kernel.git/arch/x86/boot/compressed, branch linux-5.4.y

x86/boot/compressed: prefer cc-option for CFLAGS additions

2025-06-27T10:02:50Z

commit 994f5f7816ff963f49269cfc97f63cb2e4edb84f upstream. as-option tests new options using KBUILD_CFLAGS, which causes problems when using as-option to update KBUILD_AFLAGS because many compiler options are not valid assembler options. This will be fixed in a follow up patch. Before doing so, move the assembler test for -Wa,-mrelax-relocations=no from using as-option to cc-option. Link: https://lore.kernel.org/llvm/CAK7LNATcHt7GcXZ=jMszyH=+M_LC9Qr6yeAGRCBbE6xriLxtUQ@mail.gmail.com/ Suggested-by: Masahiro Yamada Reviewed-by: Nathan Chancellor Tested-by: Nathan Chancellor Signed-off-by: Nick Desaulniers Signed-off-by: Nathan Chancellor Tested-by: Linux Kernel Functional Testing Tested-by: Anders Roxell Signed-off-by: Masahiro Yamada Signed-off-by: Nathan Chancellor Signed-off-by: Greg Kroah-Hartman

x86/decompressor: Don't rely on upper 32 bits of GPRs being preserved

2023-09-23T08:59:41Z

[ Upstream commit 264b82fdb4989cf6a44a2bcd0c6ea05e8026b2ac ] The 4-to-5 level mode switch trampoline disables long mode and paging in order to be able to flick the LA57 bit. According to section 3.4.1.1 of the x86 architecture manual [0], 64-bit GPRs might not retain the upper 32 bits of their contents across such a mode switch. Given that RBP, RBX and RSI are live at this point, preserve them on the stack, along with the return address that might be above 4G as well. [0] Intel® 64 and IA-32 Architectures Software Developer’s Manual, Volume 1: Basic Architecture "Because the upper 32 bits of 64-bit general-purpose registers are undefined in 32-bit modes, the upper 32 bits of any general-purpose register are not preserved when switching from 64-bit mode to a 32-bit mode (to protected mode or compatibility mode). Software must not depend on these bits to maintain a value after a 64-bit to 32-bit mode switch." Fixes: 194a9749c73d650c ("x86/boot/compressed/64: Handle 5-level paging boot if kernel is above 4G") Signed-off-by: Ard Biesheuvel Signed-off-by: Borislav Petkov (AMD) Link: https://lore.kernel.org/r/20230807162720.545787-2-ardb@kernel.org Signed-off-by: Sasha Levin

x86/boot: Annotate local functions

2023-09-23T08:59:41Z

[ Upstream commit deff8a24e1021fb39dddf5f6bc5832e0e3a632ea ] .Lrelocated, .Lpaging_enabled, .Lno_longmode, and .Lin_pm32 are self-standing local functions, annotate them as such and preserve "no alignment". The annotations do not generate anything yet. Signed-off-by: Jiri Slaby Signed-off-by: Borislav Petkov Cc: Cao jin Cc: Greg Kroah-Hartman Cc: "H. Peter Anvin" Cc: Ingo Molnar Cc: Kate Stewart Cc: "Kirill A. Shutemov" Cc: linux-arch@vger.kernel.org Cc: Thomas Gleixner Cc: Wei Huang Cc: x86-ml Cc: Xiaoyao Li Link: https://lkml.kernel.org/r/20191011115108.12392-8-jslaby@suse.cz Stable-dep-of: 264b82fdb498 ("x86/decompressor: Don't rely on upper 32 bits of GPRs being preserved") Signed-off-by: Sasha Levin

x86: link vdso and boot with -z noexecstack --no-warn-rwx-segments

2022-08-25T09:17:17Z

commit ffcf9c5700e49c0aee42dcba9a12ba21338e8136 upstream. Users of GNU ld (BFD) from binutils 2.39+ will observe multiple instances of a new warning when linking kernels in the form: ld: warning: arch/x86/boot/pmjump.o: missing .note.GNU-stack section implies executable stack ld: NOTE: This behaviour is deprecated and will be removed in a future version of the linker ld: warning: arch/x86/boot/compressed/vmlinux has a LOAD segment with RWX permissions Generally, we would like to avoid the stack being executable. Because there could be a need for the stack to be executable, assembler sources have to opt-in to this security feature via explicit creation of the .note.GNU-stack feature (which compilers create by default) or command line flag --noexecstack. Or we can simply tell the linker the production of such sections is irrelevant and to link the stack as --noexecstack. LLVM's LLD linker defaults to -z noexecstack, so this flag isn't strictly necessary when linking with LLD, only BFD, but it doesn't hurt to be explicit here for all linkers IMO. --no-warn-rwx-segments is currently BFD specific and only available in the current latest release, so it's wrapped in an ld-option check. While the kernel makes extensive usage of ELF sections, it doesn't use permissions from ELF segments. Link: https://lore.kernel.org/linux-block/3af4127a-f453-4cf7-f133-a181cce06f73@kernel.dk/ Link: https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=ba951afb99912da01a6e8434126b8fac7aa75107 Link: https://github.com/llvm/llvm-project/issues/57009 Reported-and-tested-by: Jens Axboe Suggested-by: Fangrui Song Signed-off-by: Nick Desaulniers Signed-off-by: Linus Torvalds Signed-off-by: Greg Kroah-Hartman

x86/asm: Replace __force_order with a memory clobber

2020-10-29T08:58:01Z

[ Upstream commit aa5cacdc29d76a005cbbee018a47faa6e724dd2d ] The CRn accessor functions use __force_order as a dummy operand to prevent the compiler from reordering CRn reads/writes with respect to each other. The fact that the asm is volatile should be enough to prevent this: volatile asm statements should be executed in program order. However GCC 4.9.x and 5.x have a bug that might result in reordering. This was fixed in 8.1, 7.3 and 6.5. Versions prior to these, including 5.x and 4.9.x, may reorder volatile asm statements with respect to each other. There are some issues with __force_order as implemented: - It is used only as an input operand for the write functions, and hence doesn't do anything additional to prevent reordering writes. - It allows memory accesses to be cached/reordered across write functions, but CRn writes affect the semantics of memory accesses, so this could be dangerous. - __force_order is not actually defined in the kernel proper, but the LLVM toolchain can in some cases require a definition: LLVM (as well as GCC 4.9) requires it for PIE code, which is why the compressed kernel has a definition, but also the clang integrated assembler may consider the address of __force_order to be significant, resulting in a reference that requires a definition. Fix this by: - Using a memory clobber for the write functions to additionally prevent caching/reordering memory accesses across CRn writes. - Using a dummy input operand with an arbitrary constant address for the read functions, instead of a global variable. This will prevent reads from being reordered across writes, while allowing memory loads to be cached/reordered across CRn reads, which should be safe. Signed-off-by: Arvind Sankar Signed-off-by: Borislav Petkov Reviewed-by: Kees Cook Reviewed-by: Miguel Ojeda Tested-by: Nathan Chancellor Tested-by: Sedat Dilek Link: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82602 Link: https://lore.kernel.org/lkml/20200527135329.1172644-1-arnd@arndb.de/ Link: https://lkml.kernel.org/r/20200902232152.3709896-1-nivedita@alum.mit.edu Signed-off-by: Sasha Levin

x86/boot/compressed: Disable relocation relaxation

2020-09-23T10:40:46Z

commit 09e43968db40c33a73e9ddbfd937f46d5c334924 upstream. The x86-64 psABI [0] specifies special relocation types (R_X86_64_[REX_]GOTPCRELX) for indirection through the Global Offset Table, semantically equivalent to R_X86_64_GOTPCREL, which the linker can take advantage of for optimization (relaxation) at link time. This is supported by LLD and binutils versions 2.26 onwards. The compressed kernel is position-independent code, however, when using LLD or binutils versions before 2.27, it must be linked without the -pie option. In this case, the linker may optimize certain instructions into a non-position-independent form, by converting foo@GOTPCREL(%rip) to $foo. This potential issue has been present with LLD and binutils-2.26 for a long time, but it has never manifested itself before now: - LLD and binutils-2.26 only relax movq foo@GOTPCREL(%rip), %reg to leaq foo(%rip), %reg which is still position-independent, rather than mov $foo, %reg which is permitted by the psABI when -pie is not enabled. - GCC happens to only generate GOTPCREL relocations on mov instructions. - CLang does generate GOTPCREL relocations on non-mov instructions, but when building the compressed kernel, it uses its integrated assembler (due to the redefinition of KBUILD_CFLAGS dropping -no-integrated-as), which has so far defaulted to not generating the GOTPCRELX relocations. Nick Desaulniers reports [1,2]: "A recent change [3] to a default value of configuration variable (ENABLE_X86_RELAX_RELOCATIONS OFF -> ON) in LLVM now causes Clang's integrated assembler to emit R_X86_64_GOTPCRELX/R_X86_64_REX_GOTPCRELX relocations. LLD will relax instructions with these relocations based on whether the image is being linked as position independent or not. When not, then LLD will relax these instructions to use absolute addressing mode (R_RELAX_GOT_PC_NOPIC). This causes kernels built with Clang and linked with LLD to fail to boot." Patch series [4] is a solution to allow the compressed kernel to be linked with -pie unconditionally, but even if merged is unlikely to be backported. As a simple solution that can be applied to stable as well, prevent the assembler from generating the relaxed relocation types using the -mrelax-relocations=no option. For ease of backporting, do this unconditionally. [0] https://gitlab.com/x86-psABIs/x86-64-ABI/-/blob/master/x86-64-ABI/linker-optimization.tex#L65 [1] https://lore.kernel.org/lkml/20200807194100.3570838-1-ndesaulniers@google.com/ [2] https://github.com/ClangBuiltLinux/linux/issues/1121 [3] https://reviews.llvm.org/rGc41a18cf61790fc898dcda1055c3efbf442c14c0 [4] https://lore.kernel.org/lkml/20200731202738.2577854-1-nivedita@alum.mit.edu/ Reported-by: Nick Desaulniers Signed-off-by: Arvind Sankar Signed-off-by: Ingo Molnar Tested-by: Nick Desaulniers Tested-by: Sedat Dilek Acked-by: Ard Biesheuvel Reviewed-by: Nick Desaulniers Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20200812004308.1448603-1-nivedita@alum.mit.edu Signed-off-by: Greg Kroah-Hartman

x86/boot: kbuild: allow readelf executable to be specified

2020-08-26T08:40:46Z

commit eefb8c124fd969e9a174ff2bedff86aa305a7438 upstream. Introduce a new READELF variable to top-level Makefile, so the name of readelf binary can be specified. Before this change the name of the binary was hardcoded to "$(CROSS_COMPILE)readelf" which might not be present for every toolchain. This allows to build with LLVM Object Reader by using make parameter READELF=llvm-readelf. Link: https://github.com/ClangBuiltLinux/linux/issues/771 Signed-off-by: Dmitry Golovin Reviewed-by: Nick Desaulniers Signed-off-by: Masahiro Yamada Signed-off-by: Nick Desaulniers Signed-off-by: Greg Kroah-Hartman

x86/boot: Correct relocation destination on old linkers

2020-06-22T07:31:07Z

[ Upstream commit 5214028dd89e49ba27007c3ee475279e584261f0 ] For the 32-bit kernel, as described in 6d92bc9d483a ("x86/build: Build compressed x86 kernels as PIE"), pre-2.26 binutils generates R_386_32 relocations in PIE mode. Since the startup code does not perform relocation, any reloc entry with R_386_32 will remain as 0 in the executing code. Commit 974f221c84b0 ("x86/boot: Move compressed kernel to the end of the decompression buffer") added a new symbol _end but did not mark it hidden, which doesn't give the correct offset on older linkers. This causes the compressed kernel to be copied beyond the end of the decompression buffer, rather than flush against it. This region of memory may be reserved or already allocated for other purposes by the bootloader. Mark _end as hidden to fix. This changes the relocation from R_386_32 to R_386_RELATIVE even on the pre-2.26 binutils. For 64-bit, this is not strictly necessary, as the 64-bit kernel is only built as PIE if the linker supports -z noreloc-overflow, which implies binutils-2.27+, but for consistency, mark _end as hidden here too. The below illustrates the before/after impact of the patch using binutils-2.25 and gcc-4.6.4 (locally compiled from source) and QEMU. Disassembly before patch: 48: 8b 86 60 02 00 00 mov 0x260(%esi),%eax 4e: 2d 00 00 00 00 sub $0x0,%eax 4f: R_386_32 _end Disassembly after patch: 48: 8b 86 60 02 00 00 mov 0x260(%esi),%eax 4e: 2d 00 f0 76 00 sub $0x76f000,%eax 4f: R_386_RELATIVE *ABS* Dump from extract_kernel before patch: early console in extract_kernel input_data: 0x0207c098 <--- this is at output + init_size input_len: 0x0074fef1 output: 0x01000000 output_len: 0x00fa63d0 kernel_total_size: 0x0107c000 needed_size: 0x0107c000 Dump from extract_kernel after patch: early console in extract_kernel input_data: 0x0190d098 <--- this is at output + init_size - _end input_len: 0x0074fef1 output: 0x01000000 output_len: 0x00fa63d0 kernel_total_size: 0x0107c000 needed_size: 0x0107c000 Fixes: 974f221c84b0 ("x86/boot: Move compressed kernel to the end of the decompression buffer") Signed-off-by: Arvind Sankar Signed-off-by: Borislav Petkov Link: https://lkml.kernel.org/r/20200207214926.3564079-1-nivedita@alum.mit.edu Signed-off-by: Sasha Levin

x86/boot: Use unsigned comparison for addresses

2020-04-17T08:50:03Z

[ Upstream commit 81a34892c2c7c809f9c4e22c5ac936ae673fb9a2 ] The load address is compared with LOAD_PHYSICAL_ADDR using a signed comparison currently (using jge instruction). When loading a 64-bit kernel using the new efi32_pe_entry() point added by: 97aa276579b2 ("efi/x86: Add true mixed mode entry point into .compat section") using Qemu with -m 3072, the firmware actually loads us above 2Gb, resulting in a very early crash. Use the JAE instruction to perform a unsigned comparison instead, as physical addresses should be considered unsigned. Signed-off-by: Arvind Sankar Signed-off-by: Ard Biesheuvel Signed-off-by: Ingo Molnar Link: https://lore.kernel.org/r/20200301230436.2246909-6-nivedita@alum.mit.edu Link: https://lore.kernel.org/r/20200308080859.21568-14-ardb@kernel.org Signed-off-by: Sasha Levin

x86/boot/compressed: Don't declare __force_order in kaslr_64.c

2020-03-12T12:00:15Z

[ Upstream commit df6d4f9db79c1a5d6f48b59db35ccd1e9ff9adfc ] GCC 10 changed the default to -fno-common, which leads to LD arch/x86/boot/compressed/vmlinux ld: arch/x86/boot/compressed/pgtable_64.o:(.bss+0x0): multiple definition of `__force_order'; \ arch/x86/boot/compressed/kaslr_64.o:(.bss+0x0): first defined here make[2]: *** [arch/x86/boot/compressed/Makefile:119: arch/x86/boot/compressed/vmlinux] Error 1 Since __force_order is already provided in pgtable_64.c, there is no need to declare __force_order in kaslr_64.c. Signed-off-by: H.J. Lu Signed-off-by: Borislav Petkov Link: https://lkml.kernel.org/r/20200124181811.4780-1-hjl.tools@gmail.com Signed-off-by: Sasha Levin