kernel.git/arch/arm/include/asm/uaccess.h, branch linux-6.0.y

uaccess: generalize access_ok()

2022-02-25T08:36:05Z

There are many different ways that access_ok() is defined across architectures, but in the end, they all just compare against the user_addr_max() value or they accept anything. Provide one definition that works for most architectures, checking against TASK_SIZE_MAX for user processes or skipping the check inside of uaccess_kernel() sections. For architectures without CONFIG_SET_FS(), this should be the fastest check, as it comes down to a single comparison of a pointer against a compile-time constant, while the architecture specific versions tend to do something more complex for historic reasons or get something wrong. Type checking for __user annotations is handled inconsistently across architectures, but this is easily simplified as well by using an inline function that takes a 'const void __user *' argument. A handful of callers need an extra __user annotation for this. Some architectures had trick to use 33-bit or 65-bit arithmetic on the addresses to calculate the overflow, however this simpler version uses fewer registers, which means it can produce better object code in the end despite needing a second (statically predicted) branch. Reviewed-by: Christoph Hellwig Acked-by: Mark Rutland [arm64, asm-generic] Acked-by: Geert Uytterhoeven Acked-by: Stafford Horne Acked-by: Dinh Nguyen Signed-off-by: Arnd Bergmann

uaccess: add generic __{get,put}_kernel_nofault

2022-02-25T08:36:05Z

Nine architectures are still missing __{get,put}_kernel_nofault: alpha, ia64, microblaze, nds32, nios2, openrisc, sh, sparc32, xtensa. Add a generic version that lets everything use the normal copy_{from,to}_kernel_nofault() code based on these, removing the last use of get_fs()/set_fs() from architecture-independent code. Reviewed-by: Christoph Hellwig Acked-by: Geert Uytterhoeven Signed-off-by: Arnd Bergmann

ARM: 9179/1: uaccess: avoid alignment faults in copy_[from|to]_kernel_nofault

2022-01-19T11:10:53Z

The helpers that are used to implement copy_from_kernel_nofault() and copy_to_kernel_nofault() cast a void* to a pointer to a wider type, which may result in alignment faults on ARM if the compiler decides to use double-word or multiple-word load/store instructions. Only configurations that define CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS=y are affected, given that commit 2423de2e6f4d ("ARM: 9115/1: mm/maccess: fix unaligned copy_{from,to}_kernel_nofault") ensures that dst and src are sufficiently aligned otherwise. So use the unaligned accessors for accessing dst and src in cases where they may be misaligned. Cc: # depends on 2423de2e6f4d Fixes: 2df4c9a741a0 ("ARM: 9112/1: uaccess: add __{get,put}_kernel_nofault") Reviewed-by: Arnd Bergmann Signed-off-by: Ard Biesheuvel Signed-off-by: Russell King (Oracle)

ARM: 9132/1: Fix __get_user_check failure with ARM KASAN images

2021-10-19T09:37:35Z

ARM: kasan: Fix __get_user_check failure with kasan In macro __get_user_check defined in arch/arm/include/asm/uaccess.h, error code is store in register int __e(r0). When kasan is enabled, assigning value to kernel address might trigger kasan check, which unexpectedly overwrites r0 and causes undefined behavior on arm kasan images. One example is failure in do_futex and results in process soft lockup. Log: watchdog: BUG: soft lockup - CPU#0 stuck for 62946ms! [rs:main Q:Reg:1151] ... (__asan_store4) from (futex_wait_setup+0xf8/0x2b4) (futex_wait_setup) from (futex_wait+0x138/0x394) (futex_wait) from (do_futex+0x164/0xe40) (do_futex) from (sys_futex_time32+0x178/0x230) (sys_futex_time32) from (ret_fast_syscall+0x0/0x50) The soft lockup happens in function futex_wait_setup. The reason is function get_futex_value_locked always return EINVAL, thus pc jump back to retry label and causes looping. This line in function get_futex_value_locked ret = __get_user(*dest, from); is expanded to *dest = (typeof(*(p))) __r2; , in macro __get_user_check. Writing to pointer dest triggers kasan check and overwrites the return value of __get_user_x function. The assembly code of get_futex_value_locked in kernel/futex.c: ... c01f6dc8: eb0b020e bl c04b7608 <__get_user_4> // "x = (typeof(*(p))) __r2;" triggers kasan check and r0 is overwritten c01f6dCc: e1a00007 mov r0, r7 c01f6dd0: e1a05002 mov r5, r2 c01f6dd4: eb04f1e6 bl c0333574 <__asan_store4> c01f6dd8: e5875000 str r5, [r7] // save ret value of __get_user(*dest, from), which is dest address now c01f6ddc: e1a05000 mov r5, r0 ... // checking return value of __get_user failed c01f6e00: e3550000 cmp r5, #0 ... c01f6e0c: 01a00005 moveq r0, r5 // assign return value to EINVAL c01f6e10: 13e0000d mvnne r0, #13 Return value is the destination address of get_user thus certainly non-zero, so get_futex_value_locked always return EINVAL. Fix it by using a tmp vairable to store the error code before the assignment. This fix has no effects to non-kasan images thanks to compiler optimization. It only affects cases that overwrite r0 due to kasan check. This should fix bug discussed in Link: [1] https://lore.kernel.org/linux-arm-kernel/0ef7c2a5-5d8b-c5e0-63fa-31693fd4495c@gmail.com/ Fixes: 421015713b30 ("ARM: 9017/2: Enable KASan for ARM") Signed-off-by: Lexi Shao Signed-off-by: Russell King (Oracle)

ARM: 9113/1: uaccess: remove set_fs() implementation

2021-08-20T10:39:27Z

There are no remaining callers of set_fs(), so just remove it along with all associated code that operates on thread_info->addr_limit. There are still further optimizations that can be done: - In get_user(), the address check could be moved entirely into the out of line code, rather than passing a constant as an argument, - I assume the DACR handling can be simplified as we now only change it during user access when CONFIG_CPU_SW_DOMAIN_PAN is set, but not during set_fs(). Acked-by: Christoph Hellwig Signed-off-by: Arnd Bergmann Signed-off-by: Russell King (Oracle)

ARM: 9112/1: uaccess: add __{get,put}_kernel_nofault

2021-08-20T10:39:27Z

These mimic the behavior of get_user and put_user, except for domain switching, address limit checking and handling of mismatched sizes, none of which are relevant here. To work with pre-Armv6 kernels, this has to avoid TUSER() inside of the new macros, the new approach passes the "t" string along with the opcode, which is a bit uglier but avoids duplicating more code. As there is no __get_user_asm_dword(), I work around it by copying 32 bit at a time, which is possible because the output size is known. Acked-by: Christoph Hellwig Signed-off-by: Arnd Bergmann Signed-off-by: Russell King (Oracle)

uaccess: remove segment_eq

2020-08-12T17:57:58Z

segment_eq is only used to implement uaccess_kernel. Just open code uaccess_kernel in the arch uaccess headers and remove one layer of indirection. Signed-off-by: Christoph Hellwig Signed-off-by: Andrew Morton Acked-by: Linus Torvalds Acked-by: Greentime Hu Acked-by: Geert Uytterhoeven Cc: Nick Hu Cc: Vincent Chen Cc: Paul Walmsley Cc: Palmer Dabbelt Link: http://lkml.kernel.org/r/20200710135706.537715-5-hch@lst.de Signed-off-by: Linus Torvalds

ARM: uaccess: add further explanation of __range_ok()

2020-07-22T08:50:55Z

Detail the success return condition, and that we rely on KERNEL_DS being zero for this to operate correctly. Signed-off-by: Russell King

ARM: 8908/1: add __always_inline to functions called from __get_user_check()

2019-10-10T21:23:19Z

KernelCI reports that bcm2835_defconfig is no longer booting since commit ac7c3e4ff401 ("compiler: enable CONFIG_OPTIMIZE_INLINING forcibly") (https://lkml.org/lkml/2019/9/26/825). I also received a regression report from Nicolas Saenz Julienne (https://lkml.org/lkml/2019/9/27/263). This problem has cropped up on bcm2835_defconfig because it enables CONFIG_CC_OPTIMIZE_FOR_SIZE. The compiler tends to prefer not inlining functions with -Os. I was able to reproduce it with other boards and defconfig files by manually enabling CONFIG_CC_OPTIMIZE_FOR_SIZE. The __get_user_check() specifically uses r0, r1, r2 registers. So, uaccess_save_and_enable() and uaccess_restore() must be inlined. Otherwise, those register assignments would be entirely dropped, according to my analysis of the disassembly. Prior to commit 9012d011660e ("compiler: allow all arches to enable CONFIG_OPTIMIZE_INLINING"), the 'inline' marker was always enough for inlining functions, except on x86. Since that commit, all architectures can enable CONFIG_OPTIMIZE_INLINING. So, __always_inline is now the only guaranteed way of forcible inlining. I added __always_inline to 4 functions in the call-graph from the __get_user_check() macro. Fixes: 9012d011660e ("compiler: allow all arches to enable CONFIG_OPTIMIZE_INLINING") Reported-by: "kernelci.org bot" Reported-by: Nicolas Saenz Julienne Signed-off-by: Masahiro Yamada Tested-by: Nicolas Saenz Julienne Signed-off-by: Russell King

treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 500

2019-06-19T15:09:55Z

Based on 2 normalized pattern(s): this program is free software you can redistribute it and or modify it under the terms of the gnu general public license version 2 as published by the free software foundation this program is free software you can redistribute it and or modify it under the terms of the gnu general public license version 2 as published by the free software foundation # extracted by the scancode license scanner the SPDX license identifier GPL-2.0-only has been chosen to replace the boilerplate/reference in 4122 file(s). Signed-off-by: Thomas Gleixner Reviewed-by: Enrico Weigelt Reviewed-by: Kate Stewart Reviewed-by: Allison Randal Cc: linux-spdx@vger.kernel.org Link: https://lkml.kernel.org/r/20190604081206.933168790@linutronix.de Signed-off-by: Greg Kroah-Hartman