm-v3-2-0d855e79314d@kernel.org
Closes: https://github.com/ClangBuiltLinux/linux/issues/1364

x86/tools: objdump_reformat.awk: Ensure regex matches fwait

2023-11-30T08:38:09+00:00

If there is "wait" mnemonic in the line being parsed, it is incorrectly
handled by the script, and an extra line of "fwait" in
objdump_reformat's output is inserted. As insn_decoder_test relies upon
the formatted output, the test fails.

This is reproducible when disassembling with llvm-objdump:

Pre-processed lines:

  ffffffff81033e72: 9b                    wait
  ffffffff81033e73: 48 c7 c7 89 50 42 82  movq

After objdump_reformat.awk:

  ffffffff81033e72:       9b      fwait
  ffffffff81033e72:                               wait
  ffffffff81033e73:       48 c7 c7 89 50 42 82    movq

The regex match now accepts spaces or tabs, along with the "fwait"
instruction.

Signed-off-by: Samuel Zeter 
Signed-off-by: Nathan Chancellor 
Signed-off-by: Ingo Molnar 
Tested-by: Nathan Chancellor 
Tested-by: Kees Cook 
Reviewed-by: Kees Cook 
Acked-by: Masami Hiramatsu 
Link: https://lore.kernel.org/r/20231129-objdump-reformat-llvm-v3-1-0d855e79314d@kernel.org
Closes: https://github.com/ClangBuiltLinux/linux/issues/1364

ELF: fix all "Elf" typos

2023-04-08T20:45:37+00:00

ELF is acronym and therefore should be spelled in all caps.

I left one exception at Documentation/arm/nwfpe/nwfpe.rst which looks like
being written in the first person.

Link: https://lkml.kernel.org/r/Y/3wGWQviIOkyLJW@p183
Signed-off-by: Alexey Dobriyan 
Signed-off-by: Andrew Morton

kbuild: allow to combine multiple V= levels

2023-01-22T14:43:32+00:00

Commit a6de553da01c ("kbuild: Allow to combine multiple W= levels")
supported W=123 to enable all the extra warning groups.

I think a similar idea is applicable to the V= option.

  V=1 echos the whole command
  V=2 prints the reason for rebuilding

These are orthogonal, and can be enabled at the same time.

This commit supports V=12 to enable both of them.

Signed-off-by: Masahiro Yamada 
Tested-by: Nicolas Schier 
Reviewed-by: Nicolas Schier

x86/tools/relocs: Ignore __kcfi_typeid_ relocations

2022-09-26T17:13:15+00:00

The compiler generates __kcfi_typeid_ symbols for annotating assembly
functions with type information. These are constants that can be
referenced in assembly code and are resolved by the linker. Ignore
them in relocs.

Signed-off-by: Sami Tolvanen 
Reviewed-by: Kees Cook 
Tested-by: Kees Cook 
Tested-by: Nathan Chancellor 
Acked-by: Peter Zijlstra (Intel) 
Tested-by: Peter Zijlstra (Intel) 
Signed-off-by: Kees Cook 
Link: https://lore.kernel.org/r/20220908215504.3686827-20-samitolvanen@google.com