Skip to contentSkip to conversation

: sparc64: Add ftrace support.

PATCH Applied David Miller 2 participants 5 messages May 14, 2008 ftrace

lore
Share & tools
lore
#0 lore raw

PATCH 2/2 : sparc64: Add ftrace support.

---
 arch/sparc64/Kconfig         |    1 +
 arch/sparc64/Kconfig.debug   |    2 +-
 arch/sparc64/kernel/Makefile |    1 +
 arch/sparc64/kernel/ftrace.c |   99 ++++++++++++++++++++++++++++++++++++++++++
 arch/sparc64/lib/mcount.S    |   58 +++++++++++++++++++++++--
 5 files changed, 156 insertions(+), 5 deletions(-)
 create mode 100644 arch/sparc64/kernel/ftrace.c
arch/sparc64/Kconfig +1 -0
index 4c40862..56344b1 100644
@@ -15,6 +15,7 @@ config SPARC64
 	select HAVE_LMB
 	select HAVE_ARCH_KGDB
 	select HAVE_IMMEDIATE
+	select HAVE_FTRACE
 
 config GENERIC_TIME
 	bool
arch/sparc64/Kconfig.debug +1 -1
index 6a4d28a..d6d32d1 100644
@@ -33,7 +33,7 @@ config DEBUG_PAGEALLOC
 
 config MCOUNT
 	bool
-	depends on STACK_DEBUG
+	depends on STACK_DEBUG || FTRACE
 	default y
 
 config FRAME_POINTER
arch/sparc64/kernel/Makefile +1 -0
index 311c797..52e933f 100644
@@ -14,6 +14,7 @@ obj-y		:= process.o setup.o cpu.o idprom.o \
 		   power.o sbus.o sparc64_ksyms.o chmc.o \
 		   visemul.o prom.o of_device.o hvapi.o sstate.o mdesc.o
 
+obj-$(CONFIG_DYNAMIC_FTRACE) += ftrace.o
 obj-$(CONFIG_STACKTRACE) += stacktrace.o
 obj-$(CONFIG_PCI)	 += ebus.o pci_common.o \
 			    pci_psycho.o pci_sabre.o pci_schizo.o \
arch/sparc64/kernel/ftrace.c new file +99 -0

99 lines not on this page · raw message

arch/sparc64/lib/mcount.S +54 -4

70 lines not on this page · raw message

Signature · 1 line
1.5.5.rc0.16.g02b00
Acked-by on v1 2/2
#1 lore raw
Quoted 53 lines · David Miller
---
 arch/sparc64/Kconfig         |    1 +
 arch/sparc64/Kconfig.debug   |    2 +-
 arch/sparc64/kernel/Makefile |    1 +
 arch/sparc64/kernel/ftrace.c |   99 ++++++++++++++++++++++++++++++++++++++++++
 arch/sparc64/lib/mcount.S    |   58 +++++++++++++++++++++++--
 5 files changed, 156 insertions(+), 5 deletions(-)
 create mode 100644 arch/sparc64/kernel/ftrace.c
arch/sparc64/Kconfig +1 -0
index 4c40862..56344b1 100644
@@ -15,6 +15,7 @@ config SPARC64
 	select HAVE_LMB
 	select HAVE_ARCH_KGDB
 	select HAVE_IMMEDIATE
+	select HAVE_FTRACE
 
 config GENERIC_TIME
 	bool
arch/sparc64/Kconfig.debug +1 -1
index 6a4d28a..d6d32d1 100644
@@ -33,7 +33,7 @@ config DEBUG_PAGEALLOC
 
 config MCOUNT
 	bool
-	depends on STACK_DEBUG
+	depends on STACK_DEBUG || FTRACE
 	default y
 
 config FRAME_POINTER
arch/sparc64/kernel/Makefile +1 -0
index 311c797..52e933f 100644
@@ -14,6 +14,7 @@ obj-y		:= process.o setup.o cpu.o idprom.o \
 		   power.o sbus.o sparc64_ksyms.o chmc.o \
 		   visemul.o prom.o of_device.o hvapi.o sstate.o mdesc.o
 
+obj-$(CONFIG_DYNAMIC_FTRACE) += ftrace.o
 obj-$(CONFIG_STACKTRACE) += stacktrace.o
 obj-$(CONFIG_PCI)	 += ebus.o pci_common.o \
 			    pci_psycho.o pci_sabre.o pci_schizo.o \
arch/sparc64/kernel/ftrace.c new file +0 -0
new file mode 100644
index 0000000..f449e6d

arch/sparc64/kernel/ftrace.c line 10

Quoted 10 lines
arch/sparc64/kernel/ftrace.c new file +10 -0
@@ -0,0 +1,99 @@
+#include <linux/spinlock.h>
+#include <linux/hardirq.h>
+#include <linux/ftrace.h>
+#include <linux/percpu.h>
+#include <linux/init.h>
+#include <linux/list.h>
+
+static const u32 ftrace_nop = 0x01000000;
+
+notrace int ftrace_ip_converted(unsigned long ip)

FYI,

If you look in the kernel/Makefile you will see a way to have the whole file not be traced. This will eliminate the need for the "notrace" annotation on all functions. I haven't update the x86 version of ftrace to do this so you probably just copied the old method from us.

If you add the following in arch/sparc64/kernel/Makefile

ifdef CONFIG_DYNAMIC_FTRACE
obj-y += ftrace.o
# Do not trace the ftrace code itself
ORIG_CFLAGS := $(KBUILD_CFLAGS)

KBUILD_CFLAGS = $(if $(filter-out ftrace, $(basename $(notdir $@))), \
	$(ORIG_CFLAGS), \
	$(subst -pg,,$(ORIG_CFLAGS)))

endif

Then you wont need the "notrace" annotations on the functions.

/me needs to send Ingo a patch to do this for x86 as well.

Quoted 39 lines
arch/sparc64/kernel/ftrace.c new file +39 -0
+{
+	u32 insn = *(u32 *) ip;
+
+	return (insn == ftrace_nop);
+}
+
+notrace unsigned char *ftrace_nop_replace(void)
+{
+	return (char *)&ftrace_nop;
+}
+
+notrace unsigned char *ftrace_call_replace(unsigned long ip, unsigned long addr)
+{
+	static u32 call;
+	s32 off;
+
+	off = ((s32)addr - (s32)ip);
+	call = 0x40000000 | ((u32)off >> 2);
+
+	return (unsigned char *) &call;
+}
+
+notrace int
+ftrace_modify_code(unsigned long ip, unsigned char *old_code,
+		   unsigned char *new_code)
+{
+	u32 old = *(u32 *)old_code;
+	u32 new = *(u32 *)new_code;
+	u32 replaced;
+	int faulted;
+
+	__asm__ __volatile__(
+	"1:	cas	[%[ip]], %[old], %[new]\n"
+	"	flush	%[ip]\n"
+	"	mov	0, %[faulted]\n"
+	"2:\n"
+	"	.section .fixup,#alloc,#execinstr\n"
+	"	.align	4\n"
+	"3:	sethi	%%hi(2b), %[faulted]\n"

arch/sparc64/kernel/ftrace.c line 59

Quoted 10 lines
arch/sparc64/kernel/ftrace.c new file +10 -0
+	"	jmpl	%[faulted] + %%lo(2b), %%g0\n"
+	"	 mov	1, %[faulted]\n"
+	"	.previous\n"
+	"	.section __ex_table,\"a\"\n"
+	"	.align	4\n"
+	"	.word	1b, 3b\n"
+	"	.previous\n"
+	: "=r" (replaced), [faulted] "=r" (faulted)
+	: [new] "0" (new), [old] "r" (old), [ip] "r" (ip)
+	: "memory");

I'm so use to the %1 %2 and %3's I should update the x86 code to use the names as well. (nice)

arch/sparc64/kernel/ftrace.c line 62

Quoted 3 lines
arch/sparc64/kernel/ftrace.c new file +3 -0
+
+	if (replaced != old && replaced != new)
+		faulted = 2;

Thanks for putting in the "2", I may plan on doing something different later if the code didn't "fault" but something else happened. I should also make MACROS out of that and not uses 1 and 2.

Quoted 61 lines
arch/sparc64/kernel/ftrace.c new file +37 -0
+
+	return faulted;
+}
+
+notrace int ftrace_update_ftrace_func(ftrace_func_t func)
+{
+	unsigned long ip = (unsigned long)(&ftrace_call);
+	unsigned char old[4], *new;
+
+	memcpy(old, &ftrace_call, 4);
+	new = ftrace_call_replace(ip, (unsigned long)func);
+	return ftrace_modify_code(ip, old, new);
+}
+
+notrace int ftrace_mcount_set(unsigned long *data)
+{
+	unsigned long ip = (long)(&mcount_call);
+	unsigned long *addr = data;
+	unsigned char old[4], *new;
+
+	/*
+	 * Replace the mcount stub with a pointer to the
+	 * ip recorder function.
+	 */
+	memcpy(old, &mcount_call, 4);
+	new = ftrace_call_replace(ip, *addr);
+	*addr = ftrace_modify_code(ip, old, new);
+
+	return 0;
+}
+
+
+int __init ftrace_dyn_arch_init(void *data)
+{
+	ftrace_mcount_set(data);
+	return 0;
+}
arch/sparc64/lib/mcount.S +6 -3
index 9e4534b..7735a7a 100644
@@ -28,10 +28,13 @@ ovstack:
 	.skip		OVSTACKSIZE
 #endif
 	.text
-	.align 32
-	.globl mcount, _mcount
-mcount:
+	.align		32
+	.globl		_mcount
+	.type		_mcount,#function
+	.globl		mcount
+	.type		mcount,#function
 _mcount:
+mcount:
 #ifdef CONFIG_STACK_DEBUG
 	/*
 	 * Check whether %sp is dangerously low.
@@ -55,6 +58,53 @@ _mcount:
 	 or		%g3, %lo(panicstring), %o0
 	call		prom_halt

arch/sparc64/lib/mcount.S _mcount: · line 69

Quoted 10 lines
arch/sparc64/lib/mcount.S +9 -0
 	 nop
+1:
+#endif
+#ifdef CONFIG_FTRACE
+#ifdef CONFIG_DYNAMIC_FTRACE
+	mov		%o7, %o0
+	.globl		mcount_call
+mcount_call:
+	call		ftrace_stub
+	 mov		%o0, %o7

WOW! Sparc64 has some really efficient context saving!

Quoted 32 lines
arch/sparc64/lib/mcount.S +29 -1
+#else
+	sethi		%hi(ftrace_trace_function), %g1
+	sethi		%hi(ftrace_stub), %g2
+	ldx		[%g1 + %lo(ftrace_trace_function)], %g1
+	or		%g2, %lo(ftrace_stub), %g2
+	cmp		%g1, %g2
+	be,pn		%icc, 1f
+	 mov		%i7, %o1
+	jmpl		%g1, %g0
+	 mov		%o7, %o0
+	/* not reached */
+1:
 #endif
-1:	retl
+#endif
+	retl
 	 nop
+	.size		_mcount,.-_mcount
+	.size		mcount,.-mcount
+
+#ifdef CONFIG_FTRACE
+	.globl		ftrace_stub
+	.type		ftrace_stub,#function
+ftrace_stub:
+	retl
+	 nop
+	.size		ftrace_stub,.-ftrace_stub
+#ifdef CONFIG_DYNAMIC_FTRACE
+	.globl		ftrace_caller
+	.type		ftrace_caller,#function
+ftrace_caller:
+	mov		%i7, %o1

arch/sparc64/lib/mcount.S _mcount: · line 110

Quoted 10 lines
arch/sparc64/lib/mcount.S +10 -0
+	mov		%o7, %o0
+	.globl		ftrace_call
+ftrace_call:
+	call		ftrace_stub
+	 mov		%o0, %o7
+	retl
+	 nop
+	.size		ftrace_caller,.-ftrace_caller
+#endif
+#endif

Looks great!

-- Steve

#2 lore raw

From: Steven Rostedt <[email protected]>
Date: Wed, 14 May 2008 09:32:18 -0400

Quoted 6 lines

If you add the following in arch/sparc64/kernel/Makefile

ifdef CONFIG_DYNAMIC_FTRACE
obj-y += ftrace.o
# Do not trace the ftrace code itself
ORIG_CFLAGS := $(KBUILD_CFLAGS)

Quoted 6 lines
KBUILD_CFLAGS = $(if $(filter-out ftrace, $(basename $(notdir $@))), \
	$(ORIG_CFLAGS), \
	$(subst -pg,,$(ORIG_CFLAGS)))

endif

Then you wont need the "notrace" annotations on the functions.

That frankly looks like crap compared to the simple "notrace" annotations in the *.c file.

This Makefile turd is in fact seemingly multiplying in various places in the ftrace tree, please reverse this trend.

#3 lore raw
Quoted 14 lines · David Miller

From: Steven Rostedt <[email protected]>
Date: Wed, 14 May 2008 09:32:18 -0400

Quoted 11 lines

If you add the following in arch/sparc64/kernel/Makefile

ifdef CONFIG_DYNAMIC_FTRACE
obj-y += ftrace.o
# Do not trace the ftrace code itself
ORIG_CFLAGS := $(KBUILD_CFLAGS)

KBUILD_CFLAGS = $(if $(filter-out ftrace, $(basename $(notdir $@))), \
	$(ORIG_CFLAGS), \
	$(subst -pg,,$(ORIG_CFLAGS)))

endif

Quoted 7 lines
Quoted 1 line

Then you wont need the "notrace" annotations on the functions.

That frankly looks like crap compared to the simple "notrace" annotations in the *.c file.

This Makefile turd is in fact seemingly multiplying in various places in the ftrace tree, please reverse this trend.

I wish it wasn't so ugly. The Makefile scripts allow you to add more CFLAGS to specific files but does not have a way to remove generic flags. If there was a better way then I would love to do it.

The problem is that adding "notrace" on ever function that needs it, also becomes quite messy. Especially when you have files where all functions in the file needs a notrace and there is 40 or so small functions that this applies to.

To add flags to file foo.c one simply needs to do

CFLAGS_foo.o = -newflags

It would be nice to have something like

CFLAGS_REMOVE_foo.o = -pg

Perhaps that is possible to do. Then the above "Makefile turd" wont be so crappy.

-- Steve

#4 lore raw

From: Steven Rostedt <[email protected]>
Date: Wed, 14 May 2008 17:58:13 -0400

Quoted 6 lines

It would be nice to have something like

CFLAGS_REMOVE_foo.o = -pg

Perhaps that is possible to do. Then the above "Makefile turd" wont be so crappy.

That would be much nicer.