linux_dsm_epyc7002/arch/arm64/kernel/vmlinux.lds.S
Mark Rutland ca2ef4ffab arm64: insn: consistently handle exit text
A kernel built with KASAN && FTRACE_WITH_REGS && !MODULES, produces a
boot-time splat in the bowels of ftrace:

| [    0.000000] ftrace: allocating 32281 entries in 127 pages
| [    0.000000] ------------[ cut here ]------------
| [    0.000000] WARNING: CPU: 0 PID: 0 at kernel/trace/ftrace.c:2019 ftrace_bug+0x27c/0x328
| [    0.000000] CPU: 0 PID: 0 Comm: swapper Not tainted 5.4.0-rc3-00008-g7f08ae53a7e3 #13
| [    0.000000] Hardware name: linux,dummy-virt (DT)
| [    0.000000] pstate: 60000085 (nZCv daIf -PAN -UAO)
| [    0.000000] pc : ftrace_bug+0x27c/0x328
| [    0.000000] lr : ftrace_init+0x640/0x6cc
| [    0.000000] sp : ffffa000120e7e00
| [    0.000000] x29: ffffa000120e7e00 x28: ffff00006ac01b10
| [    0.000000] x27: ffff00006ac898c0 x26: dfffa00000000000
| [    0.000000] x25: ffffa000120ef290 x24: ffffa0001216df40
| [    0.000000] x23: 000000000000018d x22: ffffa0001244c700
| [    0.000000] x21: ffffa00011bf393c x20: ffff00006ac898c0
| [    0.000000] x19: 00000000ffffffff x18: 0000000000001584
| [    0.000000] x17: 0000000000001540 x16: 0000000000000007
| [    0.000000] x15: 0000000000000000 x14: ffffa00010432770
| [    0.000000] x13: ffff940002483519 x12: 1ffff40002483518
| [    0.000000] x11: 1ffff40002483518 x10: ffff940002483518
| [    0.000000] x9 : dfffa00000000000 x8 : 0000000000000001
| [    0.000000] x7 : ffff940002483519 x6 : ffffa0001241a8c0
| [    0.000000] x5 : ffff940002483519 x4 : ffff940002483519
| [    0.000000] x3 : ffffa00011780870 x2 : 0000000000000001
| [    0.000000] x1 : 1fffe0000d591318 x0 : 0000000000000000
| [    0.000000] Call trace:
| [    0.000000]  ftrace_bug+0x27c/0x328
| [    0.000000]  ftrace_init+0x640/0x6cc
| [    0.000000]  start_kernel+0x27c/0x654
| [    0.000000] random: get_random_bytes called from print_oops_end_marker+0x30/0x60 with crng_init=0
| [    0.000000] ---[ end trace 0000000000000000 ]---
| [    0.000000] ftrace faulted on writing
| [    0.000000] [<ffffa00011bf393c>] _GLOBAL__sub_D_65535_0___tracepoint_initcall_level+0x4/0x28
| [    0.000000] Initializing ftrace call sites
| [    0.000000] ftrace record flags: 0
| [    0.000000]  (0)
| [    0.000000]  expected tramp: ffffa000100b3344

This is due to an unfortunate combination of several factors.

Building with KASAN results in the compiler generating anonymous
functions to register/unregister global variables against the shadow
memory. These functions are placed in .text.startup/.text.exit, and
given mangled names like _GLOBAL__sub_{I,D}_65535_0_$OTHER_SYMBOL. The
kernel linker script places these in .init.text and .exit.text
respectively, which are both discarded at runtime as part of initmem.

Building with FTRACE_WITH_REGS uses -fpatchable-function-entry=2, which
also instruments KASAN's anonymous functions. When these are discarded
with the rest of initmem, ftrace removes dangling references to these
call sites.

Building without MODULES implicitly disables STRICT_MODULE_RWX, and
causes arm64's patch_map() function to treat any !core_kernel_text()
symbol as something that can be modified in-place. As core_kernel_text()
is only true for .text and .init.text, with the latter depending on
system_state < SYSTEM_RUNNING, we'll treat .exit.text as something that
can be patched in-place. However, .exit.text is mapped read-only.

Hence in this configuration the ftrace init code blows up while trying
to patch one of the functions generated by KASAN.

We could try to filter out the call sites in .exit.text rather than
initializing them, but this would be inconsistent with how we handle
.init.text, and requires hooking into core bits of ftrace. The behaviour
of patch_map() is also inconsistent today, so instead let's clean that
up and have it consistently handle .exit.text.

This patch teaches patch_map() to handle .exit.text at init time,
preventing the boot-time splat above. The flow of patch_map() is
reworked to make the logic clearer and minimize redundant
conditionality.

Fixes: 3b23e4991f ("arm64: implement ftrace with regs")
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Amit Daniel Kachhap <amit.kachhap@arm.com>
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc: Torsten Duwe <duwe@suse.de>
Cc: Will Deacon <will@kernel.org>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2019-12-04 11:32:20 +00:00

279 lines
6.2 KiB
ArmAsm

/* SPDX-License-Identifier: GPL-2.0 */
/*
* ld script to make ARM Linux kernel
* taken from the i386 version by Russell King
* Written by Martin Mares <mj@atrey.karlin.mff.cuni.cz>
*/
#include <asm-generic/vmlinux.lds.h>
#include <asm/cache.h>
#include <asm/kernel-pgtable.h>
#include <asm/thread_info.h>
#include <asm/memory.h>
#include <asm/page.h>
#include <asm/pgtable.h>
#include "image.h"
/* .exit.text needed in case of alternative patching */
#define ARM_EXIT_KEEP(x) x
#define ARM_EXIT_DISCARD(x)
OUTPUT_ARCH(aarch64)
ENTRY(_text)
jiffies = jiffies_64;
#define HYPERVISOR_TEXT \
/* \
* Align to 4 KB so that \
* a) the HYP vector table is at its minimum \
* alignment of 2048 bytes \
* b) the HYP init code will not cross a page \
* boundary if its size does not exceed \
* 4 KB (see related ASSERT() below) \
*/ \
. = ALIGN(SZ_4K); \
__hyp_idmap_text_start = .; \
*(.hyp.idmap.text) \
__hyp_idmap_text_end = .; \
__hyp_text_start = .; \
*(.hyp.text) \
__hyp_text_end = .;
#define IDMAP_TEXT \
. = ALIGN(SZ_4K); \
__idmap_text_start = .; \
*(.idmap.text) \
__idmap_text_end = .;
#ifdef CONFIG_HIBERNATION
#define HIBERNATE_TEXT \
. = ALIGN(SZ_4K); \
__hibernate_exit_text_start = .; \
*(.hibernate_exit.text) \
__hibernate_exit_text_end = .;
#else
#define HIBERNATE_TEXT
#endif
#ifdef CONFIG_UNMAP_KERNEL_AT_EL0
#define TRAMP_TEXT \
. = ALIGN(PAGE_SIZE); \
__entry_tramp_text_start = .; \
*(.entry.tramp.text) \
. = ALIGN(PAGE_SIZE); \
__entry_tramp_text_end = .;
#else
#define TRAMP_TEXT
#endif
/*
* The size of the PE/COFF section that covers the kernel image, which
* runs from stext to _edata, must be a round multiple of the PE/COFF
* FileAlignment, which we set to its minimum value of 0x200. 'stext'
* itself is 4 KB aligned, so padding out _edata to a 0x200 aligned
* boundary should be sufficient.
*/
PECOFF_FILE_ALIGNMENT = 0x200;
#ifdef CONFIG_EFI
#define PECOFF_EDATA_PADDING \
.pecoff_edata_padding : { BYTE(0); . = ALIGN(PECOFF_FILE_ALIGNMENT); }
#else
#define PECOFF_EDATA_PADDING
#endif
SECTIONS
{
/*
* XXX: The linker does not define how output sections are
* assigned to input sections when there are multiple statements
* matching the same input section name. There is no documented
* order of matching.
*/
/DISCARD/ : {
ARM_EXIT_DISCARD(EXIT_TEXT)
ARM_EXIT_DISCARD(EXIT_DATA)
EXIT_CALL
*(.discard)
*(.discard.*)
*(.interp .dynamic)
*(.dynsym .dynstr .hash .gnu.hash)
*(.eh_frame)
}
. = KIMAGE_VADDR + TEXT_OFFSET;
.head.text : {
_text = .;
HEAD_TEXT
}
.text : { /* Real text segment */
_stext = .; /* Text and read-only data */
IRQENTRY_TEXT
SOFTIRQENTRY_TEXT
ENTRY_TEXT
TEXT_TEXT
SCHED_TEXT
CPUIDLE_TEXT
LOCK_TEXT
KPROBES_TEXT
HYPERVISOR_TEXT
IDMAP_TEXT
HIBERNATE_TEXT
TRAMP_TEXT
*(.fixup)
*(.gnu.warning)
. = ALIGN(16);
*(.got) /* Global offset table */
}
. = ALIGN(SEGMENT_ALIGN);
_etext = .; /* End of text section */
RO_DATA(PAGE_SIZE) /* everything from this point to */
EXCEPTION_TABLE(8) /* __init_begin will be marked RO NX */
NOTES
. = ALIGN(PAGE_SIZE);
idmap_pg_dir = .;
. += IDMAP_DIR_SIZE;
#ifdef CONFIG_UNMAP_KERNEL_AT_EL0
tramp_pg_dir = .;
. += PAGE_SIZE;
#endif
#ifdef CONFIG_ARM64_SW_TTBR0_PAN
reserved_ttbr0 = .;
. += RESERVED_TTBR0_SIZE;
#endif
swapper_pg_dir = .;
. += PAGE_SIZE;
swapper_pg_end = .;
. = ALIGN(SEGMENT_ALIGN);
__init_begin = .;
__inittext_begin = .;
INIT_TEXT_SECTION(8)
__exittext_begin = .;
.exit.text : {
ARM_EXIT_KEEP(EXIT_TEXT)
}
__exittext_end = .;
. = ALIGN(4);
.altinstructions : {
__alt_instructions = .;
*(.altinstructions)
__alt_instructions_end = .;
}
.altinstr_replacement : {
*(.altinstr_replacement)
}
. = ALIGN(PAGE_SIZE);
__inittext_end = .;
__initdata_begin = .;
.init.data : {
INIT_DATA
INIT_SETUP(16)
INIT_CALLS
CON_INITCALL
INIT_RAM_FS
*(.init.rodata.* .init.bss) /* from the EFI stub */
}
.exit.data : {
ARM_EXIT_KEEP(EXIT_DATA)
}
PERCPU_SECTION(L1_CACHE_BYTES)
.rela.dyn : ALIGN(8) {
*(.rela .rela*)
}
__rela_offset = ABSOLUTE(ADDR(.rela.dyn) - KIMAGE_VADDR);
__rela_size = SIZEOF(.rela.dyn);
#ifdef CONFIG_RELR
.relr.dyn : ALIGN(8) {
*(.relr.dyn)
}
__relr_offset = ABSOLUTE(ADDR(.relr.dyn) - KIMAGE_VADDR);
__relr_size = SIZEOF(.relr.dyn);
#endif
. = ALIGN(SEGMENT_ALIGN);
__initdata_end = .;
__init_end = .;
_data = .;
_sdata = .;
RW_DATA_SECTION(L1_CACHE_BYTES, PAGE_SIZE, THREAD_ALIGN)
/*
* Data written with the MMU off but read with the MMU on requires
* cache lines to be invalidated, discarding up to a Cache Writeback
* Granule (CWG) of data from the cache. Keep the section that
* requires this type of maintenance to be in its own Cache Writeback
* Granule (CWG) area so the cache maintenance operations don't
* interfere with adjacent data.
*/
.mmuoff.data.write : ALIGN(SZ_2K) {
__mmuoff_data_start = .;
*(.mmuoff.data.write)
}
. = ALIGN(SZ_2K);
.mmuoff.data.read : {
*(.mmuoff.data.read)
__mmuoff_data_end = .;
}
PECOFF_EDATA_PADDING
__pecoff_data_rawsize = ABSOLUTE(. - __initdata_begin);
_edata = .;
BSS_SECTION(0, 0, 0)
. = ALIGN(PAGE_SIZE);
init_pg_dir = .;
. += INIT_DIR_SIZE;
init_pg_end = .;
__pecoff_data_size = ABSOLUTE(. - __initdata_begin);
_end = .;
STABS_DEBUG
HEAD_SYMBOLS
}
#include "image-vars.h"
/*
* The HYP init code and ID map text can't be longer than a page each,
* and should not cross a page boundary.
*/
ASSERT(__hyp_idmap_text_end - (__hyp_idmap_text_start & ~(SZ_4K - 1)) <= SZ_4K,
"HYP init code too big or misaligned")
ASSERT(__idmap_text_end - (__idmap_text_start & ~(SZ_4K - 1)) <= SZ_4K,
"ID map text too big or misaligned")
#ifdef CONFIG_HIBERNATION
ASSERT(__hibernate_exit_text_end - (__hibernate_exit_text_start & ~(SZ_4K - 1))
<= SZ_4K, "Hibernate exit text too big or misaligned")
#endif
#ifdef CONFIG_UNMAP_KERNEL_AT_EL0
ASSERT((__entry_tramp_text_end - __entry_tramp_text_start) == PAGE_SIZE,
"Entry trampoline text too big")
#endif
/*
* If padding is applied before .head.text, virt<->phys conversions will fail.
*/
ASSERT(_text == (KIMAGE_VADDR + TEXT_OFFSET), "HEAD is misaligned")