From 44e6a2069122dbec29d25b8ced3f59f5626a97d7 Mon Sep 17 00:00:00 2001
From: Michal Marek <mmarek@suse.cz>
Date: Tue, 17 May 2011 15:36:46 +0200
Subject: [PATCH 01/15] ftrace/kbuild: Add recordmcount files to force full
 build

Modifications to recordmcount must be performed on all object
files to stay consistent with what the kernel code may expect.
Add the recordmcount files to the main dependencies to make sure
any change to them causes a full recompile.

Signed-off-by: Michal Marek <mmarek@suse.cz>
Link: http://lkml.kernel.org/r/20110517133646.GP13293@sepie.suse.cz
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
 scripts/Makefile.build | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/scripts/Makefile.build b/scripts/Makefile.build
index d5f925abe4d2..7d3f9036eee2 100644
--- a/scripts/Makefile.build
+++ b/scripts/Makefile.build
@@ -252,6 +252,8 @@ sub_cmd_record_mcount =					\
 	if [ $(@) != "scripts/mod/empty.o" ]; then	\
 		$(objtree)/scripts/recordmcount "$(@)";	\
 	fi;
+recordmcount_source := $(srctree)/scripts/recordmcount.c \
+		    $(srctree)/scripts/recordmcount.h
 else
 sub_cmd_record_mcount = set -e ; perl $(srctree)/scripts/recordmcount.pl "$(ARCH)" \
 	"$(if $(CONFIG_CPU_BIG_ENDIAN),big,little)" \
@@ -259,6 +261,7 @@ sub_cmd_record_mcount = set -e ; perl $(srctree)/scripts/recordmcount.pl "$(ARCH
 	"$(OBJDUMP)" "$(OBJCOPY)" "$(CC) $(KBUILD_CFLAGS)" \
 	"$(LD)" "$(NM)" "$(RM)" "$(MV)" \
 	"$(if $(part-of-module),1,0)" "$(@)";
+recordmcount_source := $(srctree)/scripts/recordmcount.pl
 endif
 cmd_record_mcount = 						\
 	if [ "$(findstring -pg,$(_c_flags))" = "-pg" ]; then	\
@@ -279,13 +282,13 @@ define rule_cc_o_c
 endef
 
 # Built-in and composite module parts
-$(obj)/%.o: $(src)/%.c FORCE
+$(obj)/%.o: $(src)/%.c $(recordmcount_source) FORCE
 	$(call cmd,force_checksrc)
 	$(call if_changed_rule,cc_o_c)
 
 # Single-part modules are special since we need to mark them in $(MODVERDIR)
 
-$(single-used-m): $(obj)/%.o: $(src)/%.c FORCE
+$(single-used-m): $(obj)/%.o: $(src)/%.c $(recordmcount_source) FORCE
 	$(call cmd,force_checksrc)
 	$(call if_changed_rule,cc_o_c)
 	@{ echo $(@:.o=.ko); echo $@; } > $(MODVERDIR)/$(@F:.o=.mod)

From c6fece27dfa46abc8c4b87d94a8c2f8cd6c5d8b1 Mon Sep 17 00:00:00 2001
From: Steven Rostedt <srostedt@redhat.com>
Date: Wed, 6 Apr 2011 13:21:17 -0400
Subject: [PATCH 02/15] ftrace/trivial: Clean up recordmcount.c to use Linux
 style comparisons

The Linux ftrace subsystem style for comparing is:

  var == 1
  var > 0

and not:

  1 == var
  0 < var

It is considered that Linux developers are smart enough not to do the

  if (var = 1)

mistake.

Cc: John Reiser <jreiser@bitwagon.com>
Link: http://lkml.kernel.org/r/20110421023737.290712238@goodmis.org
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
 scripts/recordmcount.c | 48 +++++++++++++++++++++---------------------
 scripts/recordmcount.h | 16 +++++++-------
 2 files changed, 32 insertions(+), 32 deletions(-)

diff --git a/scripts/recordmcount.c b/scripts/recordmcount.c
index f9f6f52db772..37afe0ecacb2 100644
--- a/scripts/recordmcount.c
+++ b/scripts/recordmcount.c
@@ -78,7 +78,7 @@ static off_t
 ulseek(int const fd, off_t const offset, int const whence)
 {
 	off_t const w = lseek(fd, offset, whence);
-	if ((off_t)-1 == w) {
+	if (w == (off_t)-1) {
 		perror("lseek");
 		fail_file();
 	}
@@ -111,7 +111,7 @@ static void *
 umalloc(size_t size)
 {
 	void *const addr = malloc(size);
-	if (0 == addr) {
+	if (addr == 0) {
 		fprintf(stderr, "malloc failed: %zu bytes\n", size);
 		fail_file();
 	}
@@ -136,7 +136,7 @@ static void *mmap_file(char const *fname)
 	void *addr;
 
 	fd_map = open(fname, O_RDWR);
-	if (0 > fd_map || 0 > fstat(fd_map, &sb)) {
+	if (fd_map < 0 || fstat(fd_map, &sb) < 0) {
 		perror(fname);
 		fail_file();
 	}
@@ -147,7 +147,7 @@ static void *mmap_file(char const *fname)
 	addr = mmap(0, sb.st_size, PROT_READ|PROT_WRITE, MAP_PRIVATE,
 		    fd_map, 0);
 	mmap_failed = 0;
-	if (MAP_FAILED == addr) {
+	if (addr == MAP_FAILED) {
 		mmap_failed = 1;
 		addr = umalloc(sb.st_size);
 		uread(fd_map, addr, sb.st_size);
@@ -206,12 +206,12 @@ static uint32_t (*w2)(uint16_t);
 static int
 is_mcounted_section_name(char const *const txtname)
 {
-	return 0 == strcmp(".text",           txtname) ||
-		0 == strcmp(".ref.text",      txtname) ||
-		0 == strcmp(".sched.text",    txtname) ||
-		0 == strcmp(".spinlock.text", txtname) ||
-		0 == strcmp(".irqentry.text", txtname) ||
-		0 == strcmp(".text.unlikely", txtname);
+	return strcmp(".text",           txtname) == 0 ||
+		strcmp(".ref.text",      txtname) == 0 ||
+		strcmp(".sched.text",    txtname) == 0 ||
+		strcmp(".spinlock.text", txtname) == 0 ||
+		strcmp(".irqentry.text", txtname) == 0 ||
+		strcmp(".text.unlikely", txtname) == 0;
 }
 
 /* 32 bit and 64 bit are very similar */
@@ -270,7 +270,7 @@ do_file(char const *const fname)
 		fail_file();
 	} break;
 	case ELFDATA2LSB: {
-		if (1 != *(unsigned char const *)&endian) {
+		if (*(unsigned char const *)&endian != 1) {
 			/* main() is big endian, file.o is little endian. */
 			w = w4rev;
 			w2 = w2rev;
@@ -278,7 +278,7 @@ do_file(char const *const fname)
 		}
 	} break;
 	case ELFDATA2MSB: {
-		if (0 != *(unsigned char const *)&endian) {
+		if (*(unsigned char const *)&endian != 0) {
 			/* main() is little endian, file.o is big endian. */
 			w = w4rev;
 			w2 = w2rev;
@@ -286,9 +286,9 @@ do_file(char const *const fname)
 		}
 	} break;
 	}  /* end switch */
-	if (0 != memcmp(ELFMAG, ehdr->e_ident, SELFMAG)
-	||  ET_REL != w2(ehdr->e_type)
-	||  EV_CURRENT != ehdr->e_ident[EI_VERSION]) {
+	if (memcmp(ELFMAG, ehdr->e_ident, SELFMAG) != 0
+	||  w2(ehdr->e_type) != ET_REL
+	||  ehdr->e_ident[EI_VERSION] != EV_CURRENT) {
 		fprintf(stderr, "unrecognized ET_REL file %s\n", fname);
 		fail_file();
 	}
@@ -321,15 +321,15 @@ do_file(char const *const fname)
 		fail_file();
 	} break;
 	case ELFCLASS32: {
-		if (sizeof(Elf32_Ehdr) != w2(ehdr->e_ehsize)
-		||  sizeof(Elf32_Shdr) != w2(ehdr->e_shentsize)) {
+		if (w2(ehdr->e_ehsize) != sizeof(Elf32_Ehdr)
+		||  w2(ehdr->e_shentsize) != sizeof(Elf32_Shdr)) {
 			fprintf(stderr,
 				"unrecognized ET_REL file: %s\n", fname);
 			fail_file();
 		}
-		if (EM_S390 == w2(ehdr->e_machine))
+		if (w2(ehdr->e_machine) == EM_S390)
 			reltype = R_390_32;
-		if (EM_MIPS == w2(ehdr->e_machine)) {
+		if (w2(ehdr->e_machine) == EM_MIPS) {
 			reltype = R_MIPS_32;
 			is_fake_mcount32 = MIPS32_is_fake_mcount;
 		}
@@ -337,15 +337,15 @@ do_file(char const *const fname)
 	} break;
 	case ELFCLASS64: {
 		Elf64_Ehdr *const ghdr = (Elf64_Ehdr *)ehdr;
-		if (sizeof(Elf64_Ehdr) != w2(ghdr->e_ehsize)
-		||  sizeof(Elf64_Shdr) != w2(ghdr->e_shentsize)) {
+		if (w2(ghdr->e_ehsize) != sizeof(Elf64_Ehdr)
+		||  w2(ghdr->e_shentsize) != sizeof(Elf64_Shdr)) {
 			fprintf(stderr,
 				"unrecognized ET_REL file: %s\n", fname);
 			fail_file();
 		}
-		if (EM_S390 == w2(ghdr->e_machine))
+		if (w2(ghdr->e_machine) == EM_S390)
 			reltype = R_390_64;
-		if (EM_MIPS == w2(ghdr->e_machine)) {
+		if (w2(ghdr->e_machine) == EM_MIPS) {
 			reltype = R_MIPS_64;
 			Elf64_r_sym = MIPS64_r_sym;
 			Elf64_r_info = MIPS64_r_info;
@@ -371,7 +371,7 @@ main(int argc, char const *argv[])
 	}
 
 	/* Process each file in turn, allowing deep failure. */
-	for (--argc, ++argv; 0 < argc; --argc, ++argv) {
+	for (--argc, ++argv; argc > 0; --argc, ++argv) {
 		int const sjval = setjmp(jmpenv);
 		int len;
 
diff --git a/scripts/recordmcount.h b/scripts/recordmcount.h
index baf187bee983..ac7b3303cb31 100644
--- a/scripts/recordmcount.h
+++ b/scripts/recordmcount.h
@@ -275,12 +275,12 @@ static uint_t *sift_rel_mcount(uint_t *mlocp,
 			Elf_Sym const *const symp =
 				&sym0[Elf_r_sym(relp)];
 			char const *symname = &str0[w(symp->st_name)];
-			char const *mcount = '_' == gpfx ? "_mcount" : "mcount";
+			char const *mcount = gpfx == '_' ? "_mcount" : "mcount";
 
-			if ('.' == symname[0])
+			if (symname[0] == '.')
 				++symname;  /* ppc64 hack */
-			if (0 == strcmp(mcount, symname) ||
-			    (altmcount && 0 == strcmp(altmcount, symname)))
+			if (strcmp(mcount, symname) == 0 ||
+			    (altmcount && strcmp(altmcount, symname) == 0))
 				mcountsym = Elf_r_sym(relp);
 		}
 
@@ -290,7 +290,7 @@ static uint_t *sift_rel_mcount(uint_t *mlocp,
 			mrelp->r_offset = _w(offbase
 				+ ((void *)mlocp - (void *)mloc0));
 			Elf_r_info(mrelp, recsym, reltype);
-			if (sizeof(Elf_Rela) == rel_entsize) {
+			if (rel_entsize == sizeof(Elf_Rela)) {
 				((Elf_Rela *)mrelp)->r_addend = addend;
 				*mlocp++ = 0;
 			} else
@@ -354,12 +354,12 @@ __has_rel_mcount(Elf_Shdr const *const relhdr,  /* is SHT_REL or SHT_RELA */
 	Elf_Shdr const *const txthdr = &shdr0[w(relhdr->sh_info)];
 	char const *const txtname = &shstrtab[w(txthdr->sh_name)];
 
-	if (0 == strcmp("__mcount_loc", txtname)) {
+	if (strcmp("__mcount_loc", txtname) == 0) {
 		fprintf(stderr, "warning: __mcount_loc already exists: %s\n",
 			fname);
 		succeed_file();
 	}
-	if (SHT_PROGBITS != w(txthdr->sh_type) ||
+	if (w(txthdr->sh_type) != SHT_PROGBITS ||
 	    !is_mcounted_section_name(txtname))
 		return NULL;
 	return txtname;
@@ -370,7 +370,7 @@ static char const *has_rel_mcount(Elf_Shdr const *const relhdr,
 				  char const *const shstrtab,
 				  char const *const fname)
 {
-	if (SHT_REL  != w(relhdr->sh_type) && SHT_RELA != w(relhdr->sh_type))
+	if (w(relhdr->sh_type) != SHT_REL && w(relhdr->sh_type) != SHT_RELA)
 		return NULL;
 	return __has_rel_mcount(relhdr, shdr0, shstrtab, fname);
 }

From d7e8623ae9cf93a9b517336085a86cabb014ea77 Mon Sep 17 00:00:00 2001
From: Steven Rostedt <srostedt@redhat.com>
Date: Wed, 6 Apr 2011 13:32:24 -0400
Subject: [PATCH 03/15] ftrace/trivial: Clean up record mcount to use Linux
 switch style

The Linux style for switch statements is:

	switch (var) {
	case x:
		[...]
		break;
	}

Not:
	switch (var) {
	case x: {
		[...]
	} break;

Cc: John Reiser <jreiser@bitwagon.com>
Link: http://lkml.kernel.org/r/20110421023737.523968644@goodmis.org
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
 scripts/recordmcount.c | 43 +++++++++++++++++++++---------------------
 1 file changed, 22 insertions(+), 21 deletions(-)

diff --git a/scripts/recordmcount.c b/scripts/recordmcount.c
index 37afe0ecacb2..4ebd8399cb33 100644
--- a/scripts/recordmcount.c
+++ b/scripts/recordmcount.c
@@ -264,27 +264,27 @@ do_file(char const *const fname)
 	w8 = w8nat;
 	switch (ehdr->e_ident[EI_DATA]) {
 		static unsigned int const endian = 1;
-	default: {
+	default:
 		fprintf(stderr, "unrecognized ELF data encoding %d: %s\n",
 			ehdr->e_ident[EI_DATA], fname);
 		fail_file();
-	} break;
-	case ELFDATA2LSB: {
+		break;
+	case ELFDATA2LSB:
 		if (*(unsigned char const *)&endian != 1) {
 			/* main() is big endian, file.o is little endian. */
 			w = w4rev;
 			w2 = w2rev;
 			w8 = w8rev;
 		}
-	} break;
-	case ELFDATA2MSB: {
+		break;
+	case ELFDATA2MSB:
 		if (*(unsigned char const *)&endian != 0) {
 			/* main() is little endian, file.o is big endian. */
 			w = w4rev;
 			w2 = w2rev;
 			w8 = w8rev;
 		}
-	} break;
+		break;
 	}  /* end switch */
 	if (memcmp(ELFMAG, ehdr->e_ident, SELFMAG) != 0
 	||  w2(ehdr->e_type) != ET_REL
@@ -295,11 +295,11 @@ do_file(char const *const fname)
 
 	gpfx = 0;
 	switch (w2(ehdr->e_machine)) {
-	default: {
+	default:
 		fprintf(stderr, "unrecognized e_machine %d %s\n",
 			w2(ehdr->e_machine), fname);
 		fail_file();
-	} break;
+		break;
 	case EM_386:	 reltype = R_386_32;                   break;
 	case EM_ARM:	 reltype = R_ARM_ABS32;
 			 altmcount = "__gnu_mcount_nc";
@@ -315,12 +315,12 @@ do_file(char const *const fname)
 	}  /* end switch */
 
 	switch (ehdr->e_ident[EI_CLASS]) {
-	default: {
+	default:
 		fprintf(stderr, "unrecognized ELF class %d %s\n",
 			ehdr->e_ident[EI_CLASS], fname);
 		fail_file();
-	} break;
-	case ELFCLASS32: {
+		break;
+	case ELFCLASS32:
 		if (w2(ehdr->e_ehsize) != sizeof(Elf32_Ehdr)
 		||  w2(ehdr->e_shentsize) != sizeof(Elf32_Shdr)) {
 			fprintf(stderr,
@@ -334,7 +334,7 @@ do_file(char const *const fname)
 			is_fake_mcount32 = MIPS32_is_fake_mcount;
 		}
 		do32(ehdr, fname, reltype);
-	} break;
+		break;
 	case ELFCLASS64: {
 		Elf64_Ehdr *const ghdr = (Elf64_Ehdr *)ehdr;
 		if (w2(ghdr->e_ehsize) != sizeof(Elf64_Ehdr)
@@ -352,7 +352,8 @@ do_file(char const *const fname)
 			is_fake_mcount64 = MIPS64_is_fake_mcount;
 		}
 		do64(ghdr, fname, reltype);
-	} break;
+		break;
+	}
 	}  /* end switch */
 
 	cleanup();
@@ -386,23 +387,23 @@ main(int argc, char const *argv[])
 			continue;
 
 		switch (sjval) {
-		default: {
+		default:
 			fprintf(stderr, "internal error: %s\n", argv[0]);
 			exit(1);
-		} break;
-		case SJ_SETJMP: {  /* normal sequence */
+			break;
+		case SJ_SETJMP:    /* normal sequence */
 			/* Avoid problems if early cleanup() */
 			fd_map = -1;
 			ehdr_curr = NULL;
 			mmap_failed = 1;
 			do_file(argv[0]);
-		} break;
-		case SJ_FAIL: {  /* error in do_file or below */
+			break;
+		case SJ_FAIL:    /* error in do_file or below */
 			++n_error;
-		} break;
-		case SJ_SUCCEED: {  /* premature success */
+			break;
+		case SJ_SUCCEED:    /* premature success */
 			/* do nothing */
-		} break;
+			break;
 		}  /* end switch */
 	}
 	return !!n_error;

From bde66c3f45d777082a0df01e64870f47ba2a5055 Mon Sep 17 00:00:00 2001
From: Steven Rostedt <srostedt@redhat.com>
Date: Wed, 6 Apr 2011 14:10:22 -0400
Subject: [PATCH 04/15] ftrace: Add .kprobe.text section to whitelist for
 recordmcount.c

The .kprobe.text section is safe to modify mcount to nop and tracing.
Add it to the whitelist in recordmcount.c and recordmcount.pl.

Cc: John Reiser <jreiser@bitwagon.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Link: http://lkml.kernel.org/r/20110421023737.743350547@goodmis.org
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
 scripts/recordmcount.c  | 1 +
 scripts/recordmcount.pl | 1 +
 2 files changed, 2 insertions(+)

diff --git a/scripts/recordmcount.c b/scripts/recordmcount.c
index 4ebd8399cb33..37c59654c133 100644
--- a/scripts/recordmcount.c
+++ b/scripts/recordmcount.c
@@ -211,6 +211,7 @@ is_mcounted_section_name(char const *const txtname)
 		strcmp(".sched.text",    txtname) == 0 ||
 		strcmp(".spinlock.text", txtname) == 0 ||
 		strcmp(".irqentry.text", txtname) == 0 ||
+		strcmp(".kprobes.text", txtname) == 0 ||
 		strcmp(".text.unlikely", txtname) == 0;
 }
 
diff --git a/scripts/recordmcount.pl b/scripts/recordmcount.pl
index 4be0deea71ca..a871cd414055 100755
--- a/scripts/recordmcount.pl
+++ b/scripts/recordmcount.pl
@@ -134,6 +134,7 @@ my %text_sections = (
      ".sched.text" => 1,
      ".spinlock.text" => 1,
      ".irqentry.text" => 1,
+     ".kprobes.text" => 1,
      ".text.unlikely" => 1,
 );
 

From df2ccb69454d022ce99e3a3b7ee7f9fb4a4e9563 Mon Sep 17 00:00:00 2001
From: Steven Rostedt <srostedt@redhat.com>
Date: Wed, 13 Apr 2011 13:31:08 -0400
Subject: [PATCH 05/15] ftrace/recordmcount: Modify only executable sections

PROGBITS is not enough to determine if the section should be modified
or not. Only process sections that are marked as executable.

Cc: John Reiser <jreiser@bitwagon.com>
Link: http://lkml.kernel.org/r/20110421023737.991485123@goodmis.org
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
 scripts/recordmcount.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/scripts/recordmcount.h b/scripts/recordmcount.h
index ac7b3303cb31..7f8d5c4c780f 100644
--- a/scripts/recordmcount.h
+++ b/scripts/recordmcount.h
@@ -360,6 +360,7 @@ __has_rel_mcount(Elf_Shdr const *const relhdr,  /* is SHT_REL or SHT_RELA */
 		succeed_file();
 	}
 	if (w(txthdr->sh_type) != SHT_PROGBITS ||
+	    !(w(txthdr->sh_flags) & SHF_EXECINSTR) ||
 	    !is_mcounted_section_name(txtname))
 		return NULL;
 	return txtname;

From 68eb1d2e5370c6da4c871c80fdccbe315dc2f3c1 Mon Sep 17 00:00:00 2001
From: Steven Rostedt <srostedt@redhat.com>
Date: Fri, 8 Apr 2011 03:58:48 -0400
Subject: [PATCH 06/15] ftrace/recordmcount: Make ignored mcount calls into
 nops at compile time

There are sections that are ignored by ftrace for the function tracing because
the text is in a section that can be removed without notice. The mcount calls
in these sections are ignored and ftrace never sees them. The downside of this
is that the functions in these sections still call mcount. Although the mcount
function is defined in assembly simply as a return, this added overhead is
unnecessary.

The solution is to convert these callers into nops at compile time.
A better solution is to add 'notrace' to the section markers, but as new sections
come up all the time, it would be nice that they are delt with when they
are created.

Later patches will deal with finding these sections and doing the proper solution.

Thanks to H. Peter Anvin for giving me the right nops to use for x86.

Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: John Reiser <jreiser@bitwagon.com>
Link: http://lkml.kernel.org/r/20110421023738.237101176@goodmis.org
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
 scripts/recordmcount.c | 40 +++++++++++++++++++--
 scripts/recordmcount.h | 82 +++++++++++++++++++++++++++++++++++++++---
 2 files changed, 116 insertions(+), 6 deletions(-)

diff --git a/scripts/recordmcount.c b/scripts/recordmcount.c
index 37c59654c133..78054a41d134 100644
--- a/scripts/recordmcount.c
+++ b/scripts/recordmcount.c
@@ -118,6 +118,34 @@ umalloc(size_t size)
 	return addr;
 }
 
+static unsigned char ideal_nop5_x86_64[5] = { 0x0f, 0x1f, 0x44, 0x00, 0x00 };
+static unsigned char ideal_nop5_x86_32[5] = { 0x3e, 0x8d, 0x74, 0x26, 0x00 };
+static unsigned char *ideal_nop;
+
+static char rel_type_nop;
+
+static int (*make_nop)(void *map, size_t const offset);
+
+static int make_nop_x86(void *map, size_t const offset)
+{
+	uint32_t *ptr;
+	unsigned char *op;
+
+	/* Confirm we have 0xe8 0x0 0x0 0x0 0x0 */
+	ptr = map + offset;
+	if (*ptr != 0)
+		return -1;
+
+	op = map + offset - 1;
+	if (*op != 0xe8)
+		return -1;
+
+	/* convert to nop */
+	ulseek(fd_map, offset - 1, SEEK_SET);
+	uwrite(fd_map, ideal_nop, 5);
+	return 0;
+}
+
 /*
  * Get the whole file as a programming convenience in order to avoid
  * malloc+lseek+read+free of many pieces.  If successful, then mmap
@@ -301,7 +329,11 @@ do_file(char const *const fname)
 			w2(ehdr->e_machine), fname);
 		fail_file();
 		break;
-	case EM_386:	 reltype = R_386_32;                   break;
+	case EM_386:
+		reltype = R_386_32;
+		make_nop = make_nop_x86;
+		ideal_nop = ideal_nop5_x86_32;
+		break;
 	case EM_ARM:	 reltype = R_ARM_ABS32;
 			 altmcount = "__gnu_mcount_nc";
 			 break;
@@ -312,7 +344,11 @@ do_file(char const *const fname)
 	case EM_S390:    /* reltype: e_class    */ gpfx = '_'; break;
 	case EM_SH:	 reltype = R_SH_DIR32;                 break;
 	case EM_SPARCV9: reltype = R_SPARC_64;     gpfx = '_'; break;
-	case EM_X86_64:	 reltype = R_X86_64_64;                break;
+	case EM_X86_64:
+		make_nop = make_nop_x86;
+		ideal_nop = ideal_nop5_x86_64;
+		reltype = R_X86_64_64;
+		break;
 	}  /* end switch */
 
 	switch (ehdr->e_ident[EI_CLASS]) {
diff --git a/scripts/recordmcount.h b/scripts/recordmcount.h
index 7f8d5c4c780f..657dbedd1c7f 100644
--- a/scripts/recordmcount.h
+++ b/scripts/recordmcount.h
@@ -23,6 +23,7 @@
 #undef fn_is_fake_mcount
 #undef MIPS_is_fake_mcount
 #undef sift_rel_mcount
+#undef nop_mcount
 #undef find_secsym_ndx
 #undef __has_rel_mcount
 #undef has_rel_mcount
@@ -49,6 +50,7 @@
 #ifdef RECORD_MCOUNT_64
 # define append_func		append64
 # define sift_rel_mcount	sift64_rel_mcount
+# define nop_mcount		nop_mcount_64
 # define find_secsym_ndx	find64_secsym_ndx
 # define __has_rel_mcount	__has64_rel_mcount
 # define has_rel_mcount		has64_rel_mcount
@@ -77,6 +79,7 @@
 #else
 # define append_func		append32
 # define sift_rel_mcount	sift32_rel_mcount
+# define nop_mcount		nop_mcount_32
 # define find_secsym_ndx	find32_secsym_ndx
 # define __has_rel_mcount	__has32_rel_mcount
 # define has_rel_mcount		has32_rel_mcount
@@ -304,6 +307,70 @@ static uint_t *sift_rel_mcount(uint_t *mlocp,
 	return mlocp;
 }
 
+/*
+ * Read the relocation table again, but this time its called on sections
+ * that are not going to be traced. The mcount calls here will be converted
+ * into nops.
+ */
+static void nop_mcount(Elf_Shdr const *const relhdr,
+		       Elf_Ehdr const *const ehdr)
+{
+	Elf_Shdr *const shdr0 = (Elf_Shdr *)(_w(ehdr->e_shoff)
+		+ (void *)ehdr);
+	unsigned const symsec_sh_link = w(relhdr->sh_link);
+	Elf_Shdr const *const symsec = &shdr0[symsec_sh_link];
+	Elf_Sym const *const sym0 = (Elf_Sym const *)(_w(symsec->sh_offset)
+		+ (void *)ehdr);
+
+	Elf_Shdr const *const strsec = &shdr0[w(symsec->sh_link)];
+	char const *const str0 = (char const *)(_w(strsec->sh_offset)
+		+ (void *)ehdr);
+
+	Elf_Rel const *const rel0 = (Elf_Rel const *)(_w(relhdr->sh_offset)
+		+ (void *)ehdr);
+	unsigned rel_entsize = _w(relhdr->sh_entsize);
+	unsigned const nrel = _w(relhdr->sh_size) / rel_entsize;
+	Elf_Rel const *relp = rel0;
+
+	Elf_Shdr const *const shdr = &shdr0[w(relhdr->sh_info)];
+
+	unsigned mcountsym = 0;
+	unsigned t;
+
+	for (t = nrel; t; --t) {
+		int ret = -1;
+
+		if (!mcountsym) {
+			Elf_Sym const *const symp =
+				&sym0[Elf_r_sym(relp)];
+			char const *symname = &str0[w(symp->st_name)];
+			char const *mcount = gpfx == '_' ? "_mcount" : "mcount";
+
+			if (symname[0] == '.')
+				++symname;  /* ppc64 hack */
+			if (strcmp(mcount, symname) == 0 ||
+			    (altmcount && strcmp(altmcount, symname) == 0))
+				mcountsym = Elf_r_sym(relp);
+		}
+
+		if (mcountsym == Elf_r_sym(relp) && !is_fake_mcount(relp))
+			ret = make_nop((void *)ehdr, shdr->sh_offset + relp->r_offset);
+
+		/*
+		 * If we successfully removed the mcount, mark the relocation
+		 * as a nop (don't do anything with it).
+		 */
+		if (!ret) {
+			Elf_Rel rel;
+			rel = *(Elf_Rel *)relp;
+			Elf_r_info(&rel, Elf_r_sym(relp), rel_type_nop);
+			ulseek(fd_map, (void *)relp - (void *)ehdr, SEEK_SET);
+			uwrite(fd_map, &rel, sizeof(rel));
+		}
+		relp = (Elf_Rel const *)(rel_entsize + (void *)relp);
+	}
+}
+
 
 /*
  * Find a symbol in the given section, to be used as the base for relocating
@@ -360,8 +427,7 @@ __has_rel_mcount(Elf_Shdr const *const relhdr,  /* is SHT_REL or SHT_RELA */
 		succeed_file();
 	}
 	if (w(txthdr->sh_type) != SHT_PROGBITS ||
-	    !(w(txthdr->sh_flags) & SHF_EXECINSTR) ||
-	    !is_mcounted_section_name(txtname))
+	    !(w(txthdr->sh_flags) & SHF_EXECINSTR))
 		return NULL;
 	return txtname;
 }
@@ -384,9 +450,11 @@ static unsigned tot_relsize(Elf_Shdr const *const shdr0,
 {
 	unsigned totrelsz = 0;
 	Elf_Shdr const *shdrp = shdr0;
+	char const *txtname;
 
 	for (; nhdr; --nhdr, ++shdrp) {
-		if (has_rel_mcount(shdrp, shdr0, shstrtab, fname))
+		txtname = has_rel_mcount(shdrp, shdr0, shstrtab, fname);
+		if (txtname && is_mcounted_section_name(txtname))
 			totrelsz += _w(shdrp->sh_size);
 	}
 	return totrelsz;
@@ -422,7 +490,7 @@ do_func(Elf_Ehdr *const ehdr, char const *const fname, unsigned const reltype)
 	for (relhdr = shdr0, k = nhdr; k; --k, ++relhdr) {
 		char const *const txtname = has_rel_mcount(relhdr, shdr0,
 			shstrtab, fname);
-		if (txtname) {
+		if (txtname && is_mcounted_section_name(txtname)) {
 			uint_t recval = 0;
 			unsigned const recsym = find_secsym_ndx(
 				w(relhdr->sh_info), txtname, &recval,
@@ -433,6 +501,12 @@ do_func(Elf_Ehdr *const ehdr, char const *const fname, unsigned const reltype)
 			mlocp = sift_rel_mcount(mlocp,
 				(void *)mlocp - (void *)mloc0, &mrelp,
 				relhdr, ehdr, recsym, recval, reltype);
+		} else if (make_nop && txtname) {
+			/*
+			 * This section is ignored by ftrace, but still
+			 * has mcount calls. Convert them to nops now.
+			 */
+			nop_mcount(relhdr, ehdr);
 		}
 	}
 	if (mloc0 != mlocp) {

From 06fd6f6301cce9af4a03dd9e40bfca167e931ff1 Mon Sep 17 00:00:00 2001
From: Steven Rostedt <srostedt@redhat.com>
Date: Tue, 12 Apr 2011 18:53:25 -0400
Subject: [PATCH 07/15] ftrace/recordmcount: Add warning logic to warn on
 mcount not recorded

There's some sections that should not have mcount recorded and should not have
modifications to the that code. But currently they waste some time by calling
mcount anyway (which simply returns). As the real answer should be to
either whitelist the section or have gcc ignore it fully.

This change adds a option to recordmcount to warn when it finds a section
that is ignored by ftrace but still contains mcount callers. This is not on
by default as developers may not know if the section should be completely
ignored or added to the whitelist.

Cc: John Reiser <jreiser@bitwagon.com>
Link: http://lkml.kernel.org/r/20110421023738.476989377@goodmis.org
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
 scripts/recordmcount.c | 32 ++++++++++++++++++++++++--------
 scripts/recordmcount.h | 22 +++++++++++++++++-----
 2 files changed, 41 insertions(+), 13 deletions(-)

diff --git a/scripts/recordmcount.c b/scripts/recordmcount.c
index 78054a41d134..0e18975824f7 100644
--- a/scripts/recordmcount.c
+++ b/scripts/recordmcount.c
@@ -24,6 +24,7 @@
 #include <sys/types.h>
 #include <sys/mman.h>
 #include <sys/stat.h>
+#include <getopt.h>
 #include <elf.h>
 #include <fcntl.h>
 #include <setjmp.h>
@@ -39,6 +40,7 @@ static char gpfx;	/* prefix for global symbol name (sometimes '_') */
 static struct stat sb;	/* Remember .st_size, etc. */
 static jmp_buf jmpenv;	/* setjmp/longjmp per-file error escape */
 static const char *altmcount;	/* alternate mcount symbol name */
+static int warn_on_notrace_sect; /* warn when section has mcount not being recorded */
 
 /* setjmp() return values */
 enum {
@@ -397,19 +399,33 @@ do_file(char const *const fname)
 }
 
 int
-main(int argc, char const *argv[])
+main(int argc, char *argv[])
 {
 	const char ftrace[] = "/ftrace.o";
 	int ftrace_size = sizeof(ftrace) - 1;
 	int n_error = 0;  /* gcc-4.3.0 false positive complaint */
+	int c;
+	int i;
 
-	if (argc <= 1) {
-		fprintf(stderr, "usage: recordmcount file.o...\n");
+	while ((c = getopt(argc, argv, "w")) >= 0) {
+		switch (c) {
+		case 'w':
+			warn_on_notrace_sect = 1;
+			break;
+		default:
+			fprintf(stderr, "usage: recordmcount [-w] file.o...\n");
+			return 0;
+		}
+	}
+
+	if ((argc - optind) < 1) {
+		fprintf(stderr, "usage: recordmcount [-w] file.o...\n");
 		return 0;
 	}
 
 	/* Process each file in turn, allowing deep failure. */
-	for (--argc, ++argv; argc > 0; --argc, ++argv) {
+	for (i = optind; i < argc; i++) {
+		char *file = argv[i];
 		int const sjval = setjmp(jmpenv);
 		int len;
 
@@ -418,14 +434,14 @@ main(int argc, char const *argv[])
 		 * function but does not call it. Since ftrace.o should
 		 * not be traced anyway, we just skip it.
 		 */
-		len = strlen(argv[0]);
+		len = strlen(file);
 		if (len >= ftrace_size &&
-		    strcmp(argv[0] + (len - ftrace_size), ftrace) == 0)
+		    strcmp(file + (len - ftrace_size), ftrace) == 0)
 			continue;
 
 		switch (sjval) {
 		default:
-			fprintf(stderr, "internal error: %s\n", argv[0]);
+			fprintf(stderr, "internal error: %s\n", file);
 			exit(1);
 			break;
 		case SJ_SETJMP:    /* normal sequence */
@@ -433,7 +449,7 @@ main(int argc, char const *argv[])
 			fd_map = -1;
 			ehdr_curr = NULL;
 			mmap_failed = 1;
-			do_file(argv[0]);
+			do_file(file);
 			break;
 		case SJ_FAIL:    /* error in do_file or below */
 			++n_error;
diff --git a/scripts/recordmcount.h b/scripts/recordmcount.h
index 657dbedd1c7f..22033d563bcd 100644
--- a/scripts/recordmcount.h
+++ b/scripts/recordmcount.h
@@ -313,7 +313,8 @@ static uint_t *sift_rel_mcount(uint_t *mlocp,
  * into nops.
  */
 static void nop_mcount(Elf_Shdr const *const relhdr,
-		       Elf_Ehdr const *const ehdr)
+		       Elf_Ehdr const *const ehdr,
+		       const char *const txtname)
 {
 	Elf_Shdr *const shdr0 = (Elf_Shdr *)(_w(ehdr->e_shoff)
 		+ (void *)ehdr);
@@ -336,6 +337,7 @@ static void nop_mcount(Elf_Shdr const *const relhdr,
 
 	unsigned mcountsym = 0;
 	unsigned t;
+	int once = 0;
 
 	for (t = nrel; t; --t) {
 		int ret = -1;
@@ -353,8 +355,18 @@ static void nop_mcount(Elf_Shdr const *const relhdr,
 				mcountsym = Elf_r_sym(relp);
 		}
 
-		if (mcountsym == Elf_r_sym(relp) && !is_fake_mcount(relp))
-			ret = make_nop((void *)ehdr, shdr->sh_offset + relp->r_offset);
+		if (mcountsym == Elf_r_sym(relp) && !is_fake_mcount(relp)) {
+			if (make_nop)
+				ret = make_nop((void *)ehdr, shdr->sh_offset + relp->r_offset);
+			if (warn_on_notrace_sect && !once) {
+				printf("Section %s has mcount callers being ignored\n",
+				       txtname);
+				once = 1;
+				/* just warn? */
+				if (!make_nop)
+					return;
+			}
+		}
 
 		/*
 		 * If we successfully removed the mcount, mark the relocation
@@ -501,12 +513,12 @@ do_func(Elf_Ehdr *const ehdr, char const *const fname, unsigned const reltype)
 			mlocp = sift_rel_mcount(mlocp,
 				(void *)mlocp - (void *)mloc0, &mrelp,
 				relhdr, ehdr, recsym, recval, reltype);
-		} else if (make_nop && txtname) {
+		} else if (txtname && (warn_on_notrace_sect || make_nop)) {
 			/*
 			 * This section is ignored by ftrace, but still
 			 * has mcount calls. Convert them to nops now.
 			 */
-			nop_mcount(relhdr, ehdr);
+			nop_mcount(relhdr, ehdr, txtname);
 		}
 	}
 	if (mloc0 != mlocp) {

From a3a8350ab2f588f3a7a08dc86658bf90773f9a52 Mon Sep 17 00:00:00 2001
From: Steven Rostedt <srostedt@redhat.com>
Date: Tue, 12 Apr 2011 18:59:10 -0400
Subject: [PATCH 08/15] kbuild/recordmcount: Add RECORDMCOUNT_WARN to warn
 about mcount callers

When mcount is called in a section that ftrace will not modify it into
a nop, we want to warn about this. But not warn about this always. Now
if the user builds the kernel with the option RECORDMCOUNT_WARN=1 then
the build will warn about mcount callers that are ignored and will just
waste execution time.

Acked-by: Michal Marek <mmarek@suse.cz>
Cc: linux-kbuild@vger.kernel.org
Link: http://lkml.kernel.org/r/20110421023738.714956282@goodmis.org
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
 Makefile               | 1 +
 scripts/Makefile.build | 5 ++++-
 2 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/Makefile b/Makefile
index 41ea6fbec55a..e7d01adaf692 100644
--- a/Makefile
+++ b/Makefile
@@ -1268,6 +1268,7 @@ help:
 	@echo  '  make C=1   [targets] Check all c source with $$CHECK (sparse by default)'
 	@echo  '  make C=2   [targets] Force check of all c source with $$CHECK'
 	@echo  '  make W=1   [targets] Enable extra gcc checks'
+	@echo  '  make RECORDMCOUNT_WARN=1 [targets] Warn about ignored mcount sections'
 	@echo  ''
 	@echo  'Execute "make" or "make all" to build all targets marked with [*] '
 	@echo  'For further info see the ./README file'
diff --git a/scripts/Makefile.build b/scripts/Makefile.build
index 7d3f9036eee2..6165622c3e29 100644
--- a/scripts/Makefile.build
+++ b/scripts/Makefile.build
@@ -244,13 +244,16 @@ endif
 
 ifdef CONFIG_FTRACE_MCOUNT_RECORD
 ifdef BUILD_C_RECORDMCOUNT
+ifeq ("$(origin RECORDMCOUNT_WARN)", "command line")
+  RECORDMCOUNT_FLAGS = -w
+endif
 # Due to recursion, we must skip empty.o.
 # The empty.o file is created in the make process in order to determine
 #  the target endianness and word size. It is made before all other C
 #  files, including recordmcount.
 sub_cmd_record_mcount =					\
 	if [ $(@) != "scripts/mod/empty.o" ]; then	\
-		$(objtree)/scripts/recordmcount "$(@)";	\
+		$(objtree)/scripts/recordmcount $(RECORDMCOUNT_FLAGS) "$(@)";	\
 	fi;
 recordmcount_source := $(srctree)/scripts/recordmcount.c \
 		    $(srctree)/scripts/recordmcount.h

From 5f9b93c39239507d2af03f79c8b0df9ebb9be768 Mon Sep 17 00:00:00 2001
From: Steven Rostedt <srostedt@redhat.com>
Date: Tue, 12 Apr 2011 19:06:39 -0400
Subject: [PATCH 09/15] ftrace: Avoid recording mcount on .init sections
 directly

The init and exit sections should not be traced and adding a call to
mcount to them is a waste of text and instruction cache. Have the
macro section attributes include notrace to ignore these functions
for tracing from the build.

Link: http://lkml.kernel.org/r/20110421023738.953028219@goodmis.org
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
 include/linux/init.h | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/include/linux/init.h b/include/linux/init.h
index 577671c55153..9146f39cdddf 100644
--- a/include/linux/init.h
+++ b/include/linux/init.h
@@ -79,29 +79,29 @@
 #define __exitused  __used
 #endif
 
-#define __exit          __section(.exit.text) __exitused __cold
+#define __exit          __section(.exit.text) __exitused __cold notrace
 
 /* Used for HOTPLUG */
-#define __devinit        __section(.devinit.text) __cold
+#define __devinit        __section(.devinit.text) __cold notrace
 #define __devinitdata    __section(.devinit.data)
 #define __devinitconst   __section(.devinit.rodata)
-#define __devexit        __section(.devexit.text) __exitused __cold
+#define __devexit        __section(.devexit.text) __exitused __cold notrace
 #define __devexitdata    __section(.devexit.data)
 #define __devexitconst   __section(.devexit.rodata)
 
 /* Used for HOTPLUG_CPU */
-#define __cpuinit        __section(.cpuinit.text) __cold
+#define __cpuinit        __section(.cpuinit.text) __cold notrace
 #define __cpuinitdata    __section(.cpuinit.data)
 #define __cpuinitconst   __section(.cpuinit.rodata)
-#define __cpuexit        __section(.cpuexit.text) __exitused __cold
+#define __cpuexit        __section(.cpuexit.text) __exitused __cold notrace
 #define __cpuexitdata    __section(.cpuexit.data)
 #define __cpuexitconst   __section(.cpuexit.rodata)
 
 /* Used for MEMORY_HOTPLUG */
-#define __meminit        __section(.meminit.text) __cold
+#define __meminit        __section(.meminit.text) __cold notrace
 #define __meminitdata    __section(.meminit.data)
 #define __meminitconst   __section(.meminit.rodata)
-#define __memexit        __section(.memexit.text) __exitused __cold
+#define __memexit        __section(.memexit.text) __exitused __cold notrace
 #define __memexitdata    __section(.memexit.data)
 #define __memexitconst   __section(.memexit.rodata)
 

From 406a733db0052fed66d17008430b6047ecea7b92 Mon Sep 17 00:00:00 2001
From: Steven Rostedt <srostedt@redhat.com>
Date: Wed, 13 Apr 2011 16:43:29 -0400
Subject: [PATCH 10/15] ftrace/x86: Do not trace .discard.text section

The section called .discard.text has tracing attached to it and is
currently ignored by ftrace. But it does include a call to the mcount
stub. Adding a notrace to the code keeps gcc from adding the useless
mcount caller to it.

Link: http://lkml.kernel.org/r/20110421023739.243651696@goodmis.org
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
 arch/x86/include/asm/setup.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86/include/asm/setup.h b/arch/x86/include/asm/setup.h
index db8aa19a08a2..647d8a06ce4f 100644
--- a/arch/x86/include/asm/setup.h
+++ b/arch/x86/include/asm/setup.h
@@ -88,7 +88,7 @@ void *extend_brk(size_t size, size_t align);
  * executable.)
  */
 #define RESERVE_BRK(name,sz)						\
-	static void __section(.discard.text) __used			\
+	static void __section(.discard.text) __used notrace		\
 	__brk_reservation_fn_##name##__(void) {				\
 		asm volatile (						\
 			".pushsection .brk_reservation,\"aw\",@nobits;" \

From f6d4f08194d77c86fe3af6e2e4b9184347014926 Mon Sep 17 00:00:00 2001
From: Steven Rostedt <srostedt@redhat.com>
Date: Wed, 20 Apr 2011 20:47:34 -0400
Subject: [PATCH 11/15] ftrace/recordmcount: Remove duplicate code to find
 mcount symbol

The code in sift_rel_mcount() and nop_mcount() to get the mcount symbol
number is identical. Replace the two locations with a call to a function
that does the work.

Cc: John Reiser <jreiser@bitwagon.com>
Link: http://lkml.kernel.org/r/20110421023739.488093407@goodmis.org
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
 scripts/recordmcount.h | 51 ++++++++++++++++++++++--------------------
 1 file changed, 27 insertions(+), 24 deletions(-)

diff --git a/scripts/recordmcount.h b/scripts/recordmcount.h
index 22033d563bcd..deb6a517638c 100644
--- a/scripts/recordmcount.h
+++ b/scripts/recordmcount.h
@@ -28,6 +28,7 @@
 #undef __has_rel_mcount
 #undef has_rel_mcount
 #undef tot_relsize
+#undef get_mcountsym
 #undef do_func
 #undef Elf_Addr
 #undef Elf_Ehdr
@@ -56,6 +57,7 @@
 # define has_rel_mcount		has64_rel_mcount
 # define tot_relsize		tot64_relsize
 # define do_func		do64
+# define get_mcountsym		get_mcountsym_64
 # define is_fake_mcount		is_fake_mcount64
 # define fn_is_fake_mcount	fn_is_fake_mcount64
 # define MIPS_is_fake_mcount	MIPS64_is_fake_mcount
@@ -85,6 +87,7 @@
 # define has_rel_mcount		has32_rel_mcount
 # define tot_relsize		tot32_relsize
 # define do_func		do32
+# define get_mcountsym		get_mcountsym_32
 # define is_fake_mcount		is_fake_mcount32
 # define fn_is_fake_mcount	fn_is_fake_mcount32
 # define MIPS_is_fake_mcount	MIPS32_is_fake_mcount
@@ -237,6 +240,26 @@ static void append_func(Elf_Ehdr *const ehdr,
 	uwrite(fd_map, ehdr, sizeof(*ehdr));
 }
 
+static unsigned get_mcountsym(Elf_Sym const *const sym0,
+			      Elf_Rel const *relp,
+			      char const *const str0)
+{
+	unsigned mcountsym = 0;
+
+	Elf_Sym const *const symp =
+		&sym0[Elf_r_sym(relp)];
+	char const *symname = &str0[w(symp->st_name)];
+	char const *mcount = gpfx == '_' ? "_mcount" : "mcount";
+
+	if (symname[0] == '.')
+		++symname;  /* ppc64 hack */
+	if (strcmp(mcount, symname) == 0 ||
+	    (altmcount && strcmp(altmcount, symname) == 0))
+		mcountsym = Elf_r_sym(relp);
+
+	return mcountsym;
+}
+
 /*
  * Look at the relocations in order to find the calls to mcount.
  * Accumulate the section offsets that are found, and their relocation info,
@@ -274,18 +297,8 @@ static uint_t *sift_rel_mcount(uint_t *mlocp,
 	unsigned t;
 
 	for (t = nrel; t; --t) {
-		if (!mcountsym) {
-			Elf_Sym const *const symp =
-				&sym0[Elf_r_sym(relp)];
-			char const *symname = &str0[w(symp->st_name)];
-			char const *mcount = gpfx == '_' ? "_mcount" : "mcount";
-
-			if (symname[0] == '.')
-				++symname;  /* ppc64 hack */
-			if (strcmp(mcount, symname) == 0 ||
-			    (altmcount && strcmp(altmcount, symname) == 0))
-				mcountsym = Elf_r_sym(relp);
-		}
+		if (!mcountsym)
+			mcountsym = get_mcountsym(sym0, relp, str0);
 
 		if (mcountsym == Elf_r_sym(relp) && !is_fake_mcount(relp)) {
 			uint_t const addend = _w(_w(relp->r_offset) - recval);
@@ -342,18 +355,8 @@ static void nop_mcount(Elf_Shdr const *const relhdr,
 	for (t = nrel; t; --t) {
 		int ret = -1;
 
-		if (!mcountsym) {
-			Elf_Sym const *const symp =
-				&sym0[Elf_r_sym(relp)];
-			char const *symname = &str0[w(symp->st_name)];
-			char const *mcount = gpfx == '_' ? "_mcount" : "mcount";
-
-			if (symname[0] == '.')
-				++symname;  /* ppc64 hack */
-			if (strcmp(mcount, symname) == 0 ||
-			    (altmcount && strcmp(altmcount, symname) == 0))
-				mcountsym = Elf_r_sym(relp);
-		}
+		if (!mcountsym)
+			mcountsym = get_mcountsym(sym0, relp, str0);
 
 		if (mcountsym == Elf_r_sym(relp) && !is_fake_mcount(relp)) {
 			if (make_nop)

From 0286d2d12d78ea8aaf0722473dd20cb609a99709 Mon Sep 17 00:00:00 2001
From: Steven Rostedt <srostedt@redhat.com>
Date: Wed, 20 Apr 2011 21:13:06 -0400
Subject: [PATCH 12/15] ftrace/recordmcount: Add helper function
 get_sym_str_and_relp()

The code to get the symbol, string, and relp pointers in the two functions
sift_rel_mcount() and nop_mcount() are identical and also non-trivial.
Moving this duplicate code into a single helper function makes the code
easier to read and more maintainable.

Cc: John Reiser <jreiser@bitwagon.com>
Link: http://lkml.kernel.org/r/20110421023739.723658553@goodmis.org
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
 scripts/recordmcount.h | 67 +++++++++++++++++++++++-------------------
 1 file changed, 37 insertions(+), 30 deletions(-)

diff --git a/scripts/recordmcount.h b/scripts/recordmcount.h
index deb6a517638c..3c00faba745f 100644
--- a/scripts/recordmcount.h
+++ b/scripts/recordmcount.h
@@ -29,6 +29,7 @@
 #undef has_rel_mcount
 #undef tot_relsize
 #undef get_mcountsym
+#undef get_sym_str_and_relp
 #undef do_func
 #undef Elf_Addr
 #undef Elf_Ehdr
@@ -56,6 +57,7 @@
 # define __has_rel_mcount	__has64_rel_mcount
 # define has_rel_mcount		has64_rel_mcount
 # define tot_relsize		tot64_relsize
+# define get_sym_str_and_relp	get_sym_str_and_relp_64
 # define do_func		do64
 # define get_mcountsym		get_mcountsym_64
 # define is_fake_mcount		is_fake_mcount64
@@ -86,6 +88,7 @@
 # define __has_rel_mcount	__has32_rel_mcount
 # define has_rel_mcount		has32_rel_mcount
 # define tot_relsize		tot32_relsize
+# define get_sym_str_and_relp	get_sym_str_and_relp_32
 # define do_func		do32
 # define get_mcountsym		get_mcountsym_32
 # define is_fake_mcount		is_fake_mcount32
@@ -260,6 +263,29 @@ static unsigned get_mcountsym(Elf_Sym const *const sym0,
 	return mcountsym;
 }
 
+static void get_sym_str_and_relp(Elf_Shdr const *const relhdr,
+				 Elf_Ehdr const *const ehdr,
+				 Elf_Sym const **sym0,
+				 char const **str0,
+				 Elf_Rel const **relp)
+{
+	Elf_Shdr *const shdr0 = (Elf_Shdr *)(_w(ehdr->e_shoff)
+		+ (void *)ehdr);
+	unsigned const symsec_sh_link = w(relhdr->sh_link);
+	Elf_Shdr const *const symsec = &shdr0[symsec_sh_link];
+	Elf_Shdr const *const strsec = &shdr0[w(symsec->sh_link)];
+	Elf_Rel const *const rel0 = (Elf_Rel const *)(_w(relhdr->sh_offset)
+		+ (void *)ehdr);
+
+	*sym0 = (Elf_Sym const *)(_w(symsec->sh_offset)
+				  + (void *)ehdr);
+
+	*str0 = (char const *)(_w(strsec->sh_offset)
+			       + (void *)ehdr);
+
+	*relp = rel0;
+}
+
 /*
  * Look at the relocations in order to find the calls to mcount.
  * Accumulate the section offsets that are found, and their relocation info,
@@ -276,26 +302,16 @@ static uint_t *sift_rel_mcount(uint_t *mlocp,
 {
 	uint_t *const mloc0 = mlocp;
 	Elf_Rel *mrelp = *mrelpp;
-	Elf_Shdr *const shdr0 = (Elf_Shdr *)(_w(ehdr->e_shoff)
-		+ (void *)ehdr);
-	unsigned const symsec_sh_link = w(relhdr->sh_link);
-	Elf_Shdr const *const symsec = &shdr0[symsec_sh_link];
-	Elf_Sym const *const sym0 = (Elf_Sym const *)(_w(symsec->sh_offset)
-		+ (void *)ehdr);
-
-	Elf_Shdr const *const strsec = &shdr0[w(symsec->sh_link)];
-	char const *const str0 = (char const *)(_w(strsec->sh_offset)
-		+ (void *)ehdr);
-
-	Elf_Rel const *const rel0 = (Elf_Rel const *)(_w(relhdr->sh_offset)
-		+ (void *)ehdr);
+	Elf_Sym const *sym0;
+	char const *str0;
+	Elf_Rel const *relp;
 	unsigned rel_entsize = _w(relhdr->sh_entsize);
 	unsigned const nrel = _w(relhdr->sh_size) / rel_entsize;
-	Elf_Rel const *relp = rel0;
-
 	unsigned mcountsym = 0;
 	unsigned t;
 
+	get_sym_str_and_relp(relhdr, ehdr, &sym0, &str0, &relp);
+
 	for (t = nrel; t; --t) {
 		if (!mcountsym)
 			mcountsym = get_mcountsym(sym0, relp, str0);
@@ -331,27 +347,18 @@ static void nop_mcount(Elf_Shdr const *const relhdr,
 {
 	Elf_Shdr *const shdr0 = (Elf_Shdr *)(_w(ehdr->e_shoff)
 		+ (void *)ehdr);
-	unsigned const symsec_sh_link = w(relhdr->sh_link);
-	Elf_Shdr const *const symsec = &shdr0[symsec_sh_link];
-	Elf_Sym const *const sym0 = (Elf_Sym const *)(_w(symsec->sh_offset)
-		+ (void *)ehdr);
-
-	Elf_Shdr const *const strsec = &shdr0[w(symsec->sh_link)];
-	char const *const str0 = (char const *)(_w(strsec->sh_offset)
-		+ (void *)ehdr);
-
-	Elf_Rel const *const rel0 = (Elf_Rel const *)(_w(relhdr->sh_offset)
-		+ (void *)ehdr);
+	Elf_Sym const *sym0;
+	char const *str0;
+	Elf_Rel const *relp;
+	Elf_Shdr const *const shdr = &shdr0[w(relhdr->sh_info)];
 	unsigned rel_entsize = _w(relhdr->sh_entsize);
 	unsigned const nrel = _w(relhdr->sh_size) / rel_entsize;
-	Elf_Rel const *relp = rel0;
-
-	Elf_Shdr const *const shdr = &shdr0[w(relhdr->sh_info)];
-
 	unsigned mcountsym = 0;
 	unsigned t;
 	int once = 0;
 
+	get_sym_str_and_relp(relhdr, ehdr, &sym0, &str0, &relp);
+
 	for (t = nrel; t; --t) {
 		int ret = -1;
 

From f215efc50a3c1430735e89915fae188f927d8773 Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
Date: Tue, 10 May 2011 10:10:40 +0200
Subject: [PATCH 13/15] ftrace/recordmcount: mcount address adjustment

Introduce mcount_adjust{,_32,_64} to the C implementation of
recordmcount analog to $mcount_adjust in the perl script.
The adjustment is added to the address of the relocations
against the mcount symbol. If this adjustment is done by
recordmcount at compile time the ftrace_call_adjust function
can be turned into a nop.

Cc: John Reiser <jreiser@bitwagon.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
 scripts/recordmcount.h | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/scripts/recordmcount.h b/scripts/recordmcount.h
index 3c00faba745f..4be60364a405 100644
--- a/scripts/recordmcount.h
+++ b/scripts/recordmcount.h
@@ -22,6 +22,7 @@
 #undef is_fake_mcount
 #undef fn_is_fake_mcount
 #undef MIPS_is_fake_mcount
+#undef mcount_adjust
 #undef sift_rel_mcount
 #undef nop_mcount
 #undef find_secsym_ndx
@@ -63,6 +64,7 @@
 # define is_fake_mcount		is_fake_mcount64
 # define fn_is_fake_mcount	fn_is_fake_mcount64
 # define MIPS_is_fake_mcount	MIPS64_is_fake_mcount
+# define mcount_adjust		mcount_adjust_64
 # define Elf_Addr		Elf64_Addr
 # define Elf_Ehdr		Elf64_Ehdr
 # define Elf_Shdr		Elf64_Shdr
@@ -94,6 +96,7 @@
 # define is_fake_mcount		is_fake_mcount32
 # define fn_is_fake_mcount	fn_is_fake_mcount32
 # define MIPS_is_fake_mcount	MIPS32_is_fake_mcount
+# define mcount_adjust		mcount_adjust_32
 # define Elf_Addr		Elf32_Addr
 # define Elf_Ehdr		Elf32_Ehdr
 # define Elf_Shdr		Elf32_Shdr
@@ -132,6 +135,8 @@ static void fn_ELF_R_INFO(Elf_Rel *const rp, unsigned sym, unsigned type)
 }
 static void (*Elf_r_info)(Elf_Rel *const rp, unsigned sym, unsigned type) = fn_ELF_R_INFO;
 
+static int mcount_adjust = 0;
+
 /*
  * MIPS mcount long call has 2 _mcount symbols, only the position of the 1st
  * _mcount symbol is needed for dynamic function tracer, with it, to disable
@@ -317,8 +322,8 @@ static uint_t *sift_rel_mcount(uint_t *mlocp,
 			mcountsym = get_mcountsym(sym0, relp, str0);
 
 		if (mcountsym == Elf_r_sym(relp) && !is_fake_mcount(relp)) {
-			uint_t const addend = _w(_w(relp->r_offset) - recval);
-
+			uint_t const addend =
+				_w(_w(relp->r_offset) - recval + mcount_adjust);
 			mrelp->r_offset = _w(offbase
 				+ ((void *)mlocp - (void *)mloc0));
 			Elf_r_info(mrelp, recsym, reltype);

From 414eb44dda6b9893b212b2306ccb423ab01eef8e Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
Date: Tue, 10 May 2011 10:10:41 +0200
Subject: [PATCH 14/15] ftrace/x86: mcount offset calculation

Do the mcount offset adjustment in the recordmcount.pl/recordmcount.[ch]
at compile time and not in ftrace_call_adjust at run time.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
 arch/x86/include/asm/ftrace.h | 7 +++----
 scripts/recordmcount.c        | 2 ++
 scripts/recordmcount.pl       | 2 ++
 3 files changed, 7 insertions(+), 4 deletions(-)

diff --git a/arch/x86/include/asm/ftrace.h b/arch/x86/include/asm/ftrace.h
index db24c2278be0..268c783ab1c0 100644
--- a/arch/x86/include/asm/ftrace.h
+++ b/arch/x86/include/asm/ftrace.h
@@ -38,11 +38,10 @@ extern void mcount(void);
 static inline unsigned long ftrace_call_adjust(unsigned long addr)
 {
 	/*
-	 * call mcount is "e8 <4 byte offset>"
-	 * The addr points to the 4 byte offset and the caller of this
-	 * function wants the pointer to e8. Simply subtract one.
+	 * addr is the address of the mcount call instruction.
+	 * recordmcount does the necessary offset calculation.
 	 */
-	return addr - 1;
+	return addr;
 }
 
 #ifdef CONFIG_DYNAMIC_FTRACE
diff --git a/scripts/recordmcount.c b/scripts/recordmcount.c
index 0e18975824f7..7648a5d11154 100644
--- a/scripts/recordmcount.c
+++ b/scripts/recordmcount.c
@@ -335,6 +335,7 @@ do_file(char const *const fname)
 		reltype = R_386_32;
 		make_nop = make_nop_x86;
 		ideal_nop = ideal_nop5_x86_32;
+		mcount_adjust_32 = -1;
 		break;
 	case EM_ARM:	 reltype = R_ARM_ABS32;
 			 altmcount = "__gnu_mcount_nc";
@@ -350,6 +351,7 @@ do_file(char const *const fname)
 		make_nop = make_nop_x86;
 		ideal_nop = ideal_nop5_x86_64;
 		reltype = R_X86_64_64;
+		mcount_adjust_64 = -1;
 		break;
 	}  /* end switch */
 
diff --git a/scripts/recordmcount.pl b/scripts/recordmcount.pl
index a871cd414055..414e7f5e42ec 100755
--- a/scripts/recordmcount.pl
+++ b/scripts/recordmcount.pl
@@ -223,6 +223,7 @@ if ($arch eq "x86_64") {
     $mcount_regex = "^\\s*([0-9a-fA-F]+):.*\\smcount([+-]0x[0-9a-zA-Z]+)?\$";
     $type = ".quad";
     $alignment = 8;
+    $mcount_adjust = -1;
 
     # force flags for this arch
     $ld .= " -m elf_x86_64";
@@ -232,6 +233,7 @@ if ($arch eq "x86_64") {
 
 } elsif ($arch eq "i386") {
     $alignment = 4;
+    $mcount_adjust = -1;
 
     # force flags for this arch
     $ld .= " -m elf_i386";

From 0ef5bf1cd3bd2e3eacb11fca80efe3674927ed1a Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
Date: Tue, 10 May 2011 10:10:43 +0200
Subject: [PATCH 15/15] ftrace/s390: mcount offset calculation

Do the mcount offset adjustment in the recordmcount.pl/recordmcount.[ch]
at compile time and not in ftrace_call_adjust at run time.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
 arch/s390/include/asm/ftrace.h | 4 +---
 scripts/recordmcount.c         | 8 ++++++--
 scripts/recordmcount.pl        | 2 ++
 3 files changed, 9 insertions(+), 5 deletions(-)

diff --git a/arch/s390/include/asm/ftrace.h b/arch/s390/include/asm/ftrace.h
index 3c29be4836ed..b7931faaef6d 100644
--- a/arch/s390/include/asm/ftrace.h
+++ b/arch/s390/include/asm/ftrace.h
@@ -11,15 +11,13 @@ struct dyn_arch_ftrace { };
 
 #ifdef CONFIG_64BIT
 #define MCOUNT_INSN_SIZE  12
-#define MCOUNT_OFFSET	   8
 #else
 #define MCOUNT_INSN_SIZE  20
-#define MCOUNT_OFFSET	   4
 #endif
 
 static inline unsigned long ftrace_call_adjust(unsigned long addr)
 {
-	return addr - MCOUNT_OFFSET;
+	return addr;
 }
 
 #endif /* __ASSEMBLY__ */
diff --git a/scripts/recordmcount.c b/scripts/recordmcount.c
index 7648a5d11154..ee52cb8e17ad 100644
--- a/scripts/recordmcount.c
+++ b/scripts/recordmcount.c
@@ -368,8 +368,10 @@ do_file(char const *const fname)
 				"unrecognized ET_REL file: %s\n", fname);
 			fail_file();
 		}
-		if (w2(ehdr->e_machine) == EM_S390)
+		if (w2(ehdr->e_machine) == EM_S390) {
 			reltype = R_390_32;
+			mcount_adjust_32 = -4;
+		}
 		if (w2(ehdr->e_machine) == EM_MIPS) {
 			reltype = R_MIPS_32;
 			is_fake_mcount32 = MIPS32_is_fake_mcount;
@@ -384,8 +386,10 @@ do_file(char const *const fname)
 				"unrecognized ET_REL file: %s\n", fname);
 			fail_file();
 		}
-		if (w2(ghdr->e_machine) == EM_S390)
+		if (w2(ghdr->e_machine) == EM_S390) {
 			reltype = R_390_64;
+			mcount_adjust_64 = -8;
+		}
 		if (w2(ghdr->e_machine) == EM_MIPS) {
 			reltype = R_MIPS_64;
 			Elf64_r_sym = MIPS64_r_sym;
diff --git a/scripts/recordmcount.pl b/scripts/recordmcount.pl
index 414e7f5e42ec..858966ab019c 100755
--- a/scripts/recordmcount.pl
+++ b/scripts/recordmcount.pl
@@ -243,12 +243,14 @@ if ($arch eq "x86_64") {
 
 } elsif ($arch eq "s390" && $bits == 32) {
     $mcount_regex = "^\\s*([0-9a-fA-F]+):\\s*R_390_32\\s+_mcount\$";
+    $mcount_adjust = -4;
     $alignment = 4;
     $ld .= " -m elf_s390";
     $cc .= " -m31";
 
 } elsif ($arch eq "s390" && $bits == 64) {
     $mcount_regex = "^\\s*([0-9a-fA-F]+):\\s*R_390_(PC|PLT)32DBL\\s+_mcount\\+0x2\$";
+    $mcount_adjust = -8;
     $alignment = 8;
     $type = ".quad";
     $ld .= " -m elf64_s390";