LoongArch: Add ORC stack unwinder support

The kernel CONFIG_UNWINDER_ORC option enables the ORC unwinder, which is
similar in concept to a DWARF unwinder. The difference is that the format
of the ORC data is much simpler than DWARF, which in turn allows the ORC
unwinder to be much simpler and faster.

The ORC data consists of unwind tables which are generated by objtool.
After analyzing all the code paths of a .o file, it determines information
about the stack state at each instruction address in the file and outputs
that information to the .orc_unwind and .orc_unwind_ip sections.

The per-object ORC sections are combined at link time and are sorted and
post-processed at boot time. The unwinder uses the resulting data to
correlate instruction addresses with their stack states at run time.

Most of the logic are similar with x86, in order to get ra info before ra
is saved into stack, add ra_reg and ra_offset into orc_entry. At the same
time, modify some arch-specific code to silence the objtool warnings.

Co-developed-by: Jinyang He <hejinyang@loongson.cn>
Signed-off-by: Jinyang He <hejinyang@loongson.cn>
Co-developed-by: Youling Tang <tangyouling@loongson.cn>
Signed-off-by: Youling Tang <tangyouling@loongson.cn>
Signed-off-by: Tiezhu Yang <yangtiezhu@loongson.cn>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>

Author: seehearfeel

arch/loongarch/Kconfig

@@ -136,6 +136,7 @@ config LOONGARCH
 	select HAVE_KVM
 	select HAVE_MOD_ARCH_SPECIFIC
 	select HAVE_NMI
+	select HAVE_OBJTOOL if AS_HAS_EXPLICIT_RELOCS
 	select HAVE_PCI
 	select HAVE_PERF_EVENTS
 	select HAVE_PERF_REGS
@@ -148,6 +149,7 @@ config LOONGARCH
 	select HAVE_SAMPLE_FTRACE_DIRECT
 	select HAVE_SAMPLE_FTRACE_DIRECT_MULTI
 	select HAVE_SETUP_PER_CPU_AREA if NUMA
+	select HAVE_STACK_VALIDATION if HAVE_OBJTOOL
 	select HAVE_STACKPROTECTOR
 	select HAVE_SYSCALL_TRACEPOINTS
 	select HAVE_TIF_NOHZ

arch/loongarch/Kconfig.debug

@@ -26,4 +26,15 @@ config UNWINDER_PROLOGUE
 	  Some of the addresses it reports may be incorrect (but better than the
 	  Guess unwinder).
 
+config UNWINDER_ORC
+	bool "ORC unwinder"
+	select OBJTOOL
+	help
+	  This option enables the ORC (Oops Rewind Capability) unwinder for
+	  unwinding kernel stack traces.  It uses a custom data format which is
+	  a simplified version of the DWARF Call Frame Information standard.
+
+	  Enabling this option will increase the kernel's runtime memory usage
+	  by roughly 2-4MB, depending on your kernel config.
+
 endchoice

arch/loongarch/Makefile

@@ -26,6 +26,18 @@ endif
 32bit-emul		= elf32loongarch
 64bit-emul		= elf64loongarch
 
+ifdef CONFIG_UNWINDER_ORC
+orc_hash_h := arch/$(SRCARCH)/include/generated/asm/orc_hash.h
+orc_hash_sh := $(srctree)/scripts/orc_hash.sh
+targets += $(orc_hash_h)
+quiet_cmd_orc_hash = GEN     $@
+      cmd_orc_hash = mkdir -p $(dir $@); \
+		     $(CONFIG_SHELL) $(orc_hash_sh) < $< > $@
+$(orc_hash_h): $(srctree)/arch/loongarch/include/asm/orc_types.h $(orc_hash_sh) FORCE
+	$(call if_changed,orc_hash)
+archprepare: $(orc_hash_h)
+endif
+
 ifdef CONFIG_DYNAMIC_FTRACE
 KBUILD_CPPFLAGS += -DCC_USING_PATCHABLE_FUNCTION_ENTRY
 CC_FLAGS_FTRACE := -fpatchable-function-entry=2
@@ -72,8 +84,6 @@ KBUILD_CFLAGS_KERNEL		+= $(call cc-option,-mdirect-extern-access)
 KBUILD_CFLAGS_KERNEL		+= $(call cc-option,-fdirect-access-external-data)
 KBUILD_AFLAGS_MODULE		+= $(call cc-option,-fno-direct-access-external-data)
 KBUILD_CFLAGS_MODULE		+= $(call cc-option,-fno-direct-access-external-data)
-KBUILD_AFLAGS_MODULE		+= $(call cc-option,-mno-relax) $(call cc-option,-Wa$(comma)-mno-relax)
-KBUILD_CFLAGS_MODULE		+= $(call cc-option,-mno-relax) $(call cc-option,-Wa$(comma)-mno-relax)
 else
 cflags-y			+= $(call cc-option,-mno-explicit-relocs)
 KBUILD_AFLAGS_KERNEL		+= -Wa,-mla-global-with-pcrel
@@ -82,6 +92,15 @@ KBUILD_AFLAGS_MODULE		+= -Wa,-mla-global-with-abs
 KBUILD_CFLAGS_MODULE		+= -fplt -Wa,-mla-global-with-abs,-mla-local-with-abs
 endif
 
+KBUILD_AFLAGS			+= $(call cc-option,-mno-relax) $(call cc-option,-Wa$(comma)-mno-relax)
+KBUILD_CFLAGS			+= $(call cc-option,-mno-relax) $(call cc-option,-Wa$(comma)-mno-relax)
+KBUILD_AFLAGS			+= $(call cc-option,-mthin-add-sub) $(call cc-option,-Wa$(comma)-mthin-add-sub)
+KBUILD_CFLAGS			+= $(call cc-option,-mthin-add-sub) $(call cc-option,-Wa$(comma)-mthin-add-sub)
+
+ifdef CONFIG_OBJTOOL
+KBUILD_CFLAGS			+= -fno-jump-tables
+endif
+
 KBUILD_RUSTFLAGS_MODULE		+= -Crelocation-model=pic
 
 ifeq ($(CONFIG_RELOCATABLE),y)

arch/loongarch/include/asm/Kbuild

@@ -1,4 +1,6 @@
 # SPDX-License-Identifier: GPL-2.0
+generated-y += orc_hash.h
+
 generic-y += dma-contiguous.h
 generic-y += mcs_spinlock.h
 generic-y += parport.h

arch/loongarch/include/asm/bug.h

@@ -44,6 +44,7 @@
 do {								\
 	instrumentation_begin();				\
 	__BUG_FLAGS(BUGFLAG_WARNING|(flags));			\
+	annotate_reachable();					\
 	instrumentation_end();					\
 } while (0)
 

arch/loongarch/include/asm/exception.h

@@ -6,6 +6,8 @@
 #include <asm/ptrace.h>
 #include <linux/kprobes.h>
 
+extern void *exception_table[];
+
 void show_registers(struct pt_regs *regs);
 
 asmlinkage void cache_parity_error(void);

arch/loongarch/include/asm/module.h

@@ -6,6 +6,7 @@
 #define _ASM_MODULE_H
 
 #include <asm/inst.h>
+#include <asm/orc_types.h>
 #include <asm-generic/module.h>
 
 #define RELA_STACK_DEPTH 16
@@ -21,6 +22,12 @@ struct mod_arch_specific {
 	struct mod_section plt;
 	struct mod_section plt_idx;
 
+#ifdef CONFIG_UNWINDER_ORC
+	unsigned int num_orcs;
+	int *orc_unwind_ip;
+	struct orc_entry *orc_unwind;
+#endif
+
 	/* For CONFIG_DYNAMIC_FTRACE */
 	struct plt_entry *ftrace_trampolines;
 };

arch/loongarch/include/asm/orc_header.h

@@ -0,0 +1,18 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+
+#ifndef _ORC_HEADER_H
+#define _ORC_HEADER_H
+
+#include <linux/types.h>
+#include <linux/compiler.h>
+#include <asm/orc_hash.h>
+
+/*
+ * The header is currently a 20-byte hash of the ORC entry definition; see
+ * scripts/orc_hash.sh.
+ */
+#define ORC_HEADER					\
+	__used __section(".orc_header") __aligned(4)	\
+	static const u8 orc_header[] = { ORC_HASH }
+
+#endif /* _ORC_HEADER_H */

arch/loongarch/include/asm/orc_lookup.h

@@ -0,0 +1,31 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+#ifndef _ORC_LOOKUP_H
+#define _ORC_LOOKUP_H
+
+/*
+ * This is a lookup table for speeding up access to the .orc_unwind table.
+ * Given an input address offset, the corresponding lookup table entry
+ * specifies a subset of the .orc_unwind table to search.
+ *
+ * Each block represents the end of the previous range and the start of the
+ * next range.  An extra block is added to give the last range an end.
+ *
+ * The block size should be a power of 2 to avoid a costly 'div' instruction.
+ *
+ * A block size of 256 was chosen because it roughly doubles unwinder
+ * performance while only adding ~5% to the ORC data footprint.
+ */
+#define LOOKUP_BLOCK_ORDER	8
+#define LOOKUP_BLOCK_SIZE	(1 << LOOKUP_BLOCK_ORDER)
+
+#ifndef LINKER_SCRIPT
+
+extern unsigned int orc_lookup[];
+extern unsigned int orc_lookup_end[];
+
+#define LOOKUP_START_IP		(unsigned long)_stext
+#define LOOKUP_STOP_IP		(unsigned long)_etext
+
+#endif /* LINKER_SCRIPT */
+
+#endif /* _ORC_LOOKUP_H */

arch/loongarch/include/asm/orc_types.h

@@ -0,0 +1,58 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+#ifndef _ORC_TYPES_H
+#define _ORC_TYPES_H
+
+#include <linux/types.h>
+
+/*
+ * The ORC_REG_* registers are base registers which are used to find other
+ * registers on the stack.
+ *
+ * ORC_REG_PREV_SP, also known as DWARF Call Frame Address (CFA), is the
+ * address of the previous frame: the caller's SP before it called the current
+ * function.
+ *
+ * ORC_REG_UNDEFINED means the corresponding register's value didn't change in
+ * the current frame.
+ *
+ * The most commonly used base registers are SP and FP -- which the previous SP
+ * is usually based on -- and PREV_SP and UNDEFINED -- which the previous FP is
+ * usually based on.
+ *
+ * The rest of the base registers are needed for special cases like entry code
+ * and GCC realigned stacks.
+ */
+#define ORC_REG_UNDEFINED		0
+#define ORC_REG_PREV_SP			1
+#define ORC_REG_SP			2
+#define ORC_REG_FP			3
+#define ORC_REG_MAX			4
+
+#define ORC_TYPE_UNDEFINED		0
+#define ORC_TYPE_END_OF_STACK		1
+#define ORC_TYPE_CALL			2
+#define ORC_TYPE_REGS			3
+#define ORC_TYPE_REGS_PARTIAL		4
+
+#ifndef __ASSEMBLY__
+/*
+ * This struct is more or less a vastly simplified version of the DWARF Call
+ * Frame Information standard.  It contains only the necessary parts of DWARF
+ * CFI, simplified for ease of access by the in-kernel unwinder.  It tells the
+ * unwinder how to find the previous SP and FP (and sometimes entry regs) on
+ * the stack for a given code address.  Each instance of the struct corresponds
+ * to one or more code locations.
+ */
+struct orc_entry {
+	s16		sp_offset;
+	s16		fp_offset;
+	s16		ra_offset;
+	unsigned int	sp_reg:4;
+	unsigned int	fp_reg:4;
+	unsigned int	ra_reg:4;
+	unsigned int	type:3;
+	unsigned int	signal:1;
+};
+#endif /* __ASSEMBLY__ */
+
+#endif /* _ORC_TYPES_H */