Merge branch 'core-objtool-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull objtool updates from Ingo Molnar:
 "The main changes are to move the ORC unwind table sorting from early
  init to build-time - this speeds up booting.

  No change in functionality intended"

* 'core-objtool-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/unwind/orc: Fix !CONFIG_MODULES build warning
  x86/unwind/orc: Remove boot-time ORC unwind tables sorting
  scripts/sorttable: Implement build-time ORC unwind table sorting
  scripts/sorttable: Rename 'sortextable' to 'sorttable'
  scripts/sortextable: Refactor the do_func() function
  scripts/sortextable: Remove dead code
  scripts/sortextable: Clean up the code to meet the kernel coding style better
  scripts/sortextable: Rewrite error/success handling

arch/arc/Kconfig

@@ -13,7 +13,7 @@ config ARC
 	select ARCH_HAS_SYNC_DMA_FOR_DEVICE
 	select ARCH_SUPPORTS_ATOMIC_RMW if ARC_HAS_LLSC
 	select ARCH_32BIT_OFF_T
-	select BUILDTIME_EXTABLE_SORT
+	select BUILDTIME_TABLE_SORT
 	select CLONE_BACKWARDS
 	select COMMON_CLK
 	select DMA_DIRECT_REMAP

arch/arm/Kconfig

@@ -36,7 +36,7 @@ config ARM
 	select ARCH_WANT_DEFAULT_TOPDOWN_MMAP_LAYOUT if MMU
 	select ARCH_WANT_IPC_PARSE_VERSION
 	select BINFMT_FLAT_ARGVP_ENVP_ON_STACK
-	select BUILDTIME_EXTABLE_SORT if MMU
+	select BUILDTIME_TABLE_SORT if MMU
 	select CLONE_BACKWARDS
 	select CPU_PM if SUSPEND || CPU_IDLE
 	select DCACHE_WORD_ACCESS if HAVE_EFFICIENT_UNALIGNED_ACCESS

arch/arm64/Kconfig

@@ -81,7 +81,7 @@ config ARM64
 	select ARM_GIC_V3
 	select ARM_GIC_V3_ITS if PCI
 	select ARM_PSCI_FW
-	select BUILDTIME_EXTABLE_SORT
+	select BUILDTIME_TABLE_SORT
 	select CLONE_BACKWARDS
 	select COMMON_CLK
 	select CPU_PM if (SUSPEND || CPU_IDLE)

arch/microblaze/Kconfig

@@ -11,7 +11,7 @@ config MICROBLAZE
 	select ARCH_HAS_UNCACHED_SEGMENT if !MMU
 	select ARCH_MIGHT_HAVE_PC_PARPORT
 	select ARCH_WANT_IPC_PARSE_VERSION
-	select BUILDTIME_EXTABLE_SORT
+	select BUILDTIME_TABLE_SORT
 	select TIMER_OF
 	select CLONE_BACKWARDS3
 	select COMMON_CLK

arch/mips/Kconfig

@@ -15,7 +15,7 @@ config MIPS
 	select ARCH_USE_QUEUED_SPINLOCKS
 	select ARCH_WANT_DEFAULT_TOPDOWN_MMAP_LAYOUT if MMU
 	select ARCH_WANT_IPC_PARSE_VERSION
-	select BUILDTIME_EXTABLE_SORT
+	select BUILDTIME_TABLE_SORT
 	select CLONE_BACKWARDS
 	select CPU_NO_EFFICIENT_FFS if (TARGET_ISA_REV < 1)
 	select CPU_PM if CPU_IDLE

arch/parisc/Kconfig

@@ -18,7 +18,7 @@ config PARISC
 	select RTC_DRV_GENERIC
 	select INIT_ALL_POSSIBLE
 	select BUG
-	select BUILDTIME_EXTABLE_SORT
+	select BUILDTIME_TABLE_SORT
 	select HAVE_PCI
 	select HAVE_PERF_EVENTS
 	select HAVE_KERNEL_BZIP2

arch/powerpc/Kconfig

@@ -149,7 +149,7 @@ config PPC
 	select ARCH_WANT_IPC_PARSE_VERSION
 	select ARCH_WEAK_RELEASE_ACQUIRE
 	select BINFMT_ELF
-	select BUILDTIME_EXTABLE_SORT
+	select BUILDTIME_TABLE_SORT
 	select CLONE_BACKWARDS
 	select DCACHE_WORD_ACCESS		if PPC64 && CPU_LITTLE_ENDIAN
 	select DYNAMIC_FTRACE			if FUNCTION_TRACER

arch/s390/Kconfig

@@ -110,7 +110,7 @@ config S390
 	select ARCH_USE_CMPXCHG_LOCKREF
 	select ARCH_WANTS_DYNAMIC_TASK_STRUCT
 	select ARCH_WANT_IPC_PARSE_VERSION
-	select BUILDTIME_EXTABLE_SORT
+	select BUILDTIME_TABLE_SORT
 	select CLONE_BACKWARDS2
 	select DYNAMIC_FTRACE if FUNCTION_TRACER
 	select GENERIC_CLOCKEVENTS

arch/x86/Kconfig

@@ -96,7 +96,7 @@ config X86
 	select ARCH_WANTS_DYNAMIC_TASK_STRUCT
 	select ARCH_WANT_HUGE_PMD_SHARE
 	select ARCH_WANTS_THP_SWAP		if X86_64
-	select BUILDTIME_EXTABLE_SORT
+	select BUILDTIME_TABLE_SORT
 	select CLKEVT_I8253
 	select CLOCKSOURCE_VALIDATE_LAST_CYCLE
 	select CLOCKSOURCE_WATCHDOG

arch/x86/kernel/unwind_orc.c

@@ -187,6 +187,8 @@ static struct orc_entry *orc_find(unsigned long ip)
 	return orc_ftrace_find(ip);
 }
 
+#ifdef CONFIG_MODULES
+
 static void orc_sort_swap(void *_a, void *_b, int size)
 {
 	struct orc_entry *orc_a, *orc_b;
@@ -229,7 +231,6 @@ static int orc_sort_cmp(const void *_a, const void *_b)
 	return orc_a->sp_reg == ORC_REG_UNDEFINED && !orc_a->end ? -1 : 1;
 }
 
-#ifdef CONFIG_MODULES
 void unwind_module_init(struct module *mod, void *_orc_ip, size_t orc_ip_size,
 			void *_orc, size_t orc_size)
 {
@@ -273,9 +274,11 @@ void __init unwind_init(void)
 		return;
 	}
 
-	/* Sort the .orc_unwind and .orc_unwind_ip tables: */
-	sort(__start_orc_unwind_ip, num_entries, sizeof(int), orc_sort_cmp,
-	     orc_sort_swap);
+	/*
+	 * Note, the orc_unwind and orc_unwind_ip tables were already
+	 * sorted at build time via the 'sorttable' tool.
+	 * It's ready for binary search straight away, no need to sort it.
+	 */
 
 	/* Initialize the fast lookup table: */
 	lookup_num_blocks = orc_lookup_end - orc_lookup;

arch/xtensa/Kconfig

@@ -11,7 +11,7 @@ config XTENSA
 	select ARCH_USE_QUEUED_SPINLOCKS
 	select ARCH_WANT_FRAME_POINTERS
 	select ARCH_WANT_IPC_PARSE_VERSION
-	select BUILDTIME_EXTABLE_SORT
+	select BUILDTIME_TABLE_SORT
 	select CLONE_BACKWARDS
 	select COMMON_CLK
 	select DMA_REMAP if MMU

init/Kconfig

@@ -58,7 +58,7 @@ config CONSTRUCTORS
 config IRQ_WORK
 	bool
 
-config BUILDTIME_EXTABLE_SORT
+config BUILDTIME_TABLE_SORT
 	bool
 
 config THREAD_INFO_IN_TASK

scripts/.gitignore

@@ -6,7 +6,7 @@ conmakehash
 kallsyms
 unifdef
 recordmcount
-sortextable
+sorttable
 asn1_compiler
 extract-cert
 sign-file

scripts/Makefile

@@ -13,17 +13,26 @@ hostprogs-$(CONFIG_BUILD_BIN2C)  += bin2c
 hostprogs-$(CONFIG_KALLSYMS)     += kallsyms
 hostprogs-$(CONFIG_VT)           += conmakehash
 hostprogs-$(BUILD_C_RECORDMCOUNT) += recordmcount
-hostprogs-$(CONFIG_BUILDTIME_EXTABLE_SORT) += sortextable
+hostprogs-$(CONFIG_BUILDTIME_TABLE_SORT) += sorttable
 hostprogs-$(CONFIG_ASN1)	 += asn1_compiler
 hostprogs-$(CONFIG_MODULE_SIG_FORMAT) += sign-file
 hostprogs-$(CONFIG_SYSTEM_TRUSTED_KEYRING) += extract-cert
 hostprogs-$(CONFIG_SYSTEM_EXTRA_CERTIFICATE) += insert-sys-cert
 
-HOSTCFLAGS_sortextable.o = -I$(srctree)/tools/include
+HOSTCFLAGS_sorttable.o = -I$(srctree)/tools/include
 HOSTCFLAGS_asn1_compiler.o = -I$(srctree)/include
 HOSTLDLIBS_sign-file = -lcrypto
 HOSTLDLIBS_extract-cert = -lcrypto
 
+ifdef CONFIG_UNWINDER_ORC
+ifeq ($(ARCH),x86_64)
+ARCH := x86
+endif
+HOSTCFLAGS_sorttable.o += -I$(srctree)/tools/arch/x86/include
+HOSTCFLAGS_sorttable.o += -DUNWINDER_ORC_ENABLED
+HOSTLDLIBS_sorttable = -lpthread
+endif
+
 always		:= $(hostprogs-y) $(hostprogs-m)
 
 # The following hostprogs-y programs are only build on demand

scripts/link-vmlinux.sh

@@ -180,9 +180,9 @@ mksysmap()
 	${CONFIG_SHELL} "${srctree}/scripts/mksysmap" ${1} ${2}
 }
 
-sortextable()
+sorttable()
 {
-	${objtree}/scripts/sortextable ${1}
+	${objtree}/scripts/sorttable ${1}
 }
 
 # Delete output files in case of error
@@ -304,9 +304,12 @@ fi
 
 vmlinux_link vmlinux "${kallsymso}" ${btf_vmlinux_bin_o}
 
-if [ -n "${CONFIG_BUILDTIME_EXTABLE_SORT}" ]; then
-	info SORTEX vmlinux
-	sortextable vmlinux
+if [ -n "${CONFIG_BUILDTIME_TABLE_SORT}" ]; then
+	info SORTTAB vmlinux
+	if ! sorttable vmlinux; then
+		echo >&2 Failed to sort kernel tables
+		exit 1
+	fi
 fi
 
 info SYSMAP System.map

scripts/sortextable.c → scripts/sorttable.c

 // SPDX-License-Identifier: GPL-2.0-only
 /*
- * sortextable.c: Sort the kernel's exception table
+ * sorttable.c: Sort the kernel's table
+ *
+ * Added ORC unwind tables sort support and other updates:
+ * Copyright (C) 1999-2019 Alibaba Group Holding Limited. by:
+ * Shile Zhang <shile.zhang@linux.alibaba.com>
  *
  * Copyright 2011 - 2012 Cavium, Inc.
  *
@@ -9,7 +13,7 @@
  * Copyright 2009 John F. Reiser <jreiser@BitWagon.com>.  All rights reserved.
  *
  * Restructured to fit Linux format, as well as other updates:
- *  Copyright 2010 Steven Rostedt <srostedt@redhat.com>, Red Hat Inc.
+ * Copyright 2010 Steven Rostedt <srostedt@redhat.com>, Red Hat Inc.
  */
 
 /*
@@ -22,7 +26,6 @@
 #include <getopt.h>
 #include <elf.h>
 #include <fcntl.h>
-#include <setjmp.h>
 #include <stdio.h>
 #include <stdlib.h>
 #include <string.h>
@@ -51,34 +54,13 @@
 #define EM_ARCV2	195
 #endif
 
-static int fd_map;	/* File descriptor for file being modified. */
-static int mmap_failed; /* Boolean flag. */
-static void *ehdr_curr; /* current ElfXX_Ehdr *  for resource cleanup */
-static struct stat sb;	/* Remember .st_size, etc. */
-static jmp_buf jmpenv;	/* setjmp/longjmp per-file error escape */
-
-/* setjmp() return values */
-enum {
-	SJ_SETJMP = 0,  /* hardwired first return */
-	SJ_FAIL,
-	SJ_SUCCEED
-};
-
-/* Per-file resource cleanup when multiple files. */
-static void
-cleanup(void)
-{
-	if (!mmap_failed)
-		munmap(ehdr_curr, sb.st_size);
-	close(fd_map);
-}
-
-static void __attribute__((noreturn))
-fail_file(void)
-{
-	cleanup();
-	longjmp(jmpenv, SJ_FAIL);
-}
+static uint32_t (*r)(const uint32_t *);
+static uint16_t (*r2)(const uint16_t *);
+static uint64_t (*r8)(const uint64_t *);
+static void (*w)(uint32_t, uint32_t *);
+static void (*w2)(uint16_t, uint16_t *);
+static void (*w8)(uint64_t, uint64_t *);
+typedef void (*table_sort_t)(char *, int);
 
 /*
  * Get the whole file as a programming convenience in order to avoid
@@ -86,87 +68,98 @@ fail_file(void)
  * avoids copying unused pieces; else just read the whole file.
  * Open for both read and write.
  */
-static void *mmap_file(char const *fname)
+static void *mmap_file(char const *fname, size_t *size)
 {
-	void *addr;
+	int fd;
+	struct stat sb;
+	void *addr = NULL;
 
-	fd_map = open(fname, O_RDWR);
-	if (fd_map < 0 || fstat(fd_map, &sb) < 0) {
+	fd = open(fname, O_RDWR);
+	if (fd < 0) {
 		perror(fname);
-		fail_file();
+		return NULL;
+	}
+	if (fstat(fd, &sb) < 0) {
+		perror(fname);
+		goto out;
 	}
 	if (!S_ISREG(sb.st_mode)) {
 		fprintf(stderr, "not a regular file: %s\n", fname);
-		fail_file();
+		goto out;
 	}
-	addr = mmap(0, sb.st_size, PROT_READ|PROT_WRITE, MAP_SHARED,
-		    fd_map, 0);
+
+	addr = mmap(0, sb.st_size, PROT_READ|PROT_WRITE, MAP_SHARED, fd, 0);
 	if (addr == MAP_FAILED) {
-		mmap_failed = 1;
 		fprintf(stderr, "Could not mmap file: %s\n", fname);
-		fail_file();
+		goto out;
 	}
+
+	*size = sb.st_size;
+
+out:
+	close(fd);
 	return addr;
 }
 
-static uint64_t r8be(const uint64_t *x)
-{
-	return get_unaligned_be64(x);
-}
 static uint32_t rbe(const uint32_t *x)
 {
 	return get_unaligned_be32(x);
 }
+
 static uint16_t r2be(const uint16_t *x)
 {
 	return get_unaligned_be16(x);
 }
-static uint64_t r8le(const uint64_t *x)
+
+static uint64_t r8be(const uint64_t *x)
 {
-	return get_unaligned_le64(x);
+	return get_unaligned_be64(x);
 }
+
 static uint32_t rle(const uint32_t *x)
 {
 	return get_unaligned_le32(x);
 }
+
 static uint16_t r2le(const uint16_t *x)
 {
 	return get_unaligned_le16(x);
 }
 
-static void w8be(uint64_t val, uint64_t *x)
+static uint64_t r8le(const uint64_t *x)
 {
-	put_unaligned_be64(val, x);
+	return get_unaligned_le64(x);
 }
+
 static void wbe(uint32_t val, uint32_t *x)
 {
 	put_unaligned_be32(val, x);
 }
+
 static void w2be(uint16_t val, uint16_t *x)
 {
 	put_unaligned_be16(val, x);
 }
-static void w8le(uint64_t val, uint64_t *x)
+
+static void w8be(uint64_t val, uint64_t *x)
 {
-	put_unaligned_le64(val, x);
+	put_unaligned_be64(val, x);
 }
+
 static void wle(uint32_t val, uint32_t *x)
 {
 	put_unaligned_le32(val, x);
 }
+
 static void w2le(uint16_t val, uint16_t *x)
 {
 	put_unaligned_le16(val, x);
 }
 
-static uint64_t (*r8)(const uint64_t *);
-static uint32_t (*r)(const uint32_t *);
-static uint16_t (*r2)(const uint16_t *);
-static void (*w8)(uint64_t, uint64_t *);
-static void (*w)(uint32_t, uint32_t *);
-static void (*w2)(uint16_t, uint16_t *);
-
-typedef void (*table_sort_t)(char *, int);
+static void w8le(uint64_t val, uint64_t *x)
+{
+	put_unaligned_le64(val, x);
+}
 
 /*
  * Move reserved section indices SHN_LORESERVE..SHN_HIRESERVE out of
@@ -193,9 +186,9 @@ static inline unsigned int get_secindex(unsigned int shndx,
 }
 
 /* 32 bit and 64 bit are very similar */
-#include "sortextable.h"
-#define SORTEXTABLE_64
-#include "sortextable.h"
+#include "sorttable.h"
+#define SORTTABLE_64
+#include "sorttable.h"
 
 static int compare_relative_table(const void *a, const void *b)
 {
@@ -209,110 +202,100 @@ static int compare_relative_table(const void *a, const void *b)
 	return 0;
 }
 
-static void x86_sort_relative_table(char *extab_image, int image_size)
+static void sort_relative_table(char *extab_image, int image_size)
 {
-	int i;
+	int i = 0;
 
-	i = 0;
+	/*
+	 * Do the same thing the runtime sort does, first normalize to
+	 * being relative to the start of the section.
+	 */
 	while (i < image_size) {
 		uint32_t *loc = (uint32_t *)(extab_image + i);
-
 		w(r(loc) + i, loc);
-		w(r(loc + 1) + i + 4, loc + 1);
-		w(r(loc + 2) + i + 8, loc + 2);
-
-		i += sizeof(uint32_t) * 3;
+		i += 4;
 	}
 
-	qsort(extab_image, image_size / 12, 12, compare_relative_table);
+	qsort(extab_image, image_size / 8, 8, compare_relative_table);
 
+	/* Now denormalize. */
 	i = 0;
 	while (i < image_size) {
 		uint32_t *loc = (uint32_t *)(extab_image + i);
-
 		w(r(loc) - i, loc);
-		w(r(loc + 1) - (i + 4), loc + 1);
-		w(r(loc + 2) - (i + 8), loc + 2);
-
-		i += sizeof(uint32_t) * 3;
+		i += 4;
 	}
 }
 
-static void sort_relative_table(char *extab_image, int image_size)
+static void x86_sort_relative_table(char *extab_image, int image_size)
 {
-	int i;
+	int i = 0;
 
-	/*
-	 * Do the same thing the runtime sort does, first normalize to
-	 * being relative to the start of the section.
-	 */
-	i = 0;
 	while (i < image_size) {
 		uint32_t *loc = (uint32_t *)(extab_image + i);
+
 		w(r(loc) + i, loc);
-		i += 4;
+		w(r(loc + 1) + i + 4, loc + 1);
+		w(r(loc + 2) + i + 8, loc + 2);
+
+		i += sizeof(uint32_t) * 3;
 	}
 
-	qsort(extab_image, image_size / 8, 8, compare_relative_table);
+	qsort(extab_image, image_size / 12, 12, compare_relative_table);
 
-	/* Now denormalize. */
 	i = 0;
 	while (i < image_size) {
 		uint32_t *loc = (uint32_t *)(extab_image + i);
+
 		w(r(loc) - i, loc);
-		i += 4;
+		w(r(loc + 1) - (i + 4), loc + 1);
+		w(r(loc + 2) - (i + 8), loc + 2);
+
+		i += sizeof(uint32_t) * 3;
 	}
 }
 
-static void
-do_file(char const *const fname)
+static int do_file(char const *const fname, void *addr)
 {
-	table_sort_t custom_sort;
-	Elf32_Ehdr *ehdr = mmap_file(fname);
+	int rc = -1;
+	Elf32_Ehdr *ehdr = addr;
+	table_sort_t custom_sort = NULL;
 
-	ehdr_curr = ehdr;
 	switch (ehdr->e_ident[EI_DATA]) {
-	default:
-		fprintf(stderr, "unrecognized ELF data encoding %d: %s\n",
-			ehdr->e_ident[EI_DATA], fname);
-		fail_file();
-		break;
 	case ELFDATA2LSB:
-		r = rle;
-		r2 = r2le;
-		r8 = r8le;
-		w = wle;
-		w2 = w2le;
-		w8 = w8le;
+		r	= rle;
+		r2	= r2le;
+		r8	= r8le;
+		w	= wle;
+		w2	= w2le;
+		w8	= w8le;
 		break;
 	case ELFDATA2MSB:
-		r = rbe;
-		r2 = r2be;
-		r8 = r8be;
-		w = wbe;
-		w2 = w2be;
-		w8 = w8be;
+		r	= rbe;
+		r2	= r2be;
+		r8	= r8be;
+		w	= wbe;
+		w2	= w2be;
+		w8	= w8be;
 		break;
-	}  /* end switch */
-	if (memcmp(ELFMAG, ehdr->e_ident, SELFMAG) != 0
-	||  (r2(&ehdr->e_type) != ET_EXEC && r2(&ehdr->e_type) != ET_DYN)
-	||  ehdr->e_ident[EI_VERSION] != EV_CURRENT) {
+	default:
+		fprintf(stderr, "unrecognized ELF data encoding %d: %s\n",
+			ehdr->e_ident[EI_DATA], fname);
+		return -1;
+	}
+
+	if (memcmp(ELFMAG, ehdr->e_ident, SELFMAG) != 0 ||
+	    (r2(&ehdr->e_type) != ET_EXEC && r2(&ehdr->e_type) != ET_DYN) ||
+	    ehdr->e_ident[EI_VERSION] != EV_CURRENT) {
 		fprintf(stderr, "unrecognized ET_EXEC/ET_DYN file %s\n", fname);
-		fail_file();
+		return -1;
 	}
 
-	custom_sort = NULL;
 	switch (r2(&ehdr->e_machine)) {
-	default:
-		fprintf(stderr, "unrecognized e_machine %d %s\n",
-			r2(&ehdr->e_machine), fname);
-		fail_file();
-		break;
 	case EM_386:
 	case EM_X86_64:
 		custom_sort = x86_sort_relative_table;
 		break;
-
 	case EM_S390:
 	case EM_AARCH64:
 	case EM_PARISC:
@@ -327,74 +310,68 @@ do_file(char const *const fname)
 	case EM_MIPS:
 	case EM_XTENSA:
 		break;
-	}  /* end switch */
+	default:
+		fprintf(stderr, "unrecognized e_machine %d %s\n",
+			r2(&ehdr->e_machine), fname);
+		return -1;
+	}
 
 	switch (ehdr->e_ident[EI_CLASS]) {
-	default:
-		fprintf(stderr, "unrecognized ELF class %d %s\n",
-			ehdr->e_ident[EI_CLASS], fname);
-		fail_file();
-		break;
 	case ELFCLASS32:
-		if (r2(&ehdr->e_ehsize) != sizeof(Elf32_Ehdr)
-		||  r2(&ehdr->e_shentsize) != sizeof(Elf32_Shdr)) {
+		if (r2(&ehdr->e_ehsize) != sizeof(Elf32_Ehdr) ||
+		    r2(&ehdr->e_shentsize) != sizeof(Elf32_Shdr)) {
 			fprintf(stderr,
 				"unrecognized ET_EXEC/ET_DYN file: %s\n", fname);
-			fail_file();
+			break;
 		}
-		do32(ehdr, fname, custom_sort);
+		rc = do_sort_32(ehdr, fname, custom_sort);
 		break;
-	case ELFCLASS64: {
+	case ELFCLASS64:
+		{
 		Elf64_Ehdr *const ghdr = (Elf64_Ehdr *)ehdr;
-		if (r2(&ghdr->e_ehsize) != sizeof(Elf64_Ehdr)
-		||  r2(&ghdr->e_shentsize) != sizeof(Elf64_Shdr)) {
+		if (r2(&ghdr->e_ehsize) != sizeof(Elf64_Ehdr) ||
+		    r2(&ghdr->e_shentsize) != sizeof(Elf64_Shdr)) {
 			fprintf(stderr,
-				"unrecognized ET_EXEC/ET_DYN file: %s\n", fname);
-			fail_file();
+				"unrecognized ET_EXEC/ET_DYN file: %s\n",
+				fname);
+			break;
+		}
+		rc = do_sort_64(ghdr, fname, custom_sort);
 		}
-		do64(ghdr, fname, custom_sort);
+		break;
+	default:
+		fprintf(stderr, "unrecognized ELF class %d %s\n",
+			ehdr->e_ident[EI_CLASS], fname);
 		break;
 	}
-	}  /* end switch */
 
-	cleanup();
+	return rc;
 }
 
-int
-main(int argc, char *argv[])
+int main(int argc, char *argv[])
 {
-	int n_error = 0;  /* gcc-4.3.0 false positive complaint */
-	int i;
+	int i, n_error = 0;  /* gcc-4.3.0 false positive complaint */
+	size_t size = 0;
+	void *addr = NULL;
 
 	if (argc < 2) {
-		fprintf(stderr, "usage: sortextable vmlinux...\n");
+		fprintf(stderr, "usage: sorttable vmlinux...\n");
 		return 0;
 	}
 
 	/* Process each file in turn, allowing deep failure. */
 	for (i = 1; i < argc; i++) {
-		char *file = argv[i];
-		int const sjval = setjmp(jmpenv);
+		addr = mmap_file(argv[i], &size);
+		if (!addr) {
+			++n_error;
+			continue;
+		}
 
-		switch (sjval) {
-		default:
-			fprintf(stderr, "internal error: %s\n", file);
-			exit(1);
-			break;
-		case SJ_SETJMP:    /* normal sequence */
-			/* Avoid problems if early cleanup() */
-			fd_map = -1;
-			ehdr_curr = NULL;
-			mmap_failed = 1;
-			do_file(file);
-			break;
-		case SJ_FAIL:    /* error in do_file or below */
+		if (do_file(argv[i], addr))
 			++n_error;
-			break;
-		case SJ_SUCCEED:    /* premature success */
-			/* do nothing */
-			break;
-		}  /* end switch */
+
+		munmap(addr, size);
 	}
+
 	return !!n_error;
 }

scripts/sortextable.h → scripts/sorttable.h

 /* SPDX-License-Identifier: GPL-2.0-only */
 /*
- * sortextable.h
+ * sorttable.h
+ *
+ * Added ORC unwind tables sort support and other updates:
+ * Copyright (C) 1999-2019 Alibaba Group Holding Limited. by:
+ * Shile Zhang <shile.zhang@linux.alibaba.com>
  *
  * Copyright 2011 - 2012 Cavium, Inc.
  *
+ * Some of code was taken out of arch/x86/kernel/unwind_orc.c, written by:
+ * Copyright (C) 2017 Josh Poimboeuf <jpoimboe@redhat.com>
+ *
  * Some of this code was taken out of recordmcount.h written by:
  *
- * Copyright 2009 John F. Reiser <jreiser@BitWagon.com>.  All rights reserved.
+ * Copyright 2009 John F. Reiser <jreiser@BitWagon.com>. All rights reserved.
  * Copyright 2010 Steven Rostedt <srostedt@redhat.com>, Red Hat Inc.
  */
 
 #undef extable_ent_size
 #undef compare_extable
-#undef do_func
+#undef do_sort
 #undef Elf_Addr
 #undef Elf_Ehdr
 #undef Elf_Shdr
@@ -31,10 +38,10 @@
 #undef _r
 #undef _w
 
-#ifdef SORTEXTABLE_64
+#ifdef SORTTABLE_64
 # define extable_ent_size	16
 # define compare_extable	compare_extable_64
-# define do_func		do64
+# define do_sort		do_sort_64
 # define Elf_Addr		Elf64_Addr
 # define Elf_Ehdr		Elf64_Ehdr
 # define Elf_Shdr		Elf64_Shdr
@@ -55,7 +62,7 @@
 #else
 # define extable_ent_size	8
 # define compare_extable	compare_extable_32
-# define do_func		do32
+# define do_sort		do_sort_32
 # define Elf_Addr		Elf32_Addr
 # define Elf_Ehdr		Elf32_Ehdr
 # define Elf_Shdr		Elf32_Shdr
@@ -75,6 +82,104 @@
 # define _w			w
 #endif
 
+#if defined(SORTTABLE_64) && defined(UNWINDER_ORC_ENABLED)
+/* ORC unwinder only support X86_64 */
+#include <errno.h>
+#include <pthread.h>
+#include <asm/orc_types.h>
+
+#define ERRSTR_MAXSZ	256
+
+char g_err[ERRSTR_MAXSZ];
+int *g_orc_ip_table;
+struct orc_entry *g_orc_table;
+
+pthread_t orc_sort_thread;
+
+static inline unsigned long orc_ip(const int *ip)
+{
+	return (unsigned long)ip + *ip;
+}
+
+static int orc_sort_cmp(const void *_a, const void *_b)
+{
+	struct orc_entry *orc_a;
+	const int *a = g_orc_ip_table + *(int *)_a;
+	const int *b = g_orc_ip_table + *(int *)_b;
+	unsigned long a_val = orc_ip(a);
+	unsigned long b_val = orc_ip(b);
+
+	if (a_val > b_val)
+		return 1;
+	if (a_val < b_val)
+		return -1;
+
+	/*
+	 * The "weak" section terminator entries need to always be on the left
+	 * to ensure the lookup code skips them in favor of real entries.
+	 * These terminator entries exist to handle any gaps created by
+	 * whitelisted .o files which didn't get objtool generation.
+	 */
+	orc_a = g_orc_table + (a - g_orc_ip_table);
+	return orc_a->sp_reg == ORC_REG_UNDEFINED && !orc_a->end ? -1 : 1;
+}
+
+static void *sort_orctable(void *arg)
+{
+	int i;
+	int *idxs = NULL;
+	int *tmp_orc_ip_table = NULL;
+	struct orc_entry *tmp_orc_table = NULL;
+	unsigned int *orc_ip_size = (unsigned int *)arg;
+	unsigned int num_entries = *orc_ip_size / sizeof(int);
+	unsigned int orc_size = num_entries * sizeof(struct orc_entry);
+
+	idxs = (int *)malloc(*orc_ip_size);
+	if (!idxs) {
+		snprintf(g_err, ERRSTR_MAXSZ, "malloc idxs: %s",
+			 strerror(errno));
+		pthread_exit(g_err);
+	}
+
+	tmp_orc_ip_table = (int *)malloc(*orc_ip_size);
+	if (!tmp_orc_ip_table) {
+		snprintf(g_err, ERRSTR_MAXSZ, "malloc tmp_orc_ip_table: %s",
+			 strerror(errno));
+		pthread_exit(g_err);
+	}
+
+	tmp_orc_table = (struct orc_entry *)malloc(orc_size);
+	if (!tmp_orc_table) {
+		snprintf(g_err, ERRSTR_MAXSZ, "malloc tmp_orc_table: %s",
+			 strerror(errno));
+		pthread_exit(g_err);
+	}
+
+	/* initialize indices array, convert ip_table to absolute address */
+	for (i = 0; i < num_entries; i++) {
+		idxs[i] = i;
+		tmp_orc_ip_table[i] = g_orc_ip_table[i] + i * sizeof(int);
+	}
+	memcpy(tmp_orc_table, g_orc_table, orc_size);
+
+	qsort(idxs, num_entries, sizeof(int), orc_sort_cmp);
+
+	for (i = 0; i < num_entries; i++) {
+		if (idxs[i] == i)
+			continue;
+
+		/* convert back to relative address */
+		g_orc_ip_table[i] = tmp_orc_ip_table[idxs[i]] - i * sizeof(int);
+		g_orc_table[i] = tmp_orc_table[idxs[i]];
+	}
+
+	free(idxs);
+	free(tmp_orc_ip_table);
+	free(tmp_orc_table);
+	pthread_exit(NULL);
+}
+#endif
+
 static int compare_extable(const void *a, const void *b)
 {
 	Elf_Addr av = _r(a);
@@ -87,80 +192,127 @@ static int compare_extable(const void *a, const void *b)
 	return 0;
 }
 
-static void
-do_func(Elf_Ehdr *ehdr, char const *const fname, table_sort_t custom_sort)
+static int do_sort(Elf_Ehdr *ehdr,
+		   char const *const fname,
+		   table_sort_t custom_sort)
 {
-	Elf_Shdr *shdr;
-	Elf_Shdr *shstrtab_sec;
+	int rc = -1;
+	Elf_Shdr *s, *shdr = (Elf_Shdr *)((char *)ehdr + _r(&ehdr->e_shoff));
 	Elf_Shdr *strtab_sec = NULL;
 	Elf_Shdr *symtab_sec = NULL;
 	Elf_Shdr *extab_sec = NULL;
 	Elf_Sym *sym;
 	const Elf_Sym *symtab;
-	Elf32_Word *symtab_shndx_start = NULL;
-	Elf_Sym *sort_needed_sym;
+	Elf32_Word *symtab_shndx = NULL;
+	Elf_Sym *sort_needed_sym = NULL;
 	Elf_Shdr *sort_needed_sec;
 	Elf_Rel *relocs = NULL;
 	int relocs_size = 0;
-	uint32_t *sort_done_location;
-	const char *secstrtab;
+	uint32_t *sort_needed_loc;
+	const char *secstrings;
 	const char *strtab;
 	char *extab_image;
 	int extab_index = 0;
 	int i;
 	int idx;
-	unsigned int num_sections;
-	unsigned int secindex_strings;
-
-	shdr = (Elf_Shdr *)((char *)ehdr + _r(&ehdr->e_shoff));
+	unsigned int shnum;
+	unsigned int shstrndx;
+#if defined(SORTTABLE_64) && defined(UNWINDER_ORC_ENABLED)
+	unsigned int orc_ip_size = 0;
+	unsigned int orc_size = 0;
+	unsigned int orc_num_entries = 0;
+#endif
 
-	num_sections = r2(&ehdr->e_shnum);
-	if (num_sections == SHN_UNDEF)
-		num_sections = _r(&shdr[0].sh_size);
+	shstrndx = r2(&ehdr->e_shstrndx);
+	if (shstrndx == SHN_XINDEX)
+		shstrndx = r(&shdr[0].sh_link);
+	secstrings = (const char *)ehdr + _r(&shdr[shstrndx].sh_offset);
 
-	secindex_strings = r2(&ehdr->e_shstrndx);
-	if (secindex_strings == SHN_XINDEX)
-		secindex_strings = r(&shdr[0].sh_link);
+	shnum = r2(&ehdr->e_shnum);
+	if (shnum == SHN_UNDEF)
+		shnum = _r(&shdr[0].sh_size);
 
-	shstrtab_sec = shdr + secindex_strings;
-	secstrtab = (const char *)ehdr + _r(&shstrtab_sec->sh_offset);
-	for (i = 0; i < num_sections; i++) {
-		idx = r(&shdr[i].sh_name);
-		if (strcmp(secstrtab + idx, "__ex_table") == 0) {
-			extab_sec = shdr + i;
+	for (i = 0, s = shdr; s < shdr + shnum; i++, s++) {
+		idx = r(&s->sh_name);
+		if (!strcmp(secstrings + idx, "__ex_table")) {
+			extab_sec = s;
 			extab_index = i;
 		}
-		if ((r(&shdr[i].sh_type) == SHT_REL ||
-		     r(&shdr[i].sh_type) == SHT_RELA) &&
-		    r(&shdr[i].sh_info) == extab_index) {
-			relocs = (void *)ehdr + _r(&shdr[i].sh_offset);
-			relocs_size = _r(&shdr[i].sh_size);
+		if (!strcmp(secstrings + idx, ".symtab"))
+			symtab_sec = s;
+		if (!strcmp(secstrings + idx, ".strtab"))
+			strtab_sec = s;
+
+		if ((r(&s->sh_type) == SHT_REL ||
+		     r(&s->sh_type) == SHT_RELA) &&
+		    r(&s->sh_info) == extab_index) {
+			relocs = (void *)ehdr + _r(&s->sh_offset);
+			relocs_size = _r(&s->sh_size);
+		}
+		if (r(&s->sh_type) == SHT_SYMTAB_SHNDX)
+			symtab_shndx = (Elf32_Word *)((const char *)ehdr +
+						      _r(&s->sh_offset));
+
+#if defined(SORTTABLE_64) && defined(UNWINDER_ORC_ENABLED)
+		/* locate the ORC unwind tables */
+		if (!strcmp(secstrings + idx, ".orc_unwind_ip")) {
+			orc_ip_size = s->sh_size;
+			g_orc_ip_table = (int *)((void *)ehdr +
+						   s->sh_offset);
+		}
+		if (!strcmp(secstrings + idx, ".orc_unwind")) {
+			orc_size = s->sh_size;
+			g_orc_table = (struct orc_entry *)((void *)ehdr +
+							     s->sh_offset);
 		}
-		if (strcmp(secstrtab + idx, ".symtab") == 0)
-			symtab_sec = shdr + i;
-		if (strcmp(secstrtab + idx, ".strtab") == 0)
-			strtab_sec = shdr + i;
-		if (r(&shdr[i].sh_type) == SHT_SYMTAB_SHNDX)
-			symtab_shndx_start = (Elf32_Word *)(
-				(const char *)ehdr + _r(&shdr[i].sh_offset));
+#endif
+	} /* for loop */
+
+#if defined(SORTTABLE_64) && defined(UNWINDER_ORC_ENABLED)
+	if (!g_orc_ip_table || !g_orc_table) {
+		fprintf(stderr,
+			"incomplete ORC unwind tables in file: %s\n", fname);
+		goto out;
+	}
+
+	orc_num_entries = orc_ip_size / sizeof(int);
+	if (orc_ip_size % sizeof(int) != 0 ||
+	    orc_size % sizeof(struct orc_entry) != 0 ||
+	    orc_num_entries != orc_size / sizeof(struct orc_entry)) {
+		fprintf(stderr,
+			"inconsistent ORC unwind table entries in file: %s\n",
+			fname);
+		goto out;
 	}
-	if (strtab_sec == NULL) {
-		fprintf(stderr,	"no .strtab in  file: %s\n", fname);
-		fail_file();
+
+	/* create thread to sort ORC unwind tables concurrently */
+	if (pthread_create(&orc_sort_thread, NULL,
+			   sort_orctable, &orc_ip_size)) {
+		fprintf(stderr,
+			"pthread_create orc_sort_thread failed '%s': %s\n",
+			strerror(errno), fname);
+		goto out;
 	}
-	if (symtab_sec == NULL) {
-		fprintf(stderr,	"no .symtab in  file: %s\n", fname);
-		fail_file();
+#endif
+	if (!extab_sec) {
+		fprintf(stderr,	"no __ex_table in file: %s\n", fname);
+		goto out;
 	}
-	symtab = (const Elf_Sym *)((const char *)ehdr +
-				   _r(&symtab_sec->sh_offset));
-	if (extab_sec == NULL) {
-		fprintf(stderr,	"no __ex_table in  file: %s\n", fname);
-		fail_file();
+
+	if (!symtab_sec) {
+		fprintf(stderr,	"no .symtab in file: %s\n", fname);
+		goto out;
+	}
+
+	if (!strtab_sec) {
+		fprintf(stderr,	"no .strtab in file: %s\n", fname);
+		goto out;
 	}
-	strtab = (const char *)ehdr + _r(&strtab_sec->sh_offset);
 
 	extab_image = (void *)ehdr + _r(&extab_sec->sh_offset);
+	strtab = (const char *)ehdr + _r(&strtab_sec->sh_offset);
+	symtab = (const Elf_Sym *)((const char *)ehdr +
+						  _r(&symtab_sec->sh_offset));
 
 	if (custom_sort) {
 		custom_sort(extab_image, _r(&extab_sec->sh_size));
@@ -169,41 +321,60 @@ do_func(Elf_Ehdr *ehdr, char const *const fname, table_sort_t custom_sort)
 		qsort(extab_image, num_entries,
 		      extable_ent_size, compare_extable);
 	}
+
 	/* If there were relocations, we no longer need them. */
 	if (relocs)
 		memset(relocs, 0, relocs_size);
 
-	/* find main_extable_sort_needed */
-	sort_needed_sym = NULL;
-	for (i = 0; i < _r(&symtab_sec->sh_size) / sizeof(Elf_Sym); i++) {
-		sym = (void *)ehdr + _r(&symtab_sec->sh_offset);
-		sym += i;
+	/* find the flag main_extable_sort_needed */
+	for (sym = (void *)ehdr + _r(&symtab_sec->sh_offset);
+	     sym < sym + _r(&symtab_sec->sh_size) / sizeof(Elf_Sym);
+	     sym++) {
 		if (ELF_ST_TYPE(sym->st_info) != STT_OBJECT)
 			continue;
-		idx = r(&sym->st_name);
-		if (strcmp(strtab + idx, "main_extable_sort_needed") == 0) {
+		if (!strcmp(strtab + r(&sym->st_name),
+			    "main_extable_sort_needed")) {
 			sort_needed_sym = sym;
 			break;
 		}
 	}
-	if (sort_needed_sym == NULL) {
+
+	if (!sort_needed_sym) {
 		fprintf(stderr,
-			"no main_extable_sort_needed symbol in  file: %s\n",
+			"no main_extable_sort_needed symbol in file: %s\n",
 			fname);
-		fail_file();
+		goto out;
 	}
+
 	sort_needed_sec = &shdr[get_secindex(r2(&sym->st_shndx),
 					     sort_needed_sym - symtab,
-					     symtab_shndx_start)];
-	sort_done_location = (void *)ehdr +
+					     symtab_shndx)];
+	sort_needed_loc = (void *)ehdr +
 		_r(&sort_needed_sec->sh_offset) +
 		_r(&sort_needed_sym->st_value) -
 		_r(&sort_needed_sec->sh_addr);
 
-#if 0
-	printf("sort done marker at %lx\n",
-	       (unsigned long)((char *)sort_done_location - (char *)ehdr));
+	/* extable has been sorted, clear the flag */
+	w(0, sort_needed_loc);
+	rc = 0;
+
+out:
+#if defined(SORTTABLE_64) && defined(UNWINDER_ORC_ENABLED)
+	if (orc_sort_thread) {
+		void *retval = NULL;
+		/* wait for ORC tables sort done */
+		rc = pthread_join(orc_sort_thread, &retval);
+		if (rc)
+			fprintf(stderr,
+				"pthread_join failed '%s': %s\n",
+				strerror(errno), fname);
+		else if (retval) {
+			rc = -1;
+			fprintf(stderr,
+				"failed to sort ORC tables '%s': %s\n",
+				(char *)retval, fname);
+		}
+	}
 #endif
-	/* We sorted it, clear the flag. */
-	w(0, sort_done_location);
+	return rc;
 }