kernel/linux.git/tools/objtool/check.c, branch v6.18.21

objtool: Handle Clang RSP musical chairs

2026-04-02T11:23:01+00:00

[ Upstream commit 7fdaa640c810cb42090a182c33f905bcc47a616a ] For no apparent reason (possibly related to CONFIG_KMSAN), Clang can randomly pass the value of RSP to other registers and then back again to RSP. Handle that accordingly. Fixes the following warnings: drivers/input/misc/uinput.o: warning: objtool: uinput_str_to_user+0x165: undefined stack state drivers/input/misc/uinput.o: warning: objtool: uinput_str_to_user+0x165: unknown CFA base reg -1 Reported-by: Arnd Bergmann Closes: https://lore.kernel.org/90956545-2066-46e3-b547-10c884582eb0@app.fastmail.com Link: https://patch.msgid.link/240e6a172cc73292499334a3724d02ccb3247fc7.1772818491.git.jpoimboe@kernel.org Signed-off-by: Josh Poimboeuf Signed-off-by: Sasha Levin

objtool: Fix standalone --hacks=jump_label

2025-12-18T13:02:28+00:00

[ Upstream commit be8374a5ba7cbab6b97df94b4ffe0b92f5c8a6d2 ] The objtool command line 'objtool --hacks=jump_label foo.o' on its own should be expected to rewrite jump labels to NOPs. This means the add_special_section_alts() code path needs to run when only this option is provided. This is mainly relevant in certain debugging situations, but could potentially also fix kernel builds in which objtool is run with --hacks=jump_label but without --orc, --stackval, --uaccess, or --hacks=noinstr. Fixes: de6fbcedf5ab ("objtool: Read special sections with alts only when specific options are selected") Signed-off-by: Dylan Hatch Signed-off-by: Josh Poimboeuf Signed-off-by: Sasha Levin

objtool: Fix skip_alt_group() for non-alternative STAC/CLAC

2025-11-01T06:43:20+00:00

If an insn->alt points to a STAC/CLAC instruction, skip_alt_group() assumes it's part of an alternative ("alt group") as opposed to some other kind of "alt" such as an exception fixup. While that assumption may hold true in the current code base, Linus has an out-of-tree patch which breaks that assumption by replacing the STAC/CLAC alternatives with raw STAC/CLAC instructions. Make skip_alt_group() more robust by making sure it's actually an alt group before continuing. Reported-by: Linus Torvalds Fixes: 2d12c6fb7875 ("objtool: Remove ANNOTATE_IGNORE_ALTERNATIVE from CLAC/STAC") Closes: https://lore.kernel.org/CAHk-=wi6goUT36sR8GE47_P-aVrd5g38=VTRHpktWARbyE-0ow@mail.gmail.com Signed-off-by: Josh Poimboeuf Signed-off-by: Ingo Molnar Tested-by: Linus Torvalds Link: https://patch.msgid.link/3d22415f7b8e06a64e0873b21f48389290eeaa49.1761767616.git.jpoimboe@kernel.org

objtool: Fix failure when being compiled on x32 system

2025-10-22T13:21:55+00:00

Fix compilation failure when compiling the kernel with the x32 toolchain. In file included from check.c:16: check.c: In function ¡check_abs_references¢: /usr/src/git/linux-2.6/tools/objtool/include/objtool/warn.h:47:17: error: format ¡%lx¢ expects argument of type ¡long unsigned int¢, but argument 7 has type ¡u64¢ {aka ¡long long unsigned int¢} [-Werror=format=] 47 | "%s%s%s: objtool" extra ": " format "\n", \ | ^~~~~~~~~~~~~~~~~ /usr/src/git/linux-2.6/tools/objtool/include/objtool/warn.h:54:9: note: in expansion of macro ¡___WARN¢ 54 | ___WARN(severity, "", format, ##__VA_ARGS__) | ^~~~~~~ /usr/src/git/linux-2.6/tools/objtool/include/objtool/warn.h:74:27: note: in expansion of macro ¡__WARN¢ 74 | #define WARN(format, ...) __WARN(WARN_STR, format, ##__VA_ARGS__) | ^~~~~~ check.c:4713:33: note: in expansion of macro ¡WARN¢ 4713 | WARN("section %s has absolute relocation at offset 0x%lx", | ^~~~ Fixes: 0d6e4563fc03 ("objtool: Add action to check for absence of absolute relocations") Signed-off-by: Mikulas Patocka Signed-off-by: Peter Zijlstra (Intel) Acked-by: Ard Biesheuvel Link: https://patch.msgid.link/1ac32fff-2e67-5155-f570-69aad5bf5412@redhat.com

objtool/rust: add one more `noreturn` Rust function

2025-10-22T13:21:54+00:00

Between Rust 1.79 and 1.86, under `CONFIG_RUST_KERNEL_DOCTESTS=y`, `objtool` may report: rust/doctests_kernel_generated.o: warning: objtool: rust_doctest_kernel_alloc_kbox_rs_13() falls through to next function rust_doctest_kernel_alloc_kvec_rs_0() (as well as in rust_doctest_kernel_alloc_kvec_rs_0) due to calls to the `noreturn` symbol: core::option::expect_failed from code added in commits 779db37373a3 ("rust: alloc: kvec: implement AsPageIter for VVec") and 671618432f46 ("rust: alloc: kbox: implement AsPageIter for VBox"). Thus add the mangled one to the list so that `objtool` knows it is actually `noreturn`. This can be reproduced as well in other versions by tweaking the code, such as the latest stable Rust (1.90.0). Stable does not have code that triggers this, but it could have it in the future. Downstream forks could too. Thus tag it for backport. See commit 56d680dd23c3 ("objtool/rust: list `noreturn` Rust functions") for more details. Signed-off-by: Miguel Ojeda Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Alice Ryhl Cc: stable@vger.kernel.org # Needed in 6.12.y and later. Link: https://patch.msgid.link/20251020020714.2511718-1-ojeda@kernel.org

Merge tag 'x86_core_for_v6.18_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

2025-10-11T18:19:16+00:00

Pull more x86 updates from Borislav Petkov: - Remove a bunch of asm implementing condition flags testing in KVM's emulator in favor of int3_emulate_jcc() which is written in C - Replace KVM fastops with C-based stubs which avoids problems with the fastop infra related to latter not adhering to the C ABI due to their special calling convention and, more importantly, bypassing compiler control-flow integrity checking because they're written in asm - Remove wrongly used static branches and other ugliness accumulated over time in hyperv's hypercall implementation with a proper static function call to the correct hypervisor call variant - Add some fixes and modifications to allow running FRED-enabled kernels in KVM even on non-FRED hardware - Add kCFI improvements like validating indirect calls and prepare for enabling kCFI with GCC. Add cmdline params documentation and other code cleanups - Use the single-byte 0xd6 insn as the official #UD single-byte undefined opcode instruction as agreed upon by both x86 vendors - Other smaller cleanups and touchups all over the place * tag 'x86_core_for_v6.18_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (24 commits) x86,retpoline: Optimize patch_retpoline() x86,ibt: Use UDB instead of 0xEA x86/cfi: Remove __noinitretpoline and __noretpoline x86/cfi: Add "debug" option to "cfi=" bootparam x86/cfi: Standardize on common "CFI:" prefix for CFI reports x86/cfi: Document the "cfi=" bootparam options x86/traps: Clarify KCFI instruction layout compiler_types.h: Move __nocfi out of compiler-specific header objtool: Validate kCFI calls x86/fred: KVM: VMX: Always use FRED for IRQs when CONFIG_X86_FRED=y x86/fred: Play nice with invoking asm_fred_entry_from_kvm() on non-FRED hardware x86/fred: Install system vector handlers even if FRED isn't fully enabled x86/hyperv: Use direct call to hypercall-page x86/hyperv: Clean up hv_do_hypercall() KVM: x86: Remove fastops KVM: x86: Convert em_salc() to C KVM: x86: Introduce EM_ASM_3WCL KVM: x86: Introduce EM_ASM_1SRC2 KVM: x86: Introduce EM_ASM_2CL KVM: x86: Introduce EM_ASM_2W ...

Merge tag 'kbuild-6.18-1' of git://git.kernel.org/pub/scm/linux/kernel/git/kbuild/linux

2025-10-02T03:58:51+00:00

Pull Kbuild updates from Nathan Chancellor: - Extend modules.builtin.modinfo to include module aliases from MODULE_DEVICE_TABLE for builtin modules so that userspace tools (such as kmod) can verify that a particular module alias will be handled by a builtin module - Bump the minimum version of LLVM for building the kernel to 15.0.0 - Upgrade several userspace API checks in headers_check.pl to errors - Unify and consolidate CONFIG_WERROR / W=e handling - Turn assembler and linker warnings into errors with CONFIG_WERROR / W=e - Respect CONFIG_WERROR / W=e when building userspace programs (userprogs) - Enable -Werror unconditionally when building host programs (hostprogs) - Support copy_file_range() and data segment alignment in gen_init_cpio to improve performance on filesystems that support reflinks such as btrfs and XFS - Miscellaneous small changes to scripts and configuration files * tag 'kbuild-6.18-1' of git://git.kernel.org/pub/scm/linux/kernel/git/kbuild/linux: (47 commits) modpost: Initialize builtin_modname to stop SIGSEGVs Documentation: kbuild: note CONFIG_DEBUG_EFI in reproducible builds kbuild: vmlinux.unstripped should always depend on .vmlinux.export.o modpost: Create modalias for builtin modules modpost: Add modname to mod_device_table alias scsi: Always define blogic_pci_tbl structure kbuild: extract modules.builtin.modinfo from vmlinux.unstripped kbuild: keep .modinfo section in vmlinux.unstripped kbuild: always create intermediate vmlinux.unstripped s390: vmlinux.lds.S: Reorder sections KMSAN: Remove tautological checks objtool: Drop noinstr hack for KCSAN_WEAK_MEMORY lib/Kconfig.debug: Drop CLANG_VERSION check from DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT riscv: Remove ld.lld version checks from many TOOLCHAIN_HAS configs riscv: Unconditionally use linker relaxation riscv: Remove version check for LTO_CLANG selects powerpc: Drop unnecessary initializations in __copy_inst_from_kernel_nofault() mips: Unconditionally select ARCH_HAS_CURRENT_STACK_POINTER arm64: Remove tautological LLVM Kconfig conditions ARM: Clean up definition of ARM_HAS_GROUP_RELOCS ...

objtool: Ignore pi_cfi_ prefixed symbols

2025-09-09T10:02:01+00:00

When building with CONFIG_CFI_CLANG=y after the recent series to separate the x86 startup code, there are objtool warnings along the lines of: vmlinux.o: warning: objtool: __pi___cfi_startup_64_load_idt() falls through to next function __pi_startup_64_load_idt() vmlinux.o: warning: objtool: __pi___cfi_startup_64_setup_gdt_idt() falls through to next function __pi_startup_64_setup_gdt_idt() vmlinux.o: warning: objtool: __pi___cfi___startup_64() falls through to next function __pi___startup_64() As the comment in validate_branch() states, this is expected, so ignore these symbols in the same way that __cfi_ and __pfx_ symbols are already ignored for the rest of the kernel. Fixes: 7b38dec3c5af ("x86/boot: Create a confined code area for startup code") Signed-off-by: Nathan Chancellor Signed-off-by: Borislav Petkov (AMD) Acked-by: Ard Biesheuvel

x86/boot: Create a confined code area for startup code

2025-09-03T16:00:01+00:00

In order to be able to have tight control over which code may execute from the early 1:1 mapping of memory, but still link vmlinux as a single executable, prefix all symbol references in startup code with __pi_, and invoke it from outside using the __pi_ prefix. Use objtool to check that no absolute symbol references are present in the startup code, as these cannot be used from code running from the 1:1 mapping. Note that this also requires disabling the latent-entropy GCC plugin, as the global symbol references that it injects would require explicit exports, and given that the startup code rarely executes more than once, it is not a useful source of entropy anyway. Signed-off-by: Ard Biesheuvel Signed-off-by: Borislav Petkov (AMD) Link: https://lore.kernel.org/20250828102202.1849035-43-ardb+git@google.com

objtool: Add action to check for absence of absolute relocations

2025-09-03T15:59:51+00:00

The x86 startup code must not use absolute references to code or data, as it executes before the kernel virtual mapping is up. Add an action to objtool to check all allocatable sections (with the exception of __patchable_function_entries, which uses absolute references for nebulous reasons) and raise an error if any absolute references are found. Note that debug sections typically contain lots of absolute references too, but those are not allocatable so they will be ignored. Signed-off-by: Ard Biesheuvel Signed-off-by: Borislav Petkov (AMD) Acked-by: Peter Zijlstra (Intel) Link: https://lore.kernel.org/20250828102202.1849035-39-ardb+git@google.com