kernel/linux.git/arch/arm/boot/compressed, branch v5.10.257

ARM: 9159/1: decompressor: Avoid UNPREDICTABLE NOP encoding

2022-01-27T09:53:57+00:00

[ Upstream commit a92882a4d270fbcc021ee6848de5e48b7f0d27f3 ] In the decompressor's head.S we need to start with an instruction that is some kind of NOP, but also mimics as the PE/COFF header, when the kernel is linked as an UEFI application. The clever solution here is "tstne r0, #0x4d000", which in the worst case just clobbers the condition flags, and bears the magic "MZ" signature in the lowest 16 bits. However the encoding used (0x13105a4d) is actually not valid, since bits [15:12] are supposed to be 0 (written as "(0)" in the ARM ARM). Violating this is UNPREDICTABLE, and *can* trigger an UNDEFINED exception. Common Cortex cores seem to ignore those bits, but QEMU chooses to trap, so the code goes fishing because of a missing exception handler at this point. We are just saved by the fact that commonly (with -kernel or when running from U-Boot) the "Z" bit is set, so the instruction is never executed. See [0] for more details. To make things more robust and avoid UNPREDICTABLE behaviour in the kernel code, lets replace this with a "two-instruction NOP": The first instruction is an exclusive OR, the effect of which the second instruction reverts. This does not leave any trace, neither in a register nor in the condition flags. Also it's a perfectly valid encoding. Kudos to Peter Maydell for coming up with this gem. [0] https://lore.kernel.org/qemu-devel/YTPIdbUCmwagL5%2FD@os.inf.tu-dresden.de/T/ Link: https://lore.kernel.org/linux-arm-kernel/20210908162617.104962-1-andre.przywara@arm.com/T/ Fixes: 81a0bc39ea19 ("ARM: add UEFI stub support") Signed-off-by: Andre Przywara Reported-by: Adam Lackorzynski Suggested-by: Peter Maydell Reviewed-by: Ard Biesheuvel Reviewed-by: Linus Walleij Signed-off-by: Russell King (Oracle) Signed-off-by: Sasha Levin

ARM: 9134/1: remove duplicate memcpy() definition

2021-11-02T18:48:18+00:00

commit eaf6cc7165c9c5aa3c2f9faa03a98598123d0afb upstream. Both the decompressor code and the kasan logic try to override the memcpy() and memmove() definitions, which leading to a clash in a KASAN-enabled kernel with XZ decompression: arch/arm/boot/compressed/decompress.c:50:9: error: 'memmove' macro redefined [-Werror,-Wmacro-redefined] #define memmove memmove ^ arch/arm/include/asm/string.h:59:9: note: previous definition is here #define memmove(dst, src, len) __memmove(dst, src, len) ^ arch/arm/boot/compressed/decompress.c:51:9: error: 'memcpy' macro redefined [-Werror,-Wmacro-redefined] #define memcpy memcpy ^ arch/arm/include/asm/string.h:58:9: note: previous definition is here #define memcpy(dst, src, len) __memcpy(dst, src, len) ^ Here we want the set of functions from the decompressor, so undefine the other macros before the override. Link: https://lore.kernel.org/linux-arm-kernel/CACRpkdZYJogU_SN3H9oeVq=zJkRgRT1gDz3xp59gdqWXxw-B=w@mail.gmail.com/ Link: https://lore.kernel.org/lkml/202105091112.F5rmd4By-lkp@intel.com/ Fixes: d6d51a96c7d6 ("ARM: 9014/2: Replace string mem* functions for KASan") Fixes: a7f464f3db93 ("ARM: 7001/2: Wire up support for the XZ decompressor") Reported-by: kernel test robot Reviewed-by: Linus Walleij Signed-off-by: Arnd Bergmann Signed-off-by: Russell King (Oracle) Signed-off-by: Greg Kroah-Hartman

ARM: 9105/1: atags_to_fdt: don't warn about stack size

2021-09-18T11:40:09+00:00

commit b30d0289de72c62516df03fdad8d53f552c69839 upstream. The merge_fdt_bootargs() function by definition consumes more than 1024 bytes of stack because it has a 1024 byte command line on the stack, meaning that we always get a warning when building this file: arch/arm/boot/compressed/atags_to_fdt.c: In function 'merge_fdt_bootargs': arch/arm/boot/compressed/atags_to_fdt.c:98:1: warning: the frame size of 1032 bytes is larger than 1024 bytes [-Wframe-larger-than=] However, as this is the decompressor and we know that it has a very shallow call chain, and we do not actually risk overflowing the kernel stack at runtime here. This just shuts up the warning by disabling the warning flag for this file. Tested on Nexus 7 2012 builds. Acked-by: Nicolas Pitre Signed-off-by: David Heidelberg Signed-off-by: Arnd Bergmann Cc: Signed-off-by: Russell King (Oracle) Signed-off-by: Greg Kroah-Hartman

ARM: 9056/1: decompressor: fix BSS size calculation for LLVM ld.lld

2021-05-11T12:47:12+00:00

commit c4e792d1acce31c2eb7b9193ab06ab94de05bf42 upstream. The LLVM ld.lld linker uses a different symbol type for __bss_start, resulting in the calculation of KBSS_SZ to be thrown off. Up until now, this has gone unnoticed as it only affects the appended DTB case, but pending changes for ARM in the way the decompressed kernel is cleaned from the caches has uncovered this problem. On a ld.lld build: $ nm vmlinux |grep bss_ c1c22034 D __bss_start c1c86e98 B __bss_stop resulting in $ readelf -s arch/arm/boot/compressed/vmlinux | grep bss_size 433: c1c86e98 0 NOTYPE GLOBAL DEFAULT ABS _kernel_bss_size which is obviously incorrect, and may cause the cache clean to access unmapped memory, or cause the size calculation to wrap, resulting in no cache clean to be performed at all. Fix this by updating the sed regex to take D type symbols into account. Link: https://lore.kernel.org/linux-arm-kernel/6c65bcef-d4e7-25fa-43cf-2c435bb61bb9@collabora.com/ Link: https://lore.kernel.org/linux-arm-kernel/20210205085220.31232-1-ardb@kernel.org/ Cc: # v4.19+ Reviewed-by: Nick Desaulniers Tested-by: Nick Desaulniers Reported-by: Guillaume Tucker Reported-by: "kernelci.org bot" Signed-off-by: Ard Biesheuvel Signed-off-by: Russell King Signed-off-by: Greg Kroah-Hartman

ARM: efistub: replace adrl pseudo-op with adr_l macro invocation

2021-03-17T16:06:26+00:00

commit 67e3f828bd4bf5e4eb4214dc4eb227d8f1c8a877 upstream. The ARM 'adrl' pseudo instruction is a bit problematic, as it does not exist in Thumb mode, and it is not implemented by Clang either. Since the Thumb variant has a slightly bigger range, it is sometimes necessary to emit the 'adrl' variant in ARM mode where Thumb mode can use adr just fine. However, that still leaves the Clang issue, which does not appear to be supporting this any time soon. So let's switch to the adr_l macro, which works for both ARM and Thumb, and has unlimited range. Reviewed-by: Nicolas Pitre Signed-off-by: Ard Biesheuvel Signed-off-by: Nick Desaulniers Signed-off-by: Sasha Levin

ARM: 9046/1: decompressor: Do not clear SCTLR.nTLSMD for ARMv7+ cores

2021-03-04T10:38:02+00:00

[ Upstream commit 2acb909750431030b65a0a2a17fd8afcbd813a84 ] It was observed that decompressor running on hardware implementing ARM v8.2 Load/Store Multiple Atomicity and Ordering Control (LSMAOC), say, as guest, would stuck just after: Uncompressing Linux... done, booting the kernel. The reason is that it clears nTLSMD bit when disabling caches: nTLSMD, bit [3] When ARMv8.2-LSMAOC is implemented: No Trap Load Multiple and Store Multiple to Device-nGRE/Device-nGnRE/Device-nGnRnE memory. 0b0 All memory accesses by A32 and T32 Load Multiple and Store Multiple at EL1 or EL0 that are marked at stage 1 as Device-nGRE/Device-nGnRE/Device-nGnRnE memory are trapped and generate a stage 1 Alignment fault. 0b1 All memory accesses by A32 and T32 Load Multiple and Store Multiple at EL1 or EL0 that are marked at stage 1 as Device-nGRE/Device-nGnRE/Device-nGnRnE memory are not trapped. This bit is permitted to be cached in a TLB. This field resets to 1. Otherwise: Reserved, RES1 So as effect we start getting traps we are not quite ready for. Looking into history it seems that mask used for SCTLR clear came from the similar code for ARMv4, where bit[3] is the enable/disable bit for the write buffer. That not applicable to ARMv7 and onwards, so retire that bit from the masks. Fixes: 7d09e85448dfa78e3e58186c934449aaf6d49b50 ("[ARM] 4393/2: ARMv7: Add uncompressing code for the new CPU Id format") Signed-off-by: Vladimir Murzin Signed-off-by: Russell King Signed-off-by: Sasha Levin

ARM: zImage: atags_to_fdt: Fix node names on added root nodes

2021-02-03T22:28:44+00:00

commit 30596ae0547dbda469d31a2678d9072fb0a3fa27 upstream. Commit 7536c7e03e74 ("of/fdt: Remove redundant kbasename function call") exposed a bug creating DT nodes in the ATAGS to DT fixup code. Non-existent nodes would mistaken get created with a leading '/'. The problem was fdt_path_offset() takes a full path while creating a node with fdt_add_subnode() takes just the basename. Since this we only add root child nodes, we can just skip over the '/'. Fixes: 7536c7e03e74 ("of/fdt: Remove redundant kbasename function call") Reported-by: Chris Packham Cc: Qi Zheng Cc: Russell King Signed-off-by: Rob Herring Tested-by: Chris Packham Link: https://lore.kernel.org/r/20210126023905.1631161-1-robh@kernel.org Signed-off-by: Greg Kroah-Hartman

ARM: 9036/1: uncompress: Fix dbgadtb size parameter name

2020-12-30T10:54:02+00:00

[ Upstream commit 1ecec38547d415054fdb63a231234f44396b6d06 ] The dbgadtb macro is passed the size of the appended DTB, not the end address. Fixes: c03e41470e901123 ("ARM: 9010/1: uncompress: Print the location of appended DTB") Signed-off-by: Geert Uytterhoeven Reviewed-by: Linus Walleij Signed-off-by: Russell King Signed-off-by: Sasha Levin

kbuild: Hoist '--orphan-handling' into Kconfig

2020-12-01T13:45:36+00:00

Currently, '--orphan-handling=warn' is spread out across four different architectures in their respective Makefiles, which makes it a little unruly to deal with in case it needs to be disabled for a specific linker version (in this case, ld.lld 10.0.1). To make it easier to control this, hoist this warning into Kconfig and the main Makefile so that disabling it is simpler, as the warning will only be enabled in a couple places (main Makefile and a couple of compressed boot folders that blow away LDFLAGS_vmlinx) and making it conditional is easier due to Kconfig syntax. One small additional benefit of this is saving a call to ld-option on incremental builds because we will have already evaluated it for CONFIG_LD_ORPHAN_WARN. To keep the list of supported architectures the same, introduce CONFIG_ARCH_WANT_LD_ORPHAN_WARN, which an architecture can select to gain this automatically after all of the sections are specified and size asserted. A special thanks to Kees Cook for the help text on this config. Link: https://github.com/ClangBuiltLinux/linux/issues/1187 Acked-by: Kees Cook Acked-by: Michael Ellerman (powerpc) Reviewed-by: Nick Desaulniers Tested-by: Nick Desaulniers Signed-off-by: Nathan Chancellor Signed-off-by: Masahiro Yamada

efi/arm: set HSCTLR Thumb2 bit correctly for HVC calls from HYP

2020-10-26T07:02:11+00:00

Commit db227c19e68db353 ("ARM: 8985/1: efi/decompressor: deal with HYP mode boot gracefully") updated the EFI entry code to permit firmware to invoke the EFI stub loader in HYP mode, with the MMU either enabled or disabled, neither of which is permitted by the EFI spec, but which does happen in the field. In the MMU on case, we remain in HYP mode as configured by the firmware, and rely on the fact that any HVC instruction issued in this mode will be dispatched via the SVC slot in the HYP vector table. However, this slot will point to a Thumb2 symbol if the kernel is built in Thumb2 mode, and so we have to configure HSCTLR to ensure that the exception handlers are invoked in Thumb2 mode as well. Fixes: db227c19e68db353 ("ARM: 8985/1: efi/decompressor: deal with HYP mode boot gracefully") Signed-off-by: Ard Biesheuvel