* [PATCH -tip v14 03/12] kprobes: checks probe address is instruction boudary on x86
2009-08-13 20:31 [PATCH -tip v14 00/12] tracing: kprobe-based event tracer and x86 instruction decoder Masami Hiramatsu
@ 2009-08-13 20:31 ` Masami Hiramatsu
[not found] ` <20090818230340.GF5231@nowhere>
2009-08-13 20:31 ` [PATCH -tip v14 01/12] x86: instruction decoder API Masami Hiramatsu
` (12 subsequent siblings)
13 siblings, 1 reply; 29+ messages in thread
From: Masami Hiramatsu @ 2009-08-13 20:31 UTC (permalink / raw)
To: Ingo Molnar, Steven Rostedt, lkml
Cc: systemtap, kvm, DLE, Masami Hiramatsu,
Ananth N Mavinakayanahalli, Avi Kivity, Andi Kleen,
Christoph Hellwig, Frank Ch. Eigler, Frederic Weisbecker,
H. Peter Anvin, Ingo Molnar, Jason Baron, Jim Keniston, K.Prasad,
Lai Jiangshan, Li Zefan, PrzemysławPawełczyk,
Roland McGrath, Sam Ravnborg, Srikar Dronamraju, Steven Rostedt,
Tom Zanussi, Vegard Nossum
Ensure safeness of inserting kprobes by checking whether the specified
address is at the first byte of a instruction on x86.
This is done by decoding probed function from its head to the probe point.
Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
Acked-by: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Avi Kivity <avi@redhat.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Frank Ch. Eigler <fche@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Jason Baron <jbaron@redhat.com>
Cc: Jim Keniston <jkenisto@us.ibm.com>
Cc: K.Prasad <prasad@linux.vnet.ibm.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Przemysław Pawełczyk <przemyslaw@pawelczyk.it>
Cc: Roland McGrath <roland@redhat.com>
Cc: Sam Ravnborg <sam@ravnborg.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Tom Zanussi <tzanussi@gmail.com>
Cc: Vegard Nossum <vegard.nossum@gmail.com>
---
arch/x86/kernel/kprobes.c | 69 +++++++++++++++++++++++++++++++++++++++++++++
1 files changed, 69 insertions(+), 0 deletions(-)
diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c
index b5b1848..80d493f 100644
--- a/arch/x86/kernel/kprobes.c
+++ b/arch/x86/kernel/kprobes.c
@@ -48,6 +48,7 @@
#include <linux/preempt.h>
#include <linux/module.h>
#include <linux/kdebug.h>
+#include <linux/kallsyms.h>
#include <asm/cacheflush.h>
#include <asm/desc.h>
@@ -55,6 +56,7 @@
#include <asm/uaccess.h>
#include <asm/alternative.h>
#include <asm/debugreg.h>
+#include <asm/insn.h>
void jprobe_return_end(void);
@@ -245,6 +247,71 @@ retry:
}
}
+/* Recover the probed instruction at addr for further analysis. */
+static int recover_probed_instruction(kprobe_opcode_t *buf, unsigned long addr)
+{
+ struct kprobe *kp;
+ kp = get_kprobe((void *)addr);
+ if (!kp)
+ return -EINVAL;
+
+ /*
+ * Basically, kp->ainsn.insn has an original instruction.
+ * However, RIP-relative instruction can not do single-stepping
+ * at different place, fix_riprel() tweaks the displacement of
+ * that instruction. In that case, we can't recover the instruction
+ * from the kp->ainsn.insn.
+ *
+ * On the other hand, kp->opcode has a copy of the first byte of
+ * the probed instruction, which is overwritten by int3. And
+ * the instruction at kp->addr is not modified by kprobes except
+ * for the first byte, we can recover the original instruction
+ * from it and kp->opcode.
+ */
+ memcpy(buf, kp->addr, MAX_INSN_SIZE * sizeof(kprobe_opcode_t));
+ buf[0] = kp->opcode;
+ return 0;
+}
+
+/* Dummy buffers for kallsyms_lookup */
+static char __dummy_buf[KSYM_NAME_LEN];
+
+/* Check if paddr is at an instruction boundary */
+static int __kprobes can_probe(unsigned long paddr)
+{
+ int ret;
+ unsigned long addr, offset = 0;
+ struct insn insn;
+ kprobe_opcode_t buf[MAX_INSN_SIZE];
+
+ if (!kallsyms_lookup(paddr, NULL, &offset, NULL, __dummy_buf))
+ return 0;
+
+ /* Decode instructions */
+ addr = paddr - offset;
+ while (addr < paddr) {
+ kernel_insn_init(&insn, (void *)addr);
+ insn_get_opcode(&insn);
+
+ /* Check if the instruction has been modified. */
+ if (insn.opcode.bytes[0] == BREAKPOINT_INSTRUCTION) {
+ ret = recover_probed_instruction(buf, addr);
+ if (ret)
+ /*
+ * Another debugging subsystem might insert
+ * this breakpoint. In that case, we can't
+ * recover it.
+ */
+ return 0;
+ kernel_insn_init(&insn, buf);
+ }
+ insn_get_length(&insn);
+ addr += insn.length;
+ }
+
+ return (addr == paddr);
+}
+
/*
* Returns non-zero if opcode modifies the interrupt flag.
*/
@@ -360,6 +427,8 @@ static void __kprobes arch_copy_kprobe(struct kprobe *p)
int __kprobes arch_prepare_kprobe(struct kprobe *p)
{
+ if (!can_probe((unsigned long)p->addr))
+ return -EILSEQ;
/* insn: must be on special executable page on x86. */
p->ainsn.insn = get_insn_slot();
if (!p->ainsn.insn)
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America), Inc.
Software Solutions Division
e-mail: mhiramat@redhat.com
^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH -tip v14 04/12] kprobes: cleanup fix_riprel() using insn decoder on x86
2009-08-13 20:31 [PATCH -tip v14 00/12] tracing: kprobe-based event tracer and x86 instruction decoder Masami Hiramatsu
` (2 preceding siblings ...)
2009-08-13 20:31 ` [PATCH -tip v14 05/12] x86: add pt_regs register and stack access APIs Masami Hiramatsu
@ 2009-08-13 20:31 ` Masami Hiramatsu
2009-08-13 20:31 ` [PATCH -tip v14 02/12] x86: x86 instruction decoder build-time selftest Masami Hiramatsu
` (9 subsequent siblings)
13 siblings, 0 replies; 29+ messages in thread
From: Masami Hiramatsu @ 2009-08-13 20:31 UTC (permalink / raw)
To: Ingo Molnar, Steven Rostedt, lkml
Cc: systemtap, kvm, DLE, Masami Hiramatsu,
Ananth N Mavinakayanahalli, Avi Kivity, Andi Kleen,
Christoph Hellwig, Frank Ch. Eigler, Frederic Weisbecker,
H. Peter Anvin, Ingo Molnar, Jason Baron, Jim Keniston, K.Prasad,
Lai Jiangshan, Li Zefan, PrzemysławPawełczyk,
Roland McGrath, Sam Ravnborg, Srikar Dronamraju, Steven Rostedt,
Tom Zanussi, Vegard Nossum
Cleanup fix_riprel() in arch/x86/kernel/kprobes.c by using x86 instruction
decoder.
Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Avi Kivity <avi@redhat.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Frank Ch. Eigler <fche@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Jason Baron <jbaron@redhat.com>
Cc: Jim Keniston <jkenisto@us.ibm.com>
Cc: K.Prasad <prasad@linux.vnet.ibm.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Przemysław Pawełczyk <przemyslaw@pawelczyk.it>
Cc: Roland McGrath <roland@redhat.com>
Cc: Sam Ravnborg <sam@ravnborg.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Tom Zanussi <tzanussi@gmail.com>
Cc: Vegard Nossum <vegard.nossum@gmail.com>
---
arch/x86/kernel/kprobes.c | 128 ++++++++-------------------------------------
1 files changed, 23 insertions(+), 105 deletions(-)
diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c
index 80d493f..98f48d0 100644
--- a/arch/x86/kernel/kprobes.c
+++ b/arch/x86/kernel/kprobes.c
@@ -109,50 +109,6 @@ static const u32 twobyte_is_boostable[256 / 32] = {
/* ----------------------------------------------- */
/* 0 1 2 3 4 5 6 7 8 9 a b c d e f */
};
-static const u32 onebyte_has_modrm[256 / 32] = {
- /* 0 1 2 3 4 5 6 7 8 9 a b c d e f */
- /* ----------------------------------------------- */
- W(0x00, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) | /* 00 */
- W(0x10, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) , /* 10 */
- W(0x20, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) | /* 20 */
- W(0x30, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0) , /* 30 */
- W(0x40, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 40 */
- W(0x50, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 50 */
- W(0x60, 0, 0, 1, 1, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0) | /* 60 */
- W(0x70, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 70 */
- W(0x80, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* 80 */
- W(0x90, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 90 */
- W(0xa0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* a0 */
- W(0xb0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* b0 */
- W(0xc0, 1, 1, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0) | /* c0 */
- W(0xd0, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1) , /* d0 */
- W(0xe0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* e0 */
- W(0xf0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1) /* f0 */
- /* ----------------------------------------------- */
- /* 0 1 2 3 4 5 6 7 8 9 a b c d e f */
-};
-static const u32 twobyte_has_modrm[256 / 32] = {
- /* 0 1 2 3 4 5 6 7 8 9 a b c d e f */
- /* ----------------------------------------------- */
- W(0x00, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1) | /* 0f */
- W(0x10, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0) , /* 1f */
- W(0x20, 1, 1, 1, 1, 1, 0, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1) | /* 2f */
- W(0x30, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) , /* 3f */
- W(0x40, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* 4f */
- W(0x50, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* 5f */
- W(0x60, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* 6f */
- W(0x70, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 1, 1, 1, 1) , /* 7f */
- W(0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) | /* 8f */
- W(0x90, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* 9f */
- W(0xa0, 0, 0, 0, 1, 1, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1) | /* af */
- W(0xb0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 1, 1, 1, 1, 1) , /* bf */
- W(0xc0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0) | /* cf */
- W(0xd0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) , /* df */
- W(0xe0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1) | /* ef */
- W(0xf0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0) /* ff */
- /* ----------------------------------------------- */
- /* 0 1 2 3 4 5 6 7 8 9 a b c d e f */
-};
#undef W
struct kretprobe_blackpoint kretprobe_blacklist[] = {
@@ -345,68 +301,30 @@ static int __kprobes is_IF_modifier(kprobe_opcode_t *insn)
static void __kprobes fix_riprel(struct kprobe *p)
{
#ifdef CONFIG_X86_64
- u8 *insn = p->ainsn.insn;
- s64 disp;
- int need_modrm;
-
- /* Skip legacy instruction prefixes. */
- while (1) {
- switch (*insn) {
- case 0x66:
- case 0x67:
- case 0x2e:
- case 0x3e:
- case 0x26:
- case 0x64:
- case 0x65:
- case 0x36:
- case 0xf0:
- case 0xf3:
- case 0xf2:
- ++insn;
- continue;
- }
- break;
- }
+ struct insn insn;
+ kernel_insn_init(&insn, p->ainsn.insn);
- /* Skip REX instruction prefix. */
- if (is_REX_prefix(insn))
- ++insn;
-
- if (*insn == 0x0f) {
- /* Two-byte opcode. */
- ++insn;
- need_modrm = test_bit(*insn,
- (unsigned long *)twobyte_has_modrm);
- } else
- /* One-byte opcode. */
- need_modrm = test_bit(*insn,
- (unsigned long *)onebyte_has_modrm);
-
- if (need_modrm) {
- u8 modrm = *++insn;
- if ((modrm & 0xc7) == 0x05) {
- /* %rip+disp32 addressing mode */
- /* Displacement follows ModRM byte. */
- ++insn;
- /*
- * The copied instruction uses the %rip-relative
- * addressing mode. Adjust the displacement for the
- * difference between the original location of this
- * instruction and the location of the copy that will
- * actually be run. The tricky bit here is making sure
- * that the sign extension happens correctly in this
- * calculation, since we need a signed 32-bit result to
- * be sign-extended to 64 bits when it's added to the
- * %rip value and yield the same 64-bit result that the
- * sign-extension of the original signed 32-bit
- * displacement would have given.
- */
- disp = (u8 *) p->addr + *((s32 *) insn) -
- (u8 *) p->ainsn.insn;
- BUG_ON((s64) (s32) disp != disp); /* Sanity check. */
- *(s32 *)insn = (s32) disp;
- }
+ if (insn_rip_relative(&insn)) {
+ s64 newdisp;
+ u8 *disp;
+ insn_get_displacement(&insn);
+ /*
+ * The copied instruction uses the %rip-relative addressing
+ * mode. Adjust the displacement for the difference between
+ * the original location of this instruction and the location
+ * of the copy that will actually be run. The tricky bit here
+ * is making sure that the sign extension happens correctly in
+ * this calculation, since we need a signed 32-bit result to
+ * be sign-extended to 64 bits when it's added to the %rip
+ * value and yield the same 64-bit result that the sign-
+ * extension of the original signed 32-bit displacement would
+ * have given.
+ */
+ newdisp = (u8 *) p->addr + (s64) insn.displacement.value -
+ (u8 *) p->ainsn.insn;
+ BUG_ON((s64) (s32) newdisp != newdisp); /* Sanity check. */
+ disp = (u8 *) p->ainsn.insn + insn_offset_displacement(&insn);
+ *(s32 *) disp = (s32) newdisp;
}
#endif
}
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America), Inc.
Software Solutions Division
e-mail: mhiramat@redhat.com
^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH -tip v14 02/12] x86: x86 instruction decoder build-time selftest
2009-08-13 20:31 [PATCH -tip v14 00/12] tracing: kprobe-based event tracer and x86 instruction decoder Masami Hiramatsu
` (3 preceding siblings ...)
2009-08-13 20:31 ` [PATCH -tip v14 04/12] kprobes: cleanup fix_riprel() using insn decoder on x86 Masami Hiramatsu
@ 2009-08-13 20:31 ` Masami Hiramatsu
2009-08-13 20:32 ` [PATCH -tip v14 08/12] tracing: add kprobe-based event tracer Masami Hiramatsu
` (8 subsequent siblings)
13 siblings, 0 replies; 29+ messages in thread
From: Masami Hiramatsu @ 2009-08-13 20:31 UTC (permalink / raw)
To: Ingo Molnar, Steven Rostedt, lkml
Cc: systemtap, kvm, DLE, Masami Hiramatsu, Jim Keniston,
Ananth N Mavinakayanahalli, Avi Kivity, Andi Kleen,
Christoph Hellwig, Frank Ch. Eigler, Frederic Weisbecker,
H. Peter Anvin, Ingo Molnar, Jason Baron, K.Prasad,
Lai Jiangshan, Li Zefan, PrzemysławPawełczyk,
Roland McGrath, Sam Ravnborg, Srikar Dronamraju, Steven Rostedt,
Tom Zanussi, Vegard Nossum
Add a user-space selftest of x86 instruction decoder at kernel build time.
When CONFIG_X86_DECODER_SELFTEST=y, Kbuild builds a test harness of x86
instruction decoder and performs it after building vmlinux.
The test compares the results of objdump and x86 instruction decoder
code and check there are no differences.
Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
Signed-off-by: Jim Keniston <jkenisto@us.ibm.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Avi Kivity <avi@redhat.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Frank Ch. Eigler <fche@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Jason Baron <jbaron@redhat.com>
Cc: K.Prasad <prasad@linux.vnet.ibm.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Przemysław Pawełczyk <przemyslaw@pawelczyk.it>
Cc: Roland McGrath <roland@redhat.com>
Cc: Sam Ravnborg <sam@ravnborg.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Tom Zanussi <tzanussi@gmail.com>
Cc: Vegard Nossum <vegard.nossum@gmail.com>
---
arch/x86/Kconfig.debug | 9 +++
arch/x86/Makefile | 3 +
arch/x86/tools/Makefile | 15 +++++
arch/x86/tools/distill.awk | 42 +++++++++++++++
arch/x86/tools/test_get_len.c | 113 +++++++++++++++++++++++++++++++++++++++++
5 files changed, 182 insertions(+), 0 deletions(-)
create mode 100644 arch/x86/tools/Makefile
create mode 100644 arch/x86/tools/distill.awk
create mode 100644 arch/x86/tools/test_get_len.c
diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug
index d105f29..7d0b681 100644
--- a/arch/x86/Kconfig.debug
+++ b/arch/x86/Kconfig.debug
@@ -186,6 +186,15 @@ config X86_DS_SELFTEST
config HAVE_MMIOTRACE_SUPPORT
def_bool y
+config X86_DECODER_SELFTEST
+ bool "x86 instruction decoder selftest"
+ depends on DEBUG_KERNEL
+ ---help---
+ Perform x86 instruction decoder selftests at build time.
+ This option is useful for checking the sanity of x86 instruction
+ decoder code.
+ If unsure, say "N".
+
#
# IO delay types:
#
diff --git a/arch/x86/Makefile b/arch/x86/Makefile
index 1f3851a..f79580c 100644
--- a/arch/x86/Makefile
+++ b/arch/x86/Makefile
@@ -154,6 +154,9 @@ all: bzImage
KBUILD_IMAGE := $(boot)/bzImage
bzImage: vmlinux
+ifeq ($(CONFIG_X86_DECODER_SELFTEST),y)
+ $(Q)$(MAKE) $(build)=arch/x86/tools posttest
+endif
$(Q)$(MAKE) $(build)=$(boot) $(KBUILD_IMAGE)
$(Q)mkdir -p $(objtree)/arch/$(UTS_MACHINE)/boot
$(Q)ln -fsn ../../x86/boot/bzImage $(objtree)/arch/$(UTS_MACHINE)/boot/$@
diff --git a/arch/x86/tools/Makefile b/arch/x86/tools/Makefile
new file mode 100644
index 0000000..3dd626b
--- /dev/null
+++ b/arch/x86/tools/Makefile
@@ -0,0 +1,15 @@
+PHONY += posttest
+quiet_cmd_posttest = TEST $@
+ cmd_posttest = $(OBJDUMP) -d $(objtree)/vmlinux | awk -f $(srctree)/arch/x86/tools/distill.awk | $(obj)/test_get_len
+
+posttest: $(obj)/test_get_len vmlinux
+ $(call cmd,posttest)
+
+hostprogs-y := test_get_len
+
+# -I needed for generated C source and C source which in the kernel tree.
+HOSTCFLAGS_test_get_len.o := -Wall -I$(objtree)/arch/x86/lib/ -I$(srctree)/arch/x86/include/ -I$(srctree)/arch/x86/lib/
+
+# Dependancies are also needed.
+$(obj)/test_get_len.o: $(srctree)/arch/x86/lib/insn.c $(srctree)/arch/x86/lib/inat.c $(srctree)/arch/x86/include/asm/inat_types.h $(srctree)/arch/x86/include/asm/inat.h $(srctree)/arch/x86/include/asm/insn.h $(objtree)/arch/x86/lib/inat-tables.c
+
diff --git a/arch/x86/tools/distill.awk b/arch/x86/tools/distill.awk
new file mode 100644
index 0000000..d433619
--- /dev/null
+++ b/arch/x86/tools/distill.awk
@@ -0,0 +1,42 @@
+#!/bin/awk -f
+# Usage: objdump -d a.out | awk -f distill.awk | ./test_get_len
+# Distills the disassembly as follows:
+# - Removes all lines except the disassembled instructions.
+# - For instructions that exceed 1 line (7 bytes), crams all the hex bytes
+# into a single line.
+# - Remove bad(or prefix only) instructions
+
+BEGIN {
+ prev_addr = ""
+ prev_hex = ""
+ prev_mnemonic = ""
+ bad_expr = "(\\(bad\\)|^rex|^.byte|^rep(z|nz)$|^lock$|^es$|^cs$|^ss$|^ds$|^fs$|^gs$|^data(16|32)$|^addr(16|32|64))"
+ fwait_expr = "^9b "
+ fwait_str="9b\tfwait"
+}
+
+/^ *[0-9a-f]+:/ {
+ if (split($0, field, "\t") < 3) {
+ # This is a continuation of the same insn.
+ prev_hex = prev_hex field[2]
+ } else {
+ # Skip bad instructions
+ if (match(prev_mnemonic, bad_expr))
+ prev_addr = ""
+ # Split fwait from other f* instructions
+ if (match(prev_hex, fwait_expr) && prev_mnemonic != "fwait") {
+ printf "%s\t%s\n", prev_addr, fwait_str
+ sub(fwait_expr, "", prev_hex)
+ }
+ if (prev_addr != "")
+ printf "%s\t%s\t%s\n", prev_addr, prev_hex, prev_mnemonic
+ prev_addr = field[1]
+ prev_hex = field[2]
+ prev_mnemonic = field[3]
+ }
+}
+
+END {
+ if (prev_addr != "")
+ printf "%s\t%s\t%s\n", prev_addr, prev_hex, prev_mnemonic
+}
diff --git a/arch/x86/tools/test_get_len.c b/arch/x86/tools/test_get_len.c
new file mode 100644
index 0000000..1e81adb
--- /dev/null
+++ b/arch/x86/tools/test_get_len.c
@@ -0,0 +1,113 @@
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ * Copyright (C) IBM Corporation, 2009
+ */
+
+#include <stdlib.h>
+#include <stdio.h>
+#include <string.h>
+#include <assert.h>
+
+#ifdef __x86_64__
+#define CONFIG_X86_64
+#else
+#define CONFIG_X86_32
+#endif
+#define unlikely(cond) (cond)
+
+#include <asm/insn.h>
+#include <inat.c>
+#include <insn.c>
+
+/*
+ * Test of instruction analysis in general and insn_get_length() in
+ * particular. See if insn_get_length() and the disassembler agree
+ * on the length of each instruction in an elf disassembly.
+ *
+ * Usage: objdump -d a.out | awk -f distill.awk | ./test_get_len
+ */
+
+const char *prog;
+
+static void usage(void)
+{
+ fprintf(stderr, "Usage: objdump -d a.out | awk -f distill.awk |"
+ " ./test_get_len\n");
+ exit(1);
+}
+
+static void malformed_line(const char *line, int line_nr)
+{
+ fprintf(stderr, "%s: malformed line %d:\n%s", prog, line_nr, line);
+ exit(3);
+}
+
+#define BUFSIZE 256
+
+int main(int argc, char **argv)
+{
+ char line[BUFSIZE];
+ unsigned char insn_buf[16];
+ struct insn insn;
+ int insns = 0;
+
+ prog = argv[0];
+ if (argc > 1)
+ usage();
+
+ while (fgets(line, BUFSIZE, stdin)) {
+ char copy[BUFSIZE], *s, *tab1, *tab2;
+ int nb = 0;
+ unsigned int b;
+
+ insns++;
+ memset(insn_buf, 0, 16);
+ strcpy(copy, line);
+ tab1 = strchr(copy, '\t');
+ if (!tab1)
+ malformed_line(line, insns);
+ s = tab1 + 1;
+ s += strspn(s, " ");
+ tab2 = strchr(s, '\t');
+ if (!tab2)
+ malformed_line(line, insns);
+ *tab2 = '\0'; /* Characters beyond tab2 aren't examined */
+ while (s < tab2) {
+ if (sscanf(s, "%x", &b) == 1) {
+ insn_buf[nb++] = (unsigned char) b;
+ s += 3;
+ } else
+ break;
+ }
+ /* Decode an instruction */
+#ifdef __x86_64__
+ insn_init(&insn, insn_buf, 1);
+#else
+ insn_init(&insn, insn_buf, 0);
+#endif
+ insn_get_length(&insn);
+ if (insn.length != nb) {
+ fprintf(stderr, "Error: %s", line);
+ fprintf(stderr, "Error: objdump says %d bytes, but "
+ "insn_get_length() says %d (attr:%x)\n", nb,
+ insn.length, insn.attr);
+ exit(2);
+ }
+ }
+ fprintf(stderr, "Succeed: decoded and checked %d instructions\n",
+ insns);
+ return 0;
+}
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America), Inc.
Software Solutions Division
e-mail: mhiramat@redhat.com
^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH -tip v14 01/12] x86: instruction decoder API
2009-08-13 20:31 [PATCH -tip v14 00/12] tracing: kprobe-based event tracer and x86 instruction decoder Masami Hiramatsu
2009-08-13 20:31 ` [PATCH -tip v14 03/12] kprobes: checks probe address is instruction boudary on x86 Masami Hiramatsu
@ 2009-08-13 20:31 ` Masami Hiramatsu
[not found] ` <20090819234227.GJ4972@nowhere>
2009-08-13 20:31 ` [PATCH -tip v14 05/12] x86: add pt_regs register and stack access APIs Masami Hiramatsu
` (11 subsequent siblings)
13 siblings, 1 reply; 29+ messages in thread
From: Masami Hiramatsu @ 2009-08-13 20:31 UTC (permalink / raw)
To: Ingo Molnar, Steven Rostedt, lkml
Cc: systemtap, kvm, DLE, Masami Hiramatsu, Jim Keniston,
H. Peter Anvin, Ananth N Mavinakayanahalli, Avi Kivity,
Andi Kleen, Christoph Hellwig, Frank Ch. Eigler,
Frederic Weisbecker, Ingo Molnar, Jason Baron, K.Prasad,
Lai Jiangshan, Li Zefan, PrzemysławPawełczyk,
Roland McGrath, Sam Ravnborg, Srikar Dronamraju, Steven Rostedt,
Tom Zanussi, Vegard Nossum
Add x86 instruction decoder to arch-specific libraries. This decoder
can decode x86 instructions used in kernel into prefix, opcode, modrm,
sib, displacement and immediates. This can also show the length of
instructions.
This version introduces instruction attributes for decoding instructions.
The instruction attribute tables are generated from the opcode map file
(x86-opcode-map.txt) by the generator script(gen-insn-attr-x86.awk).
Currently, the opcode maps are based on opcode maps in Intel(R) 64 and
IA-32 Architectures Software Developers Manual Vol.2: Appendix.A,
and consist of below two types of opcode tables.
1-byte/2-bytes/3-bytes opcodes, which has 256 elements, are
written as below;
Table: table-name
Referrer: escaped-name
opcode: mnemonic|GrpXXX [operand1[,operand2...]] [(extra1)[,(extra2)...] [| 2nd-mnemonic ...]
(or)
opcode: escape # escaped-name
EndTable
Group opcodes, which has 8 elements, are written as below;
GrpTable: GrpXXX
reg: mnemonic [operand1[,operand2...]] [(extra1)[,(extra2)...] [| 2nd-mnemonic ...]
EndTable
These opcode maps include a few SSE and FP opcodes (for setup), because
those opcodes are used in the kernel.
Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
Signed-off-by: Jim Keniston <jkenisto@us.ibm.com>
Acked-by: H. Peter Anvin <hpa@zytor.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Avi Kivity <avi@redhat.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Frank Ch. Eigler <fche@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Jason Baron <jbaron@redhat.com>
Cc: K.Prasad <prasad@linux.vnet.ibm.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Przemysław Pawełczyk <przemyslaw@pawelczyk.it>
Cc: Roland McGrath <roland@redhat.com>
Cc: Sam Ravnborg <sam@ravnborg.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Tom Zanussi <tzanussi@gmail.com>
Cc: Vegard Nossum <vegard.nossum@gmail.com>
---
arch/x86/include/asm/inat.h | 188 +++++++++
arch/x86/include/asm/inat_types.h | 29 +
arch/x86/include/asm/insn.h | 143 +++++++
arch/x86/lib/Makefile | 13 +
arch/x86/lib/inat.c | 78 ++++
arch/x86/lib/insn.c | 464 ++++++++++++++++++++++
arch/x86/lib/x86-opcode-map.txt | 719 ++++++++++++++++++++++++++++++++++
arch/x86/tools/gen-insn-attr-x86.awk | 314 +++++++++++++++
8 files changed, 1948 insertions(+), 0 deletions(-)
create mode 100644 arch/x86/include/asm/inat.h
create mode 100644 arch/x86/include/asm/inat_types.h
create mode 100644 arch/x86/include/asm/insn.h
create mode 100644 arch/x86/lib/inat.c
create mode 100644 arch/x86/lib/insn.c
create mode 100644 arch/x86/lib/x86-opcode-map.txt
create mode 100644 arch/x86/tools/gen-insn-attr-x86.awk
diff --git a/arch/x86/include/asm/inat.h b/arch/x86/include/asm/inat.h
new file mode 100644
index 0000000..2866fdd
--- /dev/null
+++ b/arch/x86/include/asm/inat.h
@@ -0,0 +1,188 @@
+#ifndef _ASM_X86_INAT_H
+#define _ASM_X86_INAT_H
+/*
+ * x86 instruction attributes
+ *
+ * Written by Masami Hiramatsu <mhiramat@redhat.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ */
+#include <asm/inat_types.h>
+
+/*
+ * Internal bits. Don't use bitmasks directly, because these bits are
+ * unstable. You should use checking functions.
+ */
+
+#define INAT_OPCODE_TABLE_SIZE 256
+#define INAT_GROUP_TABLE_SIZE 8
+
+/* Legacy instruction prefixes */
+#define INAT_PFX_OPNDSZ 1 /* 0x66 */ /* LPFX1 */
+#define INAT_PFX_REPNE 2 /* 0xF2 */ /* LPFX2 */
+#define INAT_PFX_REPE 3 /* 0xF3 */ /* LPFX3 */
+#define INAT_PFX_LOCK 4 /* 0xF0 */
+#define INAT_PFX_CS 5 /* 0x2E */
+#define INAT_PFX_DS 6 /* 0x3E */
+#define INAT_PFX_ES 7 /* 0x26 */
+#define INAT_PFX_FS 8 /* 0x64 */
+#define INAT_PFX_GS 9 /* 0x65 */
+#define INAT_PFX_SS 10 /* 0x36 */
+#define INAT_PFX_ADDRSZ 11 /* 0x67 */
+
+#define INAT_LPREFIX_MAX 3
+
+/* Immediate size */
+#define INAT_IMM_BYTE 1
+#define INAT_IMM_WORD 2
+#define INAT_IMM_DWORD 3
+#define INAT_IMM_QWORD 4
+#define INAT_IMM_PTR 5
+#define INAT_IMM_VWORD32 6
+#define INAT_IMM_VWORD 7
+
+/* Legacy prefix */
+#define INAT_PFX_OFFS 0
+#define INAT_PFX_BITS 4
+#define INAT_PFX_MAX ((1 << INAT_PFX_BITS) - 1)
+#define INAT_PFX_MASK (INAT_PFX_MAX << INAT_PFX_OFFS)
+/* Escape opcodes */
+#define INAT_ESC_OFFS (INAT_PFX_OFFS + INAT_PFX_BITS)
+#define INAT_ESC_BITS 2
+#define INAT_ESC_MAX ((1 << INAT_ESC_BITS) - 1)
+#define INAT_ESC_MASK (INAT_ESC_MAX << INAT_ESC_OFFS)
+/* Group opcodes (1-16) */
+#define INAT_GRP_OFFS (INAT_ESC_OFFS + INAT_ESC_BITS)
+#define INAT_GRP_BITS 5
+#define INAT_GRP_MAX ((1 << INAT_GRP_BITS) - 1)
+#define INAT_GRP_MASK (INAT_GRP_MAX << INAT_GRP_OFFS)
+/* Immediates */
+#define INAT_IMM_OFFS (INAT_GRP_OFFS + INAT_GRP_BITS)
+#define INAT_IMM_BITS 3
+#define INAT_IMM_MASK (((1 << INAT_IMM_BITS) - 1) << INAT_IMM_OFFS)
+/* Flags */
+#define INAT_FLAG_OFFS (INAT_IMM_OFFS + INAT_IMM_BITS)
+#define INAT_REXPFX (1 << INAT_FLAG_OFFS)
+#define INAT_MODRM (1 << (INAT_FLAG_OFFS + 1))
+#define INAT_FORCE64 (1 << (INAT_FLAG_OFFS + 2))
+#define INAT_SCNDIMM (1 << (INAT_FLAG_OFFS + 3))
+#define INAT_MOFFSET (1 << (INAT_FLAG_OFFS + 4))
+#define INAT_VARIANT (1 << (INAT_FLAG_OFFS + 5))
+/* Attribute making macros for attribute tables */
+#define INAT_MAKE_PREFIX(pfx) (pfx << INAT_PFX_OFFS)
+#define INAT_MAKE_ESCAPE(esc) (esc << INAT_ESC_OFFS)
+#define INAT_MAKE_GROUP(grp) ((grp << INAT_GRP_OFFS) | INAT_MODRM)
+#define INAT_MAKE_IMM(imm) (imm << INAT_IMM_OFFS)
+
+/* Attribute search APIs */
+extern insn_attr_t inat_get_opcode_attribute(insn_byte_t opcode);
+extern insn_attr_t inat_get_escape_attribute(insn_byte_t opcode,
+ insn_byte_t last_pfx,
+ insn_attr_t esc_attr);
+extern insn_attr_t inat_get_group_attribute(insn_byte_t modrm,
+ insn_byte_t last_pfx,
+ insn_attr_t esc_attr);
+
+/* Attribute checking functions */
+static inline int inat_is_prefix(insn_attr_t attr)
+{
+ return attr & INAT_PFX_MASK;
+}
+
+static inline int inat_is_address_size_prefix(insn_attr_t attr)
+{
+ return (attr & INAT_PFX_MASK) == INAT_PFX_ADDRSZ;
+}
+
+static inline int inat_is_operand_size_prefix(insn_attr_t attr)
+{
+ return (attr & INAT_PFX_MASK) == INAT_PFX_OPNDSZ;
+}
+
+static inline int inat_last_prefix_id(insn_attr_t attr)
+{
+ if ((attr & INAT_PFX_MASK) > INAT_LPREFIX_MAX)
+ return 0;
+ else
+ return attr & INAT_PFX_MASK;
+}
+
+static inline int inat_is_escape(insn_attr_t attr)
+{
+ return attr & INAT_ESC_MASK;
+}
+
+static inline int inat_escape_id(insn_attr_t attr)
+{
+ return (attr & INAT_ESC_MASK) >> INAT_ESC_OFFS;
+}
+
+static inline int inat_is_group(insn_attr_t attr)
+{
+ return attr & INAT_GRP_MASK;
+}
+
+static inline int inat_group_id(insn_attr_t attr)
+{
+ return (attr & INAT_GRP_MASK) >> INAT_GRP_OFFS;
+}
+
+static inline int inat_group_common_attribute(insn_attr_t attr)
+{
+ return attr & ~INAT_GRP_MASK;
+}
+
+static inline int inat_has_immediate(insn_attr_t attr)
+{
+ return attr & INAT_IMM_MASK;
+}
+
+static inline int inat_immediate_size(insn_attr_t attr)
+{
+ return (attr & INAT_IMM_MASK) >> INAT_IMM_OFFS;
+}
+
+static inline int inat_is_rex_prefix(insn_attr_t attr)
+{
+ return attr & INAT_REXPFX;
+}
+
+static inline int inat_has_modrm(insn_attr_t attr)
+{
+ return attr & INAT_MODRM;
+}
+
+static inline int inat_is_force64(insn_attr_t attr)
+{
+ return attr & INAT_FORCE64;
+}
+
+static inline int inat_has_second_immediate(insn_attr_t attr)
+{
+ return attr & INAT_SCNDIMM;
+}
+
+static inline int inat_has_moffset(insn_attr_t attr)
+{
+ return attr & INAT_MOFFSET;
+}
+
+static inline int inat_has_variant(insn_attr_t attr)
+{
+ return attr & INAT_VARIANT;
+}
+
+#endif
diff --git a/arch/x86/include/asm/inat_types.h b/arch/x86/include/asm/inat_types.h
new file mode 100644
index 0000000..cb3c20c
--- /dev/null
+++ b/arch/x86/include/asm/inat_types.h
@@ -0,0 +1,29 @@
+#ifndef _ASM_X86_INAT_TYPES_H
+#define _ASM_X86_INAT_TYPES_H
+/*
+ * x86 instruction attributes
+ *
+ * Written by Masami Hiramatsu <mhiramat@redhat.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ */
+
+/* Instruction attributes */
+typedef unsigned int insn_attr_t;
+typedef unsigned char insn_byte_t;
+typedef signed int insn_value_t;
+
+#endif
diff --git a/arch/x86/include/asm/insn.h b/arch/x86/include/asm/insn.h
new file mode 100644
index 0000000..12b4e37
--- /dev/null
+++ b/arch/x86/include/asm/insn.h
@@ -0,0 +1,143 @@
+#ifndef _ASM_X86_INSN_H
+#define _ASM_X86_INSN_H
+/*
+ * x86 instruction analysis
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ * Copyright (C) IBM Corporation, 2009
+ */
+
+/* insn_attr_t is defined in inat.h */
+#include <asm/inat.h>
+
+struct insn_field {
+ union {
+ insn_value_t value;
+ insn_byte_t bytes[4];
+ };
+ /* !0 if we've run insn_get_xxx() for this field */
+ unsigned char got;
+ unsigned char nbytes;
+};
+
+struct insn {
+ struct insn_field prefixes; /*
+ * Prefixes
+ * prefixes.bytes[3]: last prefix
+ */
+ struct insn_field rex_prefix; /* REX prefix */
+ struct insn_field opcode; /*
+ * opcode.bytes[0]: opcode1
+ * opcode.bytes[1]: opcode2
+ * opcode.bytes[2]: opcode3
+ */
+ struct insn_field modrm;
+ struct insn_field sib;
+ struct insn_field displacement;
+ union {
+ struct insn_field immediate;
+ struct insn_field moffset1; /* for 64bit MOV */
+ struct insn_field immediate1; /* for 64bit imm or off16/32 */
+ };
+ union {
+ struct insn_field moffset2; /* for 64bit MOV */
+ struct insn_field immediate2; /* for 64bit imm or seg16 */
+ };
+
+ insn_attr_t attr;
+ unsigned char opnd_bytes;
+ unsigned char addr_bytes;
+ unsigned char length;
+ unsigned char x86_64;
+
+ const insn_byte_t *kaddr; /* kernel address of insn to analyze */
+ const insn_byte_t *next_byte;
+};
+
+#define X86_MODRM_MOD(modrm) (((modrm) & 0xc0) >> 6)
+#define X86_MODRM_REG(modrm) (((modrm) & 0x38) >> 3)
+#define X86_MODRM_RM(modrm) ((modrm) & 0x07)
+
+#define X86_SIB_SCALE(sib) (((sib) & 0xc0) >> 6)
+#define X86_SIB_INDEX(sib) (((sib) & 0x38) >> 3)
+#define X86_SIB_BASE(sib) ((sib) & 0x07)
+
+#define X86_REX_W(rex) ((rex) & 8)
+#define X86_REX_R(rex) ((rex) & 4)
+#define X86_REX_X(rex) ((rex) & 2)
+#define X86_REX_B(rex) ((rex) & 1)
+
+/* The last prefix is needed for two-byte and three-byte opcodes */
+static inline insn_byte_t insn_last_prefix(struct insn *insn)
+{
+ return insn->prefixes.bytes[3];
+}
+
+extern void insn_init(struct insn *insn, const void *kaddr, int x86_64);
+extern void insn_get_prefixes(struct insn *insn);
+extern void insn_get_opcode(struct insn *insn);
+extern void insn_get_modrm(struct insn *insn);
+extern void insn_get_sib(struct insn *insn);
+extern void insn_get_displacement(struct insn *insn);
+extern void insn_get_immediate(struct insn *insn);
+extern void insn_get_length(struct insn *insn);
+
+/* Attribute will be determined after getting ModRM (for opcode groups) */
+static inline void insn_get_attribute(struct insn *insn)
+{
+ insn_get_modrm(insn);
+}
+
+/* Instruction uses RIP-relative addressing */
+extern int insn_rip_relative(struct insn *insn);
+
+/* Init insn for kernel text */
+static inline void kernel_insn_init(struct insn *insn, const void *kaddr)
+{
+#ifdef CONFIG_X86_64
+ insn_init(insn, kaddr, 1);
+#else /* CONFIG_X86_32 */
+ insn_init(insn, kaddr, 0);
+#endif
+}
+
+/* Offset of each field from kaddr */
+static inline int insn_offset_rex_prefix(struct insn *insn)
+{
+ return insn->prefixes.nbytes;
+}
+static inline int insn_offset_opcode(struct insn *insn)
+{
+ return insn_offset_rex_prefix(insn) + insn->rex_prefix.nbytes;
+}
+static inline int insn_offset_modrm(struct insn *insn)
+{
+ return insn_offset_opcode(insn) + insn->opcode.nbytes;
+}
+static inline int insn_offset_sib(struct insn *insn)
+{
+ return insn_offset_modrm(insn) + insn->modrm.nbytes;
+}
+static inline int insn_offset_displacement(struct insn *insn)
+{
+ return insn_offset_sib(insn) + insn->sib.nbytes;
+}
+static inline int insn_offset_immediate(struct insn *insn)
+{
+ return insn_offset_displacement(insn) + insn->displacement.nbytes;
+}
+
+#endif /* _ASM_X86_INSN_H */
diff --git a/arch/x86/lib/Makefile b/arch/x86/lib/Makefile
index 07c3189..c77f8a7 100644
--- a/arch/x86/lib/Makefile
+++ b/arch/x86/lib/Makefile
@@ -2,12 +2,25 @@
# Makefile for x86 specific library files.
#
+inat_tables_script = $(srctree)/arch/x86/tools/gen-insn-attr-x86.awk
+inat_tables_maps = $(srctree)/arch/x86/lib/x86-opcode-map.txt
+quiet_cmd_inat_tables = GEN $@
+ cmd_inat_tables = $(AWK) -f $(inat_tables_script) $(inat_tables_maps) > $@
+
+$(obj)/inat-tables.c: $(inat_tables_script) $(inat_tables_maps)
+ $(call cmd,inat_tables)
+
+$(obj)/inat.o: $(obj)/inat-tables.c
+
+clean-files := inat-tables.c
+
obj-$(CONFIG_SMP) := msr.o
lib-y := delay.o
lib-y += thunk_$(BITS).o
lib-y += usercopy_$(BITS).o getuser.o putuser.o
lib-y += memcpy_$(BITS).o
+lib-y += insn.o inat.o
ifeq ($(CONFIG_X86_32),y)
obj-y += atomic64_32.o
diff --git a/arch/x86/lib/inat.c b/arch/x86/lib/inat.c
new file mode 100644
index 0000000..054656a
--- /dev/null
+++ b/arch/x86/lib/inat.c
@@ -0,0 +1,78 @@
+/*
+ * x86 instruction attribute tables
+ *
+ * Written by Masami Hiramatsu <mhiramat@redhat.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ */
+#include <asm/insn.h>
+
+/* Attribute tables are generated from opcode map */
+#include "inat-tables.c"
+
+/* Attribute search APIs */
+insn_attr_t inat_get_opcode_attribute(insn_byte_t opcode)
+{
+ return inat_primary_table[opcode];
+}
+
+insn_attr_t inat_get_escape_attribute(insn_byte_t opcode, insn_byte_t last_pfx,
+ insn_attr_t esc_attr)
+{
+ const insn_attr_t *table;
+ insn_attr_t lpfx_attr;
+ int n, m = 0;
+
+ n = inat_escape_id(esc_attr);
+ if (last_pfx) {
+ lpfx_attr = inat_get_opcode_attribute(last_pfx);
+ m = inat_last_prefix_id(lpfx_attr);
+ }
+ table = inat_escape_tables[n][0];
+ if (!table)
+ return 0;
+ if (inat_has_variant(table[opcode]) && m) {
+ table = inat_escape_tables[n][m];
+ if (!table)
+ return 0;
+ }
+ return table[opcode];
+}
+
+insn_attr_t inat_get_group_attribute(insn_byte_t modrm, insn_byte_t last_pfx,
+ insn_attr_t grp_attr)
+{
+ const insn_attr_t *table;
+ insn_attr_t lpfx_attr;
+ int n, m = 0;
+
+ n = inat_group_id(grp_attr);
+ if (last_pfx) {
+ lpfx_attr = inat_get_opcode_attribute(last_pfx);
+ m = inat_last_prefix_id(lpfx_attr);
+ }
+ table = inat_group_tables[n][0];
+ if (!table)
+ return inat_group_common_attribute(grp_attr);
+ if (inat_has_variant(table[X86_MODRM_REG(modrm)]) && m) {
+ table = inat_escape_tables[n][m];
+ if (!table)
+ return inat_group_common_attribute(grp_attr);
+ }
+ return table[X86_MODRM_REG(modrm)] |
+ inat_group_common_attribute(grp_attr);
+}
+
diff --git a/arch/x86/lib/insn.c b/arch/x86/lib/insn.c
new file mode 100644
index 0000000..dfd56a3
--- /dev/null
+++ b/arch/x86/lib/insn.c
@@ -0,0 +1,464 @@
+/*
+ * x86 instruction analysis
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ * Copyright (C) IBM Corporation, 2002, 2004, 2009
+ */
+
+#include <linux/string.h>
+#include <asm/inat.h>
+#include <asm/insn.h>
+
+#define get_next(t, insn) \
+ ({t r; r = *(t*)insn->next_byte; insn->next_byte += sizeof(t); r; })
+
+#define peek_next(t, insn) \
+ ({t r; r = *(t*)insn->next_byte; r; })
+
+/**
+ * insn_init() - initialize struct insn
+ * @insn: &struct insn to be initialized
+ * @kaddr: address (in kernel memory) of instruction (or copy thereof)
+ * @x86_64: !0 for 64-bit kernel or 64-bit app
+ */
+void insn_init(struct insn *insn, const void *kaddr, int x86_64)
+{
+ memset(insn, 0, sizeof(*insn));
+ insn->kaddr = kaddr;
+ insn->next_byte = kaddr;
+ insn->x86_64 = x86_64 ? 1 : 0;
+ insn->opnd_bytes = 4;
+ if (x86_64)
+ insn->addr_bytes = 8;
+ else
+ insn->addr_bytes = 4;
+}
+
+/**
+ * insn_get_prefixes - scan x86 instruction prefix bytes
+ * @insn: &struct insn containing instruction
+ *
+ * Populates the @insn->prefixes bitmap, and updates @insn->next_byte
+ * to point to the (first) opcode. No effect if @insn->prefixes.got
+ * is already set.
+ */
+void insn_get_prefixes(struct insn *insn)
+{
+ struct insn_field *prefixes = &insn->prefixes;
+ insn_attr_t attr;
+ insn_byte_t b, lb;
+ int i, nb;
+
+ if (prefixes->got)
+ return;
+
+ nb = 0;
+ lb = 0;
+ b = peek_next(insn_byte_t, insn);
+ attr = inat_get_opcode_attribute(b);
+ while (inat_is_prefix(attr)) {
+ /* Skip if same prefix */
+ for (i = 0; i < nb; i++)
+ if (prefixes->bytes[i] == b)
+ goto found;
+ if (nb == 4)
+ /* Invalid instruction */
+ break;
+ prefixes->bytes[nb++] = b;
+ if (inat_is_address_size_prefix(attr)) {
+ /* address size switches 2/4 or 4/8 */
+ if (insn->x86_64)
+ insn->addr_bytes ^= 12;
+ else
+ insn->addr_bytes ^= 6;
+ } else if (inat_is_operand_size_prefix(attr)) {
+ /* oprand size switches 2/4 */
+ insn->opnd_bytes ^= 6;
+ }
+found:
+ prefixes->nbytes++;
+ insn->next_byte++;
+ lb = b;
+ b = peek_next(insn_byte_t, insn);
+ attr = inat_get_opcode_attribute(b);
+ }
+ /* Set the last prefix */
+ if (lb && lb != insn->prefixes.bytes[3]) {
+ if (unlikely(insn->prefixes.bytes[3])) {
+ /* Swap the last prefix */
+ b = insn->prefixes.bytes[3];
+ for (i = 0; i < nb; i++)
+ if (prefixes->bytes[i] == lb)
+ prefixes->bytes[i] = b;
+ }
+ insn->prefixes.bytes[3] = lb;
+ }
+
+ if (insn->x86_64) {
+ b = peek_next(insn_byte_t, insn);
+ attr = inat_get_opcode_attribute(b);
+ if (inat_is_rex_prefix(attr)) {
+ insn->rex_prefix.value = b;
+ insn->rex_prefix.nbytes = 1;
+ insn->next_byte++;
+ if (X86_REX_W(b))
+ /* REX.W overrides opnd_size */
+ insn->opnd_bytes = 8;
+ }
+ }
+ insn->rex_prefix.got = 1;
+ prefixes->got = 1;
+ return;
+}
+
+/**
+ * insn_get_opcode - collect opcode(s)
+ * @insn: &struct insn containing instruction
+ *
+ * Populates @insn->opcode, updates @insn->next_byte to point past the
+ * opcode byte(s), and set @insn->attr (except for groups).
+ * If necessary, first collects any preceding (prefix) bytes.
+ * Sets @insn->opcode.value = opcode1. No effect if @insn->opcode.got
+ * is already 1.
+ */
+void insn_get_opcode(struct insn *insn)
+{
+ struct insn_field *opcode = &insn->opcode;
+ insn_byte_t op, pfx;
+ if (opcode->got)
+ return;
+ if (!insn->prefixes.got)
+ insn_get_prefixes(insn);
+
+ /* Get first opcode */
+ op = get_next(insn_byte_t, insn);
+ opcode->bytes[0] = op;
+ opcode->nbytes = 1;
+ insn->attr = inat_get_opcode_attribute(op);
+ while (inat_is_escape(insn->attr)) {
+ /* Get escaped opcode */
+ op = get_next(insn_byte_t, insn);
+ opcode->bytes[opcode->nbytes++] = op;
+ pfx = insn_last_prefix(insn);
+ insn->attr = inat_get_escape_attribute(op, pfx, insn->attr);
+ }
+ opcode->got = 1;
+}
+
+/**
+ * insn_get_modrm - collect ModRM byte, if any
+ * @insn: &struct insn containing instruction
+ *
+ * Populates @insn->modrm and updates @insn->next_byte to point past the
+ * ModRM byte, if any. If necessary, first collects the preceding bytes
+ * (prefixes and opcode(s)). No effect if @insn->modrm.got is already 1.
+ */
+void insn_get_modrm(struct insn *insn)
+{
+ struct insn_field *modrm = &insn->modrm;
+ insn_byte_t pfx, mod;
+ if (modrm->got)
+ return;
+ if (!insn->opcode.got)
+ insn_get_opcode(insn);
+
+ if (inat_has_modrm(insn->attr)) {
+ mod = get_next(insn_byte_t, insn);
+ modrm->value = mod;
+ modrm->nbytes = 1;
+ if (inat_is_group(insn->attr)) {
+ pfx = insn_last_prefix(insn);
+ insn->attr = inat_get_group_attribute(mod, pfx,
+ insn->attr);
+ }
+ }
+
+ if (insn->x86_64 && inat_is_force64(insn->attr))
+ insn->opnd_bytes = 8;
+ modrm->got = 1;
+}
+
+
+/**
+ * insn_rip_relative() - Does instruction use RIP-relative addressing mode?
+ * @insn: &struct insn containing instruction
+ *
+ * If necessary, first collects the instruction up to and including the
+ * ModRM byte. No effect if @insn->x86_64 is 0.
+ */
+int insn_rip_relative(struct insn *insn)
+{
+ struct insn_field *modrm = &insn->modrm;
+
+ if (!insn->x86_64)
+ return 0;
+ if (!modrm->got)
+ insn_get_modrm(insn);
+ /*
+ * For rip-relative instructions, the mod field (top 2 bits)
+ * is zero and the r/m field (bottom 3 bits) is 0x5.
+ */
+ return (modrm->nbytes && (modrm->value & 0xc7) == 0x5);
+}
+
+/**
+ * insn_get_sib() - Get the SIB byte of instruction
+ * @insn: &struct insn containing instruction
+ *
+ * If necessary, first collects the instruction up to and including the
+ * ModRM byte.
+ */
+void insn_get_sib(struct insn *insn)
+{
+ insn_byte_t modrm;
+
+ if (insn->sib.got)
+ return;
+ if (!insn->modrm.got)
+ insn_get_modrm(insn);
+ if (insn->modrm.nbytes) {
+ modrm = (insn_byte_t)insn->modrm.value;
+ if (insn->addr_bytes != 2 &&
+ X86_MODRM_MOD(modrm) != 3 && X86_MODRM_RM(modrm) == 4) {
+ insn->sib.value = get_next(insn_byte_t, insn);
+ insn->sib.nbytes = 1;
+ }
+ }
+ insn->sib.got = 1;
+}
+
+
+/**
+ * insn_get_displacement() - Get the displacement of instruction
+ * @insn: &struct insn containing instruction
+ *
+ * If necessary, first collects the instruction up to and including the
+ * SIB byte.
+ * Displacement value is sign-expanded.
+ */
+void insn_get_displacement(struct insn *insn)
+{
+ insn_byte_t mod, rm, base;
+
+ if (insn->displacement.got)
+ return;
+ if (!insn->sib.got)
+ insn_get_sib(insn);
+ if (insn->modrm.nbytes) {
+ /*
+ * Interpreting the modrm byte:
+ * mod = 00 - no displacement fields (exceptions below)
+ * mod = 01 - 1-byte displacement field
+ * mod = 10 - displacement field is 4 bytes, or 2 bytes if
+ * address size = 2 (0x67 prefix in 32-bit mode)
+ * mod = 11 - no memory operand
+ *
+ * If address size = 2...
+ * mod = 00, r/m = 110 - displacement field is 2 bytes
+ *
+ * If address size != 2...
+ * mod != 11, r/m = 100 - SIB byte exists
+ * mod = 00, SIB base = 101 - displacement field is 4 bytes
+ * mod = 00, r/m = 101 - rip-relative addressing, displacement
+ * field is 4 bytes
+ */
+ mod = X86_MODRM_MOD(insn->modrm.value);
+ rm = X86_MODRM_RM(insn->modrm.value);
+ base = X86_SIB_BASE(insn->sib.value);
+ if (mod == 3)
+ goto out;
+ if (mod == 1) {
+ insn->displacement.value = get_next(char, insn);
+ insn->displacement.nbytes = 1;
+ } else if (insn->addr_bytes == 2) {
+ if ((mod == 0 && rm == 6) || mod == 2) {
+ insn->displacement.value =
+ get_next(short, insn);
+ insn->displacement.nbytes = 2;
+ }
+ } else {
+ if ((mod == 0 && rm == 5) || mod == 2 ||
+ (mod == 0 && base == 5)) {
+ insn->displacement.value = get_next(int, insn);
+ insn->displacement.nbytes = 4;
+ }
+ }
+ }
+out:
+ insn->displacement.got = 1;
+}
+
+/* Decode moffset16/32/64 */
+static void __get_moffset(struct insn *insn)
+{
+ switch (insn->addr_bytes) {
+ case 2:
+ insn->moffset1.value = get_next(short, insn);
+ insn->moffset1.nbytes = 2;
+ break;
+ case 4:
+ insn->moffset1.value = get_next(int, insn);
+ insn->moffset1.nbytes = 4;
+ break;
+ case 8:
+ insn->moffset1.value = get_next(int, insn);
+ insn->moffset1.nbytes = 4;
+ insn->moffset2.value = get_next(int, insn);
+ insn->moffset2.nbytes = 4;
+ break;
+ }
+ insn->moffset1.got = insn->moffset2.got = 1;
+}
+
+/* Decode imm v32(Iz) */
+static void __get_immv32(struct insn *insn)
+{
+ switch (insn->opnd_bytes) {
+ case 2:
+ insn->immediate.value = get_next(short, insn);
+ insn->immediate.nbytes = 2;
+ break;
+ case 4:
+ case 8:
+ insn->immediate.value = get_next(int, insn);
+ insn->immediate.nbytes = 4;
+ break;
+ }
+}
+
+/* Decode imm v64(Iv/Ov) */
+static void __get_immv(struct insn *insn)
+{
+ switch (insn->opnd_bytes) {
+ case 2:
+ insn->immediate1.value = get_next(short, insn);
+ insn->immediate1.nbytes = 2;
+ break;
+ case 4:
+ insn->immediate1.value = get_next(int, insn);
+ insn->immediate1.nbytes = 4;
+ break;
+ case 8:
+ insn->immediate1.value = get_next(int, insn);
+ insn->immediate1.nbytes = 4;
+ insn->immediate2.value = get_next(int, insn);
+ insn->immediate2.nbytes = 4;
+ break;
+ }
+ insn->immediate1.got = insn->immediate2.got = 1;
+}
+
+/* Decode ptr16:16/32(Ap) */
+static void __get_immptr(struct insn *insn)
+{
+ switch (insn->opnd_bytes) {
+ case 2:
+ insn->immediate1.value = get_next(short, insn);
+ insn->immediate1.nbytes = 2;
+ break;
+ case 4:
+ insn->immediate1.value = get_next(int, insn);
+ insn->immediate1.nbytes = 4;
+ break;
+ case 8:
+ /* ptr16:64 is not exist (no segment) */
+ return;
+ }
+ insn->immediate2.value = get_next(unsigned short, insn);
+ insn->immediate2.nbytes = 2;
+ insn->immediate1.got = insn->immediate2.got = 1;
+}
+
+/**
+ * insn_get_immediate() - Get the immediates of instruction
+ * @insn: &struct insn containing instruction
+ *
+ * If necessary, first collects the instruction up to and including the
+ * displacement bytes.
+ * Basically, most of immediates are sign-expanded. Unsigned-value can be
+ * get by bit masking with ((1 << (nbytes * 8)) - 1)
+ */
+void insn_get_immediate(struct insn *insn)
+{
+ if (insn->immediate.got)
+ return;
+ if (!insn->displacement.got)
+ insn_get_displacement(insn);
+
+ if (inat_has_moffset(insn->attr)) {
+ __get_moffset(insn);
+ goto done;
+ }
+
+ if (!inat_has_immediate(insn->attr))
+ /* no immediates */
+ goto done;
+
+ switch (inat_immediate_size(insn->attr)) {
+ case INAT_IMM_BYTE:
+ insn->immediate.value = get_next(char, insn);
+ insn->immediate.nbytes = 1;
+ break;
+ case INAT_IMM_WORD:
+ insn->immediate.value = get_next(short, insn);
+ insn->immediate.nbytes = 2;
+ break;
+ case INAT_IMM_DWORD:
+ insn->immediate.value = get_next(int, insn);
+ insn->immediate.nbytes = 4;
+ break;
+ case INAT_IMM_QWORD:
+ insn->immediate1.value = get_next(int, insn);
+ insn->immediate1.nbytes = 4;
+ insn->immediate2.value = get_next(int, insn);
+ insn->immediate2.nbytes = 4;
+ break;
+ case INAT_IMM_PTR:
+ __get_immptr(insn);
+ break;
+ case INAT_IMM_VWORD32:
+ __get_immv32(insn);
+ break;
+ case INAT_IMM_VWORD:
+ __get_immv(insn);
+ break;
+ default:
+ break;
+ }
+ if (inat_has_second_immediate(insn->attr)) {
+ insn->immediate2.value = get_next(char, insn);
+ insn->immediate2.nbytes = 1;
+ }
+done:
+ insn->immediate.got = 1;
+}
+
+/**
+ * insn_get_length() - Get the length of instruction
+ * @insn: &struct insn containing instruction
+ *
+ * If necessary, first collects the instruction up to and including the
+ * immediates bytes.
+ */
+void insn_get_length(struct insn *insn)
+{
+ if (insn->length)
+ return;
+ if (!insn->immediate.got)
+ insn_get_immediate(insn);
+ insn->length = (unsigned char)((unsigned long)insn->next_byte
+ - (unsigned long)insn->kaddr);
+}
diff --git a/arch/x86/lib/x86-opcode-map.txt b/arch/x86/lib/x86-opcode-map.txt
new file mode 100644
index 0000000..083dd59
--- /dev/null
+++ b/arch/x86/lib/x86-opcode-map.txt
@@ -0,0 +1,719 @@
+# x86 Opcode Maps
+#
+#<Opcode maps>
+# Table: table-name
+# Referrer: escaped-name
+# opcode: mnemonic|GrpXXX [operand1[,operand2...]] [(extra1)[,(extra2)...] [| 2nd-mnemonic ...]
+# (or)
+# opcode: escape # escaped-name
+# EndTable
+#
+#<group maps>
+# GrpTable: GrpXXX
+# reg: mnemonic [operand1[,operand2...]] [(extra1)[,(extra2)...] [| 2nd-mnemonic ...]
+# EndTable
+#
+
+Table: one byte opcode
+Referrer:
+# 0x00 - 0x0f
+00: ADD Eb,Gb
+01: ADD Ev,Gv
+02: ADD Gb,Eb
+03: ADD Gv,Ev
+04: ADD AL,Ib
+05: ADD rAX,Iz
+06: PUSH ES (i64)
+07: POP ES (i64)
+08: OR Eb,Gb
+09: OR Ev,Gv
+0a: OR Gb,Eb
+0b: OR Gv,Ev
+0c: OR AL,Ib
+0d: OR rAX,Iz
+0e: PUSH CS (i64)
+0f: escape # 2-byte escape
+# 0x10 - 0x1f
+10: ADC Eb,Gb
+11: ADC Ev,Gv
+12: ADC Gb,Eb
+13: ADC Gv,Ev
+14: ADC AL,Ib
+15: ADC rAX,Iz
+16: PUSH SS (i64)
+17: POP SS (i64)
+18: SBB Eb,Gb
+19: SBB Ev,Gv
+1a: SBB Gb,Eb
+1b: SBB Gv,Ev
+1c: SBB AL,Ib
+1d: SBB rAX,Iz
+1e: PUSH DS (i64)
+1f: POP DS (i64)
+# 0x20 - 0x2f
+20: AND Eb,Gb
+21: AND Ev,Gv
+22: AND Gb,Eb
+23: AND Gv,Ev
+24: AND AL,Ib
+25: AND rAx,Iz
+26: SEG=ES (Prefix)
+27: DAA (i64)
+28: SUB Eb,Gb
+29: SUB Ev,Gv
+2a: SUB Gb,Eb
+2b: SUB Gv,Ev
+2c: SUB AL,Ib
+2d: SUB rAX,Iz
+2e: SEG=CS (Prefix)
+2f: DAS (i64)
+# 0x30 - 0x3f
+30: XOR Eb,Gb
+31: XOR Ev,Gv
+32: XOR Gb,Eb
+33: XOR Gv,Ev
+34: XOR AL,Ib
+35: XOR rAX,Iz
+36: SEG=SS (Prefix)
+37: AAA (i64)
+38: CMP Eb,Gb
+39: CMP Ev,Gv
+3a: CMP Gb,Eb
+3b: CMP Gv,Ev
+3c: CMP AL,Ib
+3d: CMP rAX,Iz
+3e: SEG=DS (Prefix)
+3f: AAS (i64)
+# 0x40 - 0x4f
+40: INC eAX (i64) | REX (o64)
+41: INC eCX (i64) | REX.B (o64)
+42: INC eDX (i64) | REX.X (o64)
+43: INC eBX (i64) | REX.XB (o64)
+44: INC eSP (i64) | REX.R (o64)
+45: INC eBP (i64) | REX.RB (o64)
+46: INC eSI (i64) | REX.RX (o64)
+47: INC eDI (i64) | REX.RXB (o64)
+48: DEC eAX (i64) | REX.W (o64)
+49: DEC eCX (i64) | REX.WB (o64)
+4a: DEC eDX (i64) | REX.WX (o64)
+4b: DEC eBX (i64) | REX.WXB (o64)
+4c: DEC eSP (i64) | REX.WR (o64)
+4d: DEC eBP (i64) | REX.WRB (o64)
+4e: DEC eSI (i64) | REX.WRX (o64)
+4f: DEC eDI (i64) | REX.WRXB (o64)
+# 0x50 - 0x5f
+50: PUSH rAX/r8 (d64)
+51: PUSH rCX/r9 (d64)
+52: PUSH rDX/r10 (d64)
+53: PUSH rBX/r11 (d64)
+54: PUSH rSP/r12 (d64)
+55: PUSH rBP/r13 (d64)
+56: PUSH rSI/r14 (d64)
+57: PUSH rDI/r15 (d64)
+58: POP rAX/r8 (d64)
+59: POP rCX/r9 (d64)
+5a: POP rDX/r10 (d64)
+5b: POP rBX/r11 (d64)
+5c: POP rSP/r12 (d64)
+5d: POP rBP/r13 (d64)
+5e: POP rSI/r14 (d64)
+5f: POP rDI/r15 (d64)
+# 0x60 - 0x6f
+60: PUSHA/PUSHAD (i64)
+61: POPA/POPAD (i64)
+62: BOUND Gv,Ma (i64)
+63: ARPL Ew,Gw (i64) | MOVSXD Gv,Ev (o64)
+64: SEG=FS (Prefix)
+65: SEG=GS (Prefix)
+66: Operand-Size (Prefix)
+67: Address-Size (Prefix)
+68: PUSH Iz (d64)
+69: IMUL Gv,Ev,Iz
+6a: PUSH Ib (d64)
+6b: IMUL Gv,Ev,Ib
+6c: INS/INSB Yb,DX
+6d: INS/INSW/INSD Yz,DX
+6e: OUTS/OUTSB DX,Xb
+6f: OUTS/OUTSW/OUTSD DX,Xz
+# 0x70 - 0x7f
+70: JO Jb
+71: JNO Jb
+72: JB/JNAE/JC Jb
+73: JNB/JAE/JNC Jb
+74: JZ/JE Jb
+75: JNZ/JNE Jb
+76: JBE/JNA Jb
+77: JNBE/JA Jb
+78: JS Jb
+79: JNS Jb
+7a: JP/JPE Jb
+7b: JNP/JPO Jb
+7c: JL/JNGE Jb
+7d: JNL/JGE Jb
+7e: JLE/JNG Jb
+7f: JNLE/JG Jb
+# 0x80 - 0x8f
+80: Grp1 Eb,Ib (1A)
+81: Grp1 Ev,Iz (1A)
+82: Grp1 Eb,Ib (1A),(i64)
+83: Grp1 Ev,Ib (1A)
+84: TEST Eb,Gb
+85: TEST Ev,Gv
+86: XCHG Eb,Gb
+87: XCHG Ev,Gv
+88: MOV Eb,Gb
+89: MOV Ev,Gv
+8a: MOV Gb,Eb
+8b: MOV Gv,Ev
+8c: MOV Ev,Sw
+8d: LEA Gv,M
+8e: MOV Sw,Ew
+8f: Grp1A (1A) | POP Ev (d64)
+# 0x90 - 0x9f
+90: NOP | PAUSE (F3) | XCHG r8,rAX
+91: XCHG rCX/r9,rAX
+92: XCHG rDX/r10,rAX
+93: XCHG rBX/r11,rAX
+94: XCHG rSP/r12,rAX
+95: XCHG rBP/r13,rAX
+96: XCHG rSI/r14,rAX
+97: XCHG rDI/r15,rAX
+98: CBW/CWDE/CDQE
+99: CWD/CDQ/CQO
+9a: CALLF Ap (i64)
+9b: FWAIT/WAIT
+9c: PUSHF/D/Q Fv (d64)
+9d: POPF/D/Q Fv (d64)
+9e: SAHF
+9f: LAHF
+# 0xa0 - 0xaf
+a0: MOV AL,Ob
+a1: MOV rAX,Ov
+a2: MOV Ob,AL
+a3: MOV Ov,rAX
+a4: MOVS/B Xb,Yb
+a5: MOVS/W/D/Q Xv,Yv
+a6: CMPS/B Xb,Yb
+a7: CMPS/W/D Xv,Yv
+a8: TEST AL,Ib
+a9: TEST rAX,Iz
+aa: STOS/B Yb,AL
+ab: STOS/W/D/Q Yv,rAX
+ac: LODS/B AL,Xb
+ad: LODS/W/D/Q rAX,Xv
+ae: SCAS/B AL,Yb
+af: SCAS/W/D/Q rAX,Xv
+# 0xb0 - 0xbf
+b0: MOV AL/R8L,Ib
+b1: MOV CL/R9L,Ib
+b2: MOV DL/R10L,Ib
+b3: MOV BL/R11L,Ib
+b4: MOV AH/R12L,Ib
+b5: MOV CH/R13L,Ib
+b6: MOV DH/R14L,Ib
+b7: MOV BH/R15L,Ib
+b8: MOV rAX/r8,Iv
+b9: MOV rCX/r9,Iv
+ba: MOV rDX/r10,Iv
+bb: MOV rBX/r11,Iv
+bc: MOV rSP/r12,Iv
+bd: MOV rBP/r13,Iv
+be: MOV rSI/r14,Iv
+bf: MOV rDI/r15,Iv
+# 0xc0 - 0xcf
+c0: Grp2 Eb,Ib (1A)
+c1: Grp2 Ev,Ib (1A)
+c2: RETN Iw (f64)
+c3: RETN
+c4: LES Gz,Mp (i64)
+c5: LDS Gz,Mp (i64)
+c6: Grp11 Eb,Ib (1A)
+c7: Grp11 Ev,Iz (1A)
+c8: ENTER Iw,Ib
+c9: LEAVE (d64)
+ca: RETF Iw
+cb: RETF
+cc: INT3
+cd: INT Ib
+ce: INTO (i64)
+cf: IRET/D/Q
+# 0xd0 - 0xdf
+d0: Grp2 Eb,1 (1A)
+d1: Grp2 Ev,1 (1A)
+d2: Grp2 Eb,CL (1A)
+d3: Grp2 Ev,CL (1A)
+d4: AAM Ib (i64)
+d5: AAD Ib (i64)
+d6:
+d7: XLAT/XLATB
+d8: ESC
+d9: ESC
+da: ESC
+db: ESC
+dc: ESC
+dd: ESC
+de: ESC
+df: ESC
+# 0xe0 - 0xef
+e0: LOOPNE/LOOPNZ Jb (f64)
+e1: LOOPE/LOOPZ Jb (f64)
+e2: LOOP Jb (f64)
+e3: JrCXZ Jb (f64)
+e4: IN AL,Ib
+e5: IN eAX,Ib
+e6: OUT Ib,AL
+e7: OUT Ib,eAX
+e8: CALL Jz (f64)
+e9: JMP-near Jz (f64)
+ea: JMP-far Ap (i64)
+eb: JMP-short Jb (f64)
+ec: IN AL,DX
+ed: IN eAX,DX
+ee: OUT DX,AL
+ef: OUT DX,eAX
+# 0xf0 - 0xff
+f0: LOCK (Prefix)
+f1:
+f2: REPNE (Prefix)
+f3: REP/REPE (Prefix)
+f4: HLT
+f5: CMC
+f6: Grp3_1 Eb (1A)
+f7: Grp3_2 Ev (1A)
+f8: CLC
+f9: STC
+fa: CLI
+fb: STI
+fc: CLD
+fd: STD
+fe: Grp4 (1A)
+ff: Grp5 (1A)
+EndTable
+
+Table: 2-byte opcode # First Byte is 0x0f
+Referrer: 2-byte escape
+# 0x0f 0x00-0x0f
+00: Grp6 (1A)
+01: Grp7 (1A)
+02: LAR Gv,Ew
+03: LSL Gv,Ew
+04:
+05: SYSCALL (o64)
+06: CLTS
+07: SYSRET (o64)
+08: INVD
+09: WBINVD
+0a:
+0b: UD2 (1B)
+0c:
+0d: NOP Ev
+0e:
+0f:
+# 0x0f 0x10-0x1f
+10:
+11:
+12:
+13:
+14:
+15:
+16:
+17:
+18: Grp16 (1A)
+19:
+1a:
+1b:
+1c:
+1d:
+1e:
+1f: NOP Ev
+# 0x0f 0x20-0x2f
+20: MOV Rd,Cd
+21: MOV Rd,Dd
+22: MOV Cd,Rd
+23: MOV Dd,Rd
+24:
+25:
+26:
+27:
+28: movaps Vps,Wps | movapd Vpd,Wpd (66)
+29: movaps Wps,Vps | movapd Wpd,Vpd (66)
+2a:
+2b:
+2c:
+2d:
+2e:
+2f:
+# 0x0f 0x30-0x3f
+30: WRMSR
+31: RDTSC
+32: RDMSR
+33: RDPMC
+34: SYSENTER
+35: SYSEXIT
+36:
+37: GETSEC
+38: escape # 3-byte escape 1
+39:
+3a: escape # 3-byte escape 2
+3b:
+3c:
+3d:
+3e:
+3f:
+# 0x0f 0x40-0x4f
+40: CMOVO Gv,Ev
+41: CMOVNO Gv,Ev
+42: CMOVB/C/NAE Gv,Ev
+43: CMOVAE/NB/NC Gv,Ev
+44: CMOVE/Z Gv,Ev
+45: CMOVNE/NZ Gv,Ev
+46: CMOVBE/NA Gv,Ev
+47: CMOVA/NBE Gv,Ev
+48: CMOVS Gv,Ev
+49: CMOVNS Gv,Ev
+4a: CMOVP/PE Gv,Ev
+4b: CMOVNP/PO Gv,Ev
+4c: CMOVL/NGE Gv,Ev
+4d: CMOVNL/GE Gv,Ev
+4e: CMOVLE/NG Gv,Ev
+4f: CMOVNLE/G Gv,Ev
+# 0x0f 0x50-0x5f
+50:
+51:
+52:
+53:
+54:
+55:
+56:
+57:
+58:
+59:
+5a:
+5b:
+5c:
+5d:
+5e:
+5f:
+# 0x0f 0x60-0x6f
+60:
+61:
+62:
+63:
+64:
+65:
+66:
+67:
+68:
+69:
+6a:
+6b:
+6c:
+6d:
+6e:
+6f:
+# 0x0f 0x70-0x7f
+70:
+71: Grp12 (1A)
+72: Grp13 (1A)
+73: Grp14 (1A)
+74:
+75:
+76:
+77:
+78: VMREAD Ed/q,Gd/q
+79: VMWRITE Gd/q,Ed/q
+7a:
+7b:
+7c:
+7d:
+7e:
+7f:
+# 0x0f 0x80-0x8f
+80: JO Jz (f64)
+81: JNO Jz (f64)
+82: JB/JNAE/JC Jz (f64)
+83: JNB/JAE/JNC Jz (f64)
+84: JZ/JE Jz (f64)
+85: JNZ/JNE Jz (f64)
+86: JBE/JNA Jz (f64)
+87: JNBE/JA Jz (f64)
+88: JS Jz (f64)
+89: JNS Jz (f64)
+8a: JP/JPE Jz (f64)
+8b: JNP/JPO Jz (f64)
+8c: JL/JNGE Jz (f64)
+8d: JNL/JGE Jz (f64)
+8e: JLE/JNG Jz (f64)
+8f: JNLE/JG Jz (f64)
+# 0x0f 0x90-0x9f
+90: SETO Eb
+91: SETNO Eb
+92: SETB/C/NAE Eb
+93: SETAE/NB/NC Eb
+94: SETE/Z Eb
+95: SETNE/NZ Eb
+96: SETBE/NA Eb
+97: SETA/NBE Eb
+98: SETS Eb
+99: SETNS Eb
+9a: SETP/PE Eb
+9b: SETNP/PO Eb
+9c: SETL/NGE Eb
+9d: SETNL/GE Eb
+9e: SETLE/NG Eb
+9f: SETNLE/G Eb
+# 0x0f 0xa0-0xaf
+a0: PUSH FS (d64)
+a1: POP FS (d64)
+a2: CPUID
+a3: BT Ev,Gv
+a4: SHLD Ev,Gv,Ib
+a5: SHLD Ev,Gv,CL
+a6:
+a7: GrpRNG
+a8: PUSH GS (d64)
+a9: POP GS (d64)
+aa: RSM
+ab: BTS Ev,Gv
+ac: SHRD Ev,Gv,Ib
+ad: SHRD Ev,Gv,CL
+ae: Grp15 (1A),(1C)
+af: IMUL Gv,Ev
+# 0x0f 0xb0-0xbf
+b0: CMPXCHG Eb,Gb
+b1: CMPXCHG Ev,Gv
+b2: LSS Gv,Mp
+b3: BTR Ev,Gv
+b4: LFS Gv,Mp
+b5: LGS Gv,Mp
+b6: MOVZX Gv,Eb
+b7: MOVZX Gv,Ew
+b8: JMPE | POPCNT Gv,Ev (F3)
+b9: Grp10 (1A)
+ba: Grp8 Ev,Ib (1A)
+bb: BTC Ev,Gv
+bc: BSF Gv,Ev
+bd: BSR Gv,Ev
+be: MOVSX Gv,Eb
+bf: MOVSX Gv,Ew
+# 0x0f 0xc0-0xcf
+c0: XADD Eb,Gb
+c1: XADD Ev,Gv
+c2:
+c3: movnti Md/q,Gd/q
+c4:
+c5:
+c6:
+c7: Grp9 (1A)
+c8: BSWAP RAX/EAX/R8/R8D
+c9: BSWAP RCX/ECX/R9/R9D
+ca: BSWAP RDX/EDX/R10/R10D
+cb: BSWAP RBX/EBX/R11/R11D
+cc: BSWAP RSP/ESP/R12/R12D
+cd: BSWAP RBP/EBP/R13/R13D
+ce: BSWAP RSI/ESI/R14/R14D
+cf: BSWAP RDI/EDI/R15/R15D
+# 0x0f 0xd0-0xdf
+d0:
+d1:
+d2:
+d3:
+d4:
+d5:
+d6:
+d7:
+d8:
+d9:
+da:
+db:
+dc:
+dd:
+de:
+df:
+# 0x0f 0xe0-0xef
+e0:
+e1:
+e2:
+e3:
+e4:
+e5:
+e6:
+e7:
+e8:
+e9:
+ea:
+eb:
+ec:
+ed:
+ee:
+ef:
+# 0x0f 0xf0-0xff
+f0:
+f1:
+f2:
+f3:
+f4:
+f5:
+f6:
+f7:
+f8:
+f9:
+fa:
+fb:
+fc:
+fd:
+fe:
+ff:
+EndTable
+
+Table: 3-byte opcode 1
+Referrer: 3-byte escape 1
+80: INVEPT Gd/q,Mdq (66)
+81: INVPID Gd/q,Mdq (66)
+f0: MOVBE Gv,Mv | CRC32 Gd,Eb (F2)
+f1: MOVBE Mv,Gv | CRC32 Gd,Ev (F2)
+EndTable
+
+Table: 3-byte opcode 2
+Referrer: 3-byte escape 2
+# all opcode is for SSE
+EndTable
+
+GrpTable: Grp1
+0: ADD
+1: OR
+2: ADC
+3: SBB
+4: AND
+5: SUB
+6: XOR
+7: CMP
+EndTable
+
+GrpTable: Grp1A
+0: POP
+EndTable
+
+GrpTable: Grp2
+0: ROL
+1: ROR
+2: RCL
+3: RCR
+4: SHL/SAL
+5: SHR
+6:
+7: SAR
+EndTable
+
+GrpTable: Grp3_1
+0: TEST Eb,Ib
+1:
+2: NOT Eb
+3: NEG Eb
+4: MUL AL,Eb
+5: IMUL AL,Eb
+6: DIV AL,Eb
+7: IDIV AL,Eb
+EndTable
+
+GrpTable: Grp3_2
+0: TEST Ev,Iz
+1:
+2: NOT Ev
+3: NEG Ev
+4: MUL rAX,Ev
+5: IMUL rAX,Ev
+6: DIV rAX,Ev
+7: IDIV rAX,Ev
+EndTable
+
+GrpTable: Grp4
+0: INC Eb
+1: DEC Eb
+EndTable
+
+GrpTable: Grp5
+0: INC Ev
+1: DEC Ev
+2: CALLN Ev (f64)
+3: CALLF Ep
+4: JMPN Ev (f64)
+5: JMPF Ep
+6: PUSH Ev (d64)
+7:
+EndTable
+
+GrpTable: Grp6
+0: SLDT Rv/Mw
+1: STR Rv/Mw
+2: LLDT Ew
+3: LTR Ew
+4: VERR Ew
+5: VERW Ew
+EndTable
+
+GrpTable: Grp7
+0: SGDT Ms | VMCALL (001),(11B) | VMLAUNCH (010),(11B) | VMRESUME (011),(11B) | VMXOFF (100),(11B)
+1: SIDT Ms | MONITOR (000),(11B) | MWAIT (001)
+2: LGDT Ms | XGETBV (000),(11B) | XSETBV (001),(11B)
+3: LIDT Ms
+4: SMSW Mw/Rv
+5:
+6: LMSW Ew
+7: INVLPG Mb | SWAPGS (o64),(000),(11B) | RDTSCP (001),(11B)
+EndTable
+
+GrpTable: Grp8
+4: BT
+5: BTS
+6: BTR
+7: BTC
+EndTable
+
+GrpTable: Grp9
+1: CMPXCHG8B/16B Mq/Mdq
+6: VMPTRLD Mq | VMCLEAR Mq (66) | VMXON Mq (F3)
+7: VMPTRST Mq
+EndTable
+
+GrpTable: Grp10
+EndTable
+
+GrpTable: Grp11
+0: MOV
+EndTable
+
+GrpTable: Grp12
+EndTable
+
+GrpTable: Grp13
+EndTable
+
+GrpTable: Grp14
+EndTable
+
+GrpTable: Grp15
+0: fxsave
+1: fxstor
+2: ldmxcsr
+3: stmxcsr
+4: XSAVE
+5: XRSTOR | lfence (11B)
+6: mfence (11B)
+7: clflush | sfence (11B)
+EndTable
+
+GrpTable: Grp16
+0: prefetch NTA
+1: prefetch T0
+2: prefetch T1
+3: prefetch T2
+EndTable
+
+GrpTable: GrpRNG
+0: xstore-rng
+1: xcrypt-ecb
+2: xcrypt-cbc
+4: xcrypt-cfb
+5: xcrypt-ofb
+EndTable
diff --git a/arch/x86/tools/gen-insn-attr-x86.awk b/arch/x86/tools/gen-insn-attr-x86.awk
new file mode 100644
index 0000000..93b62c9
--- /dev/null
+++ b/arch/x86/tools/gen-insn-attr-x86.awk
@@ -0,0 +1,314 @@
+#!/bin/awk -f
+# gen-insn-attr-x86.awk: Instruction attribute table generator
+# Written by Masami Hiramatsu <mhiramat@redhat.com>
+#
+# Usage: awk -f gen-insn-attr-x86.awk x86-opcode-map.txt > inat-tables.c
+
+BEGIN {
+ print "/* x86 opcode map generated from x86-opcode-map.txt */"
+ print "/* Do not change this code. */"
+ ggid = 1
+ geid = 1
+
+ opnd_expr = "^[[:alpha:]]"
+ ext_expr = "^\\("
+ sep_expr = "^\\|$"
+ group_expr = "^Grp[[:alnum:]]+"
+
+ imm_expr = "^[IJAO][[:lower:]]"
+ imm_flag["Ib"] = "INAT_MAKE_IMM(INAT_IMM_BYTE)"
+ imm_flag["Jb"] = "INAT_MAKE_IMM(INAT_IMM_BYTE)"
+ imm_flag["Iw"] = "INAT_MAKE_IMM(INAT_IMM_WORD)"
+ imm_flag["Id"] = "INAT_MAKE_IMM(INAT_IMM_DWORD)"
+ imm_flag["Iq"] = "INAT_MAKE_IMM(INAT_IMM_QWORD)"
+ imm_flag["Ap"] = "INAT_MAKE_IMM(INAT_IMM_PTR)"
+ imm_flag["Iz"] = "INAT_MAKE_IMM(INAT_IMM_VWORD32)"
+ imm_flag["Jz"] = "INAT_MAKE_IMM(INAT_IMM_VWORD32)"
+ imm_flag["Iv"] = "INAT_MAKE_IMM(INAT_IMM_VWORD)"
+ imm_flag["Ob"] = "INAT_MOFFSET"
+ imm_flag["Ov"] = "INAT_MOFFSET"
+
+ modrm_expr = "^([CDEGMNPQRSUVW][[:lower:]]+|NTA|T[012])"
+ force64_expr = "\\([df]64\\)"
+ rex_expr = "^REX(\\.[XRWB]+)*"
+ fpu_expr = "^ESC" # TODO
+
+ lprefix1_expr = "\\(66\\)"
+ delete lptable1
+ lprefix2_expr = "\\(F2\\)"
+ delete lptable2
+ lprefix3_expr = "\\(F3\\)"
+ delete lptable3
+ max_lprefix = 4
+
+ prefix_expr = "\\(Prefix\\)"
+ prefix_num["Operand-Size"] = "INAT_PFX_OPNDSZ"
+ prefix_num["REPNE"] = "INAT_PFX_REPNE"
+ prefix_num["REP/REPE"] = "INAT_PFX_REPE"
+ prefix_num["LOCK"] = "INAT_PFX_LOCK"
+ prefix_num["SEG=CS"] = "INAT_PFX_CS"
+ prefix_num["SEG=DS"] = "INAT_PFX_DS"
+ prefix_num["SEG=ES"] = "INAT_PFX_ES"
+ prefix_num["SEG=FS"] = "INAT_PFX_FS"
+ prefix_num["SEG=GS"] = "INAT_PFX_GS"
+ prefix_num["SEG=SS"] = "INAT_PFX_SS"
+ prefix_num["Address-Size"] = "INAT_PFX_ADDRSZ"
+
+ delete table
+ delete etable
+ delete gtable
+ eid = -1
+ gid = -1
+}
+
+function semantic_error(msg) {
+ print "Semantic error at " NR ": " msg > "/dev/stderr"
+ exit 1
+}
+
+function debug(msg) {
+ print "DEBUG: " msg
+}
+
+function array_size(arr, i,c) {
+ c = 0
+ for (i in arr)
+ c++
+ return c
+}
+
+/^Table:/ {
+ print "/* " $0 " */"
+}
+
+/^Referrer:/ {
+ if (NF == 1) {
+ # primary opcode table
+ tname = "inat_primary_table"
+ eid = -1
+ } else {
+ # escape opcode table
+ ref = ""
+ for (i = 2; i <= NF; i++)
+ ref = ref $i
+ eid = escape[ref]
+ tname = sprintf("inat_escape_table_%d", eid)
+ }
+}
+
+/^GrpTable:/ {
+ print "/* " $0 " */"
+ if (!($2 in group))
+ semantic_error("No group: " $2 )
+ gid = group[$2]
+ tname = "inat_group_table_" gid
+}
+
+function print_table(tbl,name,fmt,n)
+{
+ print "const insn_attr_t " name " = {"
+ for (i = 0; i < n; i++) {
+ id = sprintf(fmt, i)
+ if (tbl[id])
+ print " [" id "] = " tbl[id] ","
+ }
+ print "};"
+}
+
+/^EndTable/ {
+ if (gid != -1) {
+ # print group tables
+ if (array_size(table) != 0) {
+ print_table(table, tname "[INAT_GROUP_TABLE_SIZE]",
+ "0x%x", 8)
+ gtable[gid,0] = tname
+ }
+ if (array_size(lptable1) != 0) {
+ print_table(lptable1, tname "_1[INAT_GROUP_TABLE_SIZE]",
+ "0x%x", 8)
+ gtable[gid,1] = tname "_1"
+ }
+ if (array_size(lptable2) != 0) {
+ print_table(lptable2, tname "_2[INAT_GROUP_TABLE_SIZE]",
+ "0x%x", 8)
+ gtable[gid,2] = tname "_2"
+ }
+ if (array_size(lptable3) != 0) {
+ print_table(lptable3, tname "_3[INAT_GROUP_TABLE_SIZE]",
+ "0x%x", 8)
+ gtable[gid,3] = tname "_3"
+ }
+ } else {
+ # print primary/escaped tables
+ if (array_size(table) != 0) {
+ print_table(table, tname "[INAT_OPCODE_TABLE_SIZE]",
+ "0x%02x", 256)
+ etable[eid,0] = tname
+ }
+ if (array_size(lptable1) != 0) {
+ print_table(lptable1,tname "_1[INAT_OPCODE_TABLE_SIZE]",
+ "0x%02x", 256)
+ etable[eid,1] = tname "_1"
+ }
+ if (array_size(lptable2) != 0) {
+ print_table(lptable2,tname "_2[INAT_OPCODE_TABLE_SIZE]",
+ "0x%02x", 256)
+ etable[eid,2] = tname "_2"
+ }
+ if (array_size(lptable3) != 0) {
+ print_table(lptable3,tname "_3[INAT_OPCODE_TABLE_SIZE]",
+ "0x%02x", 256)
+ etable[eid,3] = tname "_3"
+ }
+ }
+ print ""
+ delete table
+ delete lptable1
+ delete lptable2
+ delete lptable3
+ gid = -1
+ eid = -1
+}
+
+function add_flags(old,new) {
+ if (old && new)
+ return old " | " new
+ else if (old)
+ return old
+ else
+ return new
+}
+
+# convert operands to flags.
+function convert_operands(opnd, i,imm,mod)
+{
+ imm = null
+ mod = null
+ for (i in opnd) {
+ i = opnd[i]
+ if (match(i, imm_expr) == 1) {
+ if (!imm_flag[i])
+ semantic_error("Unknown imm opnd: " i)
+ if (imm) {
+ if (i != "Ib")
+ semantic_error("Second IMM error")
+ imm = add_flags(imm, "INAT_SCNDIMM")
+ } else
+ imm = imm_flag[i]
+ } else if (match(i, modrm_expr))
+ mod = "INAT_MODRM"
+ }
+ return add_flags(imm, mod)
+}
+
+/^[0-9a-f]+\:/ {
+ if (NR == 1)
+ next
+ # get index
+ idx = "0x" substr($1, 1, index($1,":") - 1)
+ if (idx in table)
+ semantic_error("Redefine " idx " in " tname)
+
+ # check if escaped opcode
+ if ("escape" == $2) {
+ if ($3 != "#")
+ semantic_error("No escaped name")
+ ref = ""
+ for (i = 4; i <= NF; i++)
+ ref = ref $i
+ if (ref in escape)
+ semantic_error("Redefine escape (" ref ")")
+ escape[ref] = geid
+ geid++
+ table[idx] = "INAT_MAKE_ESCAPE(" escape[ref] ")"
+ next
+ }
+
+ variant = null
+ # converts
+ i = 2
+ while (i <= NF) {
+ opcode = $(i++)
+ delete opnds
+ ext = null
+ flags = null
+ opnd = null
+ # parse one opcode
+ if (match($i, opnd_expr)) {
+ opnd = $i
+ split($(i++), opnds, ",")
+ flags = convert_operands(opnds)
+ }
+ if (match($i, ext_expr))
+ ext = $(i++)
+ if (match($i, sep_expr))
+ i++
+ else if (i < NF)
+ semantic_error($i " is not a separator")
+
+ # check if group opcode
+ if (match(opcode, group_expr)) {
+ if (!(opcode in group)) {
+ group[opcode] = ggid
+ ggid++
+ }
+ flags = add_flags(flags, "INAT_MAKE_GROUP(" group[opcode] ")")
+ }
+ # check force(or default) 64bit
+ if (match(ext, force64_expr))
+ flags = add_flags(flags, "INAT_FORCE64")
+
+ # check REX prefix
+ if (match(opcode, rex_expr))
+ flags = add_flags(flags, "INAT_REXPFX")
+
+ # check coprocessor escape : TODO
+ if (match(opcode, fpu_expr))
+ flags = add_flags(flags, "INAT_MODRM")
+
+ # check prefixes
+ if (match(ext, prefix_expr)) {
+ if (!prefix_num[opcode])
+ semantic_error("Unknown prefix: " opcode)
+ flags = add_flags(flags, "INAT_MAKE_PREFIX(" prefix_num[opcode] ")")
+ }
+ if (length(flags) == 0)
+ continue
+ # check if last prefix
+ if (match(ext, lprefix1_expr)) {
+ lptable1[idx] = add_flags(lptable1[idx],flags)
+ variant = "INAT_VARIANT"
+ } else if (match(ext, lprefix2_expr)) {
+ lptable2[idx] = add_flags(lptable2[idx],flags)
+ variant = "INAT_VARIANT"
+ } else if (match(ext, lprefix3_expr)) {
+ lptable3[idx] = add_flags(lptable3[idx],flags)
+ variant = "INAT_VARIANT"
+ } else {
+ table[idx] = add_flags(table[idx],flags)
+ }
+ }
+ if (variant)
+ table[idx] = add_flags(table[idx],variant)
+}
+
+END {
+ # print escape opcode map's array
+ print "/* Escape opcode map array */"
+ print "const insn_attr_t const *inat_escape_tables[INAT_ESC_MAX + 1]" \
+ "[INAT_LPREFIX_MAX + 1] = {"
+ for (i = 0; i < geid; i++)
+ for (j = 0; j < max_lprefix; j++)
+ if (etable[i,j])
+ print " ["i"]["j"] = "etable[i,j]","
+ print "};\n"
+ # print group opcode map's array
+ print "/* Group opcode map array */"
+ print "const insn_attr_t const *inat_group_tables[INAT_GRP_MAX + 1]"\
+ "[INAT_LPREFIX_MAX + 1] = {"
+ for (i = 0; i < ggid; i++)
+ for (j = 0; j < max_lprefix; j++)
+ if (gtable[i,j])
+ print " ["i"]["j"] = "gtable[i,j]","
+ print "};"
+}
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America), Inc.
Software Solutions Division
e-mail: mhiramat@redhat.com
^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH -tip v14 00/12] tracing: kprobe-based event tracer and x86 instruction decoder
@ 2009-08-13 20:31 Masami Hiramatsu
2009-08-13 20:31 ` [PATCH -tip v14 03/12] kprobes: checks probe address is instruction boudary on x86 Masami Hiramatsu
` (13 more replies)
0 siblings, 14 replies; 29+ messages in thread
From: Masami Hiramatsu @ 2009-08-13 20:31 UTC (permalink / raw)
To: Ingo Molnar, Steven Rostedt, lkml
Cc: Ananth N Mavinakayanahalli, Avi Kivity, Andi Kleen,
Christoph Hellwig, Frank Ch. Eigler, Frederic Weisbecker,
H. Peter Anvin, Jason Baron, Jim Keniston, K.Prasad,
Lai Jiangshan, Li Zefan, PrzemysławPawełczyk,
Roland McGrath, Sam Ravnborg, Srikar Dronamraju, Steven Rostedt,
Tom Zanussi, Vegard Nossum, systemtap, kvm, DLE
Hi,
Here are the patches of kprobe-based event tracer for x86, version 14,
which allows you to probe various kernel events through ftrace interface.
The tracer supports per-probe filtering which allows you to set filters
on each probe and shows formats of each probe.
This version includes below fixes.
- Define remove_subsystem_dir() always (patch 6/12)
- Modify syscall_tracer because of ftrace_event_call change (patch 6/12)
- Support 'sa' argument for stack address (patch 8/12)
- Use call->data instead of container_of() macro. (patch 8/12)
- Assign new event id for each event. (patch 11/12)
Lai, this version still can not be applied on your patch ('use defined
fields to print formats') yet, since I couldn't update your patch on
the latest -tip tree.
This patchset also includes x86(-64) instruction decoder which
supports non-SSE/FP opcodes and includes x86 opcode map. The decoder
is used for finding the instruction boundaries when inserting new
kprobes. I think it will be possible to share this opcode map
with KVM's decoder.
The decoder is tested when building kernel, the test compares the
results of objdump and the decoder right after building vmlinux.
You can enable that test by CONFIG_X86_DECODER_SELFTEST=y.
This series can be applied on the latest linux-2.6.31-rc5-tip.
This supports only x86(-32/-64) (but porting it on other arch
just needs kprobes/kretprobes and register and stack access APIs).
I also made two tools for this tracer.
- Kprobe stress test script which tests kprobes on all kernel symbols to
find symbols which should be blacklisted.
- C expression to kprobes event format converter which helps you to define
kprobes events by C source code line number or function name, and local
variable name.
Enhancement ideas will be added after merging:
- .init function tracing support.
- Support primitive types(long, ulong, int, uint, etc) for args.
Kprobe-based Event Tracer
=========================
Overview
--------
This tracer is similar to the events tracer which is based on Tracepoint
infrastructure. Instead of Tracepoint, this tracer is based on kprobes(kprobe
and kretprobe). It probes anywhere where kprobes can probe(this means, all
functions body except for __kprobes functions).
Unlike the function tracer, this tracer can probe instructions inside of
kernel functions. It allows you to check which instruction has been executed.
Unlike the Tracepoint based events tracer, this tracer can add new probe points
on the fly.
Similar to the events tracer, this tracer doesn't need to be activated via
current_tracer, instead of that, just set probe points via
/sys/kernel/debug/tracing/kprobe_events. And you can set filters on each
probe events via /sys/kernel/debug/tracing/events/kprobes/<EVENT>/filter.
Synopsis of kprobe_events
-------------------------
p[:EVENT] SYMBOL[+offs|-offs]|MEMADDR [FETCHARGS] : Set a probe
r[:EVENT] SYMBOL[+0] [FETCHARGS] : Set a return probe
EVENT : Event name. If omitted, the event name is generated
based on SYMBOL+offs or MEMADDR.
SYMBOL[+offs|-offs] : Symbol+offset where the probe is inserted.
MEMADDR : Address where the probe is inserted.
FETCHARGS : Arguments. Each probe can have up to 128 args.
%REG : Fetch register REG
sN : Fetch Nth entry of stack (N >= 0)
sa : Fetch stack address.
@ADDR : Fetch memory at ADDR (ADDR should be in kernel)
@SYM[+|-offs] : Fetch memory at SYM +|- offs (SYM should be a data symbol)
aN : Fetch function argument. (N >= 0)(*)
rv : Fetch return value.(**)
ra : Fetch return address.(**)
+|-offs(FETCHARG) : fetch memory at FETCHARG +|- offs address.(***)
(*) aN may not correct on asmlinkaged functions and at the middle of
function body.
(**) only for return probe.
(***) this is useful for fetching a field of data structures.
Per-Probe Event Filtering
-------------------------
Per-probe event filtering feature allows you to set different filter on each
probe and gives you what arguments will be shown in trace buffer. If an event
name is specified right after 'p:' or 'r:' in kprobe_events, the tracer adds
an event under tracing/events/kprobes/<EVENT>, at the directory you can see
'id', 'enabled', 'format' and 'filter'.
enabled:
You can enable/disable the probe by writing 1 or 0 on it.
format:
It shows the format of this probe event. It also shows aliases of arguments
which you specified to kprobe_events.
filter:
You can write filtering rules of this event. And you can use both of aliase
names and field names for describing filters.
Event Profiling
---------------
You can check the total number of probe hits and probe miss-hits via
/sys/kernel/debug/tracing/kprobe_profile.
The first column is event name, the second is the number of probe hits,
the third is the number of probe miss-hits.
Usage examples
--------------
To add a probe as a new event, write a new definition to kprobe_events
as below.
echo p:myprobe do_sys_open a0 a1 a2 a3 > /sys/kernel/debug/tracing/kprobe_events
This sets a kprobe on the top of do_sys_open() function with recording
1st to 4th arguments as "myprobe" event.
echo r:myretprobe do_sys_open rv ra >> /sys/kernel/debug/tracing/kprobe_events
This sets a kretprobe on the return point of do_sys_open() function with
recording return value and return address as "myretprobe" event.
You can see the format of these events via
/sys/kernel/debug/tracing/events/kprobes/<EVENT>/format.
cat /sys/kernel/debug/tracing/events/kprobes/myprobe/format
name: myprobe
ID: 23
format:
field:unsigned short common_type; offset:0; size:2;
field:unsigned char common_flags; offset:2; size:1;
field:unsigned char common_preempt_count; offset:3; size:1;
field:int common_pid; offset:4; size:4;
field:int common_tgid; offset:8; size:4;
field: unsigned long ip; offset:16;tsize:8;
field: int nargs; offset:24;tsize:4;
field: unsigned long arg0; offset:32;tsize:8;
field: unsigned long arg1; offset:40;tsize:8;
field: unsigned long arg2; offset:48;tsize:8;
field: unsigned long arg3; offset:56;tsize:8;
alias: a0; original: arg0;
alias: a1; original: arg1;
alias: a2; original: arg2;
alias: a3; original: arg3;
print fmt: "%lx: 0x%lx 0x%lx 0x%lx 0x%lx", ip, arg0, arg1, arg2, arg3
You can see that the event has 4 arguments and alias expressions
corresponding to it.
echo > /sys/kernel/debug/tracing/kprobe_events
This clears all probe points. and you can see the traced information via
/sys/kernel/debug/tracing/trace.
cat /sys/kernel/debug/tracing/trace
# tracer: nop
#
# TASK-PID CPU# TIMESTAMP FUNCTION
# | | | | |
<...>-1447 [001] 1038282.286875: do_sys_open+0x0/0xd6: 0x3 0x7fffd1ec4440 0x8000 0x0
<...>-1447 [001] 1038282.286878: sys_openat+0xc/0xe <- do_sys_open: 0xfffffffffffffffe 0xffffffff81367a3a
<...>-1447 [001] 1038282.286885: do_sys_open+0x0/0xd6: 0xffffff9c 0x40413c 0x8000 0x1b6
<...>-1447 [001] 1038282.286915: sys_open+0x1b/0x1d <- do_sys_open: 0x3 0xffffffff81367a3a
<...>-1447 [001] 1038282.286969: do_sys_open+0x0/0xd6: 0xffffff9c 0x4041c6 0x98800 0x10
<...>-1447 [001] 1038282.286976: sys_open+0x1b/0x1d <- do_sys_open: 0x3 0xffffffff81367a3a
Each line shows when the kernel hits a probe, and <- SYMBOL means kernel
returns from SYMBOL(e.g. "sys_open+0x1b/0x1d <- do_sys_open" means kernel
returns from do_sys_open to sys_open+0x1b).
Thank you,
---
Masami Hiramatsu (12):
tracing: Add kprobes event profiling interface
tracing: Kprobe tracer assigns new event ids for each event
tracing: Generate names for each kprobe event automatically
tracing: Kprobe-tracer supports more than 6 arguments
tracing: add kprobe-based event tracer
tracing: Introduce TRACE_FIELD_ZERO() macro
tracing: ftrace dynamic ftrace_event_call support
x86: add pt_regs register and stack access APIs
kprobes: cleanup fix_riprel() using insn decoder on x86
kprobes: checks probe address is instruction boudary on x86
x86: x86 instruction decoder build-time selftest
x86: instruction decoder API
Documentation/trace/kprobetrace.txt | 148 ++++
arch/x86/Kconfig.debug | 9
arch/x86/Makefile | 3
arch/x86/include/asm/inat.h | 188 +++++
arch/x86/include/asm/inat_types.h | 29 +
arch/x86/include/asm/insn.h | 143 ++++
arch/x86/include/asm/ptrace.h | 62 ++
arch/x86/kernel/kprobes.c | 197 +++--
arch/x86/kernel/ptrace.c | 112 +++
arch/x86/lib/Makefile | 13
arch/x86/lib/inat.c | 78 ++
arch/x86/lib/insn.c | 464 +++++++++++++
arch/x86/lib/x86-opcode-map.txt | 719 ++++++++++++++++++++
arch/x86/tools/Makefile | 15
arch/x86/tools/distill.awk | 42 +
arch/x86/tools/gen-insn-attr-x86.awk | 314 +++++++++
arch/x86/tools/test_get_len.c | 113 +++
include/linux/ftrace_event.h | 14
include/linux/syscalls.h | 4
include/trace/ftrace.h | 19 -
include/trace/syscall.h | 8
kernel/trace/Kconfig | 12
kernel/trace/Makefile | 1
kernel/trace/trace.h | 23 +
kernel/trace/trace_event_types.h | 4
kernel/trace/trace_events.c | 119 ++-
kernel/trace/trace_export.c | 39 +
kernel/trace/trace_kprobe.c | 1234 ++++++++++++++++++++++++++++++++++
kernel/trace/trace_syscalls.c | 16
29 files changed, 3949 insertions(+), 193 deletions(-)
create mode 100644 Documentation/trace/kprobetrace.txt
create mode 100644 arch/x86/include/asm/inat.h
create mode 100644 arch/x86/include/asm/inat_types.h
create mode 100644 arch/x86/include/asm/insn.h
create mode 100644 arch/x86/lib/inat.c
create mode 100644 arch/x86/lib/insn.c
create mode 100644 arch/x86/lib/x86-opcode-map.txt
create mode 100644 arch/x86/tools/Makefile
create mode 100644 arch/x86/tools/distill.awk
create mode 100644 arch/x86/tools/gen-insn-attr-x86.awk
create mode 100644 arch/x86/tools/test_get_len.c
create mode 100644 kernel/trace/trace_kprobe.c
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America), Inc.
Software Solutions Division
e-mail: mhiramat@redhat.com
^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH -tip v14 05/12] x86: add pt_regs register and stack access APIs
2009-08-13 20:31 [PATCH -tip v14 00/12] tracing: kprobe-based event tracer and x86 instruction decoder Masami Hiramatsu
2009-08-13 20:31 ` [PATCH -tip v14 03/12] kprobes: checks probe address is instruction boudary on x86 Masami Hiramatsu
2009-08-13 20:31 ` [PATCH -tip v14 01/12] x86: instruction decoder API Masami Hiramatsu
@ 2009-08-13 20:31 ` Masami Hiramatsu
2009-08-13 20:31 ` [PATCH -tip v14 04/12] kprobes: cleanup fix_riprel() using insn decoder on x86 Masami Hiramatsu
` (10 subsequent siblings)
13 siblings, 0 replies; 29+ messages in thread
From: Masami Hiramatsu @ 2009-08-13 20:31 UTC (permalink / raw)
To: Ingo Molnar, Steven Rostedt, lkml
Cc: systemtap, kvm, DLE, Masami Hiramatsu, linux-arch,
Ananth N Mavinakayanahalli, Avi Kivity, Andi Kleen,
Christoph Hellwig, Frank Ch. Eigler, H. Peter Anvin, Ingo Molnar,
Jason Baron, Jim Keniston, K.Prasad, Lai Jiangshan, Li Zefan,
PrzemysławPawełczyk, Roland McGrath, Sam Ravnborg,
Srikar Dronamraju, Steven Rostedt, Tom Zanussi, Vegard Nossum
Add following APIs for accessing registers and stack entries from pt_regs.
These APIs are required by kprobes-based event tracer on ftrace.
Some other debugging tools might be able to use it too.
- regs_query_register_offset(const char *name)
Query the offset of "name" register.
- regs_query_register_name(unsigned int offset)
Query the name of register by its offset.
- regs_get_register(struct pt_regs *regs, unsigned int offset)
Get the value of a register by its offset.
- regs_within_kernel_stack(struct pt_regs *regs, unsigned long addr)
Check the address is in the kernel stack.
- regs_get_kernel_stack_nth(struct pt_regs *reg, unsigned int nth)
Get Nth entry of the kernel stack. (N >= 0)
- regs_get_argument_nth(struct pt_regs *reg, unsigned int nth)
Get Nth argument at function call. (N >= 0)
Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
Reviewed-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: linux-arch@vger.kernel.org
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Avi Kivity <avi@redhat.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Frank Ch. Eigler <fche@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Jason Baron <jbaron@redhat.com>
Cc: Jim Keniston <jkenisto@us.ibm.com>
Cc: K.Prasad <prasad@linux.vnet.ibm.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Przemysław Pawełczyk <przemyslaw@pawelczyk.it>
Cc: Roland McGrath <roland@redhat.com>
Cc: Sam Ravnborg <sam@ravnborg.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Tom Zanussi <tzanussi@gmail.com>
Cc: Vegard Nossum <vegard.nossum@gmail.com>
---
arch/x86/include/asm/ptrace.h | 62 +++++++++++++++++++++++
arch/x86/kernel/ptrace.c | 112 +++++++++++++++++++++++++++++++++++++++++
2 files changed, 174 insertions(+), 0 deletions(-)
diff --git a/arch/x86/include/asm/ptrace.h b/arch/x86/include/asm/ptrace.h
index 0f0d908..a3d49dd 100644
--- a/arch/x86/include/asm/ptrace.h
+++ b/arch/x86/include/asm/ptrace.h
@@ -7,6 +7,7 @@
#ifdef __KERNEL__
#include <asm/segment.h>
+#include <asm/page_types.h>
#endif
#ifndef __ASSEMBLY__
@@ -216,6 +217,67 @@ static inline unsigned long user_stack_pointer(struct pt_regs *regs)
return regs->sp;
}
+/* Query offset/name of register from its name/offset */
+extern int regs_query_register_offset(const char *name);
+extern const char *regs_query_register_name(unsigned int offset);
+#define MAX_REG_OFFSET (offsetof(struct pt_regs, ss))
+
+/**
+ * regs_get_register() - get register value from its offset
+ * @regs: pt_regs from which register value is gotten.
+ * @offset: offset number of the register.
+ *
+ * regs_get_register returns the value of a register whose offset from @regs
+ * is @offset. The @offset is the offset of the register in struct pt_regs.
+ * If @offset is bigger than MAX_REG_OFFSET, this returns 0.
+ */
+static inline unsigned long regs_get_register(struct pt_regs *regs,
+ unsigned int offset)
+{
+ if (unlikely(offset > MAX_REG_OFFSET))
+ return 0;
+ return *(unsigned long *)((unsigned long)regs + offset);
+}
+
+/**
+ * regs_within_kernel_stack() - check the address in the stack
+ * @regs: pt_regs which contains kernel stack pointer.
+ * @addr: address which is checked.
+ *
+ * regs_within_kenel_stack() checks @addr is within the kernel stack page(s).
+ * If @addr is within the kernel stack, it returns true. If not, returns false.
+ */
+static inline int regs_within_kernel_stack(struct pt_regs *regs,
+ unsigned long addr)
+{
+ return ((addr & ~(THREAD_SIZE - 1)) ==
+ (kernel_stack_pointer(regs) & ~(THREAD_SIZE - 1)));
+}
+
+/**
+ * regs_get_kernel_stack_nth() - get Nth entry of the stack
+ * @regs: pt_regs which contains kernel stack pointer.
+ * @n: stack entry number.
+ *
+ * regs_get_kernel_stack_nth() returns @n th entry of the kernel stack which
+ * is specifined by @regs. If the @n th entry is NOT in the kernel stack,
+ * this returns 0.
+ */
+static inline unsigned long regs_get_kernel_stack_nth(struct pt_regs *regs,
+ unsigned int n)
+{
+ unsigned long *addr = (unsigned long *)kernel_stack_pointer(regs);
+ addr += n;
+ if (regs_within_kernel_stack(regs, (unsigned long)addr))
+ return *addr;
+ else
+ return 0;
+}
+
+/* Get Nth argument at function call */
+extern unsigned long regs_get_argument_nth(struct pt_regs *regs,
+ unsigned int n);
+
/*
* These are defined as per linux/ptrace.h, which see.
*/
diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c
index 916082d..d270530 100644
--- a/arch/x86/kernel/ptrace.c
+++ b/arch/x86/kernel/ptrace.c
@@ -52,6 +52,118 @@ enum x86_regset {
REGSET_IOPERM32,
};
+struct pt_regs_offset {
+ const char *name;
+ int offset;
+};
+
+#define REG_OFFSET_NAME(r) {.name = #r, .offset = offsetof(struct pt_regs, r)}
+#define REG_OFFSET_END {.name = NULL, .offset = 0}
+
+static const struct pt_regs_offset regoffset_table[] = {
+#ifdef CONFIG_X86_64
+ REG_OFFSET_NAME(r15),
+ REG_OFFSET_NAME(r14),
+ REG_OFFSET_NAME(r13),
+ REG_OFFSET_NAME(r12),
+ REG_OFFSET_NAME(r11),
+ REG_OFFSET_NAME(r10),
+ REG_OFFSET_NAME(r9),
+ REG_OFFSET_NAME(r8),
+#endif
+ REG_OFFSET_NAME(bx),
+ REG_OFFSET_NAME(cx),
+ REG_OFFSET_NAME(dx),
+ REG_OFFSET_NAME(si),
+ REG_OFFSET_NAME(di),
+ REG_OFFSET_NAME(bp),
+ REG_OFFSET_NAME(ax),
+#ifdef CONFIG_X86_32
+ REG_OFFSET_NAME(ds),
+ REG_OFFSET_NAME(es),
+ REG_OFFSET_NAME(fs),
+ REG_OFFSET_NAME(gs),
+#endif
+ REG_OFFSET_NAME(orig_ax),
+ REG_OFFSET_NAME(ip),
+ REG_OFFSET_NAME(cs),
+ REG_OFFSET_NAME(flags),
+ REG_OFFSET_NAME(sp),
+ REG_OFFSET_NAME(ss),
+ REG_OFFSET_END,
+};
+
+/**
+ * regs_query_register_offset() - query register offset from its name
+ * @name: the name of a register
+ *
+ * regs_query_register_offset() returns the offset of a register in struct
+ * pt_regs from its name. If the name is invalid, this returns -EINVAL;
+ */
+int regs_query_register_offset(const char *name)
+{
+ const struct pt_regs_offset *roff;
+ for (roff = regoffset_table; roff->name != NULL; roff++)
+ if (!strcmp(roff->name, name))
+ return roff->offset;
+ return -EINVAL;
+}
+
+/**
+ * regs_query_register_name() - query register name from its offset
+ * @offset: the offset of a register in struct pt_regs.
+ *
+ * regs_query_register_name() returns the name of a register from its
+ * offset in struct pt_regs. If the @offset is invalid, this returns NULL;
+ */
+const char *regs_query_register_name(unsigned int offset)
+{
+ const struct pt_regs_offset *roff;
+ for (roff = regoffset_table; roff->name != NULL; roff++)
+ if (roff->offset == offset)
+ return roff->name;
+ return NULL;
+}
+
+static const int arg_offs_table[] = {
+#ifdef CONFIG_X86_32
+ [0] = offsetof(struct pt_regs, ax),
+ [1] = offsetof(struct pt_regs, dx),
+ [2] = offsetof(struct pt_regs, cx)
+#else /* CONFIG_X86_64 */
+ [0] = offsetof(struct pt_regs, di),
+ [1] = offsetof(struct pt_regs, si),
+ [2] = offsetof(struct pt_regs, dx),
+ [3] = offsetof(struct pt_regs, cx),
+ [4] = offsetof(struct pt_regs, r8),
+ [5] = offsetof(struct pt_regs, r9)
+#endif
+};
+
+/**
+ * regs_get_argument_nth() - get Nth argument at function call
+ * @regs: pt_regs which contains registers at function entry.
+ * @n: argument number.
+ *
+ * regs_get_argument_nth() returns @n th argument of a function call.
+ * Since usually the kernel stack will be changed right after function entry,
+ * you must use this at function entry. If the @n th entry is NOT in the
+ * kernel stack or pt_regs, this returns 0.
+ */
+unsigned long regs_get_argument_nth(struct pt_regs *regs, unsigned int n)
+{
+ if (n < ARRAY_SIZE(arg_offs_table))
+ return *((unsigned long *)regs + arg_offs_table[n]);
+ else {
+ /*
+ * The typical case: arg n is on the stack.
+ * (Note: stack[0] = return address, so skip it)
+ */
+ n -= ARRAY_SIZE(arg_offs_table);
+ return regs_get_kernel_stack_nth(regs, 1 + n);
+ }
+}
+
/*
* does not yet catch signals sent when the child dies.
* in exit.c or in signal.c.
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America), Inc.
Software Solutions Division
e-mail: mhiramat@redhat.com
^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH -tip v14 09/12] tracing: Kprobe-tracer supports more than 6 arguments
2009-08-13 20:31 [PATCH -tip v14 00/12] tracing: kprobe-based event tracer and x86 instruction decoder Masami Hiramatsu
` (6 preceding siblings ...)
2009-08-13 20:32 ` [PATCH -tip v14 10/12] tracing: Generate names for each kprobe event automatically Masami Hiramatsu
@ 2009-08-13 20:32 ` Masami Hiramatsu
2009-08-13 20:33 ` [PATCH -tip v14 12/12] tracing: Add kprobes event profiling interface Masami Hiramatsu
` (5 subsequent siblings)
13 siblings, 0 replies; 29+ messages in thread
From: Masami Hiramatsu @ 2009-08-13 20:32 UTC (permalink / raw)
To: Ingo Molnar, Steven Rostedt, lkml
Cc: systemtap, kvm, DLE, Masami Hiramatsu,
Ananth N Mavinakayanahalli, Avi Kivity, Andi Kleen,
Christoph Hellwig, Frank Ch. Eigler, Frederic Weisbecker,
H. Peter Anvin, Ingo Molnar, Jason Baron, Jim Keniston, K.Prasad,
Lai Jiangshan, Li Zefan, PrzemysławPawełczyk,
Roland McGrath, Sam Ravnborg, Srikar Dronamraju, Steven Rostedt,
Tom Zanussi, Vegard Nossum
Support up to 128 arguments for each kprobes event.
Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Avi Kivity <avi@redhat.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Frank Ch. Eigler <fche@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Jason Baron <jbaron@redhat.com>
Cc: Jim Keniston <jkenisto@us.ibm.com>
Cc: K.Prasad <prasad@linux.vnet.ibm.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Przemysław Pawełczyk <przemyslaw@pawelczyk.it>
Cc: Roland McGrath <roland@redhat.com>
Cc: Sam Ravnborg <sam@ravnborg.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Tom Zanussi <tzanussi@gmail.com>
Cc: Vegard Nossum <vegard.nossum@gmail.com>
---
Documentation/trace/kprobetrace.txt | 2 +-
kernel/trace/trace_kprobe.c | 21 +++++++++++++--------
2 files changed, 14 insertions(+), 9 deletions(-)
diff --git a/Documentation/trace/kprobetrace.txt b/Documentation/trace/kprobetrace.txt
index efff6eb..c9c09b4 100644
--- a/Documentation/trace/kprobetrace.txt
+++ b/Documentation/trace/kprobetrace.txt
@@ -32,7 +32,7 @@ Synopsis of kprobe_events
SYMBOL[+offs|-offs] : Symbol+offset where the probe is inserted.
MEMADDR : Address where the probe is inserted.
- FETCHARGS : Arguments.
+ FETCHARGS : Arguments. Each probe can have up to 128 args.
%REG : Fetch register REG
sN : Fetch Nth entry of stack (N >= 0)
sa : Fetch stack address.
diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c
index d92877a..4704e40 100644
--- a/kernel/trace/trace_kprobe.c
+++ b/kernel/trace/trace_kprobe.c
@@ -32,7 +32,7 @@
#include "trace.h"
#include "trace_output.h"
-#define TRACE_KPROBE_ARGS 6
+#define MAX_TRACE_ARGS 128
#define MAX_ARGSTR_LEN 63
/* currently, trace_kprobe only supports X86. */
@@ -184,11 +184,15 @@ struct trace_probe {
struct kretprobe rp;
};
const char *symbol; /* symbol name */
- unsigned int nr_args;
- struct fetch_func args[TRACE_KPROBE_ARGS];
struct ftrace_event_call call;
+ unsigned int nr_args;
+ struct fetch_func args[];
};
+#define SIZEOF_TRACE_PROBE(n) \
+ (offsetof(struct trace_probe, args) + \
+ (sizeof(struct fetch_func) * (n)))
+
static int kprobe_trace_func(struct kprobe *kp, struct pt_regs *regs);
static int kretprobe_trace_func(struct kretprobe_instance *ri,
struct pt_regs *regs);
@@ -263,11 +267,11 @@ static DEFINE_MUTEX(probe_lock);
static LIST_HEAD(probe_list);
static struct trace_probe *alloc_trace_probe(const char *symbol,
- const char *event)
+ const char *event, int nargs)
{
struct trace_probe *tp;
- tp = kzalloc(sizeof(struct trace_probe), GFP_KERNEL);
+ tp = kzalloc(SIZEOF_TRACE_PROBE(nargs), GFP_KERNEL);
if (!tp)
return ERR_PTR(-ENOMEM);
@@ -573,9 +577,10 @@ static int create_trace_probe(int argc, char **argv)
if (offset && is_return)
return -EINVAL;
}
+ argc -= 2; argv += 2;
/* setup a probe */
- tp = alloc_trace_probe(symbol, event);
+ tp = alloc_trace_probe(symbol, event, argc);
if (IS_ERR(tp))
return PTR_ERR(tp);
@@ -594,8 +599,8 @@ static int create_trace_probe(int argc, char **argv)
kp->addr = addr;
/* parse arguments */
- argc -= 2; argv += 2; ret = 0;
- for (i = 0; i < argc && i < TRACE_KPROBE_ARGS; i++) {
+ ret = 0;
+ for (i = 0; i < argc && i < MAX_TRACE_ARGS; i++) {
if (strlen(argv[i]) > MAX_ARGSTR_LEN) {
pr_info("Argument%d(%s) is too long.\n", i, argv[i]);
ret = -ENOSPC;
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America), Inc.
Software Solutions Division
e-mail: mhiramat@redhat.com
^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH -tip v14 10/12] tracing: Generate names for each kprobe event automatically
2009-08-13 20:31 [PATCH -tip v14 00/12] tracing: kprobe-based event tracer and x86 instruction decoder Masami Hiramatsu
` (5 preceding siblings ...)
2009-08-13 20:32 ` [PATCH -tip v14 08/12] tracing: add kprobe-based event tracer Masami Hiramatsu
@ 2009-08-13 20:32 ` Masami Hiramatsu
2009-08-13 20:32 ` [PATCH -tip v14 09/12] tracing: Kprobe-tracer supports more than 6 arguments Masami Hiramatsu
` (6 subsequent siblings)
13 siblings, 0 replies; 29+ messages in thread
From: Masami Hiramatsu @ 2009-08-13 20:32 UTC (permalink / raw)
To: Ingo Molnar, Steven Rostedt, lkml
Cc: systemtap, kvm, DLE, Masami Hiramatsu,
Ananth N Mavinakayanahalli, Avi Kivity, Andi Kleen,
Christoph Hellwig, Frank Ch. Eigler, Frederic Weisbecker,
H. Peter Anvin, Ingo Molnar, Jason Baron, Jim Keniston, K.Prasad,
Lai Jiangshan, Li Zefan, PrzemysławPawełczyk,
Roland McGrath, Sam Ravnborg, Srikar Dronamraju, Steven Rostedt,
Tom Zanussi, Vegard Nossum
Generate names for each kprobe event based on the probe point,
and remove generic k*probe event types because there is no user
of those types.
Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Avi Kivity <avi@redhat.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Frank Ch. Eigler <fche@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Jason Baron <jbaron@redhat.com>
Cc: Jim Keniston <jkenisto@us.ibm.com>
Cc: K.Prasad <prasad@linux.vnet.ibm.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Przemysław Pawełczyk <przemyslaw@pawelczyk.it>
Cc: Roland McGrath <roland@redhat.com>
Cc: Sam Ravnborg <sam@ravnborg.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Tom Zanussi <tzanussi@gmail.com>
Cc: Vegard Nossum <vegard.nossum@gmail.com>
---
Documentation/trace/kprobetrace.txt | 3 +-
kernel/trace/trace_event_types.h | 18 ----------
kernel/trace/trace_kprobe.c | 64 ++++++++++++++++++-----------------
3 files changed, 35 insertions(+), 50 deletions(-)
diff --git a/Documentation/trace/kprobetrace.txt b/Documentation/trace/kprobetrace.txt
index c9c09b4..5e59e85 100644
--- a/Documentation/trace/kprobetrace.txt
+++ b/Documentation/trace/kprobetrace.txt
@@ -28,7 +28,8 @@ Synopsis of kprobe_events
p[:EVENT] SYMBOL[+offs|-offs]|MEMADDR [FETCHARGS] : Set a probe
r[:EVENT] SYMBOL[+0] [FETCHARGS] : Set a return probe
- EVENT : Event name.
+ EVENT : Event name. If omitted, the event name is generated
+ based on SYMBOL+offs or MEMADDR.
SYMBOL[+offs|-offs] : Symbol+offset where the probe is inserted.
MEMADDR : Address where the probe is inserted.
diff --git a/kernel/trace/trace_event_types.h b/kernel/trace/trace_event_types.h
index 186b598..e74f090 100644
--- a/kernel/trace/trace_event_types.h
+++ b/kernel/trace/trace_event_types.h
@@ -175,22 +175,4 @@ TRACE_EVENT_FORMAT(kmem_free, TRACE_KMEM_FREE, kmemtrace_free_entry, ignore,
TP_RAW_FMT("type:%u call_site:%lx ptr:%p")
);
-TRACE_EVENT_FORMAT(kprobe, TRACE_KPROBE, kprobe_trace_entry, ignore,
- TRACE_STRUCT(
- TRACE_FIELD(unsigned long, ip, ip)
- TRACE_FIELD(int, nargs, nargs)
- TRACE_FIELD_ZERO(unsigned long, args)
- ),
- TP_RAW_FMT("%08lx: args:0x%lx ...")
-);
-
-TRACE_EVENT_FORMAT(kretprobe, TRACE_KRETPROBE, kretprobe_trace_entry, ignore,
- TRACE_STRUCT(
- TRACE_FIELD(unsigned long, func, func)
- TRACE_FIELD(unsigned long, ret_ip, ret_ip)
- TRACE_FIELD(int, nargs, nargs)
- TRACE_FIELD_ZERO(unsigned long, args)
- ),
- TP_RAW_FMT("%08lx <- %08lx: args:0x%lx ...")
-);
#undef TRACE_SYSTEM
diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c
index 4704e40..ec137ed 100644
--- a/kernel/trace/trace_kprobe.c
+++ b/kernel/trace/trace_kprobe.c
@@ -34,6 +34,7 @@
#define MAX_TRACE_ARGS 128
#define MAX_ARGSTR_LEN 63
+#define MAX_EVENT_NAME_LEN 64
/* currently, trace_kprobe only supports X86. */
@@ -280,11 +281,11 @@ static struct trace_probe *alloc_trace_probe(const char *symbol,
if (!tp->symbol)
goto error;
}
- if (event) {
- tp->call.name = kstrdup(event, GFP_KERNEL);
- if (!tp->call.name)
- goto error;
- }
+ if (!event)
+ goto error;
+ tp->call.name = kstrdup(event, GFP_KERNEL);
+ if (!tp->call.name)
+ goto error;
INIT_LIST_HEAD(&tp->list);
return tp;
@@ -314,7 +315,7 @@ static struct trace_probe *find_probe_event(const char *event)
struct trace_probe *tp;
list_for_each_entry(tp, &probe_list, list)
- if (tp->call.name && !strcmp(tp->call.name, event))
+ if (!strcmp(tp->call.name, event))
return tp;
return NULL;
}
@@ -330,8 +331,7 @@ static void __unregister_trace_probe(struct trace_probe *tp)
/* Unregister a trace_probe and probe_event: call with locking probe_lock */
static void unregister_trace_probe(struct trace_probe *tp)
{
- if (tp->call.name)
- unregister_probe_event(tp);
+ unregister_probe_event(tp);
__unregister_trace_probe(tp);
list_del(&tp->list);
}
@@ -360,18 +360,16 @@ static int register_trace_probe(struct trace_probe *tp)
goto end;
}
/* register as an event */
- if (tp->call.name) {
- old_tp = find_probe_event(tp->call.name);
- if (old_tp) {
- /* delete old event */
- unregister_trace_probe(old_tp);
- free_trace_probe(old_tp);
- }
- ret = register_probe_event(tp);
- if (ret) {
- pr_warning("Faild to register probe event(%d)\n", ret);
- __unregister_trace_probe(tp);
- }
+ old_tp = find_probe_event(tp->call.name);
+ if (old_tp) {
+ /* delete old event */
+ unregister_trace_probe(old_tp);
+ free_trace_probe(old_tp);
+ }
+ ret = register_probe_event(tp);
+ if (ret) {
+ pr_warning("Faild to register probe event(%d)\n", ret);
+ __unregister_trace_probe(tp);
}
list_add_tail(&tp->list, &probe_list);
end:
@@ -580,7 +578,18 @@ static int create_trace_probe(int argc, char **argv)
argc -= 2; argv += 2;
/* setup a probe */
- tp = alloc_trace_probe(symbol, event, argc);
+ if (!event) {
+ /* Make a new event name */
+ char buf[MAX_EVENT_NAME_LEN];
+ if (symbol)
+ snprintf(buf, MAX_EVENT_NAME_LEN, "%c@%s%+ld",
+ is_return ? 'r' : 'p', symbol, offset);
+ else
+ snprintf(buf, MAX_EVENT_NAME_LEN, "%c@0x%p",
+ is_return ? 'r' : 'p', addr);
+ tp = alloc_trace_probe(symbol, buf, argc);
+ } else
+ tp = alloc_trace_probe(symbol, event, argc);
if (IS_ERR(tp))
return PTR_ERR(tp);
@@ -661,8 +670,7 @@ static int probes_seq_show(struct seq_file *m, void *v)
char buf[MAX_ARGSTR_LEN + 1];
seq_printf(m, "%c", probe_is_return(tp) ? 'r' : 'p');
- if (tp->call.name)
- seq_printf(m, ":%s", tp->call.name);
+ seq_printf(m, ":%s", tp->call.name);
if (tp->symbol)
seq_printf(m, " %s%+ld", probe_symbol(tp), probe_offset(tp));
@@ -780,10 +788,7 @@ static __kprobes int kprobe_trace_func(struct kprobe *kp, struct pt_regs *regs)
struct ring_buffer_event *event;
int size, i, pc;
unsigned long irq_flags;
- struct ftrace_event_call *call = &event_kprobe;
-
- if (&tp->call.name)
- call = &tp->call;
+ struct ftrace_event_call *call = &tp->call;
local_save_flags(irq_flags);
pc = preempt_count();
@@ -815,10 +820,7 @@ static __kprobes int kretprobe_trace_func(struct kretprobe_instance *ri,
struct ring_buffer_event *event;
int size, i, pc;
unsigned long irq_flags;
- struct ftrace_event_call *call = &event_kretprobe;
-
- if (&tp->call.name)
- call = &tp->call;
+ struct ftrace_event_call *call = &tp->call;
local_save_flags(irq_flags);
pc = preempt_count();
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America), Inc.
Software Solutions Division
e-mail: mhiramat@redhat.com
^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH -tip v14 08/12] tracing: add kprobe-based event tracer
2009-08-13 20:31 [PATCH -tip v14 00/12] tracing: kprobe-based event tracer and x86 instruction decoder Masami Hiramatsu
` (4 preceding siblings ...)
2009-08-13 20:31 ` [PATCH -tip v14 02/12] x86: x86 instruction decoder build-time selftest Masami Hiramatsu
@ 2009-08-13 20:32 ` Masami Hiramatsu
2009-08-13 20:32 ` [PATCH -tip v14 10/12] tracing: Generate names for each kprobe event automatically Masami Hiramatsu
` (7 subsequent siblings)
13 siblings, 0 replies; 29+ messages in thread
From: Masami Hiramatsu @ 2009-08-13 20:32 UTC (permalink / raw)
To: Ingo Molnar, Steven Rostedt, lkml
Cc: systemtap, kvm, DLE, Masami Hiramatsu,
Ananth N Mavinakayanahalli, Avi Kivity, Andi Kleen,
Christoph Hellwig, Frank Ch. Eigler, Frederic Weisbecker,
H. Peter Anvin, Ingo Molnar, Jason Baron, Jim Keniston, K.Prasad,
Lai Jiangshan, Li Zefan, PrzemysławPawełczyk,
Roland McGrath, Sam Ravnborg, Srikar Dronamraju, Steven Rostedt,
Tom Zanussi, Vegard Nossum
Add kprobes-based event tracer on ftrace.
This tracer is similar to the events tracer which is based on Tracepoint
infrastructure. Instead of Tracepoint, this tracer is based on kprobes
(kprobe and kretprobe). It probes anywhere where kprobes can probe(this
means, all functions body except for __kprobes functions).
Similar to the events tracer, this tracer doesn't need to be activated via
current_tracer, instead of that, just set probe points via
/sys/kernel/debug/tracing/kprobe_events. And you can set filters on each
probe events via /sys/kernel/debug/tracing/events/kprobes/<EVENT>/filter.
This tracer supports following probe arguments for each probe.
%REG : Fetch register REG
sN : Fetch Nth entry of stack (N >= 0)
sa : Fetch stack address.
@ADDR : Fetch memory at ADDR (ADDR should be in kernel)
@SYM[+|-offs] : Fetch memory at SYM +|- offs (SYM should be a data symbol)
aN : Fetch function argument. (N >= 0)
rv : Fetch return value.
ra : Fetch return address.
+|-offs(FETCHARG) : fetch memory at FETCHARG +|- offs address.
See Documentation/trace/kprobetrace.txt for details.
Changes from v13:
- Support 'sa' for stack address.
- Use call->data instead of container_of() macro.
Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
Acked-by: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Avi Kivity <avi@redhat.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Frank Ch. Eigler <fche@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Jason Baron <jbaron@redhat.com>
Cc: Jim Keniston <jkenisto@us.ibm.com>
Cc: K.Prasad <prasad@linux.vnet.ibm.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Przemysław Pawełczyk <przemyslaw@pawelczyk.it>
Cc: Roland McGrath <roland@redhat.com>
Cc: Sam Ravnborg <sam@ravnborg.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Tom Zanussi <tzanussi@gmail.com>
Cc: Vegard Nossum <vegard.nossum@gmail.com>
---
Documentation/trace/kprobetrace.txt | 139 ++++
kernel/trace/Kconfig | 12
kernel/trace/Makefile | 1
kernel/trace/trace.h | 29 +
kernel/trace/trace_event_types.h | 18 +
kernel/trace/trace_kprobe.c | 1205 +++++++++++++++++++++++++++++++++++
6 files changed, 1404 insertions(+), 0 deletions(-)
create mode 100644 Documentation/trace/kprobetrace.txt
create mode 100644 kernel/trace/trace_kprobe.c
diff --git a/Documentation/trace/kprobetrace.txt b/Documentation/trace/kprobetrace.txt
new file mode 100644
index 0000000..efff6eb
--- /dev/null
+++ b/Documentation/trace/kprobetrace.txt
@@ -0,0 +1,139 @@
+ Kprobe-based Event Tracer
+ =========================
+
+ Documentation is written by Masami Hiramatsu
+
+
+Overview
+--------
+This tracer is similar to the events tracer which is based on Tracepoint
+infrastructure. Instead of Tracepoint, this tracer is based on kprobes(kprobe
+and kretprobe). It probes anywhere where kprobes can probe(this means, all
+functions body except for __kprobes functions).
+
+Unlike the function tracer, this tracer can probe instructions inside of
+kernel functions. It allows you to check which instruction has been executed.
+
+Unlike the Tracepoint based events tracer, this tracer can add and remove
+probe points on the fly.
+
+Similar to the events tracer, this tracer doesn't need to be activated via
+current_tracer, instead of that, just set probe points via
+/sys/kernel/debug/tracing/kprobe_events. And you can set filters on each
+probe events via /sys/kernel/debug/tracing/events/kprobes/<EVENT>/filter.
+
+
+Synopsis of kprobe_events
+-------------------------
+ p[:EVENT] SYMBOL[+offs|-offs]|MEMADDR [FETCHARGS] : Set a probe
+ r[:EVENT] SYMBOL[+0] [FETCHARGS] : Set a return probe
+
+ EVENT : Event name.
+ SYMBOL[+offs|-offs] : Symbol+offset where the probe is inserted.
+ MEMADDR : Address where the probe is inserted.
+
+ FETCHARGS : Arguments.
+ %REG : Fetch register REG
+ sN : Fetch Nth entry of stack (N >= 0)
+ sa : Fetch stack address.
+ @ADDR : Fetch memory at ADDR (ADDR should be in kernel)
+ @SYM[+|-offs] : Fetch memory at SYM +|- offs (SYM should be a data symbol)
+ aN : Fetch function argument. (N >= 0)(*)
+ rv : Fetch return value.(**)
+ ra : Fetch return address.(**)
+ +|-offs(FETCHARG) : fetch memory at FETCHARG +|- offs address.(***)
+
+ (*) aN may not correct on asmlinkaged functions and at the middle of
+ function body.
+ (**) only for return probe.
+ (***) this is useful for fetching a field of data structures.
+
+
+Per-Probe Event Filtering
+-------------------------
+ Per-probe event filtering feature allows you to set different filter on each
+probe and gives you what arguments will be shown in trace buffer. If an event
+name is specified right after 'p:' or 'r:' in kprobe_events, the tracer adds
+an event under tracing/events/kprobes/<EVENT>, at the directory you can see
+'id', 'enabled', 'format' and 'filter'.
+
+enabled:
+ You can enable/disable the probe by writing 1 or 0 on it.
+
+format:
+ It shows the format of this probe event. It also shows aliases of arguments
+ which you specified to kprobe_events.
+
+filter:
+ You can write filtering rules of this event. And you can use both of aliase
+ names and field names for describing filters.
+
+
+Usage examples
+--------------
+To add a probe as a new event, write a new definition to kprobe_events
+as below.
+
+ echo p:myprobe do_sys_open a0 a1 a2 a3 > /sys/kernel/debug/tracing/kprobe_events
+
+ This sets a kprobe on the top of do_sys_open() function with recording
+1st to 4th arguments as "myprobe" event.
+
+ echo r:myretprobe do_sys_open rv ra >> /sys/kernel/debug/tracing/kprobe_events
+
+ This sets a kretprobe on the return point of do_sys_open() function with
+recording return value and return address as "myretprobe" event.
+ You can see the format of these events via
+/sys/kernel/debug/tracing/events/kprobes/<EVENT>/format.
+
+ cat /sys/kernel/debug/tracing/events/kprobes/myprobe/format
+name: myprobe
+ID: 23
+format:
+ field:unsigned short common_type; offset:0; size:2;
+ field:unsigned char common_flags; offset:2; size:1;
+ field:unsigned char common_preempt_count; offset:3; size:1;
+ field:int common_pid; offset:4; size:4;
+ field:int common_tgid; offset:8; size:4;
+
+ field: unsigned long ip; offset:16;tsize:8;
+ field: int nargs; offset:24;tsize:4;
+ field: unsigned long arg0; offset:32;tsize:8;
+ field: unsigned long arg1; offset:40;tsize:8;
+ field: unsigned long arg2; offset:48;tsize:8;
+ field: unsigned long arg3; offset:56;tsize:8;
+
+ alias: a0; original: arg0;
+ alias: a1; original: arg1;
+ alias: a2; original: arg2;
+ alias: a3; original: arg3;
+
+print fmt: "%lx: 0x%lx 0x%lx 0x%lx 0x%lx", ip, arg0, arg1, arg2, arg3
+
+
+ You can see that the event has 4 arguments and alias expressions
+corresponding to it.
+
+ echo > /sys/kernel/debug/tracing/kprobe_events
+
+ This clears all probe points. and you can see the traced information via
+/sys/kernel/debug/tracing/trace.
+
+ cat /sys/kernel/debug/tracing/trace
+# tracer: nop
+#
+# TASK-PID CPU# TIMESTAMP FUNCTION
+# | | | | |
+ <...>-1447 [001] 1038282.286875: do_sys_open+0x0/0xd6: 0x3 0x7fffd1ec4440 0x8000 0x0
+ <...>-1447 [001] 1038282.286878: sys_openat+0xc/0xe <- do_sys_open: 0xfffffffffffffffe 0xffffffff81367a3a
+ <...>-1447 [001] 1038282.286885: do_sys_open+0x0/0xd6: 0xffffff9c 0x40413c 0x8000 0x1b6
+ <...>-1447 [001] 1038282.286915: sys_open+0x1b/0x1d <- do_sys_open: 0x3 0xffffffff81367a3a
+ <...>-1447 [001] 1038282.286969: do_sys_open+0x0/0xd6: 0xffffff9c 0x4041c6 0x98800 0x10
+ <...>-1447 [001] 1038282.286976: sys_open+0x1b/0x1d <- do_sys_open: 0x3 0xffffffff81367a3a
+
+
+ Each line shows when the kernel hits a probe, and <- SYMBOL means kernel
+returns from SYMBOL(e.g. "sys_open+0x1b/0x1d <- do_sys_open" means kernel
+returns from do_sys_open to sys_open+0x1b).
+
+
diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig
index 860c712..60f3401 100644
--- a/kernel/trace/Kconfig
+++ b/kernel/trace/Kconfig
@@ -445,6 +445,18 @@ config BLK_DEV_IO_TRACE
If unsure, say N.
+config KPROBE_TRACER
+ depends on KPROBES
+ depends on X86
+ bool "Trace kprobes"
+ select TRACING
+ select GENERIC_TRACER
+ help
+ This tracer probes everywhere where kprobes can probe it, and
+ records various registers and memories specified by user.
+ This also allows you to trace kprobe probe points as a dynamic
+ defined events. It provides per-probe event filtering interface.
+
config DYNAMIC_FTRACE
bool "enable/disable ftrace tracepoints dynamically"
depends on FUNCTION_TRACER
diff --git a/kernel/trace/Makefile b/kernel/trace/Makefile
index ce3b1cd..8e6884d 100644
--- a/kernel/trace/Makefile
+++ b/kernel/trace/Makefile
@@ -55,5 +55,6 @@ obj-$(CONFIG_FTRACE_SYSCALLS) += trace_syscalls.o
obj-$(CONFIG_EVENT_PROFILE) += trace_event_profile.o
obj-$(CONFIG_EVENT_TRACING) += trace_events_filter.o
obj-$(CONFIG_KSYM_TRACER) += trace_ksym.o
+obj-$(CONFIG_KPROBE_TRACER) += trace_kprobe.o
libftrace-y := ftrace.o
diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
index 606073c..4ce4525 100644
--- a/kernel/trace/trace.h
+++ b/kernel/trace/trace.h
@@ -43,6 +43,8 @@ enum trace_type {
TRACE_POWER,
TRACE_BLK,
TRACE_KSYM,
+ TRACE_KPROBE,
+ TRACE_KRETPROBE,
__TRACE_LAST_TYPE,
};
@@ -221,6 +223,29 @@ struct ksym_trace_entry {
char cmd[TASK_COMM_LEN];
};
+struct kprobe_trace_entry {
+ struct trace_entry ent;
+ unsigned long ip;
+ int nargs;
+ unsigned long args[];
+};
+
+#define SIZEOF_KPROBE_TRACE_ENTRY(n) \
+ (offsetof(struct kprobe_trace_entry, args) + \
+ (sizeof(unsigned long) * (n)))
+
+struct kretprobe_trace_entry {
+ struct trace_entry ent;
+ unsigned long func;
+ unsigned long ret_ip;
+ int nargs;
+ unsigned long args[];
+};
+
+#define SIZEOF_KRETPROBE_TRACE_ENTRY(n) \
+ (offsetof(struct kretprobe_trace_entry, args) + \
+ (sizeof(unsigned long) * (n)))
+
/*
* trace_flag_type is an enumeration that holds different
* states when a trace occurs. These are:
@@ -333,6 +358,10 @@ extern void __ftrace_bad_type(void);
IF_ASSIGN(var, ent, struct kmemtrace_free_entry, \
TRACE_KMEM_FREE); \
IF_ASSIGN(var, ent, struct ksym_trace_entry, TRACE_KSYM);\
+ IF_ASSIGN(var, ent, struct kprobe_trace_entry, \
+ TRACE_KPROBE); \
+ IF_ASSIGN(var, ent, struct kretprobe_trace_entry, \
+ TRACE_KRETPROBE); \
__ftrace_bad_type(); \
} while (0)
diff --git a/kernel/trace/trace_event_types.h b/kernel/trace/trace_event_types.h
index e74f090..186b598 100644
--- a/kernel/trace/trace_event_types.h
+++ b/kernel/trace/trace_event_types.h
@@ -175,4 +175,22 @@ TRACE_EVENT_FORMAT(kmem_free, TRACE_KMEM_FREE, kmemtrace_free_entry, ignore,
TP_RAW_FMT("type:%u call_site:%lx ptr:%p")
);
+TRACE_EVENT_FORMAT(kprobe, TRACE_KPROBE, kprobe_trace_entry, ignore,
+ TRACE_STRUCT(
+ TRACE_FIELD(unsigned long, ip, ip)
+ TRACE_FIELD(int, nargs, nargs)
+ TRACE_FIELD_ZERO(unsigned long, args)
+ ),
+ TP_RAW_FMT("%08lx: args:0x%lx ...")
+);
+
+TRACE_EVENT_FORMAT(kretprobe, TRACE_KRETPROBE, kretprobe_trace_entry, ignore,
+ TRACE_STRUCT(
+ TRACE_FIELD(unsigned long, func, func)
+ TRACE_FIELD(unsigned long, ret_ip, ret_ip)
+ TRACE_FIELD(int, nargs, nargs)
+ TRACE_FIELD_ZERO(unsigned long, args)
+ ),
+ TP_RAW_FMT("%08lx <- %08lx: args:0x%lx ...")
+);
#undef TRACE_SYSTEM
diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c
new file mode 100644
index 0000000..d92877a
--- /dev/null
+++ b/kernel/trace/trace_kprobe.c
@@ -0,0 +1,1205 @@
+/*
+ * kprobe based kernel tracer
+ *
+ * Created by Masami Hiramatsu <mhiramat@redhat.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+
+#include <linux/module.h>
+#include <linux/uaccess.h>
+#include <linux/kprobes.h>
+#include <linux/seq_file.h>
+#include <linux/slab.h>
+#include <linux/smp.h>
+#include <linux/debugfs.h>
+#include <linux/types.h>
+#include <linux/string.h>
+#include <linux/ctype.h>
+#include <linux/ptrace.h>
+
+#include "trace.h"
+#include "trace_output.h"
+
+#define TRACE_KPROBE_ARGS 6
+#define MAX_ARGSTR_LEN 63
+
+/* currently, trace_kprobe only supports X86. */
+
+struct fetch_func {
+ unsigned long (*func)(struct pt_regs *, void *);
+ void *data;
+};
+
+static __kprobes unsigned long call_fetch(struct fetch_func *f,
+ struct pt_regs *regs)
+{
+ return f->func(regs, f->data);
+}
+
+/* fetch handlers */
+static __kprobes unsigned long fetch_register(struct pt_regs *regs,
+ void *offset)
+{
+ return regs_get_register(regs, (unsigned int)((unsigned long)offset));
+}
+
+static __kprobes unsigned long fetch_stack(struct pt_regs *regs,
+ void *num)
+{
+ return regs_get_kernel_stack_nth(regs,
+ (unsigned int)((unsigned long)num));
+}
+
+static __kprobes unsigned long fetch_memory(struct pt_regs *regs, void *addr)
+{
+ unsigned long retval;
+
+ if (probe_kernel_address(addr, retval))
+ return 0;
+ return retval;
+}
+
+static __kprobes unsigned long fetch_argument(struct pt_regs *regs, void *num)
+{
+ return regs_get_argument_nth(regs, (unsigned int)((unsigned long)num));
+}
+
+static __kprobes unsigned long fetch_retvalue(struct pt_regs *regs,
+ void *dummy)
+{
+ return regs_return_value(regs);
+}
+
+static __kprobes unsigned long fetch_ip(struct pt_regs *regs, void *dummy)
+{
+ return instruction_pointer(regs);
+}
+
+static __kprobes unsigned long fetch_stack_address(struct pt_regs *regs,
+ void *dummy)
+{
+ return kernel_stack_pointer(regs);
+}
+
+/* Memory fetching by symbol */
+struct symbol_cache {
+ char *symbol;
+ long offset;
+ unsigned long addr;
+};
+
+static unsigned long update_symbol_cache(struct symbol_cache *sc)
+{
+ sc->addr = (unsigned long)kallsyms_lookup_name(sc->symbol);
+ if (sc->addr)
+ sc->addr += sc->offset;
+ return sc->addr;
+}
+
+static void free_symbol_cache(struct symbol_cache *sc)
+{
+ kfree(sc->symbol);
+ kfree(sc);
+}
+
+static struct symbol_cache *alloc_symbol_cache(const char *sym, long offset)
+{
+ struct symbol_cache *sc;
+
+ if (!sym || strlen(sym) == 0)
+ return NULL;
+ sc = kzalloc(sizeof(struct symbol_cache), GFP_KERNEL);
+ if (!sc)
+ return NULL;
+
+ sc->symbol = kstrdup(sym, GFP_KERNEL);
+ if (!sc->symbol) {
+ kfree(sc);
+ return NULL;
+ }
+ sc->offset = offset;
+
+ update_symbol_cache(sc);
+ return sc;
+}
+
+static __kprobes unsigned long fetch_symbol(struct pt_regs *regs, void *data)
+{
+ struct symbol_cache *sc = data;
+
+ if (sc->addr)
+ return fetch_memory(regs, (void *)sc->addr);
+ else
+ return 0;
+}
+
+/* Special indirect memory access interface */
+struct indirect_fetch_data {
+ struct fetch_func orig;
+ long offset;
+};
+
+static __kprobes unsigned long fetch_indirect(struct pt_regs *regs, void *data)
+{
+ struct indirect_fetch_data *ind = data;
+ unsigned long addr;
+
+ addr = call_fetch(&ind->orig, regs);
+ if (addr) {
+ addr += ind->offset;
+ return fetch_memory(regs, (void *)addr);
+ } else
+ return 0;
+}
+
+static __kprobes void free_indirect_fetch_data(struct indirect_fetch_data *data)
+{
+ if (data->orig.func == fetch_indirect)
+ free_indirect_fetch_data(data->orig.data);
+ else if (data->orig.func == fetch_symbol)
+ free_symbol_cache(data->orig.data);
+ kfree(data);
+}
+
+/**
+ * kprobe_trace_core
+ */
+
+struct trace_probe {
+ struct list_head list;
+ union {
+ struct kprobe kp;
+ struct kretprobe rp;
+ };
+ const char *symbol; /* symbol name */
+ unsigned int nr_args;
+ struct fetch_func args[TRACE_KPROBE_ARGS];
+ struct ftrace_event_call call;
+};
+
+static int kprobe_trace_func(struct kprobe *kp, struct pt_regs *regs);
+static int kretprobe_trace_func(struct kretprobe_instance *ri,
+ struct pt_regs *regs);
+
+static __kprobes int probe_is_return(struct trace_probe *tp)
+{
+ return (tp->rp.handler == kretprobe_trace_func);
+}
+
+static __kprobes const char *probe_symbol(struct trace_probe *tp)
+{
+ return tp->symbol ? tp->symbol : "unknown";
+}
+
+static __kprobes long probe_offset(struct trace_probe *tp)
+{
+ return (probe_is_return(tp)) ? tp->rp.kp.offset : tp->kp.offset;
+}
+
+static __kprobes void *probe_address(struct trace_probe *tp)
+{
+ return (probe_is_return(tp)) ? tp->rp.kp.addr : tp->kp.addr;
+}
+
+static int trace_arg_string(char *buf, size_t n, struct fetch_func *ff)
+{
+ int ret = -EINVAL;
+
+ if (ff->func == fetch_argument)
+ ret = snprintf(buf, n, "a%lu", (unsigned long)ff->data);
+ else if (ff->func == fetch_register) {
+ const char *name;
+ name = regs_query_register_name((unsigned int)((long)ff->data));
+ ret = snprintf(buf, n, "%%%s", name);
+ } else if (ff->func == fetch_stack)
+ ret = snprintf(buf, n, "s%lu", (unsigned long)ff->data);
+ else if (ff->func == fetch_memory)
+ ret = snprintf(buf, n, "@0x%p", ff->data);
+ else if (ff->func == fetch_symbol) {
+ struct symbol_cache *sc = ff->data;
+ ret = snprintf(buf, n, "@%s%+ld", sc->symbol, sc->offset);
+ } else if (ff->func == fetch_retvalue)
+ ret = snprintf(buf, n, "rv");
+ else if (ff->func == fetch_ip)
+ ret = snprintf(buf, n, "ra");
+ else if (ff->func == fetch_stack_address)
+ ret = snprintf(buf, n, "sa");
+ else if (ff->func == fetch_indirect) {
+ struct indirect_fetch_data *id = ff->data;
+ size_t l = 0;
+ ret = snprintf(buf, n, "%+ld(", id->offset);
+ if (ret >= n)
+ goto end;
+ l += ret;
+ ret = trace_arg_string(buf + l, n - l, &id->orig);
+ if (ret < 0)
+ goto end;
+ l += ret;
+ ret = snprintf(buf + l, n - l, ")");
+ ret += l;
+ }
+end:
+ if (ret >= n)
+ return -ENOSPC;
+ return ret;
+}
+
+static int register_probe_event(struct trace_probe *tp);
+static void unregister_probe_event(struct trace_probe *tp);
+
+static DEFINE_MUTEX(probe_lock);
+static LIST_HEAD(probe_list);
+
+static struct trace_probe *alloc_trace_probe(const char *symbol,
+ const char *event)
+{
+ struct trace_probe *tp;
+
+ tp = kzalloc(sizeof(struct trace_probe), GFP_KERNEL);
+ if (!tp)
+ return ERR_PTR(-ENOMEM);
+
+ if (symbol) {
+ tp->symbol = kstrdup(symbol, GFP_KERNEL);
+ if (!tp->symbol)
+ goto error;
+ }
+ if (event) {
+ tp->call.name = kstrdup(event, GFP_KERNEL);
+ if (!tp->call.name)
+ goto error;
+ }
+
+ INIT_LIST_HEAD(&tp->list);
+ return tp;
+error:
+ kfree(tp->symbol);
+ kfree(tp);
+ return ERR_PTR(-ENOMEM);
+}
+
+static void free_trace_probe(struct trace_probe *tp)
+{
+ int i;
+
+ for (i = 0; i < tp->nr_args; i++)
+ if (tp->args[i].func == fetch_symbol)
+ free_symbol_cache(tp->args[i].data);
+ else if (tp->args[i].func == fetch_indirect)
+ free_indirect_fetch_data(tp->args[i].data);
+
+ kfree(tp->call.name);
+ kfree(tp->symbol);
+ kfree(tp);
+}
+
+static struct trace_probe *find_probe_event(const char *event)
+{
+ struct trace_probe *tp;
+
+ list_for_each_entry(tp, &probe_list, list)
+ if (tp->call.name && !strcmp(tp->call.name, event))
+ return tp;
+ return NULL;
+}
+
+static void __unregister_trace_probe(struct trace_probe *tp)
+{
+ if (probe_is_return(tp))
+ unregister_kretprobe(&tp->rp);
+ else
+ unregister_kprobe(&tp->kp);
+}
+
+/* Unregister a trace_probe and probe_event: call with locking probe_lock */
+static void unregister_trace_probe(struct trace_probe *tp)
+{
+ if (tp->call.name)
+ unregister_probe_event(tp);
+ __unregister_trace_probe(tp);
+ list_del(&tp->list);
+}
+
+/* Register a trace_probe and probe_event */
+static int register_trace_probe(struct trace_probe *tp)
+{
+ struct trace_probe *old_tp;
+ int ret;
+
+ mutex_lock(&probe_lock);
+
+ if (probe_is_return(tp))
+ ret = register_kretprobe(&tp->rp);
+ else
+ ret = register_kprobe(&tp->kp);
+
+ if (ret) {
+ pr_warning("Could not insert probe(%d)\n", ret);
+ if (ret == -EILSEQ) {
+ pr_warning("Probing address(0x%p) is not an "
+ "instruction boundary.\n",
+ probe_address(tp));
+ ret = -EINVAL;
+ }
+ goto end;
+ }
+ /* register as an event */
+ if (tp->call.name) {
+ old_tp = find_probe_event(tp->call.name);
+ if (old_tp) {
+ /* delete old event */
+ unregister_trace_probe(old_tp);
+ free_trace_probe(old_tp);
+ }
+ ret = register_probe_event(tp);
+ if (ret) {
+ pr_warning("Faild to register probe event(%d)\n", ret);
+ __unregister_trace_probe(tp);
+ }
+ }
+ list_add_tail(&tp->list, &probe_list);
+end:
+ mutex_unlock(&probe_lock);
+ return ret;
+}
+
+/* Split symbol and offset. */
+static int split_symbol_offset(char *symbol, long *offset)
+{
+ char *tmp;
+ int ret;
+
+ if (!offset)
+ return -EINVAL;
+
+ tmp = strchr(symbol, '+');
+ if (!tmp)
+ tmp = strchr(symbol, '-');
+
+ if (tmp) {
+ /* skip sign because strict_strtol doesn't accept '+' */
+ ret = strict_strtol(tmp + 1, 0, offset);
+ if (ret)
+ return ret;
+ if (*tmp == '-')
+ *offset = -(*offset);
+ *tmp = '\0';
+ } else
+ *offset = 0;
+ return 0;
+}
+
+#define PARAM_MAX_ARGS 16
+#define PARAM_MAX_STACK (THREAD_SIZE / sizeof(unsigned long))
+
+static int parse_trace_arg(char *arg, struct fetch_func *ff, int is_return)
+{
+ int ret = 0;
+ unsigned long param;
+ long offset;
+ char *tmp;
+
+ switch (arg[0]) {
+ case 'a': /* argument */
+ ret = strict_strtoul(arg + 1, 10, ¶m);
+ if (ret || param > PARAM_MAX_ARGS)
+ ret = -EINVAL;
+ else {
+ ff->func = fetch_argument;
+ ff->data = (void *)param;
+ }
+ break;
+ case 'r': /* retval or retaddr */
+ if (is_return && arg[1] == 'v') {
+ ff->func = fetch_retvalue;
+ ff->data = NULL;
+ } else if (is_return && arg[1] == 'a') {
+ ff->func = fetch_ip;
+ ff->data = NULL;
+ } else
+ ret = -EINVAL;
+ break;
+ case '%': /* named register */
+ ret = regs_query_register_offset(arg + 1);
+ if (ret >= 0) {
+ ff->func = fetch_register;
+ ff->data = (void *)(unsigned long)ret;
+ ret = 0;
+ }
+ break;
+ case 's': /* stack */
+ if (arg[1] == 'a') {
+ ff->func = fetch_stack_address;
+ ff->data = NULL;
+ } else {
+ ret = strict_strtoul(arg + 1, 10, ¶m);
+ if (ret || param > PARAM_MAX_STACK)
+ ret = -EINVAL;
+ else {
+ ff->func = fetch_stack;
+ ff->data = (void *)param;
+ }
+ }
+ break;
+ case '@': /* memory or symbol */
+ if (isdigit(arg[1])) {
+ ret = strict_strtoul(arg + 1, 0, ¶m);
+ if (ret)
+ break;
+ ff->func = fetch_memory;
+ ff->data = (void *)param;
+ } else {
+ ret = split_symbol_offset(arg + 1, &offset);
+ if (ret)
+ break;
+ ff->data = alloc_symbol_cache(arg + 1,
+ offset);
+ if (ff->data)
+ ff->func = fetch_symbol;
+ else
+ ret = -EINVAL;
+ }
+ break;
+ case '+': /* indirect memory */
+ case '-':
+ tmp = strchr(arg, '(');
+ if (!tmp) {
+ ret = -EINVAL;
+ break;
+ }
+ *tmp = '\0';
+ ret = strict_strtol(arg + 1, 0, &offset);
+ if (ret)
+ break;
+ if (arg[0] == '-')
+ offset = -offset;
+ arg = tmp + 1;
+ tmp = strrchr(arg, ')');
+ if (tmp) {
+ struct indirect_fetch_data *id;
+ *tmp = '\0';
+ id = kzalloc(sizeof(struct indirect_fetch_data),
+ GFP_KERNEL);
+ if (!id)
+ return -ENOMEM;
+ id->offset = offset;
+ ret = parse_trace_arg(arg, &id->orig, is_return);
+ if (ret)
+ kfree(id);
+ else {
+ ff->func = fetch_indirect;
+ ff->data = (void *)id;
+ }
+ } else
+ ret = -EINVAL;
+ break;
+ default:
+ /* TODO: support custom handler */
+ ret = -EINVAL;
+ }
+ return ret;
+}
+
+static int create_trace_probe(int argc, char **argv)
+{
+ /*
+ * Argument syntax:
+ * - Add kprobe: p[:EVENT] SYMBOL[+OFFS|-OFFS]|ADDRESS [FETCHARGS]
+ * - Add kretprobe: r[:EVENT] SYMBOL[+0] [FETCHARGS]
+ * Fetch args:
+ * aN : fetch Nth of function argument. (N:0-)
+ * rv : fetch return value
+ * ra : fetch return address
+ * sa : fetch stack address
+ * sN : fetch Nth of stack (N:0-)
+ * @ADDR : fetch memory at ADDR (ADDR should be in kernel)
+ * @SYM[+|-offs] : fetch memory at SYM +|- offs (SYM is a data symbol)
+ * %REG : fetch register REG
+ * Indirect memory fetch:
+ * +|-offs(ARG) : fetch memory at ARG +|- offs address.
+ */
+ struct trace_probe *tp;
+ struct kprobe *kp;
+ int i, ret = 0;
+ int is_return = 0;
+ char *symbol = NULL, *event = NULL;
+ long offset = 0;
+ void *addr = NULL;
+
+ if (argc < 2)
+ return -EINVAL;
+
+ if (argv[0][0] == 'p')
+ is_return = 0;
+ else if (argv[0][0] == 'r')
+ is_return = 1;
+ else
+ return -EINVAL;
+
+ if (argv[0][1] == ':') {
+ event = &argv[0][2];
+ if (strlen(event) == 0) {
+ pr_info("Event name is not specifiled\n");
+ return -EINVAL;
+ }
+ }
+
+ if (isdigit(argv[1][0])) {
+ if (is_return)
+ return -EINVAL;
+ /* an address specified */
+ ret = strict_strtoul(&argv[0][2], 0, (unsigned long *)&addr);
+ if (ret)
+ return ret;
+ } else {
+ /* a symbol specified */
+ symbol = argv[1];
+ /* TODO: support .init module functions */
+ ret = split_symbol_offset(symbol, &offset);
+ if (ret)
+ return ret;
+ if (offset && is_return)
+ return -EINVAL;
+ }
+
+ /* setup a probe */
+ tp = alloc_trace_probe(symbol, event);
+ if (IS_ERR(tp))
+ return PTR_ERR(tp);
+
+ if (is_return) {
+ kp = &tp->rp.kp;
+ tp->rp.handler = kretprobe_trace_func;
+ } else {
+ kp = &tp->kp;
+ tp->kp.pre_handler = kprobe_trace_func;
+ }
+
+ if (tp->symbol) {
+ kp->symbol_name = tp->symbol;
+ kp->offset = offset;
+ } else
+ kp->addr = addr;
+
+ /* parse arguments */
+ argc -= 2; argv += 2; ret = 0;
+ for (i = 0; i < argc && i < TRACE_KPROBE_ARGS; i++) {
+ if (strlen(argv[i]) > MAX_ARGSTR_LEN) {
+ pr_info("Argument%d(%s) is too long.\n", i, argv[i]);
+ ret = -ENOSPC;
+ goto error;
+ }
+ ret = parse_trace_arg(argv[i], &tp->args[i], is_return);
+ if (ret)
+ goto error;
+ }
+ tp->nr_args = i;
+
+ ret = register_trace_probe(tp);
+ if (ret)
+ goto error;
+ return 0;
+
+error:
+ free_trace_probe(tp);
+ return ret;
+}
+
+static void cleanup_all_probes(void)
+{
+ struct trace_probe *tp;
+
+ mutex_lock(&probe_lock);
+ /* TODO: Use batch unregistration */
+ while (!list_empty(&probe_list)) {
+ tp = list_entry(probe_list.next, struct trace_probe, list);
+ unregister_trace_probe(tp);
+ free_trace_probe(tp);
+ }
+ mutex_unlock(&probe_lock);
+}
+
+
+/* Probes listing interfaces */
+static void *probes_seq_start(struct seq_file *m, loff_t *pos)
+{
+ mutex_lock(&probe_lock);
+ return seq_list_start(&probe_list, *pos);
+}
+
+static void *probes_seq_next(struct seq_file *m, void *v, loff_t *pos)
+{
+ return seq_list_next(v, &probe_list, pos);
+}
+
+static void probes_seq_stop(struct seq_file *m, void *v)
+{
+ mutex_unlock(&probe_lock);
+}
+
+static int probes_seq_show(struct seq_file *m, void *v)
+{
+ struct trace_probe *tp = v;
+ int i, ret;
+ char buf[MAX_ARGSTR_LEN + 1];
+
+ seq_printf(m, "%c", probe_is_return(tp) ? 'r' : 'p');
+ if (tp->call.name)
+ seq_printf(m, ":%s", tp->call.name);
+
+ if (tp->symbol)
+ seq_printf(m, " %s%+ld", probe_symbol(tp), probe_offset(tp));
+ else
+ seq_printf(m, " 0x%p", probe_address(tp));
+
+ for (i = 0; i < tp->nr_args; i++) {
+ ret = trace_arg_string(buf, MAX_ARGSTR_LEN, &tp->args[i]);
+ if (ret < 0) {
+ pr_warning("Argument%d decoding error(%d).\n", i, ret);
+ return ret;
+ }
+ seq_printf(m, " %s", buf);
+ }
+ seq_printf(m, "\n");
+ return 0;
+}
+
+static const struct seq_operations probes_seq_op = {
+ .start = probes_seq_start,
+ .next = probes_seq_next,
+ .stop = probes_seq_stop,
+ .show = probes_seq_show
+};
+
+static int probes_open(struct inode *inode, struct file *file)
+{
+ if ((file->f_mode & FMODE_WRITE) &&
+ (file->f_flags & O_TRUNC))
+ cleanup_all_probes();
+
+ return seq_open(file, &probes_seq_op);
+}
+
+static int command_trace_probe(const char *buf)
+{
+ char **argv;
+ int argc = 0, ret = 0;
+
+ argv = argv_split(GFP_KERNEL, buf, &argc);
+ if (!argv)
+ return -ENOMEM;
+
+ if (argc)
+ ret = create_trace_probe(argc, argv);
+
+ argv_free(argv);
+ return ret;
+}
+
+#define WRITE_BUFSIZE 128
+
+static ssize_t probes_write(struct file *file, const char __user *buffer,
+ size_t count, loff_t *ppos)
+{
+ char *kbuf, *tmp;
+ int ret;
+ size_t done;
+ size_t size;
+
+ kbuf = kmalloc(WRITE_BUFSIZE, GFP_KERNEL);
+ if (!kbuf)
+ return -ENOMEM;
+
+ ret = done = 0;
+ while (done < count) {
+ size = count - done;
+ if (size >= WRITE_BUFSIZE)
+ size = WRITE_BUFSIZE - 1;
+ if (copy_from_user(kbuf, buffer + done, size)) {
+ ret = -EFAULT;
+ goto out;
+ }
+ kbuf[size] = '\0';
+ tmp = strchr(kbuf, '\n');
+ if (tmp) {
+ *tmp = '\0';
+ size = tmp - kbuf + 1;
+ } else if (done + size < count) {
+ pr_warning("Line length is too long: "
+ "Should be less than %d.", WRITE_BUFSIZE);
+ ret = -EINVAL;
+ goto out;
+ }
+ done += size;
+ /* Remove comments */
+ tmp = strchr(kbuf, '#');
+ if (tmp)
+ *tmp = '\0';
+
+ ret = command_trace_probe(kbuf);
+ if (ret)
+ goto out;
+ }
+ ret = done;
+out:
+ kfree(kbuf);
+ return ret;
+}
+
+static const struct file_operations kprobe_events_ops = {
+ .owner = THIS_MODULE,
+ .open = probes_open,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = seq_release,
+ .write = probes_write,
+};
+
+/* Kprobe handler */
+static __kprobes int kprobe_trace_func(struct kprobe *kp, struct pt_regs *regs)
+{
+ struct trace_probe *tp = container_of(kp, struct trace_probe, kp);
+ struct kprobe_trace_entry *entry;
+ struct ring_buffer_event *event;
+ int size, i, pc;
+ unsigned long irq_flags;
+ struct ftrace_event_call *call = &event_kprobe;
+
+ if (&tp->call.name)
+ call = &tp->call;
+
+ local_save_flags(irq_flags);
+ pc = preempt_count();
+
+ size = SIZEOF_KPROBE_TRACE_ENTRY(tp->nr_args);
+
+ event = trace_current_buffer_lock_reserve(TRACE_KPROBE, size,
+ irq_flags, pc);
+ if (!event)
+ return 0;
+
+ entry = ring_buffer_event_data(event);
+ entry->nargs = tp->nr_args;
+ entry->ip = (unsigned long)kp->addr;
+ for (i = 0; i < tp->nr_args; i++)
+ entry->args[i] = call_fetch(&tp->args[i], regs);
+
+ if (!filter_current_check_discard(call, entry, event))
+ trace_nowake_buffer_unlock_commit(event, irq_flags, pc);
+ return 0;
+}
+
+/* Kretprobe handler */
+static __kprobes int kretprobe_trace_func(struct kretprobe_instance *ri,
+ struct pt_regs *regs)
+{
+ struct trace_probe *tp = container_of(ri->rp, struct trace_probe, rp);
+ struct kretprobe_trace_entry *entry;
+ struct ring_buffer_event *event;
+ int size, i, pc;
+ unsigned long irq_flags;
+ struct ftrace_event_call *call = &event_kretprobe;
+
+ if (&tp->call.name)
+ call = &tp->call;
+
+ local_save_flags(irq_flags);
+ pc = preempt_count();
+
+ size = SIZEOF_KRETPROBE_TRACE_ENTRY(tp->nr_args);
+
+ event = trace_current_buffer_lock_reserve(TRACE_KRETPROBE, size,
+ irq_flags, pc);
+ if (!event)
+ return 0;
+
+ entry = ring_buffer_event_data(event);
+ entry->nargs = tp->nr_args;
+ entry->func = (unsigned long)probe_address(tp);
+ entry->ret_ip = (unsigned long)ri->ret_addr;
+ for (i = 0; i < tp->nr_args; i++)
+ entry->args[i] = call_fetch(&tp->args[i], regs);
+
+ if (!filter_current_check_discard(call, entry, event))
+ trace_nowake_buffer_unlock_commit(event, irq_flags, pc);
+
+ return 0;
+}
+
+/* Event entry printers */
+enum print_line_t
+print_kprobe_event(struct trace_iterator *iter, int flags)
+{
+ struct kprobe_trace_entry *field;
+ struct trace_seq *s = &iter->seq;
+ int i;
+
+ trace_assign_type(field, iter->ent);
+
+ if (!seq_print_ip_sym(s, field->ip, flags | TRACE_ITER_SYM_OFFSET))
+ goto partial;
+
+ if (!trace_seq_puts(s, ":"))
+ goto partial;
+
+ for (i = 0; i < field->nargs; i++)
+ if (!trace_seq_printf(s, " 0x%lx", field->args[i]))
+ goto partial;
+
+ if (!trace_seq_puts(s, "\n"))
+ goto partial;
+
+ return TRACE_TYPE_HANDLED;
+partial:
+ return TRACE_TYPE_PARTIAL_LINE;
+}
+
+enum print_line_t
+print_kretprobe_event(struct trace_iterator *iter, int flags)
+{
+ struct kretprobe_trace_entry *field;
+ struct trace_seq *s = &iter->seq;
+ int i;
+
+ trace_assign_type(field, iter->ent);
+
+ if (!seq_print_ip_sym(s, field->ret_ip, flags | TRACE_ITER_SYM_OFFSET))
+ goto partial;
+
+ if (!trace_seq_puts(s, " <- "))
+ goto partial;
+
+ if (!seq_print_ip_sym(s, field->func, flags & ~TRACE_ITER_SYM_OFFSET))
+ goto partial;
+
+ if (!trace_seq_puts(s, ":"))
+ goto partial;
+
+ for (i = 0; i < field->nargs; i++)
+ if (!trace_seq_printf(s, " 0x%lx", field->args[i]))
+ goto partial;
+
+ if (!trace_seq_puts(s, "\n"))
+ goto partial;
+
+ return TRACE_TYPE_HANDLED;
+partial:
+ return TRACE_TYPE_PARTIAL_LINE;
+}
+
+static struct trace_event kprobe_trace_event = {
+ .type = TRACE_KPROBE,
+ .trace = print_kprobe_event,
+};
+
+static struct trace_event kretprobe_trace_event = {
+ .type = TRACE_KRETPROBE,
+ .trace = print_kretprobe_event,
+};
+
+static int probe_event_enable(struct ftrace_event_call *call)
+{
+ struct trace_probe *tp = (struct trace_probe *)call->data;
+
+ if (probe_is_return(tp))
+ return enable_kretprobe(&tp->rp);
+ else
+ return enable_kprobe(&tp->kp);
+}
+
+static void probe_event_disable(struct ftrace_event_call *call)
+{
+ struct trace_probe *tp = (struct trace_probe *)call->data;
+
+ if (probe_is_return(tp))
+ disable_kretprobe(&tp->rp);
+ else
+ disable_kprobe(&tp->kp);
+}
+
+static int probe_event_raw_init(struct ftrace_event_call *event_call)
+{
+ INIT_LIST_HEAD(&event_call->fields);
+ init_preds(event_call);
+ return 0;
+}
+
+#undef DEFINE_FIELD
+#define DEFINE_FIELD(type, item, name, is_signed) \
+ do { \
+ ret = trace_define_field(event_call, #type, name, \
+ offsetof(typeof(field), item), \
+ sizeof(field.item), is_signed);\
+ if (ret) \
+ return ret; \
+ } while (0)
+
+static int kprobe_event_define_fields(struct ftrace_event_call *event_call)
+{
+ int ret, i;
+ struct kprobe_trace_entry field;
+ char buf[MAX_ARGSTR_LEN + 1];
+ struct trace_probe *tp = (struct trace_probe *)event_call->data;
+
+ __common_field(int, type, 1);
+ __common_field(unsigned char, flags, 0);
+ __common_field(unsigned char, preempt_count, 0);
+ __common_field(int, pid, 1);
+ __common_field(int, tgid, 1);
+
+ DEFINE_FIELD(unsigned long, ip, "ip", 0);
+ DEFINE_FIELD(int, nargs, "nargs", 1);
+ for (i = 0; i < tp->nr_args; i++) {
+ /* Set argN as a field */
+ sprintf(buf, "arg%d", i);
+ DEFINE_FIELD(unsigned long, args[i], buf, 0);
+ /* Set argument string as an alias field */
+ ret = trace_arg_string(buf, MAX_ARGSTR_LEN, &tp->args[i]);
+ if (ret < 0)
+ return ret;
+ DEFINE_FIELD(unsigned long, args[i], buf, 0);
+ }
+ return 0;
+}
+
+static int kretprobe_event_define_fields(struct ftrace_event_call *event_call)
+{
+ int ret, i;
+ struct kretprobe_trace_entry field;
+ char buf[MAX_ARGSTR_LEN + 1];
+ struct trace_probe *tp = (struct trace_probe *)event_call->data;
+
+ __common_field(int, type, 1);
+ __common_field(unsigned char, flags, 0);
+ __common_field(unsigned char, preempt_count, 0);
+ __common_field(int, pid, 1);
+ __common_field(int, tgid, 1);
+
+ DEFINE_FIELD(unsigned long, func, "func", 0);
+ DEFINE_FIELD(unsigned long, ret_ip, "ret_ip", 0);
+ DEFINE_FIELD(int, nargs, "nargs", 1);
+ for (i = 0; i < tp->nr_args; i++) {
+ /* Set argN as a field */
+ sprintf(buf, "arg%d", i);
+ DEFINE_FIELD(unsigned long, args[i], buf, 0);
+ /* Set argument string as an alias field */
+ ret = trace_arg_string(buf, MAX_ARGSTR_LEN, &tp->args[i]);
+ if (ret < 0)
+ return ret;
+ DEFINE_FIELD(unsigned long, args[i], buf, 0);
+ }
+ return 0;
+}
+
+static int __probe_event_show_format(struct trace_seq *s,
+ struct trace_probe *tp, const char *fmt,
+ const char *arg)
+{
+ int i, ret;
+ char buf[MAX_ARGSTR_LEN + 1];
+
+ /* Show aliases */
+ for (i = 0; i < tp->nr_args; i++) {
+ ret = trace_arg_string(buf, MAX_ARGSTR_LEN, &tp->args[i]);
+ if (ret < 0)
+ return ret;
+ if (!trace_seq_printf(s, "\talias: %s;\toriginal: arg%d;\n",
+ buf, i))
+ return 0;
+ }
+ /* Show format */
+ if (!trace_seq_printf(s, "\nprint fmt: \"%s", fmt))
+ return 0;
+
+ for (i = 0; i < tp->nr_args; i++)
+ if (!trace_seq_puts(s, " 0x%lx"))
+ return 0;
+
+ if (!trace_seq_printf(s, "\", %s", arg))
+ return 0;
+
+ for (i = 0; i < tp->nr_args; i++)
+ if (!trace_seq_printf(s, ", arg%d", i))
+ return 0;
+
+ return trace_seq_puts(s, "\n");
+}
+
+#undef SHOW_FIELD
+#define SHOW_FIELD(type, item, name) \
+ do { \
+ ret = trace_seq_printf(s, "\tfield: " #type " %s;\t" \
+ "offset:%u;tsize:%u;\n", name, \
+ (unsigned int)offsetof(typeof(field), item),\
+ (unsigned int)sizeof(type)); \
+ if (!ret) \
+ return 0; \
+ } while (0)
+
+static int kprobe_event_show_format(struct ftrace_event_call *call,
+ struct trace_seq *s)
+{
+ struct kprobe_trace_entry field __attribute__((unused));
+ int ret, i;
+ char buf[8];
+ struct trace_probe *tp = (struct trace_probe *)call->data;
+
+ SHOW_FIELD(unsigned long, ip, "ip");
+ SHOW_FIELD(int, nargs, "nargs");
+
+ /* Show fields */
+ for (i = 0; i < tp->nr_args; i++) {
+ sprintf(buf, "arg%d", i);
+ SHOW_FIELD(unsigned long, args[i], buf);
+ }
+ trace_seq_puts(s, "\n");
+
+ return __probe_event_show_format(s, tp, "%lx:", "ip");
+}
+
+static int kretprobe_event_show_format(struct ftrace_event_call *call,
+ struct trace_seq *s)
+{
+ struct kretprobe_trace_entry field __attribute__((unused));
+ int ret, i;
+ char buf[8];
+ struct trace_probe *tp = (struct trace_probe *)call->data;
+
+ SHOW_FIELD(unsigned long, func, "func");
+ SHOW_FIELD(unsigned long, ret_ip, "ret_ip");
+ SHOW_FIELD(int, nargs, "nargs");
+
+ /* Show fields */
+ for (i = 0; i < tp->nr_args; i++) {
+ sprintf(buf, "arg%d", i);
+ SHOW_FIELD(unsigned long, args[i], buf);
+ }
+ trace_seq_puts(s, "\n");
+
+ return __probe_event_show_format(s, tp, "%lx <- %lx:",
+ "func, ret_ip");
+}
+
+static int register_probe_event(struct trace_probe *tp)
+{
+ struct ftrace_event_call *call = &tp->call;
+ int ret;
+
+ /* Initialize ftrace_event_call */
+ call->system = "kprobes";
+ if (probe_is_return(tp)) {
+ call->event = &kretprobe_trace_event;
+ call->id = TRACE_KRETPROBE;
+ call->raw_init = probe_event_raw_init;
+ call->show_format = kretprobe_event_show_format;
+ call->define_fields = kretprobe_event_define_fields;
+ } else {
+ call->event = &kprobe_trace_event;
+ call->id = TRACE_KPROBE;
+ call->raw_init = probe_event_raw_init;
+ call->show_format = kprobe_event_show_format;
+ call->define_fields = kprobe_event_define_fields;
+ }
+ call->enabled = 1;
+ call->regfunc = probe_event_enable;
+ call->unregfunc = probe_event_disable;
+ call->data = tp;
+ ret = trace_add_event_call(call);
+ if (ret)
+ pr_info("Failed to register kprobe event: %s\n", call->name);
+ return ret;
+}
+
+static void unregister_probe_event(struct trace_probe *tp)
+{
+ /*
+ * Prevent to unregister event itself because the event is shared
+ * among other probes.
+ */
+ tp->call.event = NULL;
+ trace_remove_event_call(&tp->call);
+}
+
+/* Make a debugfs interface for controling probe points */
+static __init int init_kprobe_trace(void)
+{
+ struct dentry *d_tracer;
+ struct dentry *entry;
+ int ret;
+
+ ret = register_ftrace_event(&kprobe_trace_event);
+ if (!ret) {
+ pr_warning("Could not register kprobe_trace_event type.\n");
+ return 0;
+ }
+ ret = register_ftrace_event(&kretprobe_trace_event);
+ if (!ret) {
+ pr_warning("Could not register kretprobe_trace_event type.\n");
+ return 0;
+ }
+
+ d_tracer = tracing_init_dentry();
+ if (!d_tracer)
+ return 0;
+
+ entry = debugfs_create_file("kprobe_events", 0644, d_tracer,
+ NULL, &kprobe_events_ops);
+
+ if (!entry)
+ pr_warning("Could not create debugfs "
+ "'kprobe_events' entry\n");
+ return 0;
+}
+fs_initcall(init_kprobe_trace);
+
+
+#ifdef CONFIG_FTRACE_STARTUP_TEST
+
+static int kprobe_trace_selftest_target(int a1, int a2, int a3,
+ int a4, int a5, int a6)
+{
+ return a1 + a2 + a3 + a4 + a5 + a6;
+}
+
+static __init int kprobe_trace_self_tests_init(void)
+{
+ int ret;
+ int (*target)(int, int, int, int, int, int);
+
+ target = kprobe_trace_selftest_target;
+
+ pr_info("Testing kprobe tracing: ");
+
+ ret = command_trace_probe("p:testprobe kprobe_trace_selftest_target "
+ "a1 a2 a3 a4 a5 a6");
+ if (WARN_ON_ONCE(ret))
+ pr_warning("error enabling function entry\n");
+
+ ret = command_trace_probe("r:testprobe2 kprobe_trace_selftest_target "
+ "ra rv");
+ if (WARN_ON_ONCE(ret))
+ pr_warning("error enabling function return\n");
+
+ ret = target(1, 2, 3, 4, 5, 6);
+
+ cleanup_all_probes();
+
+ pr_cont("OK\n");
+ return 0;
+}
+
+late_initcall(kprobe_trace_self_tests_init);
+
+#endif
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America), Inc.
Software Solutions Division
e-mail: mhiramat@redhat.com
^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH -tip v14 11/12] tracing: Kprobe tracer assigns new event ids for each event
2009-08-13 20:31 [PATCH -tip v14 00/12] tracing: kprobe-based event tracer and x86 instruction decoder Masami Hiramatsu
` (8 preceding siblings ...)
2009-08-13 20:33 ` [PATCH -tip v14 12/12] tracing: Add kprobes event profiling interface Masami Hiramatsu
@ 2009-08-13 20:33 ` Masami Hiramatsu
2009-08-13 20:47 ` [PATCH -tip v14 07/12] tracing: Introduce TRACE_FIELD_ZERO() macro Masami Hiramatsu
` (3 subsequent siblings)
13 siblings, 0 replies; 29+ messages in thread
From: Masami Hiramatsu @ 2009-08-13 20:33 UTC (permalink / raw)
To: Ingo Molnar, Steven Rostedt, lkml
Cc: systemtap, kvm, DLE, Masami Hiramatsu,
Ananth N Mavinakayanahalli, Avi Kivity, Andi Kleen,
Christoph Hellwig, Frank Ch. Eigler, Frederic Weisbecker,
H. Peter Anvin, Ingo Molnar, Jason Baron, Jim Keniston, K.Prasad,
Lai Jiangshan, Li Zefan, PrzemysławPawełczyk,
Roland McGrath, Sam Ravnborg, Srikar Dronamraju, Steven Rostedt,
Tom Zanussi, Vegard Nossum
Assigns new event ids for each kprobes event. This doesn't clear ring_buffer
when unregistering each kprobe event. Thus, if you mind 'Unknown event'
messages, clear the buffer manually after changing kprobe events.
Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Avi Kivity <avi@redhat.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Frank Ch. Eigler <fche@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Jason Baron <jbaron@redhat.com>
Cc: Jim Keniston <jkenisto@us.ibm.com>
Cc: K.Prasad <prasad@linux.vnet.ibm.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Przemysław Pawełczyk <przemyslaw@pawelczyk.it>
Cc: Roland McGrath <roland@redhat.com>
Cc: Sam Ravnborg <sam@ravnborg.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Tom Zanussi <tzanussi@gmail.com>
Cc: Vegard Nossum <vegard.nossum@gmail.com>
---
kernel/trace/trace.h | 6 -----
kernel/trace/trace_kprobe.c | 51 +++++++++++++------------------------------
2 files changed, 15 insertions(+), 42 deletions(-)
diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
index 4ce4525..0b78d76 100644
--- a/kernel/trace/trace.h
+++ b/kernel/trace/trace.h
@@ -43,8 +43,6 @@ enum trace_type {
TRACE_POWER,
TRACE_BLK,
TRACE_KSYM,
- TRACE_KPROBE,
- TRACE_KRETPROBE,
__TRACE_LAST_TYPE,
};
@@ -358,10 +356,6 @@ extern void __ftrace_bad_type(void);
IF_ASSIGN(var, ent, struct kmemtrace_free_entry, \
TRACE_KMEM_FREE); \
IF_ASSIGN(var, ent, struct ksym_trace_entry, TRACE_KSYM);\
- IF_ASSIGN(var, ent, struct kprobe_trace_entry, \
- TRACE_KPROBE); \
- IF_ASSIGN(var, ent, struct kretprobe_trace_entry, \
- TRACE_KRETPROBE); \
__ftrace_bad_type(); \
} while (0)
diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c
index ec137ed..0e8498e 100644
--- a/kernel/trace/trace_kprobe.c
+++ b/kernel/trace/trace_kprobe.c
@@ -186,6 +186,7 @@ struct trace_probe {
};
const char *symbol; /* symbol name */
struct ftrace_event_call call;
+ struct trace_event event;
unsigned int nr_args;
struct fetch_func args[];
};
@@ -795,7 +796,7 @@ static __kprobes int kprobe_trace_func(struct kprobe *kp, struct pt_regs *regs)
size = SIZEOF_KPROBE_TRACE_ENTRY(tp->nr_args);
- event = trace_current_buffer_lock_reserve(TRACE_KPROBE, size,
+ event = trace_current_buffer_lock_reserve(call->id, size,
irq_flags, pc);
if (!event)
return 0;
@@ -827,7 +828,7 @@ static __kprobes int kretprobe_trace_func(struct kretprobe_instance *ri,
size = SIZEOF_KRETPROBE_TRACE_ENTRY(tp->nr_args);
- event = trace_current_buffer_lock_reserve(TRACE_KRETPROBE, size,
+ event = trace_current_buffer_lock_reserve(call->id, size,
irq_flags, pc);
if (!event)
return 0;
@@ -853,7 +854,7 @@ print_kprobe_event(struct trace_iterator *iter, int flags)
struct trace_seq *s = &iter->seq;
int i;
- trace_assign_type(field, iter->ent);
+ field = (struct kprobe_trace_entry *)iter->ent;
if (!seq_print_ip_sym(s, field->ip, flags | TRACE_ITER_SYM_OFFSET))
goto partial;
@@ -880,7 +881,7 @@ print_kretprobe_event(struct trace_iterator *iter, int flags)
struct trace_seq *s = &iter->seq;
int i;
- trace_assign_type(field, iter->ent);
+ field = (struct kretprobe_trace_entry *)iter->ent;
if (!seq_print_ip_sym(s, field->ret_ip, flags | TRACE_ITER_SYM_OFFSET))
goto partial;
@@ -906,16 +907,6 @@ partial:
return TRACE_TYPE_PARTIAL_LINE;
}
-static struct trace_event kprobe_trace_event = {
- .type = TRACE_KPROBE,
- .trace = print_kprobe_event,
-};
-
-static struct trace_event kretprobe_trace_event = {
- .type = TRACE_KRETPROBE,
- .trace = print_kretprobe_event,
-};
-
static int probe_event_enable(struct ftrace_event_call *call)
{
struct trace_probe *tp = (struct trace_probe *)call->data;
@@ -1107,35 +1098,35 @@ static int register_probe_event(struct trace_probe *tp)
/* Initialize ftrace_event_call */
call->system = "kprobes";
if (probe_is_return(tp)) {
- call->event = &kretprobe_trace_event;
- call->id = TRACE_KRETPROBE;
+ tp->event.trace = print_kretprobe_event;
call->raw_init = probe_event_raw_init;
call->show_format = kretprobe_event_show_format;
call->define_fields = kretprobe_event_define_fields;
} else {
- call->event = &kprobe_trace_event;
- call->id = TRACE_KPROBE;
+ tp->event.trace = print_kprobe_event;
call->raw_init = probe_event_raw_init;
call->show_format = kprobe_event_show_format;
call->define_fields = kprobe_event_define_fields;
}
+ call->event = &tp->event;
+ call->id = register_ftrace_event(&tp->event);
+ if (!call->id)
+ return -ENODEV;
call->enabled = 1;
call->regfunc = probe_event_enable;
call->unregfunc = probe_event_disable;
call->data = tp;
ret = trace_add_event_call(call);
- if (ret)
+ if (ret) {
pr_info("Failed to register kprobe event: %s\n", call->name);
+ unregister_ftrace_event(&tp->event);
+ }
return ret;
}
static void unregister_probe_event(struct trace_probe *tp)
{
- /*
- * Prevent to unregister event itself because the event is shared
- * among other probes.
- */
- tp->call.event = NULL;
+ /* tp->event is unregistered in trace_remove_event_call() */
trace_remove_event_call(&tp->call);
}
@@ -1144,18 +1135,6 @@ static __init int init_kprobe_trace(void)
{
struct dentry *d_tracer;
struct dentry *entry;
- int ret;
-
- ret = register_ftrace_event(&kprobe_trace_event);
- if (!ret) {
- pr_warning("Could not register kprobe_trace_event type.\n");
- return 0;
- }
- ret = register_ftrace_event(&kretprobe_trace_event);
- if (!ret) {
- pr_warning("Could not register kretprobe_trace_event type.\n");
- return 0;
- }
d_tracer = tracing_init_dentry();
if (!d_tracer)
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America), Inc.
Software Solutions Division
e-mail: mhiramat@redhat.com
^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH -tip v14 12/12] tracing: Add kprobes event profiling interface
2009-08-13 20:31 [PATCH -tip v14 00/12] tracing: kprobe-based event tracer and x86 instruction decoder Masami Hiramatsu
` (7 preceding siblings ...)
2009-08-13 20:32 ` [PATCH -tip v14 09/12] tracing: Kprobe-tracer supports more than 6 arguments Masami Hiramatsu
@ 2009-08-13 20:33 ` Masami Hiramatsu
2009-08-13 20:33 ` [PATCH -tip v14 11/12] tracing: Kprobe tracer assigns new event ids for each event Masami Hiramatsu
` (4 subsequent siblings)
13 siblings, 0 replies; 29+ messages in thread
From: Masami Hiramatsu @ 2009-08-13 20:33 UTC (permalink / raw)
To: Ingo Molnar, Steven Rostedt, lkml
Cc: systemtap, kvm, DLE, Masami Hiramatsu,
Ananth N Mavinakayanahalli, Avi Kivity, Andi Kleen,
Christoph Hellwig, Frank Ch. Eigler, Frederic Weisbecker,
H. Peter Anvin, Ingo Molnar, Jason Baron, Jim Keniston, K.Prasad,
Lai Jiangshan, Li Zefan, PrzemysławPawełczyk,
Roland McGrath, Sam Ravnborg, Srikar Dronamraju, Steven Rostedt,
Tom Zanussi, Vegard Nossum
Add profiling interaces for each kprobes event. This interface provides
how many times each probe hit or missed.
Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Avi Kivity <avi@redhat.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Frank Ch. Eigler <fche@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Jason Baron <jbaron@redhat.com>
Cc: Jim Keniston <jkenisto@us.ibm.com>
Cc: K.Prasad <prasad@linux.vnet.ibm.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Przemysław Pawełczyk <przemyslaw@pawelczyk.it>
Cc: Roland McGrath <roland@redhat.com>
Cc: Sam Ravnborg <sam@ravnborg.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Tom Zanussi <tzanussi@gmail.com>
Cc: Vegard Nossum <vegard.nossum@gmail.com>
---
Documentation/trace/kprobetrace.txt | 8 +++++++
kernel/trace/trace_kprobe.c | 43 +++++++++++++++++++++++++++++++++++
2 files changed, 51 insertions(+), 0 deletions(-)
diff --git a/Documentation/trace/kprobetrace.txt b/Documentation/trace/kprobetrace.txt
index 5e59e85..3de7517 100644
--- a/Documentation/trace/kprobetrace.txt
+++ b/Documentation/trace/kprobetrace.txt
@@ -70,6 +70,14 @@ filter:
names and field names for describing filters.
+Event Profiling
+---------------
+ You can check the total number of probe hits and probe miss-hits via
+/sys/kernel/debug/tracing/kprobe_profile.
+ The first column is event name, the second is the number of probe hits,
+the third is the number of probe miss-hits.
+
+
Usage examples
--------------
To add a probe as a new event, write a new definition to kprobe_events
diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c
index 0e8498e..0f5d0a6 100644
--- a/kernel/trace/trace_kprobe.c
+++ b/kernel/trace/trace_kprobe.c
@@ -184,6 +184,7 @@ struct trace_probe {
struct kprobe kp;
struct kretprobe rp;
};
+ unsigned long nhit;
const char *symbol; /* symbol name */
struct ftrace_event_call call;
struct trace_event event;
@@ -781,6 +782,37 @@ static const struct file_operations kprobe_events_ops = {
.write = probes_write,
};
+/* Probes profiling interfaces */
+static int probes_profile_seq_show(struct seq_file *m, void *v)
+{
+ struct trace_probe *tp = v;
+
+ seq_printf(m, " %-44s %15lu %15lu\n", tp->call.name, tp->nhit,
+ probe_is_return(tp) ? tp->rp.kp.nmissed : tp->kp.nmissed);
+
+ return 0;
+}
+
+static const struct seq_operations profile_seq_op = {
+ .start = probes_seq_start,
+ .next = probes_seq_next,
+ .stop = probes_seq_stop,
+ .show = probes_profile_seq_show
+};
+
+static int profile_open(struct inode *inode, struct file *file)
+{
+ return seq_open(file, &profile_seq_op);
+}
+
+static const struct file_operations kprobe_profile_ops = {
+ .owner = THIS_MODULE,
+ .open = profile_open,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = seq_release,
+};
+
/* Kprobe handler */
static __kprobes int kprobe_trace_func(struct kprobe *kp, struct pt_regs *regs)
{
@@ -791,6 +823,8 @@ static __kprobes int kprobe_trace_func(struct kprobe *kp, struct pt_regs *regs)
unsigned long irq_flags;
struct ftrace_event_call *call = &tp->call;
+ tp->nhit++;
+
local_save_flags(irq_flags);
pc = preempt_count();
@@ -1143,9 +1177,18 @@ static __init int init_kprobe_trace(void)
entry = debugfs_create_file("kprobe_events", 0644, d_tracer,
NULL, &kprobe_events_ops);
+ /* Event list interface */
if (!entry)
pr_warning("Could not create debugfs "
"'kprobe_events' entry\n");
+
+ /* Profile interface */
+ entry = debugfs_create_file("kprobe_profile", 0444, d_tracer,
+ NULL, &kprobe_profile_ops);
+
+ if (!entry)
+ pr_warning("Could not create debugfs "
+ "'kprobe_profile' entry\n");
return 0;
}
fs_initcall(init_kprobe_trace);
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America), Inc.
Software Solutions Division
e-mail: mhiramat@redhat.com
^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH -tip v14 06/12] tracing: ftrace dynamic ftrace_event_call support
2009-08-13 20:31 [PATCH -tip v14 00/12] tracing: kprobe-based event tracer and x86 instruction decoder Masami Hiramatsu
` (10 preceding siblings ...)
2009-08-13 20:47 ` [PATCH -tip v14 07/12] tracing: Introduce TRACE_FIELD_ZERO() macro Masami Hiramatsu
@ 2009-08-13 20:47 ` Masami Hiramatsu
2009-08-13 20:54 ` [TOOL] kprobestest : Kprobe stress test tool Masami Hiramatsu
2009-08-13 20:56 ` [TOOL] c2kpe: C expression to kprobe event format converter Masami Hiramatsu
13 siblings, 0 replies; 29+ messages in thread
From: Masami Hiramatsu @ 2009-08-13 20:47 UTC (permalink / raw)
To: Ingo Molnar, Steven Rostedt, lkml
Cc: systemtap, kvm, DLE, Masami Hiramatsu, Frederic Weisbecker,
Ananth N Mavinakayanahalli, Avi Kivity, Andi Kleen,
Christoph Hellwig, Frank Ch. Eigler, H. Peter Anvin, Ingo Molnar,
Jason Baron, Jim Keniston, K.Prasad, Lai Jiangshan, Li Zefan,
PrzemysławPawełczyk, Roland McGrath, Sam Ravnborg,
Srikar Dronamraju, Steven Rostedt, Tom Zanussi, Vegard Nossum
Add dynamic ftrace_event_call support to ftrace. Trace engines can adds new
ftrace_event_call to ftrace on the fly. Each operator functions of the call
takes a ftrace_event_call data structure as an argument, because these
functions may be shared among several ftrace_event_calls.
Changes from v13:
- Define remove_subsystem_dir() always (revirt a2ca5e03), because
trace_remove_event_call() uses it.
- Modify syscall tracer because of ftrace_event_call change.
Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
Acked-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Avi Kivity <avi@redhat.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Frank Ch. Eigler <fche@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Jason Baron <jbaron@redhat.com>
Cc: Jim Keniston <jkenisto@us.ibm.com>
Cc: K.Prasad <prasad@linux.vnet.ibm.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Przemysław Pawełczyk <przemyslaw@pawelczyk.it>
Cc: Roland McGrath <roland@redhat.com>
Cc: Sam Ravnborg <sam@ravnborg.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Tom Zanussi <tzanussi@gmail.com>
Cc: Vegard Nossum <vegard.nossum@gmail.com>
---
include/linux/ftrace_event.h | 14 +++--
include/linux/syscalls.h | 4 +
include/trace/ftrace.h | 19 +++----
include/trace/syscall.h | 8 +--
kernel/trace/trace_events.c | 119 +++++++++++++++++++++++++++++------------
kernel/trace/trace_export.c | 23 ++++----
kernel/trace/trace_syscalls.c | 16 +++---
7 files changed, 125 insertions(+), 78 deletions(-)
diff --git a/include/linux/ftrace_event.h b/include/linux/ftrace_event.h
index 189806b..9af68ce 100644
--- a/include/linux/ftrace_event.h
+++ b/include/linux/ftrace_event.h
@@ -112,13 +112,13 @@ struct ftrace_event_call {
struct dentry *dir;
struct trace_event *event;
int enabled;
- int (*regfunc)(void *);
- void (*unregfunc)(void *);
+ int (*regfunc)(struct ftrace_event_call *);
+ void (*unregfunc)(struct ftrace_event_call *);
int id;
- int (*raw_init)(void);
- int (*show_format)(struct ftrace_event_call *call,
- struct trace_seq *s);
- int (*define_fields)(void);
+ int (*raw_init)(struct ftrace_event_call *);
+ int (*show_format)(struct ftrace_event_call *,
+ struct trace_seq *);
+ int (*define_fields)(struct ftrace_event_call *);
struct list_head fields;
int filter_active;
struct event_filter *filter;
@@ -142,6 +142,8 @@ extern int filter_current_check_discard(struct ftrace_event_call *call,
extern int trace_define_field(struct ftrace_event_call *call, char *type,
char *name, int offset, int size, int is_signed);
+extern int trace_add_event_call(struct ftrace_event_call *call);
+extern void trace_remove_event_call(struct ftrace_event_call *call);
#define is_signed_type(type) (((type)(-1)) < 0)
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index 87d06c1..be59d22 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -165,7 +165,7 @@ static void prof_sysexit_disable_##sname(struct ftrace_event_call *event_call) \
struct trace_event enter_syscall_print_##sname = { \
.trace = print_syscall_enter, \
}; \
- static int init_enter_##sname(void) \
+ static int init_enter_##sname(struct ftrace_event_call *call) \
{ \
int num, id; \
num = syscall_name_to_nr("sys"#sname); \
@@ -201,7 +201,7 @@ static void prof_sysexit_disable_##sname(struct ftrace_event_call *event_call) \
struct trace_event exit_syscall_print_##sname = { \
.trace = print_syscall_exit, \
}; \
- static int init_exit_##sname(void) \
+ static int init_exit_##sname(struct ftrace_event_call *call) \
{ \
int num, id; \
num = syscall_name_to_nr("sys"#sname); \
diff --git a/include/trace/ftrace.h b/include/trace/ftrace.h
index b250b06..6c1b5b1 100644
--- a/include/trace/ftrace.h
+++ b/include/trace/ftrace.h
@@ -294,10 +294,9 @@ ftrace_raw_output_##call(struct trace_iterator *iter, int flags) \
#undef TRACE_EVENT
#define TRACE_EVENT(call, proto, args, tstruct, func, print) \
int \
-ftrace_define_fields_##call(void) \
+ftrace_define_fields_##call(struct ftrace_event_call *event_call) \
{ \
struct ftrace_raw_##call field; \
- struct ftrace_event_call *event_call = &event_##call; \
int ret; \
\
__common_field(int, type, 1); \
@@ -411,7 +410,7 @@ static void ftrace_profile_disable_##call(struct ftrace_event_call *event_call)\
* event_trace_printk(_RET_IP_, "<call>: " <fmt>);
* }
*
- * static int ftrace_reg_event_<call>(void)
+ * static int ftrace_reg_event_<call>(struct ftrace_event_call *unused)
* {
* int ret;
*
@@ -422,7 +421,7 @@ static void ftrace_profile_disable_##call(struct ftrace_event_call *event_call)\
* return ret;
* }
*
- * static void ftrace_unreg_event_<call>(void)
+ * static void ftrace_unreg_event_<call>(struct ftrace_event_call *unused)
* {
* unregister_trace_<call>(ftrace_event_<call>);
* }
@@ -455,7 +454,7 @@ static void ftrace_profile_disable_##call(struct ftrace_event_call *event_call)\
* trace_current_buffer_unlock_commit(event, irq_flags, pc);
* }
*
- * static int ftrace_raw_reg_event_<call>(void)
+ * static int ftrace_raw_reg_event_<call>(struct ftrace_event_call *unused)
* {
* int ret;
*
@@ -466,7 +465,7 @@ static void ftrace_profile_disable_##call(struct ftrace_event_call *event_call)\
* return ret;
* }
*
- * static void ftrace_unreg_event_<call>(void)
+ * static void ftrace_unreg_event_<call>(struct ftrace_event_call *unused)
* {
* unregister_trace_<call>(ftrace_raw_event_<call>);
* }
@@ -475,7 +474,7 @@ static void ftrace_profile_disable_##call(struct ftrace_event_call *event_call)\
* .trace = ftrace_raw_output_<call>, <-- stage 2
* };
*
- * static int ftrace_raw_init_event_<call>(void)
+ * static int ftrace_raw_init_event_<call>(struct ftrace_event_call *unused)
* {
* int id;
*
@@ -569,7 +568,7 @@ static void ftrace_raw_event_##call(proto) \
trace_nowake_buffer_unlock_commit(event, irq_flags, pc); \
} \
\
-static int ftrace_raw_reg_event_##call(void *ptr) \
+static int ftrace_raw_reg_event_##call(struct ftrace_event_call *unused)\
{ \
int ret; \
\
@@ -580,7 +579,7 @@ static int ftrace_raw_reg_event_##call(void *ptr) \
return ret; \
} \
\
-static void ftrace_raw_unreg_event_##call(void *ptr) \
+static void ftrace_raw_unreg_event_##call(struct ftrace_event_call *unused)\
{ \
unregister_trace_##call(ftrace_raw_event_##call); \
} \
@@ -589,7 +588,7 @@ static struct trace_event ftrace_event_type_##call = { \
.trace = ftrace_raw_output_##call, \
}; \
\
-static int ftrace_raw_init_event_##call(void) \
+static int ftrace_raw_init_event_##call(struct ftrace_event_call *unused)\
{ \
int id; \
\
diff --git a/include/trace/syscall.h b/include/trace/syscall.h
index 0cb0362..848b4ae 100644
--- a/include/trace/syscall.h
+++ b/include/trace/syscall.h
@@ -51,10 +51,10 @@ void set_syscall_enter_id(int num, int id);
void set_syscall_exit_id(int num, int id);
extern struct trace_event event_syscall_enter;
extern struct trace_event event_syscall_exit;
-extern int reg_event_syscall_enter(void *ptr);
-extern void unreg_event_syscall_enter(void *ptr);
-extern int reg_event_syscall_exit(void *ptr);
-extern void unreg_event_syscall_exit(void *ptr);
+extern int reg_event_syscall_enter(struct ftrace_event_call *call);
+extern void unreg_event_syscall_enter(struct ftrace_event_call *call);
+extern int reg_event_syscall_exit(struct ftrace_event_call *call);
+extern void unreg_event_syscall_exit(struct ftrace_event_call *call);
extern int
ftrace_format_syscall(struct ftrace_event_call *call, struct trace_seq *s);
enum print_line_t print_syscall_enter(struct trace_iterator *iter, int flags);
diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c
index b568ade..be701d1 100644
--- a/kernel/trace/trace_events.c
+++ b/kernel/trace/trace_events.c
@@ -62,9 +62,7 @@ err:
}
EXPORT_SYMBOL_GPL(trace_define_field);
-#ifdef CONFIG_MODULES
-
-static void trace_destroy_fields(struct ftrace_event_call *call)
+void trace_destroy_fields(struct ftrace_event_call *call)
{
struct ftrace_event_field *field, *next;
@@ -76,8 +74,6 @@ static void trace_destroy_fields(struct ftrace_event_call *call)
}
}
-#endif /* CONFIG_MODULES */
-
static void ftrace_event_enable_disable(struct ftrace_event_call *call,
int enable)
{
@@ -86,14 +82,14 @@ static void ftrace_event_enable_disable(struct ftrace_event_call *call,
if (call->enabled) {
call->enabled = 0;
tracing_stop_cmdline_record();
- call->unregfunc(call->data);
+ call->unregfunc(call);
}
break;
case 1:
if (!call->enabled) {
call->enabled = 1;
tracing_start_cmdline_record();
- call->regfunc(call->data);
+ call->regfunc(call);
}
break;
}
@@ -941,7 +937,7 @@ event_create_dir(struct ftrace_event_call *call, struct dentry *d_events,
id);
if (call->define_fields) {
- ret = call->define_fields();
+ ret = call->define_fields(call);
if (ret < 0) {
pr_warning("Could not initialize trace point"
" events/%s\n", call->name);
@@ -961,27 +957,43 @@ event_create_dir(struct ftrace_event_call *call, struct dentry *d_events,
return 0;
}
-#define for_each_event(event, start, end) \
- for (event = start; \
- (unsigned long)event < (unsigned long)end; \
- event++)
+static int __trace_add_event_call(struct ftrace_event_call *call)
+{
+ struct dentry *d_events;
+ int ret;
-#ifdef CONFIG_MODULES
+ if (!call->name)
+ return -EINVAL;
-static LIST_HEAD(ftrace_module_file_list);
+ if (call->raw_init) {
+ ret = call->raw_init(call);
+ if (ret < 0) {
+ if (ret != -ENOSYS)
+ pr_warning("Could not initialize trace "
+ "events/%s\n", call->name);
+ return ret;
+ }
+ }
-/*
- * Modules must own their file_operations to keep up with
- * reference counting.
- */
-struct ftrace_module_file_ops {
- struct list_head list;
- struct module *mod;
- struct file_operations id;
- struct file_operations enable;
- struct file_operations format;
- struct file_operations filter;
-};
+ d_events = event_trace_events_dir();
+ if (!d_events)
+ return -ENOENT;
+
+ list_add(&call->list, &ftrace_events);
+ return event_create_dir(call, d_events, &ftrace_event_id_fops,
+ &ftrace_enable_fops, &ftrace_event_filter_fops,
+ &ftrace_event_format_fops);
+}
+
+/* Add an additional event_call dynamically */
+int trace_add_event_call(struct ftrace_event_call *call)
+{
+ int ret;
+ mutex_lock(&event_mutex);
+ ret = __trace_add_event_call(call);
+ mutex_unlock(&event_mutex);
+ return ret;
+}
static void remove_subsystem_dir(const char *name)
{
@@ -1009,6 +1021,48 @@ static void remove_subsystem_dir(const char *name)
}
}
+static void __trace_remove_event_call(struct ftrace_event_call *call)
+{
+ ftrace_event_enable_disable(call, 0);
+ if (call->event)
+ __unregister_ftrace_event(call->event);
+ debugfs_remove_recursive(call->dir);
+ list_del(&call->list);
+ trace_destroy_fields(call);
+ destroy_preds(call);
+ remove_subsystem_dir(call->system);
+}
+
+/* Remove an event_call */
+void trace_remove_event_call(struct ftrace_event_call *call)
+{
+ mutex_lock(&event_mutex);
+ __trace_remove_event_call(call);
+ mutex_unlock(&event_mutex);
+}
+
+#define for_each_event(event, start, end) \
+ for (event = start; \
+ (unsigned long)event < (unsigned long)end; \
+ event++)
+
+#ifdef CONFIG_MODULES
+
+static LIST_HEAD(ftrace_module_file_list);
+
+/*
+ * Modules must own their file_operations to keep up with
+ * reference counting.
+ */
+struct ftrace_module_file_ops {
+ struct list_head list;
+ struct module *mod;
+ struct file_operations id;
+ struct file_operations enable;
+ struct file_operations format;
+ struct file_operations filter;
+};
+
static struct ftrace_module_file_ops *
trace_create_file_ops(struct module *mod)
{
@@ -1066,7 +1120,7 @@ static void trace_module_add_events(struct module *mod)
if (!call->name)
continue;
if (call->raw_init) {
- ret = call->raw_init();
+ ret = call->raw_init(call);
if (ret < 0) {
if (ret != -ENOSYS)
pr_warning("Could not initialize trace "
@@ -1101,14 +1155,7 @@ static void trace_module_remove_events(struct module *mod)
list_for_each_entry_safe(call, p, &ftrace_events, list) {
if (call->mod == mod) {
found = true;
- ftrace_event_enable_disable(call, 0);
- if (call->event)
- __unregister_ftrace_event(call->event);
- debugfs_remove_recursive(call->dir);
- list_del(&call->list);
- trace_destroy_fields(call);
- destroy_preds(call);
- remove_subsystem_dir(call->system);
+ __trace_remove_event_call(call);
}
}
@@ -1226,7 +1273,7 @@ static __init int event_trace_init(void)
if (!call->name)
continue;
if (call->raw_init) {
- ret = call->raw_init();
+ ret = call->raw_init(call);
if (ret < 0) {
if (ret != -ENOSYS)
pr_warning("Could not initialize trace "
diff --git a/kernel/trace/trace_export.c b/kernel/trace/trace_export.c
index 956d4bc..71c8d7f 100644
--- a/kernel/trace/trace_export.c
+++ b/kernel/trace/trace_export.c
@@ -117,10 +117,16 @@ ftrace_format_##call(struct ftrace_event_call *unused, \
#define TRACE_FIELD_SPECIAL(type_item, item, len, cmd) \
cmd;
+static int ftrace_raw_init_event(struct ftrace_event_call *event_call)
+{
+ INIT_LIST_HEAD(&event_call->fields);
+ init_preds(event_call);
+ return 0;
+}
+
#undef TRACE_EVENT_FORMAT
#define TRACE_EVENT_FORMAT(call, proto, args, fmt, tstruct, tpfmt) \
-int ftrace_define_fields_##call(void); \
-static int ftrace_raw_init_event_##call(void); \
+int ftrace_define_fields_##call(struct ftrace_event_call *c); \
\
struct ftrace_event_call __used \
__attribute__((__aligned__(4))) \
@@ -128,16 +134,10 @@ __attribute__((section("_ftrace_events"))) event_##call = { \
.name = #call, \
.id = proto, \
.system = __stringify(TRACE_SYSTEM), \
- .raw_init = ftrace_raw_init_event_##call, \
+ .raw_init = ftrace_raw_init_event, \
.show_format = ftrace_format_##call, \
.define_fields = ftrace_define_fields_##call, \
-}; \
-static int ftrace_raw_init_event_##call(void) \
-{ \
- INIT_LIST_HEAD(&event_##call.fields); \
- init_preds(&event_##call); \
- return 0; \
-} \
+};
#undef TRACE_EVENT_FORMAT_NOFILTER
#define TRACE_EVENT_FORMAT_NOFILTER(call, proto, args, fmt, tstruct, \
@@ -184,9 +184,8 @@ __attribute__((section("_ftrace_events"))) event_##call = { \
#undef TRACE_EVENT_FORMAT
#define TRACE_EVENT_FORMAT(call, proto, args, fmt, tstruct, tpfmt) \
int \
-ftrace_define_fields_##call(void) \
+ftrace_define_fields_##call(struct ftrace_event_call *event_call) \
{ \
- struct ftrace_event_call *event_call = &event_##call; \
struct args field; \
int ret; \
\
diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c
index f837ccc..3451621 100644
--- a/kernel/trace/trace_syscalls.c
+++ b/kernel/trace/trace_syscalls.c
@@ -210,13 +210,13 @@ void ftrace_syscall_exit(struct pt_regs *regs, long ret)
trace_wake_up();
}
-int reg_event_syscall_enter(void *ptr)
+int reg_event_syscall_enter(struct ftrace_event_call *call)
{
int ret = 0;
int num;
char *name;
- name = (char *)ptr;
+ name = (char *)call->data;
num = syscall_name_to_nr(name);
if (num < 0 || num >= FTRACE_SYSCALL_MAX)
return -ENOSYS;
@@ -234,12 +234,12 @@ int reg_event_syscall_enter(void *ptr)
return ret;
}
-void unreg_event_syscall_enter(void *ptr)
+void unreg_event_syscall_enter(struct ftrace_event_call *call)
{
int num;
char *name;
- name = (char *)ptr;
+ name = (char *)call->data;
num = syscall_name_to_nr(name);
if (num < 0 || num >= FTRACE_SYSCALL_MAX)
return;
@@ -251,13 +251,13 @@ void unreg_event_syscall_enter(void *ptr)
mutex_unlock(&syscall_trace_lock);
}
-int reg_event_syscall_exit(void *ptr)
+int reg_event_syscall_exit(struct ftrace_event_call *call)
{
int ret = 0;
int num;
char *name;
- name = (char *)ptr;
+ name = (char *)call->data;
num = syscall_name_to_nr(name);
if (num < 0 || num >= FTRACE_SYSCALL_MAX)
return -ENOSYS;
@@ -275,12 +275,12 @@ int reg_event_syscall_exit(void *ptr)
return ret;
}
-void unreg_event_syscall_exit(void *ptr)
+void unreg_event_syscall_exit(struct ftrace_event_call *call)
{
int num;
char *name;
- name = (char *)ptr;
+ name = (char *)call->data;
num = syscall_name_to_nr(name);
if (num < 0 || num >= FTRACE_SYSCALL_MAX)
return;
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America), Inc.
Software Solutions Division
e-mail: mhiramat@redhat.com
^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH -tip v14 07/12] tracing: Introduce TRACE_FIELD_ZERO() macro
2009-08-13 20:31 [PATCH -tip v14 00/12] tracing: kprobe-based event tracer and x86 instruction decoder Masami Hiramatsu
` (9 preceding siblings ...)
2009-08-13 20:33 ` [PATCH -tip v14 11/12] tracing: Kprobe tracer assigns new event ids for each event Masami Hiramatsu
@ 2009-08-13 20:47 ` Masami Hiramatsu
[not found] ` <20090819010903.GI5231@nowhere>
2009-08-13 20:47 ` [PATCH -tip v14 06/12] tracing: ftrace dynamic ftrace_event_call support Masami Hiramatsu
` (2 subsequent siblings)
13 siblings, 1 reply; 29+ messages in thread
From: Masami Hiramatsu @ 2009-08-13 20:47 UTC (permalink / raw)
To: Ingo Molnar, Steven Rostedt, lkml
Cc: systemtap, kvm, DLE, Masami Hiramatsu,
Ananth N Mavinakayanahalli, Avi Kivity, Andi Kleen,
Christoph Hellwig, Frank Ch. Eigler, Frederic Weisbecker,
H. Peter Anvin, Ingo Molnar, Jason Baron, Jim Keniston, K.Prasad,
Lai Jiangshan, Li Zefan, PrzemysławPawełczyk,
Roland McGrath, Sam Ravnborg, Srikar Dronamraju, Steven Rostedt,
Tom Zanussi, Vegard Nossum
Use TRACE_FIELD_ZERO(type, item) instead of TRACE_FIELD_ZERO_CHAR(item).
This also includes a fix of TRACE_ZERO_CHAR() macro.
Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Avi Kivity <avi@redhat.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Frank Ch. Eigler <fche@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Jason Baron <jbaron@redhat.com>
Cc: Jim Keniston <jkenisto@us.ibm.com>
Cc: K.Prasad <prasad@linux.vnet.ibm.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Przemysław Pawełczyk <przemyslaw@pawelczyk.it>
Cc: Roland McGrath <roland@redhat.com>
Cc: Sam Ravnborg <sam@ravnborg.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Tom Zanussi <tzanussi@gmail.com>
Cc: Vegard Nossum <vegard.nossum@gmail.com>
---
kernel/trace/trace_event_types.h | 4 ++--
kernel/trace/trace_export.c | 16 ++++++++--------
2 files changed, 10 insertions(+), 10 deletions(-)
diff --git a/kernel/trace/trace_event_types.h b/kernel/trace/trace_event_types.h
index 6db005e..e74f090 100644
--- a/kernel/trace/trace_event_types.h
+++ b/kernel/trace/trace_event_types.h
@@ -109,7 +109,7 @@ TRACE_EVENT_FORMAT(bprint, TRACE_BPRINT, bprint_entry, ignore,
TRACE_STRUCT(
TRACE_FIELD(unsigned long, ip, ip)
TRACE_FIELD(char *, fmt, fmt)
- TRACE_FIELD_ZERO_CHAR(buf)
+ TRACE_FIELD_ZERO(char, buf)
),
TP_RAW_FMT("%08lx (%d) fmt:%p %s")
);
@@ -117,7 +117,7 @@ TRACE_EVENT_FORMAT(bprint, TRACE_BPRINT, bprint_entry, ignore,
TRACE_EVENT_FORMAT(print, TRACE_PRINT, print_entry, ignore,
TRACE_STRUCT(
TRACE_FIELD(unsigned long, ip, ip)
- TRACE_FIELD_ZERO_CHAR(buf)
+ TRACE_FIELD_ZERO(char, buf)
),
TP_RAW_FMT("%08lx (%d) fmt:%p %s")
);
diff --git a/kernel/trace/trace_export.c b/kernel/trace/trace_export.c
index 71c8d7f..b0ac92c 100644
--- a/kernel/trace/trace_export.c
+++ b/kernel/trace/trace_export.c
@@ -42,9 +42,9 @@ extern void __bad_type_size(void);
if (!ret) \
return 0;
-#undef TRACE_FIELD_ZERO_CHAR
-#define TRACE_FIELD_ZERO_CHAR(item) \
- ret = trace_seq_printf(s, "\tfield:char " #item ";\t" \
+#undef TRACE_FIELD_ZERO
+#define TRACE_FIELD_ZERO(type, item) \
+ ret = trace_seq_printf(s, "\tfield:" #type " " #item ";\t" \
"offset:%u;\tsize:0;\n", \
(unsigned int)offsetof(typeof(field), item)); \
if (!ret) \
@@ -92,9 +92,6 @@ ftrace_format_##call(struct ftrace_event_call *unused, \
#include "trace_event_types.h"
-#undef TRACE_ZERO_CHAR
-#define TRACE_ZERO_CHAR(arg)
-
#undef TRACE_FIELD
#define TRACE_FIELD(type, item, assign)\
entry->item = assign;
@@ -107,6 +104,9 @@ ftrace_format_##call(struct ftrace_event_call *unused, \
#define TRACE_FIELD_SIGN(type, item, assign, is_signed) \
TRACE_FIELD(type, item, assign)
+#undef TRACE_FIELD_ZERO
+#define TRACE_FIELD_ZERO(type, item)
+
#undef TP_CMD
#define TP_CMD(cmd...) cmd
@@ -178,8 +178,8 @@ __attribute__((section("_ftrace_events"))) event_##call = { \
if (ret) \
return ret;
-#undef TRACE_FIELD_ZERO_CHAR
-#define TRACE_FIELD_ZERO_CHAR(item)
+#undef TRACE_FIELD_ZERO
+#define TRACE_FIELD_ZERO(type, item)
#undef TRACE_EVENT_FORMAT
#define TRACE_EVENT_FORMAT(call, proto, args, fmt, tstruct, tpfmt) \
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America), Inc.
Software Solutions Division
e-mail: mhiramat@redhat.com
^ permalink raw reply [flat|nested] 29+ messages in thread
* [TOOL] kprobestest : Kprobe stress test tool
2009-08-13 20:31 [PATCH -tip v14 00/12] tracing: kprobe-based event tracer and x86 instruction decoder Masami Hiramatsu
` (11 preceding siblings ...)
2009-08-13 20:47 ` [PATCH -tip v14 06/12] tracing: ftrace dynamic ftrace_event_call support Masami Hiramatsu
@ 2009-08-13 20:54 ` Masami Hiramatsu
[not found] ` <20090820184331.GA6078@nowhere>
2009-08-13 20:56 ` [TOOL] c2kpe: C expression to kprobe event format converter Masami Hiramatsu
13 siblings, 1 reply; 29+ messages in thread
From: Masami Hiramatsu @ 2009-08-13 20:54 UTC (permalink / raw)
To: Ingo Molnar, Steven Rostedt, lkml
Cc: Ananth N Mavinakayanahalli, Avi Kivity, Andi Kleen,
Christoph Hellwig, Frank Ch. Eigler, Frederic Weisbecker,
H. Peter Anvin, Jason Baron, Jim Keniston, K.Prasad,
Lai Jiangshan, Li Zefan, PrzemysławPawełczyk,
Roland McGrath, Sam Ravnborg, Srikar Dronamraju, Tom Zanussi,
Vegard Nossum, systemtap, kvm, DLE
[-- Attachment #1: Type: text/plain, Size: 1476 bytes --]
This script tests kprobes to probe on all symbols in the kernel and finds
symbols which must be blacklisted.
Usage
-----
kprobestest [-s SYMLIST] [-b BLACKLIST] [-w WHITELIST]
Run stress test. If SYMLIST file is specified, use it as
an initial symbol list (This is useful for verifying white list
after diagnosing all symbols).
kprobestest cleanup
Cleanup all lists
How to Work
-----------
This tool list up all symbols in the kernel via /proc/kallsyms, and sorts
it into groups (each of them including 64 symbols in default). And then,
it tests each group by using kprobe-tracer. If a kernel crash occurred,
that group is moved into 'failed' dir. If the group passed the test, this
script moves it into 'passed' dir and saves kprobe_profile into
'passed/profiles/'.
After testing all groups, all 'failed' groups are merged and sorted into
smaller groups (divided by 4, in default). And those are tested again.
This loop will be repeated until all group has just 1 symbol.
Finally, the script sorts all 'passed' symbols into 'tested', 'untested',
and 'missed' based on profiles.
Note
----
- This script just gives us some clues to the blacklisted functions.
In some cases, a combination of probe points will cause a problem, but
each of them doesn't cause the problem alone.
Thank you,
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America), Inc.
Software Solutions Division
e-mail: mhiramat@redhat.com
[-- Attachment #2: kprobestest --]
[-- Type: text/plain, Size: 4945 bytes --]
#!/bin/bash
#
# kprobestest: Kprobes stress test tool
# Written by Masami Hiramatsu <mhiramat@redhat.com>
#
# Usage:
# $ kprobestest [-s SYMLIST] [-b BLACKLIST] [-w WHITELIST]
# Run stress test. If SYMLIST file is specified, use it as
# an initial symbol list (This is useful for verifying white list
# after diagnosing all symbols).
#
# $ kprobestest cleanup
# Cleanup all lists
# Configurations
DEBUGFS=/sys/kernel/debug
INITNR=64
DIV=4
SYMFILE=syms.list
FAILFILE=black.list
function do_test () {
# Do some benchmark
for i in {1..4} ; do
sleep 0.5
echo -n "."
done
}
function usage () {
echo "Usage: kprobestest [cleanup] [-s SYMLIST] [-b BLACKLIST] [-w WHITELIST]"
exit 0
}
function cleanup_test () {
echo "Cleanup all files"
rm -rf $SYMFILE failed passed testing unset
exit 0
}
# Parse arguments
WHITELIST=
BLACKLIST=
SYMLIST=
while [ "$1" ]; do
case $1 in
cleanup)
cleanup_test
;;
-s)
SYMLIST=$2
shift 1
;;
-b)
BLACKLIST=$2
shift 1
;;
-w)
WHITELIST=$2
shift 1
;;
*)
usage
;;
esac
shift 1
done
# Show configurations
echo "Kprobe stress test starting."
[ -f "$BLACKLIST" ] && echo "Blacklist: $BLACKLIST" || BLACKLIST=""
[ -f "$WHITELIST" ] && echo "Whitelist: $WHITELIST" || WHITELIST=""
[ -f "$SYMLIST" ] && echo "Symlist: $SYMLIST" || SYMLIST=""
function make_filter () {
local EXP=""
if [ -z "$WHITELIST" -a -z "$BLACKLIST" ]; then
echo "s/^$//g"
else
for i in `cat $WHITELIST $BLACKLIST` ;do
[ -z "$EXP" ] && EXP="^$i\$" || EXP="$EXP\\|^$i\$"
done ; EXP="s/$EXP//g"
echo $EXP
fi
}
function list_allsyms () {
local sym
local out=1
for sym in `sort /proc/kallsyms | egrep '[0-9a-f]+ [Tt] [^[]*$' | cut -d\ -f 3`;do
[ $sym = "__kprobes_text_start" ] && out=0 && continue
[ $sym = "__kprobes_text_end" ] && out=1 && continue
[ $sym = "_etext" ] && break
[ $out -eq 1 ] && echo $sym
done
}
function prep_testing () {
local i=0
local n=0
local NR=$1
local fname=
echo "Grouping symbols: $NR"
fname=`printf "list-%03d.%d" $i $NR`
cat $SYMFILE | while read ln; do
[ -z "$ln" ] && continue
echo "$ln" >> testing/$fname
n=$((n+1))
if [ $n -eq $NR ]; then
n=0
i=$((i+1))
fname=`printf "list-%03d.%d" $i $NR`
fi
done
sync
}
function init_first () {
local EXP
EXP=`make_filter`
if [ -f "$SYMLIST" ]; then
cat $SYMLIST | sed $EXP > $SYMFILE
else
echo -n "Generating symbol list from /proc/kallsyms..."
list_allsyms | sed $EXP > $SYMFILE
echo "done. " `wc -l $SYMFILE | cut -f1 -d\ ` "symbols listed."
fi
mkdir -p testing failed unset passed passed/profiles
prep_testing $INITNR
}
function get_max_nr () {
wc -l failed/list-* unset/list-* 2>/dev/null |\
awk '/^ *[0-9]+ .*list.*$/{ if (nr < $1) nr=$1 } BEGIN { nr=0 } END { print nr}'
}
function init_next () {
local NR
NR=`get_max_nr`
[ $NR -eq 0 ] && return 1
[ $NR -eq 1 ] && return 2
[ $NR -le $DIV ] && NR=1 || NR=`expr $NR / $DIV`
cat failed/* unset/* > $SYMFILE
rm failed/* unset/*
prep_testing $NR
return 0
}
# Initialize symbols
if [ ! -d testing ]; then
init_first
elif [ -z "`ls testing/`" ]; then
init_next
fi
function set_probes () {
local s
for s in `cat $1`; do
echo "p:$s" $s >> $DEBUGFS/tracing/kprobe_events
[ $? -ne 0 ] && return -1
done
return 0
}
function clear_probes () {
echo > $DEBUGFS/tracing/kprobe_events
}
function save_profile () {
cat $DEBUGFS/tracing/kprobe_profile > passed/profiles/$1.prof
}
clear_probes
echo "Starting tests.."
RET=0
# Main loop
while [ $RET -eq 0 ]; do
for list in `cd testing/; ls`; do
echo -n $list
# Temporary moving list into failed.
mv testing/$list failed/
sync;sync
echo -n "sync.."
set_probes failed/$list
if [ $? -ne 0 ]; then
clear_probes
sync;sync
echo "can not set"
mv failed/$list unset/
sync;sync
else
do_test
save_profile $list
clear_probes
sync;sync
echo "done"
mv failed/$list passed/
sync;sync
fi
done
init_next
RET=$?
done
if [ $RET -eq 1 ];then
# No failed symbols
echo "no failed symbols found."
else
echo "found failed symbols:"
cat failed/* | tee $FAILFILE
rm failed/*
fi
function profile_symbols () {
local s h m
rm -f tested.list missed.list untested.list
cat passed/profiles/*.prof | while read s h m ;do
if [ $h -ne 0 ]; then
echo $s >> tested.list
elif [ $m -ne 0 ]; then
echo $s >> missed.list
else
echo $s >> untested.list
fi
done
}
echo -n "Profiling symbols..."
profile_symbols
echo done
echo tested: `wc -l tested.list | cut -d\ -f1` symbols
echo missed: `wc -l missed.list | cut -d\ -f1` symbols
echo untested: `wc -l untested.list | cut -d\ -f1` symbols
^ permalink raw reply [flat|nested] 29+ messages in thread
* [TOOL] c2kpe: C expression to kprobe event format converter
2009-08-13 20:31 [PATCH -tip v14 00/12] tracing: kprobe-based event tracer and x86 instruction decoder Masami Hiramatsu
` (12 preceding siblings ...)
2009-08-13 20:54 ` [TOOL] kprobestest : Kprobe stress test tool Masami Hiramatsu
@ 2009-08-13 20:56 ` Masami Hiramatsu
[not found] ` <20090830195041.GA6189@nowhere>
13 siblings, 1 reply; 29+ messages in thread
From: Masami Hiramatsu @ 2009-08-13 20:56 UTC (permalink / raw)
To: Ingo Molnar, Steven Rostedt, lkml
Cc: Ananth N Mavinakayanahalli, Avi Kivity, Andi Kleen,
Christoph Hellwig, Frank Ch. Eigler, Frederic Weisbecker,
H. Peter Anvin, Jason Baron, Jim Keniston, K.Prasad,
Lai Jiangshan, Li Zefan, PrzemysławPawełczyk,
Roland McGrath, Sam Ravnborg, Srikar Dronamraju, Tom Zanussi,
Vegard Nossum, systemtap, kvm, DLE
[-- Attachment #1: Type: text/plain, Size: 2014 bytes --]
This program converts probe point in C expression to kprobe event
format for kprobe-based event tracer. This helps to define kprobes
events by C source line number or function name, and local variable
name. Currently, this supports only x86(32/64) kernels.
Compile
--------
Before compilation, please install libelf and libdwarf development
packages.
(e.g. elfutils-libelf-devel and libdwarf-devel on Fedora)
$ gcc -Wall -lelf -ldwarf c2kpe.c -o c2kpe
Synopsis
--------
$ c2kpe [options] FUNCTION[+OFFS][@SRC] [VAR [VAR ...]]
or
$ c2kpe [options] @SRC:LINE [VAR [VAR ...]]
FUNCTION: Probing function name.
OFFS: Offset in bytes.
SRC: Source file path.
LINE: Line number
VAR: Local variable name.
options:
-r KREL Kernel release version (e.g. 2.6.31-rc5)
-m DEBUGINFO Dwarf-format binary file (vmlinux or kmodule)
Example
-------
$ c2kpe sys_read fd buf count
sys_read+0 %di %si %dx
$ c2kpe @mm/filemap.c:339 inode pos
sync_page_range+125 -48(%bp) %r14
Example with kprobe-tracer
--------------------------
Since C expression may be converted multiple results, I recommend to use
readline.
$ c2kpe sys_read fd buf count | while read i; do \
echo "p $i" > $DEBUGFS/tracing/kprobe_events ;\
done
Note
----
- This requires a kernel compiled with CONFIG_DEBUG_INFO.
- Specifying @SRC speeds up c2kpe, because we can skip CUs which don't
include specified SRC file.
- c2kpe doesn't check whether the offset byte is correctly on the
instruction boundary. I recommend you to use @SRC:LINE expression for
tracing function body.
- This tool doesn't search kmodule file. You need to specify kmodule
file if you want to probe it.
TODO
----
- Fix bugs.
- Support multiple probepoints from stdin.
- Better kmodule support.
- Use elfutils-libdw?
- Merge into trace-cmd or perf-tools?
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America), Inc.
Software Solutions Division
e-mail: mhiramat@redhat.com
[-- Attachment #2: c2kpe.c --]
[-- Type: text/plain, Size: 21095 bytes --]
/*
* c2kpe : C expression to kprobe event converter
*
* Written by Masami Hiramatsu <mhiramat@redhat.com>
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
*
*/
#include <sys/utsname.h>
#include <sys/types.h>
#include <sys/stat.h>
#include <fcntl.h>
#include <errno.h>
#include <stdio.h>
#include <unistd.h>
#include <getopt.h>
#include <stdlib.h>
#include <string.h>
#include <libdwarf/dwarf.h>
#include <libdwarf/libdwarf.h>
/* Default vmlinux search paths */
#define NR_SEARCH_PATH 2
const char *default_search_path[NR_SEARCH_PATH] = {
"/lib/modules/%s/build/vmlinux", /* Custom build kernel */
"/usr/lib/debug/lib/modules/%s/vmlinux", /* Red Hat debuginfo */
};
#define _stringify(n) #n
#define stringify(n) _stringify(n)
#ifdef DEBUG
#define debug(fmt ...) \
fprintf(stderr, "DBG(" __FILE__ ":" stringify(__LINE__) "): " fmt)
#else
#define debug(fmt ...) do {} while (0)
#endif
#define ERR_IF(cnd) \
do { if (cnd) { \
fprintf(stderr, "Error (" __FILE__ ":" stringify(__LINE__) \
"): " stringify(cnd) "\n"); \
exit(1); \
}} while (0)
#define MAX_PATH_LEN 256
/* Dwarf_Die Linkage to parent Die */
struct die_link {
struct die_link *parent; /* Parent die */
Dwarf_Die die; /* Current die */
};
#define X86_32_MAX_REGS 8
const char *x86_32_regs_table[X86_32_MAX_REGS] = {
"%ax",
"%cx",
"%dx",
"%bx",
"sa", /* Stack address */
"%bp",
"%si",
"%di",
};
#define X86_64_MAX_REGS 16
const char *x86_64_regs_table[X86_64_MAX_REGS] = {
"%ax",
"%dx",
"%cx",
"%bx",
"%si",
"%di",
"%bp",
"%sp",
"%r8",
"%r9",
"%r10",
"%r11",
"%r12",
"%r13",
"%r14",
"%r15",
};
/* TODO: switching by dwarf address size */
#ifdef __x86_64__
#define ARCH_MAX_REGS X86_64_MAX_REGS
#define arch_regs_table x86_64_regs_table
#else
#define ARCH_MAX_REGS X86_32_MAX_REGS
#define arch_regs_table x86_32_regs_table
#endif
/* Return architecture dependent register string */
static inline const char *get_arch_regstr(unsigned int n)
{
return (n <= ARCH_MAX_REGS) ? arch_regs_table[n] : NULL;
}
struct probe_finder {
/* Inputs */
char *file; /* File name */
int line; /* Line number */
char *function; /* Function name */
int offset; /* Offset bytes */
Dwarf_Addr addr; /* Address */
int nr_args; /* Number of arguments */
char **args; /* Arguments */
/* Working area */
Dwarf_Addr cu_base; /* Current CU base address */
Dwarf_Locdesc fbloc; /* Location of Current Frame Base */
Dwarf_Unsigned fno; /* File number */
Dwarf_Off inl_offs; /* Inline offset */
const char *var; /* Current variable name */
/* Output */
int found; /* Number of found probe points */
};
/* Find a probe point */
static void find_probepoint(struct probe_finder *pf);
/* Session management structure */
static struct {
char *kver;
char *modpath;
int maxprobe;
struct probe_finder finder;
Dwarf_Debug dbg;
Dwarf_Error err;
int nfound;
} session;
static void usage(const char *msg)
{
if (msg)
printf("%s\n\n", msg);
printf("c2kpe: C expression to kprobe event converter\n");
printf("Usage: c2kpe [-r KREL] [-m mod|vmlinux] FUNC|SRC [ARG...]\n");
printf(" FUNC:\tFUNCNAME[+OFFS_BYTE][@SRCPATH]\n");
printf(" SRC:\t@SRCPATH:LINE\n");
printf(" ARG:\tLocal variable name\n");
/* TODO: @global, $vars, $params */
exit(0);
}
static void semantic_error(const char *msg)
{
fprintf(stderr, "Semantic error: %s\n", msg);
exit(1);
}
static void parse_probe_point(int argc, char *argv[], struct probe_finder *pf)
{
char *arg;
char *ptr;
arg = argv[0];
if (arg[0] == '@') {
/* Source Line */
arg++;
ptr = strchr(arg, ':');
if (!ptr || ptr[1] == '\0')
semantic_error("Line number is required.");
*ptr++ = '\0';
pf->file = arg;
if (strlen(arg) == 0)
semantic_error("No file name.");
pf->line = atoi(ptr);
debug("file:%s line:%d\n", pf->file, pf->line);
} else {
/* Function name */
pf->function = arg;
ptr = strchr(arg, '+');
if (ptr) {
if (ptr[1] == '\0' || ptr[1] == '@')
semantic_error("Offset is required.");
*ptr++ = '\0';
pf->offset = atoi(ptr);
arg = ptr;
}
ptr = strchr(arg, '@');
if (ptr) {
*ptr++ = '\0';
pf->file = ptr;
}
debug("fname:%s file:%s offset:%d\n",
pf->function, pf->file, pf->offset);
}
pf->nr_args = argc - 1;
if (pf->nr_args > 0)
pf->args = &argv[1];
debug("%d arguments\n", pf->nr_args);
}
static void parse_args(int argc, char *argv[])
{
int opt;
if (argc < 1)
usage("Need a probe point.");
while ((opt = getopt(argc, argv, "r:m:n:")) != -1) {
switch (opt) {
case 'r':
session.kver = optarg;
break;
case 'm':
session.modpath = optarg;
break;
case 'n':
session.maxprobe = atoi(optarg);
break;
default:
usage("Unexpected option found.");
}
}
if (optind >= argc)
usage("Need a probe point.");
parse_probe_point(argc - optind, &argv[optind], &session.finder);
}
static int open_default_vmlinux(const char *kver)
{
struct utsname uts;
char fname[MAX_PATH_LEN];
int fd, ret, i;
if (!kver) {
ret = uname(&uts);
if (ret) {
debug("uname() failed.\n");
return -errno;
}
kver = uts.release;
}
for (i = 0; i < NR_SEARCH_PATH; i++) {
ret = snprintf(fname, MAX_PATH_LEN,
default_search_path[i], kver);
if (ret >= MAX_PATH_LEN || ret < 0) {
debug("Filename(%d,%s) is too long.\n", i, uts.release);
errno = E2BIG;
return -E2BIG;
}
debug("try to open %s\n", fname);
fd = open(fname, O_RDONLY);
if (fd >= 0)
break;
}
return fd;
}
int main(int argc, char *argv[])
{
int fd, ret;
parse_args(argc, argv);
if (session.modpath)
fd = open(session.modpath, O_RDONLY);
else
fd = open_default_vmlinux(session.kver);
if (fd < 0) {
perror("file open");
exit(1);
}
/* TODO: handle errors */
ret = dwarf_init(fd, DW_DLC_READ, 0, 0, &session.dbg, &session.err);
if (ret != DW_DLV_OK) {
fprintf(stderr, "Failed to call dwarf_init(). "
"Maybe, not a dwarf file?\n");
exit(1);
}
find_probepoint(&session.finder);
ret = dwarf_finish(session.dbg, &session.err);
ERR_IF(ret != DW_DLV_OK);
close(fd);
return 0;
}
/*---------------------
* Dwarf Analysys Part
*---------------------*/
/*
* Compare the tail of two strings.
* Return 0 if whole of either string is same as another's tail part.
*/
static int strtailcmp(const char *s1, const char *s2)
{
int i1 = strlen(s1);
int i2 = strlen(s2);
while (--i1 > 0 && --i2 > 0) {
if (s1[i1] != s2[i2])
return s1[i1] - s2[i2];
}
return 0;
}
/* Find the fileno of the target file. */
static Dwarf_Unsigned die_get_fileno(Dwarf_Die cu_die, const char *fname)
{
Dwarf_Signed cnt, i;
Dwarf_Unsigned found = 0;
char **srcs;
int ret;
if (!fname)
return 0;
ret = dwarf_srcfiles(cu_die, &srcs, &cnt, &session.err);
if (ret == DW_DLV_OK) {
for (i = 0; i < cnt && !found; i++) {
if (strtailcmp(srcs[i], fname) == 0)
found = i + 1;
dwarf_dealloc(session.dbg, srcs[i], DW_DLA_STRING);
}
for (;i < cnt; i++)
dwarf_dealloc(session.dbg, srcs[i], DW_DLA_STRING);
dwarf_dealloc(session.dbg, srcs, DW_DLA_LIST);
}
if (found) debug("found fno: %d\n", (int)found);
return found;
}
/* Compare diename and tname */
static int die_compare_name(Dwarf_Die die, const char *tname)
{
char *name;
int ret;
ret = dwarf_diename(die, &name, &session.err);
ERR_IF(ret == DW_DLV_ERROR);
if (ret == DW_DLV_OK) {
//debug("diename: %s\n", name);
ret = strcmp(tname, name);
dwarf_dealloc(session.dbg, name, DW_DLA_STRING);
} else
ret = -1;
return ret;
}
/* Check the address is in the subprogram(function). */
static int die_within_subprogram(Dwarf_Die sp_die, Dwarf_Addr addr,
Dwarf_Signed *offs)
{
Dwarf_Addr lopc, hipc;
int ret;
ret = dwarf_lowpc(sp_die, &lopc, &session.err);
ERR_IF(ret == DW_DLV_ERROR);
if (ret == DW_DLV_NO_ENTRY)
return 0;
ret = dwarf_highpc(sp_die, &hipc, &session.err);
ERR_IF(ret != DW_DLV_OK);
if (lopc <= addr && addr < hipc) {
*offs = addr - lopc;
return 1;
} else
return 0;
}
/* Check the die is inlined function */
static Dwarf_Bool die_inlined_subprogram(Dwarf_Die die)
{
/* TODO: check strictly */
Dwarf_Bool inl;
int ret;
ret = dwarf_hasattr(die, DW_AT_inline, &inl, &session.err);
ERR_IF(ret == DW_DLV_ERROR);
return inl;
}
/* Get the offset of abstruct_origin */
static Dwarf_Off die_get_abstract_origin(Dwarf_Die die)
{
Dwarf_Attribute attr;
Dwarf_Off cu_offs;
int ret;
ret = dwarf_attr(die, DW_AT_abstract_origin, &attr, &session.err);
ERR_IF(ret != DW_DLV_OK);
ret = dwarf_formref(attr, &cu_offs, &session.err);
ERR_IF(ret != DW_DLV_OK);
dwarf_dealloc(session.dbg, attr, DW_DLA_ATTR);
return cu_offs;
}
/* Get entry pc(or low pc, 1st entry of ranges) of the die */
static Dwarf_Addr die_get_entrypc(Dwarf_Die die)
{
Dwarf_Attribute attr;
Dwarf_Addr addr;
Dwarf_Off offs;
Dwarf_Ranges *ranges;
Dwarf_Signed cnt;
int ret;
/* Try to get entry pc */
ret = dwarf_attr(die, DW_AT_entry_pc, &attr, &session.err);
ERR_IF(ret == DW_DLV_ERROR);
if (ret == DW_DLV_OK) {
ret = dwarf_formaddr(attr, &addr, &session.err);
ERR_IF(ret != DW_DLV_OK);
dwarf_dealloc(session.dbg, attr, DW_DLA_ATTR);
return addr;
}
/* Try to get low pc */
ret = dwarf_lowpc(die, &addr, &session.err);
ERR_IF(ret == DW_DLV_ERROR);
if (ret == DW_DLV_OK)
return addr;
/* Try to get ranges */
ret = dwarf_attr(die, DW_AT_ranges, &attr, &session.err);
ERR_IF(ret != DW_DLV_OK);
ret = dwarf_formref(attr, &offs, &session.err);
ERR_IF(ret != DW_DLV_OK);
ret = dwarf_get_ranges(session.dbg, offs, &ranges, &cnt, NULL,
&session.err);
ERR_IF(ret != DW_DLV_OK);
addr = ranges[0].dwr_addr1;
dwarf_ranges_dealloc(session.dbg, ranges, cnt);
return addr;
}
/*
* Search a Die from Die tree.
* Note: cur_link->die should be deallocated in this function.
*/
static int __search_die_tree(struct die_link *cur_link,
int (*die_cb)(struct die_link *, void *),
void *data)
{
Dwarf_Die new_die;
struct die_link link;
int ret;
if (!die_cb)
return 0;
/* Check current die */
while (!(ret = die_cb(cur_link, data))) {
/* Check child die */
ret = dwarf_child(cur_link->die, &new_die, &session.err);
ERR_IF(ret == DW_DLV_ERROR);
if (ret == DW_DLV_OK) {
link.parent = cur_link;
link.die = new_die;
ret = __search_die_tree(&link, die_cb, data);
if (ret)
break;
}
/* Move to next sibling */
ret = dwarf_siblingof(session.dbg, cur_link->die, &new_die,
&session.err);
ERR_IF(ret == DW_DLV_ERROR);
dwarf_dealloc(session.dbg, cur_link->die, DW_DLA_DIE);
cur_link->die = new_die;
if (ret == DW_DLV_NO_ENTRY)
return 0;
}
dwarf_dealloc(session.dbg, cur_link->die, DW_DLA_DIE);
return ret;
}
/* Search a die in its children's die tree */
static int search_die_from_children(Dwarf_Die parent_die,
int (*die_cb)(struct die_link *, void *),
void *data)
{
struct die_link link;
int ret;
link.parent = NULL;
ret = dwarf_child(parent_die, &link.die, &session.err);
ERR_IF(ret == DW_DLV_ERROR);
if (ret == DW_DLV_OK)
return __search_die_tree(&link, die_cb, data);
else
return 0;
}
/* Find a locdesc corresponding to the address */
static int attr_get_locdesc(Dwarf_Attribute attr, Dwarf_Locdesc *desc,
Dwarf_Addr addr)
{
Dwarf_Signed lcnt;
Dwarf_Locdesc **llbuf;
int ret, i;
ret = dwarf_loclist_n(attr, &llbuf, &lcnt, &session.err);
ERR_IF(ret != DW_DLV_OK);
ret = DW_DLV_NO_ENTRY;
for (i = 0; i < lcnt; ++i) {
if (llbuf[i]->ld_lopc <= addr &&
llbuf[i]->ld_hipc > addr ) {
memcpy(desc, llbuf[i], sizeof(Dwarf_Locdesc));
desc->ld_s =
malloc(sizeof(Dwarf_Loc) * llbuf[i]->ld_cents);
ERR_IF(desc->ld_s == NULL);
memcpy(desc->ld_s, llbuf[i]->ld_s,
sizeof(Dwarf_Loc) * llbuf[i]->ld_cents);
ret = DW_DLV_OK;
break;
}
dwarf_dealloc(session.dbg, llbuf[i]->ld_s, DW_DLA_LOC_BLOCK);
dwarf_dealloc(session.dbg, llbuf[i], DW_DLA_LOCDESC);
}
/* Releasing loop */
for (; i < lcnt; ++i) {
dwarf_dealloc(session.dbg, llbuf[i]->ld_s, DW_DLA_LOC_BLOCK);
dwarf_dealloc(session.dbg, llbuf[i], DW_DLA_LOCDESC);
}
dwarf_dealloc(session.dbg, llbuf, DW_DLA_LIST);
return ret;
}
/*---------------------------------
* Probe finder related functions
*-------------------------------*/
/* Show a location */
static void show_location(Dwarf_Loc *loc, struct probe_finder *pf)
{
Dwarf_Small op;
Dwarf_Unsigned regn;
Dwarf_Signed offs;
int deref = 0;
const char *regs;
op = loc->lr_atom;
/* If this is based on frame buffer, set the offset */
if (op == DW_OP_fbreg) {
deref = 1;
offs = (Dwarf_Signed)loc->lr_number;
op = pf->fbloc.ld_s[0].lr_atom;
loc = &pf->fbloc.ld_s[0];
} else
offs = 0;
if (op >= DW_OP_breg0 && op <= DW_OP_breg31) {
regn = op - DW_OP_breg0;
offs += (Dwarf_Signed)loc->lr_number;
deref = 1;
} else if (op >= DW_OP_reg0 && op <= DW_OP_reg31) {
regn = op - DW_OP_reg0;
} else if (op == DW_OP_bregx) {
regn = loc->lr_number;
offs += (Dwarf_Signed)loc->lr_number2;
deref = 1;
} else if (op == DW_OP_regx) {
regn = loc->lr_number;
} else {
fprintf(stderr, "Error: Dwarf_OP %d is not supported.\n", op);
exit(1);
}
regs = get_arch_regstr(regn);
if (!regs) {
fprintf(stderr,
"Error: %lld exceeds max register number.\n", regn);
exit(1);
}
if (deref)
printf(" %+lld(%s)", offs, regs);
else
printf(" %s", regs);
}
/* Show a variables in kprobe event format */
static void show_variable(Dwarf_Die vr_die, struct probe_finder *pf)
{
Dwarf_Attribute attr;
Dwarf_Locdesc ld;
int ret;
ret = dwarf_attr(vr_die, DW_AT_location, &attr, &session.err);
ERR_IF(ret != DW_DLV_OK);
ret = attr_get_locdesc(attr, &ld, (pf->addr - pf->cu_base));
ERR_IF(ret != DW_DLV_OK);
/* TODO? */
if (ld.ld_cents != 1) {
fprintf(stderr, "This variable type is not supported.\n");
exit(1);
}
show_location(&ld.ld_s[0], pf);
free(ld.ld_s);
dwarf_dealloc(session.dbg, attr, DW_DLA_ATTR);
}
static int variable_callback(struct die_link *link, void *data)
{
struct probe_finder *pf = (struct probe_finder *)data;
Dwarf_Half tag;
int ret;
ret = dwarf_tag(link->die, &tag, &session.err);
ERR_IF(ret == DW_DLV_ERROR);
if ((tag == DW_TAG_formal_parameter ||
tag == DW_TAG_variable) &&
(die_compare_name(link->die, pf->var) == 0)) {
show_variable(link->die, pf);
return 1;
}
/* TODO: Support struct members */
return 0;
}
/* Find a variable in a subprogram die */
static void find_variable(const char *var, Dwarf_Die sp_die,
struct probe_finder *pf)
{
int ret;
debug("Searching %s variable in context.\n", var);
pf->var = var;
/* Search child die for local variables and parameters. */
ret = search_die_from_children(sp_die, variable_callback, pf);
if (!ret) {
fprintf(stderr, "\nFailed to find %s in this function.\n", var);
exit(1);
}
}
/* Get a frame base on the address */
static void get_current_frame_base(Dwarf_Die sp_die, struct probe_finder *pf)
{
Dwarf_Attribute attr;
int ret;
ret = dwarf_attr(sp_die, DW_AT_frame_base, &attr, &session.err);
ERR_IF(ret != DW_DLV_OK);
ret = attr_get_locdesc(attr, &pf->fbloc, (pf->addr - pf->cu_base));
ERR_IF(ret != DW_DLV_OK);
dwarf_dealloc(session.dbg, attr, DW_DLA_ATTR);
}
static void free_current_frame_base(struct probe_finder *pf)
{
free(pf->fbloc.ld_s);
memset(&pf->fbloc, 0, sizeof(Dwarf_Locdesc));
}
/* Show a probe point to stdout */
static void show_probepoint(Dwarf_Die sp_die, Dwarf_Signed offs,
struct probe_finder *pf)
{
char *name;
int ret, i;
ret = dwarf_diename(sp_die, &name, &session.err);
ERR_IF(ret == DW_DLV_ERROR);
if (ret == DW_DLV_OK) {
printf("%s%+d", name, (int)offs);
dwarf_dealloc(session.dbg, name, DW_DLA_STRING);
} else {
/* This function has no name. */
printf("0x%llx", pf->addr);
}
get_current_frame_base(sp_die, pf);
for (i = 0; i < pf->nr_args; i++)
find_variable(pf->args[i], sp_die, pf);
free_current_frame_base(pf);
printf("\n");
pf->found++;
}
static int probeaddr_callback(struct die_link *link, void *data)
{
struct probe_finder *pf = (struct probe_finder *)data;
Dwarf_Half tag;
Dwarf_Signed offs;
int ret;
ret = dwarf_tag(link->die, &tag, &session.err);
ERR_IF(ret == DW_DLV_ERROR);
if (tag == DW_TAG_subprogram &&
die_within_subprogram(link->die, pf->addr, &offs)) {
show_probepoint(link->die, offs, pf);
return 1;
}
return 0;
}
/* Find probe point from its line number */
static void find_by_line(Dwarf_Die cu_die, struct probe_finder *pf)
{
Dwarf_Signed cnt, i;
Dwarf_Line *lines;
Dwarf_Unsigned lineno = 0;
Dwarf_Addr addr;
Dwarf_Unsigned fno;
int ret;
ret = dwarf_srclines(cu_die, &lines, &cnt, &session.err);
ERR_IF(ret != DW_DLV_OK);
for (i = 0; i < cnt; i++) {
ret = dwarf_line_srcfileno(lines[i], &fno, &session.err);
ERR_IF(ret != DW_DLV_OK);
if (fno != pf->fno)
continue;
ret = dwarf_lineno(lines[i], &lineno, &session.err);
ERR_IF(ret != DW_DLV_OK);
if (lineno != pf->line)
continue;
ret = dwarf_lineaddr(lines[i], &addr, &session.err);
ERR_IF(ret != DW_DLV_OK);
debug("Probe point found: 0x%llx\n", addr);
pf->addr = addr;
/* Search a real subprogram including this line, */
ret = search_die_from_children(cu_die, probeaddr_callback, pf);
if (ret == 0) {
fprintf(stderr,
"Probe point is not found in subprograms.\n");
exit(1);
}
/* Continuing, because target line might be inlined. */
}
dwarf_srclines_dealloc(session.dbg, lines, cnt);
}
/* Search function from function name */
static int probefunc_callback(struct die_link *link, void *data)
{
struct probe_finder *pf = (struct probe_finder *)data;
struct die_link *lk;
Dwarf_Signed offs;
Dwarf_Half tag;
int ret;
ret = dwarf_tag(link->die, &tag, &session.err);
ERR_IF(ret == DW_DLV_ERROR);
if (tag == DW_TAG_subprogram) {
if (die_compare_name(link->die, pf->function) == 0) {
if (die_inlined_subprogram(link->die)) {
/* Inlined function, save it. */
ret = dwarf_die_CU_offset(link->die,
&pf->inl_offs,
&session.err);
ERR_IF(ret != DW_DLV_OK);
debug("inline definition offset %lld\n",
pf->inl_offs);
return 0;
}
dwarf_lowpc(link->die, &pf->addr, &session.err);
pf->addr += pf->offset;
/* TODO: Check the address in this function */
show_probepoint(link->die, pf->offset, pf);
/* Continue to search */
}
} else if (tag == DW_TAG_inlined_subroutine && pf->inl_offs) {
if (die_get_abstract_origin(link->die) == pf->inl_offs) {
pf->addr = die_get_entrypc(link->die);
pf->addr += pf->offset;
debug("found inline addr: 0x%llx\n", pf->addr);
/* Inlined function. Get a real subprogram */
for (lk = link->parent; lk != NULL; lk = lk->parent) {
tag = 0;
dwarf_tag(lk->die, &tag, &session.err);
ERR_IF(ret == DW_DLV_ERROR);
if (tag == DW_TAG_subprogram &&
!die_inlined_subprogram(lk->die))
goto found;
}
fprintf(stderr, "Failed to find real subprogram.\n");
exit(1);
found:
ret = die_within_subprogram(lk->die, pf->addr, &offs);
ERR_IF(!ret);
show_probepoint(lk->die, offs, pf);
/* Continue to search */
}
}
return 0;
}
static void find_by_func(Dwarf_Die cu_die, struct probe_finder *pf)
{
search_die_from_children(cu_die, probefunc_callback, pf);
}
static void find_probepoint(struct probe_finder *pf)
{
Dwarf_Unsigned cuh_len = 0;
Dwarf_Half vstamp = 0;
Dwarf_Unsigned abbrev = 0;
Dwarf_Half addr_size = 0;
Dwarf_Unsigned next_cuh = 0;
Dwarf_Die cu_die = 0;
int cu_number = 0, ret;
pf->found = 0;
while (++cu_number) {
/* Search CU (Compilation Unit) */
ret = dwarf_next_cu_header(session.dbg, &cuh_len, &vstamp,
&abbrev, &addr_size, &next_cuh, &session.err);
ERR_IF(ret == DW_DLV_ERROR);
if (ret == DW_DLV_NO_ENTRY)
break;
/* Get the DIE(Debugging Information Entry) of this CU */
ret = dwarf_siblingof(session.dbg, 0, &cu_die, &session.err);
ERR_IF(ret != DW_DLV_OK);
/* Check if target file is included. */
if (pf->file)
pf->fno = die_get_fileno(cu_die, pf->file);
if (!pf->file || pf->fno) {
/* Save CU base address (for frame_base) */
ret = dwarf_lowpc(cu_die, &pf->cu_base, &session.err);
ERR_IF(ret == DW_DLV_ERROR);
if (ret == DW_DLV_NO_ENTRY)
pf->cu_base = 0;
if (pf->line)
find_by_line(cu_die, pf);
if (pf->function)
find_by_func(cu_die, pf);
}
dwarf_dealloc(session.dbg, cu_die, DW_DLA_DIE);
}
if (pf->found == 0) {
fprintf(stderr, "Probe point is not found.\n");
exit(1);
}
}
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH -tip v14 03/12] kprobes: checks probe address is instruction boudary on x86
[not found] ` <20090818230340.GF5231@nowhere>
@ 2009-08-18 23:16 ` Masami Hiramatsu
[not found] ` <20090818234341.GG5231@nowhere>
0 siblings, 1 reply; 29+ messages in thread
From: Masami Hiramatsu @ 2009-08-18 23:16 UTC (permalink / raw)
To: Frederic Weisbecker
Cc: Ingo Molnar, Steven Rostedt, lkml, systemtap, kvm, DLE,
Ananth N Mavinakayanahalli, Avi Kivity, Andi Kleen,
Christoph Hellwig, Frank Ch. Eigler, H. Peter Anvin, Jason Baron,
Jim Keniston, K.Prasad, Lai Jiangshan, Li Zefan,
PrzemysławPawełczyk, Roland McGrath, Sam Ravnborg,
Srikar Dronamraju, Tom Zanussi, Vegard Nossum
Frederic Weisbecker wrote:
> On Thu, Aug 13, 2009 at 04:34:28PM -0400, Masami Hiramatsu wrote:
>> Ensure safeness of inserting kprobes by checking whether the specified
>> address is at the first byte of a instruction on x86.
>> This is done by decoding probed function from its head to the probe point.
>>
>> Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
>> Acked-by: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
>> Cc: Avi Kivity <avi@redhat.com>
>> Cc: Andi Kleen <ak@linux.intel.com>
>> Cc: Christoph Hellwig <hch@infradead.org>
>> Cc: Frank Ch. Eigler <fche@redhat.com>
>> Cc: Frederic Weisbecker <fweisbec@gmail.com>
>> Cc: H. Peter Anvin <hpa@zytor.com>
>> Cc: Ingo Molnar <mingo@elte.hu>
>> Cc: Jason Baron <jbaron@redhat.com>
>> Cc: Jim Keniston <jkenisto@us.ibm.com>
>> Cc: K.Prasad <prasad@linux.vnet.ibm.com>
>> Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
>> Cc: Li Zefan <lizf@cn.fujitsu.com>
>> Cc: Przemysław Pawełczyk <przemyslaw@pawelczyk.it>
>> Cc: Roland McGrath <roland@redhat.com>
>> Cc: Sam Ravnborg <sam@ravnborg.org>
>> Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
>> Cc: Steven Rostedt <rostedt@goodmis.org>
>> Cc: Tom Zanussi <tzanussi@gmail.com>
>> Cc: Vegard Nossum <vegard.nossum@gmail.com>
>> ---
>>
>> arch/x86/kernel/kprobes.c | 69 +++++++++++++++++++++++++++++++++++++++++++++
>> 1 files changed, 69 insertions(+), 0 deletions(-)
>>
>> diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c
>> index b5b1848..80d493f 100644
>> --- a/arch/x86/kernel/kprobes.c
>> +++ b/arch/x86/kernel/kprobes.c
>> @@ -48,6 +48,7 @@
>> #include <linux/preempt.h>
>> #include <linux/module.h>
>> #include <linux/kdebug.h>
>> +#include <linux/kallsyms.h>
>>
>> #include <asm/cacheflush.h>
>> #include <asm/desc.h>
>> @@ -55,6 +56,7 @@
>> #include <asm/uaccess.h>
>> #include <asm/alternative.h>
>> #include <asm/debugreg.h>
>> +#include <asm/insn.h>
>>
>> void jprobe_return_end(void);
>>
>> @@ -245,6 +247,71 @@ retry:
>> }
>> }
>>
>> +/* Recover the probed instruction at addr for further analysis. */
>> +static int recover_probed_instruction(kprobe_opcode_t *buf, unsigned long addr)
>> +{
>> + struct kprobe *kp;
>> + kp = get_kprobe((void *)addr);
>> + if (!kp)
>> + return -EINVAL;
>> +
>> + /*
>> + * Basically, kp->ainsn.insn has an original instruction.
>> + * However, RIP-relative instruction can not do single-stepping
>> + * at different place, fix_riprel() tweaks the displacement of
>> + * that instruction. In that case, we can't recover the instruction
>> + * from the kp->ainsn.insn.
>> + *
>> + * On the other hand, kp->opcode has a copy of the first byte of
>> + * the probed instruction, which is overwritten by int3. And
>> + * the instruction at kp->addr is not modified by kprobes except
>> + * for the first byte, we can recover the original instruction
>> + * from it and kp->opcode.
>> + */
>> + memcpy(buf, kp->addr, MAX_INSN_SIZE * sizeof(kprobe_opcode_t));
>> + buf[0] = kp->opcode;
>> + return 0;
>> +}
>> +
>> +/* Dummy buffers for kallsyms_lookup */
>> +static char __dummy_buf[KSYM_NAME_LEN];
>> +
>> +/* Check if paddr is at an instruction boundary */
>> +static int __kprobes can_probe(unsigned long paddr)
>> +{
>> + int ret;
>> + unsigned long addr, offset = 0;
>> + struct insn insn;
>> + kprobe_opcode_t buf[MAX_INSN_SIZE];
>> +
>> + if (!kallsyms_lookup(paddr, NULL, &offset, NULL, __dummy_buf))
>> + return 0;
>> +
>> + /* Decode instructions */
>> + addr = paddr - offset;
>> + while (addr < paddr) {
>> + kernel_insn_init(&insn, (void *)addr);
>> + insn_get_opcode(&insn);
>> +
>> + /* Check if the instruction has been modified. */
>> + if (insn.opcode.bytes[0] == BREAKPOINT_INSTRUCTION) {
>> + ret = recover_probed_instruction(buf, addr);
>
>
>
> I'm confused about the reason of this recovering. Is it to remove
> kprobes behind the current setting one in the current function?
No, it recovers just an instruction which is probed by a kprobe,
because we need to know the first byte of this instruction for
decoding it.
Perhaps we'd better to have more generic interface (text_peek?)
for it because another subsystem (e.g. kgdb) may want to insert int3...
Thank you,
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America), Inc.
Software Solutions Division
e-mail: mhiramat@redhat.com
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH -tip v14 03/12] kprobes: checks probe address is instruction boudary on x86
[not found] ` <20090818234341.GG5231@nowhere>
@ 2009-08-19 0:18 ` Masami Hiramatsu
0 siblings, 0 replies; 29+ messages in thread
From: Masami Hiramatsu @ 2009-08-19 0:18 UTC (permalink / raw)
To: Frederic Weisbecker
Cc: Ingo Molnar, Steven Rostedt, lkml, systemtap, kvm, DLE,
Ananth N Mavinakayanahalli, Avi Kivity, Andi Kleen,
Christoph Hellwig, Frank Ch. Eigler, H. Peter Anvin, Jason Baron,
Jim Keniston, K.Prasad, Lai Jiangshan, Li Zefan,
PrzemysławPawełczyk, Roland McGrath, Sam Ravnborg,
Srikar Dronamraju, Tom Zanussi, Vegard Nossum
Frederic Weisbecker wrote:
> On Tue, Aug 18, 2009 at 07:17:39PM -0400, Masami Hiramatsu wrote:
>> Frederic Weisbecker wrote:
>>>> + while (addr < paddr) {
>>>> + kernel_insn_init(&insn, (void *)addr);
>>>> + insn_get_opcode(&insn);
>>>> +
>>>> + /* Check if the instruction has been modified. */
>>>> + if (insn.opcode.bytes[0] == BREAKPOINT_INSTRUCTION) {
>>>> + ret = recover_probed_instruction(buf, addr);
>>>
>>>
>>>
>>> I'm confused about the reason of this recovering. Is it to remove
>>> kprobes behind the current setting one in the current function?
>>
>> No, it recovers just an instruction which is probed by a kprobe,
>> because we need to know the first byte of this instruction for
>> decoding it.
Ah, sorry, it was not accurate. the function recovers an instruction
on the buffer(buf), not on the real kernel text. :)
>>
>> Perhaps we'd better to have more generic interface (text_peek?)
>> for it because another subsystem (e.g. kgdb) may want to insert int3...
>>
>> Thank you,
>
>
> Aah, I see now, it's to keep a sane check of the instructions
> boundaries without int 3 artifacts in the middle.
>
> But in that case, you should re-arm the breakpoint after your
> check, right?
>
> Or may be you could do the check without repatching?
Yes, it doesn't modify kernel text, just recover an original
instruction from kernel text and backup byte on a buffer.
> May be by doing a copy of insn.opcode.bytes and replacing bytes[0]
> with what a random kprobe has stolen?
Hm, no, this function is protected from other kprobes by kprobe_mutex.
Thank you,
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America), Inc.
Software Solutions Division
e-mail: mhiramat@redhat.com
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH -tip v14 07/12] tracing: Introduce TRACE_FIELD_ZERO() macro
[not found] ` <20090819010903.GI5231@nowhere>
@ 2009-08-19 2:18 ` Masami Hiramatsu
0 siblings, 0 replies; 29+ messages in thread
From: Masami Hiramatsu @ 2009-08-19 2:18 UTC (permalink / raw)
To: Frederic Weisbecker
Cc: Ingo Molnar, Steven Rostedt, lkml, systemtap, kvm, DLE,
Ananth N Mavinakayanahalli, Avi Kivity, Andi Kleen,
Christoph Hellwig, Frank Ch. Eigler, H. Peter Anvin, Jason Baron,
Jim Keniston, K.Prasad, Lai Jiangshan, Li Zefan,
PrzemysławPawełczyk, Roland McGrath, Sam Ravnborg,
Srikar Dronamraju, Tom Zanussi, Vegard Nossum
Frederic Weisbecker wrote:
> On Thu, Aug 13, 2009 at 04:35:01PM -0400, Masami Hiramatsu wrote:
>> Use TRACE_FIELD_ZERO(type, item) instead of TRACE_FIELD_ZERO_CHAR(item).
>> This also includes a fix of TRACE_ZERO_CHAR() macro.
>
>
> I can't find what the fix is about (see below)
Ah, OK. This patch actually includes two parts.
One is introducing TRACE_FIELD_ZERO which is more generic than
TRACE_FIELD_ZERO_CHAR, I think.
Another is a typo fix of TRACE_ZERO_CHAR.
>>
>> Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
>> Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
>> Cc: Avi Kivity <avi@redhat.com>
>> Cc: Andi Kleen <ak@linux.intel.com>
>> Cc: Christoph Hellwig <hch@infradead.org>
>> Cc: Frank Ch. Eigler <fche@redhat.com>
>> Cc: Frederic Weisbecker <fweisbec@gmail.com>
>> Cc: H. Peter Anvin <hpa@zytor.com>
>> Cc: Ingo Molnar <mingo@elte.hu>
>> Cc: Jason Baron <jbaron@redhat.com>
>> Cc: Jim Keniston <jkenisto@us.ibm.com>
>> Cc: K.Prasad <prasad@linux.vnet.ibm.com>
>> Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
>> Cc: Li Zefan <lizf@cn.fujitsu.com>
>> Cc: Przemysław Pawełczyk <przemyslaw@pawelczyk.it>
>> Cc: Roland McGrath <roland@redhat.com>
>> Cc: Sam Ravnborg <sam@ravnborg.org>
>> Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
>> Cc: Steven Rostedt <rostedt@goodmis.org>
>> Cc: Tom Zanussi <tzanussi@gmail.com>
>> Cc: Vegard Nossum <vegard.nossum@gmail.com>
>> ---
>>
>> kernel/trace/trace_event_types.h | 4 ++--
>> kernel/trace/trace_export.c | 16 ++++++++--------
>> 2 files changed, 10 insertions(+), 10 deletions(-)
>>
>> diff --git a/kernel/trace/trace_event_types.h b/kernel/trace/trace_event_types.h
>> index 6db005e..e74f090 100644
>> --- a/kernel/trace/trace_event_types.h
>> +++ b/kernel/trace/trace_event_types.h
>> @@ -109,7 +109,7 @@ TRACE_EVENT_FORMAT(bprint, TRACE_BPRINT, bprint_entry, ignore,
>> TRACE_STRUCT(
>> TRACE_FIELD(unsigned long, ip, ip)
>> TRACE_FIELD(char *, fmt, fmt)
>> - TRACE_FIELD_ZERO_CHAR(buf)
>> + TRACE_FIELD_ZERO(char, buf)
>> ),
>> TP_RAW_FMT("%08lx (%d) fmt:%p %s")
>> );
>> @@ -117,7 +117,7 @@ TRACE_EVENT_FORMAT(bprint, TRACE_BPRINT, bprint_entry, ignore,
>> TRACE_EVENT_FORMAT(print, TRACE_PRINT, print_entry, ignore,
>> TRACE_STRUCT(
>> TRACE_FIELD(unsigned long, ip, ip)
>> - TRACE_FIELD_ZERO_CHAR(buf)
>> + TRACE_FIELD_ZERO(char, buf)
>> ),
>> TP_RAW_FMT("%08lx (%d) fmt:%p %s")
>> );
>> diff --git a/kernel/trace/trace_export.c b/kernel/trace/trace_export.c
>> index 71c8d7f..b0ac92c 100644
>> --- a/kernel/trace/trace_export.c
>> +++ b/kernel/trace/trace_export.c
>> @@ -42,9 +42,9 @@ extern void __bad_type_size(void);
>> if (!ret) \
>> return 0;
>>
>> -#undef TRACE_FIELD_ZERO_CHAR
>> -#define TRACE_FIELD_ZERO_CHAR(item) \
>> - ret = trace_seq_printf(s, "\tfield:char " #item ";\t" \
>> +#undef TRACE_FIELD_ZERO
>> +#define TRACE_FIELD_ZERO(type, item) \
>> + ret = trace_seq_printf(s, "\tfield:" #type " " #item ";\t" \
>> "offset:%u;\tsize:0;\n", \
>> (unsigned int)offsetof(typeof(field), item)); \
>> if (!ret) \
>> @@ -92,9 +92,6 @@ ftrace_format_##call(struct ftrace_event_call *unused, \
>>
>> #include "trace_event_types.h"
>>
>> -#undef TRACE_ZERO_CHAR
>> -#define TRACE_ZERO_CHAR(arg)
>> -
>> #undef TRACE_FIELD
>> #define TRACE_FIELD(type, item, assign)\
>> entry->item = assign;
>> @@ -107,6 +104,9 @@ ftrace_format_##call(struct ftrace_event_call *unused, \
>> #define TRACE_FIELD_SIGN(type, item, assign, is_signed) \
>> TRACE_FIELD(type, item, assign)
>>
>> +#undef TRACE_FIELD_ZERO
>> +#define TRACE_FIELD_ZERO(type, item)
>> +
>
>
>
> Is it about the above moving?
> If so, could you just tell so that I can add something about
> it in the changelog.
No, I assume that TRACE_ZERO_CHAR is just a typo of TRACE_FIELD_ZERO_CHAR.
(because I couldn't find any other TRACE_ZERO_CHAR)
BTW, this patch may not be needed after applying patch 10/12, since
it removes ftrace event definitions of TRACE_KPROBE/KRETPROBE.
Perhaps, would I better merge and split those additional patches(and
remove this change)?
(It also could make the incremental review hard...)
Thank you,
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America), Inc.
Software Solutions Division
e-mail: mhiramat@redhat.com
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH -tip v14 01/12] x86: instruction decoder API
[not found] ` <20090819234227.GJ4972@nowhere>
@ 2009-08-20 14:38 ` Masami Hiramatsu
[not found] ` <20090820002113.GK4972@nowhere>
1 sibling, 0 replies; 29+ messages in thread
From: Masami Hiramatsu @ 2009-08-20 14:38 UTC (permalink / raw)
To: Frederic Weisbecker
Cc: Ingo Molnar, Steven Rostedt, lkml, systemtap, kvm, DLE,
Jim Keniston, H. Peter Anvin, Ananth N Mavinakayanahalli,
Avi Kivity, Andi Kleen, Christoph Hellwig, Frank Ch. Eigler,
Jason Baron, K.Prasad, Lai Jiangshan, Li Zefan,
PrzemysławPawełczyk, Roland McGrath, Sam Ravnborg,
Srikar Dronamraju, Tom Zanussi, Vegard Nossum
Frederic Weisbecker wrote:
> On Thu, Aug 13, 2009 at 04:34:13PM -0400, Masami Hiramatsu wrote:
>> Add x86 instruction decoder to arch-specific libraries. This decoder
>> can decode x86 instructions used in kernel into prefix, opcode, modrm,
>> sib, displacement and immediates. This can also show the length of
>> instructions.
>>
>> This version introduces instruction attributes for decoding instructions.
>> The instruction attribute tables are generated from the opcode map file
>> (x86-opcode-map.txt) by the generator script(gen-insn-attr-x86.awk).
>>
>> Currently, the opcode maps are based on opcode maps in Intel(R) 64 and
>> IA-32 Architectures Software Developers Manual Vol.2: Appendix.A,
>> and consist of below two types of opcode tables.
>>
>> 1-byte/2-bytes/3-bytes opcodes, which has 256 elements, are
>> written as below;
>>
>> Table: table-name
>> Referrer: escaped-name
>> opcode: mnemonic|GrpXXX [operand1[,operand2...]] [(extra1)[,(extra2)...] [| 2nd-mnemonic ...]
>> (or)
>> opcode: escape # escaped-name
>> EndTable
>>
>> Group opcodes, which has 8 elements, are written as below;
>>
>> GrpTable: GrpXXX
>> reg: mnemonic [operand1[,operand2...]] [(extra1)[,(extra2)...] [| 2nd-mnemonic ...]
>> EndTable
>>
>> These opcode maps include a few SSE and FP opcodes (for setup), because
>> those opcodes are used in the kernel.
>>
>
>
> I'm getting the following build error on an old K7 box:
>
> arch/x86/lib/inat.c: In function ‘inat_get_opcode_attribute’:
> arch/x86/lib/inat.c:29: erreur: ‘inat_primary_table’ undeclared (first use in this function)
> arch/x86/lib/inat.c:29: erreur: (Each undeclared identifier is reported only once
> arch/x86/lib/inat.c:29: erreur: for each function it appears in.)
Thanks for reporting!
Hmm, it seems that inat-tables.c is not correctly generated.
Could you tell me which awk you used and send the inat-tables.c?
Thank you,
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America), Inc.
Software Solutions Division
e-mail: mhiramat@redhat.com
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH -tip v14 01/12] x86: instruction decoder API
[not found] ` <20090820002113.GK4972@nowhere>
@ 2009-08-20 15:00 ` Masami Hiramatsu
[not found] ` <20090820152549.GD6389@nowhere>
0 siblings, 1 reply; 29+ messages in thread
From: Masami Hiramatsu @ 2009-08-20 15:00 UTC (permalink / raw)
To: Frederic Weisbecker
Cc: Ingo Molnar, Steven Rostedt, lkml, systemtap, kvm, DLE,
Jim Keniston, H. Peter Anvin, Ananth N Mavinakayanahalli,
Avi Kivity, Andi Kleen, Christoph Hellwig, Frank Ch. Eigler,
Jason Baron, K.Prasad, Lai Jiangshan, Li Zefan,
PrzemysławPawełczyk, Roland McGrath, Sam Ravnborg,
Srikar Dronamraju, Tom Zanussi, Vegard Nossum
Frederic Weisbecker wrote:
> On Thu, Aug 20, 2009 at 01:42:31AM +0200, Frederic Weisbecker wrote:
>> On Thu, Aug 13, 2009 at 04:34:13PM -0400, Masami Hiramatsu wrote:
>>> Add x86 instruction decoder to arch-specific libraries. This decoder
>>> can decode x86 instructions used in kernel into prefix, opcode, modrm,
>>> sib, displacement and immediates. This can also show the length of
>>> instructions.
>>>
>>> This version introduces instruction attributes for decoding instructions.
>>> The instruction attribute tables are generated from the opcode map file
>>> (x86-opcode-map.txt) by the generator script(gen-insn-attr-x86.awk).
>>>
>>> Currently, the opcode maps are based on opcode maps in Intel(R) 64 and
>>> IA-32 Architectures Software Developers Manual Vol.2: Appendix.A,
>>> and consist of below two types of opcode tables.
>>>
>>> 1-byte/2-bytes/3-bytes opcodes, which has 256 elements, are
>>> written as below;
>>>
>>> Table: table-name
>>> Referrer: escaped-name
>>> opcode: mnemonic|GrpXXX [operand1[,operand2...]] [(extra1)[,(extra2)...] [| 2nd-mnemonic ...]
>>> (or)
>>> opcode: escape # escaped-name
>>> EndTable
>>>
>>> Group opcodes, which has 8 elements, are written as below;
>>>
>>> GrpTable: GrpXXX
>>> reg: mnemonic [operand1[,operand2...]] [(extra1)[,(extra2)...] [| 2nd-mnemonic ...]
>>> EndTable
>>>
>>> These opcode maps include a few SSE and FP opcodes (for setup), because
>>> those opcodes are used in the kernel.
>>>
>>
>>
>> I'm getting the following build error on an old K7 box:
>>
>> arch/x86/lib/inat.c: In function ‘inat_get_opcode_attribute’:
>> arch/x86/lib/inat.c:29: erreur: ‘inat_primary_table’ undeclared (first use in this function)
>> arch/x86/lib/inat.c:29: erreur: (Each undeclared identifier is reported only once
>> arch/x86/lib/inat.c:29: erreur: for each function it appears in.)
>>
>>
>> I've attached my config. I haven't such problem on a dual x86-64 box.
>
>
> Actually I have the same problem in x86-64
> The content of my arch/x86/lib/inat-tables.c:
>
> /* x86 opcode map generated from x86-opcode-map.txt */
> /* Do not change this code. */
> /* Table: one byte opcode */
> /* Escape opcode map array */
> const insn_attr_t const *inat_escape_tables[INAT_ESC_MAX + 1][INAT_LPREFIX_MAX + 1] = {
> };
>
> /* Group opcode map array */
> const insn_attr_t const *inat_group_tables[INAT_GRP_MAX + 1][INAT_LPREFIX_MAX + 1] = {
> };
>
>
> I guess there is a problem with the generation of this file.
Aah, you may use mawk on Ubuntu 9.04, right?
If so, unfortunately, mawk is still under development.
http://invisible-island.net/mawk/CHANGES
> 20090727
> add check/fix to prevent gsub from recurring to modify on a substring
> of the current line when the regular expression is anchored to the
> beginning of the line; fixes gawk's anchgsub testcase.
>
> add check for implicit concatenation mistaken for exponent; fixes
> gawk's hex testcase.
>
> add character-classes to built-in regular expressions.
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Look, this means we can't use char-class expressions like
[:lower:] until this version...
And I've found another bug in mawk-1.3.3-20090728(the latest one).
it almost works, but;
$ mawk 'BEGIN {printf("0x%x\n", 0)}'
0x1
$ gawk 'BEGIN {printf("0x%x\n", 0)}'
0x0
This bug skips an array element index 0x0 in inat-tables.c :(
So, I recommend you to install gawk instead mawk until that
supports all posix-awk features, since I don't think it is
good idea to avoid all those bugs which depends on
implementation (not specification).
Thank you,
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America), Inc.
Software Solutions Division
e-mail: mhiramat@redhat.com
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH -tip v14 01/12] x86: instruction decoder API
[not found] ` <20090820152549.GD6389@nowhere>
@ 2009-08-20 16:12 ` Masami Hiramatsu
[not found] ` <20090820180737.GA6229@nowhere>
0 siblings, 1 reply; 29+ messages in thread
From: Masami Hiramatsu @ 2009-08-20 16:12 UTC (permalink / raw)
To: Frederic Weisbecker
Cc: Ingo Molnar, Steven Rostedt, lkml, systemtap, kvm, DLE,
Jim Keniston, H. Peter Anvin, Ananth N Mavinakayanahalli,
Avi Kivity, Andi Kleen, Christoph Hellwig, Frank Ch. Eigler,
Jason Baron, K.Prasad, Lai Jiangshan, Li Zefan,
PrzemysławPawełczyk, Roland McGrath, Sam Ravnborg,
Srikar Dronamraju, Tom Zanussi, Vegard Nossum
Frederic Weisbecker wrote:
> On Thu, Aug 20, 2009 at 11:03:40AM -0400, Masami Hiramatsu wrote:
>> Frederic Weisbecker wrote:
>>> On Thu, Aug 20, 2009 at 01:42:31AM +0200, Frederic Weisbecker wrote:
>>>> On Thu, Aug 13, 2009 at 04:34:13PM -0400, Masami Hiramatsu wrote:
>>>>> Add x86 instruction decoder to arch-specific libraries. This decoder
>>>>> can decode x86 instructions used in kernel into prefix, opcode, modrm,
>>>>> sib, displacement and immediates. This can also show the length of
>>>>> instructions.
>>>>>
>>>>> This version introduces instruction attributes for decoding instructions.
>>>>> The instruction attribute tables are generated from the opcode map file
>>>>> (x86-opcode-map.txt) by the generator script(gen-insn-attr-x86.awk).
>>>>>
>>>>> Currently, the opcode maps are based on opcode maps in Intel(R) 64 and
>>>>> IA-32 Architectures Software Developers Manual Vol.2: Appendix.A,
>>>>> and consist of below two types of opcode tables.
>>>>>
>>>>> 1-byte/2-bytes/3-bytes opcodes, which has 256 elements, are
>>>>> written as below;
>>>>>
>>>>> Table: table-name
>>>>> Referrer: escaped-name
>>>>> opcode: mnemonic|GrpXXX [operand1[,operand2...]] [(extra1)[,(extra2)...] [| 2nd-mnemonic ...]
>>>>> (or)
>>>>> opcode: escape # escaped-name
>>>>> EndTable
>>>>>
>>>>> Group opcodes, which has 8 elements, are written as below;
>>>>>
>>>>> GrpTable: GrpXXX
>>>>> reg: mnemonic [operand1[,operand2...]] [(extra1)[,(extra2)...] [| 2nd-mnemonic ...]
>>>>> EndTable
>>>>>
>>>>> These opcode maps include a few SSE and FP opcodes (for setup), because
>>>>> those opcodes are used in the kernel.
>>>>>
>>>>
>>>>
>>>> I'm getting the following build error on an old K7 box:
>>>>
>>>> arch/x86/lib/inat.c: In function ‘inat_get_opcode_attribute’:
>>>> arch/x86/lib/inat.c:29: erreur: ‘inat_primary_table’ undeclared (first use in this function)
>>>> arch/x86/lib/inat.c:29: erreur: (Each undeclared identifier is reported only once
>>>> arch/x86/lib/inat.c:29: erreur: for each function it appears in.)
>>>>
>>>>
>>>> I've attached my config. I haven't such problem on a dual x86-64 box.
>>>
>>>
>>> Actually I have the same problem in x86-64
>>> The content of my arch/x86/lib/inat-tables.c:
>>>
>>> /* x86 opcode map generated from x86-opcode-map.txt */
>>> /* Do not change this code. */
>>> /* Table: one byte opcode */
>>> /* Escape opcode map array */
>>> const insn_attr_t const *inat_escape_tables[INAT_ESC_MAX + 1][INAT_LPREFIX_MAX + 1] = {
>>> };
>>>
>>> /* Group opcode map array */
>>> const insn_attr_t const *inat_group_tables[INAT_GRP_MAX + 1][INAT_LPREFIX_MAX + 1] = {
>>> };
>>>
>>>
>>> I guess there is a problem with the generation of this file.
>>
>> Aah, you may use mawk on Ubuntu 9.04, right?
>> If so, unfortunately, mawk is still under development.
>>
>> http://invisible-island.net/mawk/CHANGES
>
>
>
> Aargh...
>
>
>>> 20090727
>>> add check/fix to prevent gsub from recurring to modify on a substring
>>> of the current line when the regular expression is anchored to the
>>> beginning of the line; fixes gawk's anchgsub testcase.
>>>
>>> add check for implicit concatenation mistaken for exponent; fixes
>>> gawk's hex testcase.
>>>
>>> add character-classes to built-in regular expressions.
>> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
>> Look, this means we can't use char-class expressions like
>> [:lower:] until this version...
>>
>> And I've found another bug in mawk-1.3.3-20090728(the latest one).
>> it almost works, but;
>>
>> $ mawk 'BEGIN {printf("0x%x\n", 0)}'
>> 0x1
>
>
> Ouch, indeed.
>
>
>
>> $ gawk 'BEGIN {printf("0x%x\n", 0)}'
>> 0x0
>>
>> This bug skips an array element index 0x0 in inat-tables.c :(
>>
>> So, I recommend you to install gawk instead mawk until that
>> supports all posix-awk features, since I don't think it is
>> good idea to avoid all those bugs which depends on
>> implementation (not specification).
>>
>>
>> Thank you,
>
>
>
> Yeah, indeed. May be add a warning (or build error) in case the user uses
> mawk?
Hmm, it is possible that mawk will fix those bugs and catch up soon,
so, I think checking mawk is not a good idea.
(and since there will be other awk implementations, it's not fair.)
I think what all I can do now is reporting bugs to
mawk and ubuntu people.:-)
>
> Anyway that works fine now with gawk, thanks!
> All your patches build well :-)
Thank you for testing!
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America), Inc.
Software Solutions Division
e-mail: mhiramat@redhat.com
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH -tip v14 01/12] x86: instruction decoder API
[not found] ` <20090820180737.GA6229@nowhere>
@ 2009-08-20 18:58 ` Masami Hiramatsu
0 siblings, 0 replies; 29+ messages in thread
From: Masami Hiramatsu @ 2009-08-20 18:58 UTC (permalink / raw)
To: Frederic Weisbecker
Cc: Ingo Molnar, Steven Rostedt, lkml, systemtap, kvm, DLE,
Jim Keniston, H. Peter Anvin, Ananth N Mavinakayanahalli,
Avi Kivity, Andi Kleen, Christoph Hellwig, Frank Ch. Eigler,
Jason Baron, K.Prasad, Lai Jiangshan, Li Zefan,
PrzemysławPawełczyk, Roland McGrath, Sam Ravnborg,
Srikar Dronamraju, Tom Zanussi, Vegard Nossum
Frederic Weisbecker wrote:
> On Thu, Aug 20, 2009 at 12:16:05PM -0400, Masami Hiramatsu wrote:
>> Frederic Weisbecker wrote:
>>> On Thu, Aug 20, 2009 at 11:03:40AM -0400, Masami Hiramatsu wrote:
>>>> Frederic Weisbecker wrote:
>>>>> On Thu, Aug 20, 2009 at 01:42:31AM +0200, Frederic Weisbecker wrote:
>>>>>> On Thu, Aug 13, 2009 at 04:34:13PM -0400, Masami Hiramatsu wrote:
>>>>>>> Add x86 instruction decoder to arch-specific libraries. This decoder
>>>>>>> can decode x86 instructions used in kernel into prefix, opcode, modrm,
>>>>>>> sib, displacement and immediates. This can also show the length of
>>>>>>> instructions.
>>>>>>>
>>>>>>> This version introduces instruction attributes for decoding instructions.
>>>>>>> The instruction attribute tables are generated from the opcode map file
>>>>>>> (x86-opcode-map.txt) by the generator script(gen-insn-attr-x86.awk).
>>>>>>>
>>>>>>> Currently, the opcode maps are based on opcode maps in Intel(R) 64 and
>>>>>>> IA-32 Architectures Software Developers Manual Vol.2: Appendix.A,
>>>>>>> and consist of below two types of opcode tables.
>>>>>>>
>>>>>>> 1-byte/2-bytes/3-bytes opcodes, which has 256 elements, are
>>>>>>> written as below;
>>>>>>>
>>>>>>> Table: table-name
>>>>>>> Referrer: escaped-name
>>>>>>> opcode: mnemonic|GrpXXX [operand1[,operand2...]] [(extra1)[,(extra2)...] [| 2nd-mnemonic ...]
>>>>>>> (or)
>>>>>>> opcode: escape # escaped-name
>>>>>>> EndTable
>>>>>>>
>>>>>>> Group opcodes, which has 8 elements, are written as below;
>>>>>>>
>>>>>>> GrpTable: GrpXXX
>>>>>>> reg: mnemonic [operand1[,operand2...]] [(extra1)[,(extra2)...] [| 2nd-mnemonic ...]
>>>>>>> EndTable
>>>>>>>
>>>>>>> These opcode maps include a few SSE and FP opcodes (for setup), because
>>>>>>> those opcodes are used in the kernel.
>>>>>>>
>>>>>>
>>>>>>
>>>>>> I'm getting the following build error on an old K7 box:
>>>>>>
>>>>>> arch/x86/lib/inat.c: In function ‘inat_get_opcode_attribute’:
>>>>>> arch/x86/lib/inat.c:29: erreur: ‘inat_primary_table’ undeclared (first use in this function)
>>>>>> arch/x86/lib/inat.c:29: erreur: (Each undeclared identifier is reported only once
>>>>>> arch/x86/lib/inat.c:29: erreur: for each function it appears in.)
>>>>>>
>>>>>>
>>>>>> I've attached my config. I haven't such problem on a dual x86-64 box.
>>>>>
>>>>>
>>>>> Actually I have the same problem in x86-64
>>>>> The content of my arch/x86/lib/inat-tables.c:
>>>>>
>>>>> /* x86 opcode map generated from x86-opcode-map.txt */
>>>>> /* Do not change this code. */
>>>>> /* Table: one byte opcode */
>>>>> /* Escape opcode map array */
>>>>> const insn_attr_t const *inat_escape_tables[INAT_ESC_MAX + 1][INAT_LPREFIX_MAX + 1] = {
>>>>> };
>>>>>
>>>>> /* Group opcode map array */
>>>>> const insn_attr_t const *inat_group_tables[INAT_GRP_MAX + 1][INAT_LPREFIX_MAX + 1] = {
>>>>> };
>>>>>
>>>>>
>>>>> I guess there is a problem with the generation of this file.
>>>>
>>>> Aah, you may use mawk on Ubuntu 9.04, right?
>>>> If so, unfortunately, mawk is still under development.
>>>>
>>>> http://invisible-island.net/mawk/CHANGES
>>>
>>>
>>>
>>> Aargh...
>>>
>>>
>>>>> 20090727
>>>>> add check/fix to prevent gsub from recurring to modify on a substring
>>>>> of the current line when the regular expression is anchored to the
>>>>> beginning of the line; fixes gawk's anchgsub testcase.
>>>>>
>>>>> add check for implicit concatenation mistaken for exponent; fixes
>>>>> gawk's hex testcase.
>>>>>
>>>>> add character-classes to built-in regular expressions.
>>>> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
>>>> Look, this means we can't use char-class expressions like
>>>> [:lower:] until this version...
>>>>
>>>> And I've found another bug in mawk-1.3.3-20090728(the latest one).
>>>> it almost works, but;
>>>>
>>>> $ mawk 'BEGIN {printf("0x%x\n", 0)}'
>>>> 0x1
>>>
>>>
>>> Ouch, indeed.
>>>
>>>
>>>
>>>> $ gawk 'BEGIN {printf("0x%x\n", 0)}'
>>>> 0x0
>>>>
>>>> This bug skips an array element index 0x0 in inat-tables.c :(
>>>>
>>>> So, I recommend you to install gawk instead mawk until that
>>>> supports all posix-awk features, since I don't think it is
>>>> good idea to avoid all those bugs which depends on
>>>> implementation (not specification).
>>>>
>>>>
>>>> Thank you,
>>>
>>>
>>>
>>> Yeah, indeed. May be add a warning (or build error) in case the user uses
>>> mawk?
>>
>> Hmm, it is possible that mawk will fix those bugs and catch up soon,
>> so, I think checking mawk is not a good idea.
>> (and since there will be other awk implementations, it's not fair.)
>>
>> I think what all I can do now is reporting bugs to
>> mawk and ubuntu people.:-)
>
>
>
> Yeah, but without your tip I couldn't be able to find the origin
> before some time.
> And the kernel couldn't build anyway.
>
> At least we should do something with this version of mawk.
Hm, indeed.
Maybe, we can run additional sanity check script before using
awk, like this;
---
res=`echo a | $AWK '/[[:lower:]]+/{print "OK"}'`
[ "$res" != "OK" ] && exit 1
res=`$AWK 'BEGIN {printf("%x", 0)}'`
[ "$res" != "0" ] && exit 1
exit 0
---
Thanks,
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America), Inc.
Software Solutions Division
e-mail: mhiramat@redhat.com
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [TOOL] kprobestest : Kprobe stress test tool
[not found] ` <20090820184331.GA6078@nowhere>
@ 2009-08-20 19:42 ` Masami Hiramatsu
[not found] ` <20090821000108.GC6078@nowhere>
0 siblings, 1 reply; 29+ messages in thread
From: Masami Hiramatsu @ 2009-08-20 19:42 UTC (permalink / raw)
To: Frederic Weisbecker
Cc: Ingo Molnar, Steven Rostedt, lkml, Ananth N Mavinakayanahalli,
Avi Kivity, Andi Kleen, Christoph Hellwig, Frank Ch. Eigler,
H. Peter Anvin, Jason Baron, Jim Keniston, K.Prasad,
Lai Jiangshan, Li Zefan, PrzemysławPawełczyk,
Roland McGrath, Sam Ravnborg, Srikar Dronamraju, Tom Zanussi,
Vegard Nossum, systemtap, kvm, DLE
Frederic Weisbecker wrote:
> On Thu, Aug 13, 2009 at 04:57:20PM -0400, Masami Hiramatsu wrote:
>> This script tests kprobes to probe on all symbols in the kernel and finds
>> symbols which must be blacklisted.
>>
>>
>> Usage
>> -----
>> kprobestest [-s SYMLIST] [-b BLACKLIST] [-w WHITELIST]
>> Run stress test. If SYMLIST file is specified, use it as
>> an initial symbol list (This is useful for verifying white list
>> after diagnosing all symbols).
>>
>> kprobestest cleanup
>> Cleanup all lists
>>
>>
>> How to Work
>> -----------
>> This tool list up all symbols in the kernel via /proc/kallsyms, and sorts
>> it into groups (each of them including 64 symbols in default). And then,
>> it tests each group by using kprobe-tracer. If a kernel crash occurred,
>> that group is moved into 'failed' dir. If the group passed the test, this
>> script moves it into 'passed' dir and saves kprobe_profile into
>> 'passed/profiles/'.
>> After testing all groups, all 'failed' groups are merged and sorted into
>> smaller groups (divided by 4, in default). And those are tested again.
>> This loop will be repeated until all group has just 1 symbol.
>>
>> Finally, the script sorts all 'passed' symbols into 'tested', 'untested',
>> and 'missed' based on profiles.
>>
>>
>> Note
>> ----
>> - This script just gives us some clues to the blacklisted functions.
>> In some cases, a combination of probe points will cause a problem, but
>> each of them doesn't cause the problem alone.
>>
>> Thank you,
>>
>
>
> This script makes my x86-64 dual core easily and hardly locking-up
> on the 1st batch of symbols to test.
> I have one sym list in the failed and unset directories:
>
> int_very_careful
> int_signal
> int_restore_rest
> stub_clone
> stub_fork
> stub_vfork
> stub_sigaltstack
> stub_iopl
> ptregscall_common
> stub_execve
> stub_rt_sigreturn
> irq_entries_start
> common_interrupt
> ret_from_intr
> exit_intr
> retint_with_reschedule
> retint_check
> retint_swapgs
> retint_restore_args
> restore_args
> irq_return
> retint_careful
> retint_signal
> retint_kernel
> irq_move_cleanup_interrupt
> reboot_interrupt
> apic_timer_interrupt
> generic_interrupt
> invalidate_interrupt0
> invalidate_interrupt1
> invalidate_interrupt2
> invalidate_interrupt3
> invalidate_interrupt4
> invalidate_interrupt5
> invalidate_interrupt6
> invalidate_interrupt7
> threshold_interrupt
> thermal_interrupt
> mce_self_interrupt
> call_function_single_interrupt
> call_function_interrupt
> reschedule_interrupt
> error_interrupt
> spurious_interrupt
> perf_pending_interrupt
> divide_error
> overflow
> bounds
> invalid_op
> device_not_available
> double_fault
> coprocessor_segment_overrun
> invalid_TSS
> segment_not_present
> spurious_interrupt_bug
> coprocessor_error
> alignment_check
> simd_coprocessor_error
> native_load_gs_index
> gs_change
> kernel_thread
> child_rip
> kernel_execve
> call_softirq
>
>
> I don't have a crash log because I was running with X.
> But it also happened with other batch of symbols.
Thank you for reporting, here, I also have a result
tested on KVM@x86-64.
native_read_tscp
native_read_msr_safe
native_read_msr_amd_safe
native_write_msr_safe
vmalloc_fault
spurious_fault
search_exception_tables
notify_die
trace_hardirqs_off_caller
ident_complete
lock_acquire
lock_release
bad_address
secondary_startup_64
stack_start
bad_address
restore_args
irq_return
restore
trace_hardirqs_off_thunk
init_level4_pgt
level3_ident_pgt
level3_kernel_pgt
level2_fixmap_pgt
_text
startup_64
level1_fixmap_pgt
level2_ident_pgt
level2_kernel_pgt
level2_spare_pgt
native_get_debugreg
native_set_debugreg
native_set_iopl_mask
native_load_sp0
debug_show_all_locks
debug_check_no_locks_held
valid_state
mark_lock
mark_held_locks
lockdep_trace_alloc
trace_softirqs_on
trace_hardirqs_on_caller
__down_write
__down_read
trace_hardirqs_on_thunk
lockdep_sys_exit_thunk
Most of them can be fixed just by adding __kprobes.
Some of them which are already in the another section, kprobes
should check the symbols are in the section.
> The problem is that I don't have any serial line in this
> box then I can't catch any crash log.
> My K7 testbox also died in my arms this afternoon.
>
> But I still have two other testboxes (one P2 and one P3),
> hopefully I could reproduce the problem in these boxes
> in which I can connect a serial line.
Thank you for helping me to find it!
> I've pushed your patches in the following git tree:
>
> git://git.kernel.org/pub/scm/linux/kernel/git/fgrederic/random-tracing.git \
> tracing/kprobes
>
> So you can send patches on top of this one.
Great! I've found another trivial bugs, so I'll fix those on it.
Thank you,
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America), Inc.
Software Solutions Division
e-mail: mhiramat@redhat.com
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [TOOL] kprobestest : Kprobe stress test tool
[not found] ` <20090821000108.GC6078@nowhere>
@ 2009-08-21 0:56 ` Masami Hiramatsu
2009-08-21 19:40 ` [PATCH tracing/kprobes 2/4] x86: Check awk features before generating inat-tables.c Masami Hiramatsu
` (3 subsequent siblings)
4 siblings, 0 replies; 29+ messages in thread
From: Masami Hiramatsu @ 2009-08-21 0:56 UTC (permalink / raw)
To: Frederic Weisbecker
Cc: Ingo Molnar, Steven Rostedt, lkml, Ananth N Mavinakayanahalli,
Avi Kivity, Andi Kleen, Christoph Hellwig, Frank Ch. Eigler,
H. Peter Anvin, Jason Baron, Jim Keniston, K.Prasad,
Lai Jiangshan, Li Zefan, PrzemysławPawełczyk,
Roland McGrath, Sam Ravnborg, Srikar Dronamraju, Tom Zanussi,
Vegard Nossum, systemtap, kvm, DLE
Frederic Weisbecker wrote:
>> Most of them can be fixed just by adding __kprobes.
>> Some of them which are already in the another section, kprobes
>> should check the symbols are in the section.
>
>
> You mean the blacklist?
>
> I also fear that putting bad kprobed functions into the kprobe
> section or into the blacklist may hide some kprobe internal bugs.
>
> Doing so is indeed mandatory for functions that trigger tracing
> recursion of things like that, but what if kprobe has an internal
> bug that only triggers while probe a certain class of function.
>
> Ie: it would be nice to identify the reason of the crash for
> each culprit in these lists.
>
> That may even help to find the others in advance.
Indeed, actually I've found some bugs while making jump-optimization
patches by using this stress test.
But some of them are obviously what we just forget to add __kprobes,
since those will be called from kprobes int3 handling functions.
And also, many lock-related code has been changed. I think
kprobes should use raw_*_lock, or prohibit to probe lock monitoring
functions like lockdep, because it will cause recursive call.
>
> Also kprobes seems to be a very fragile feature (that's what
> this selftest unearthes at least for me).
> And it really needs a recursion detection that stops every kprobing
> while reaching a given threshold of recursion. Something
> that would dump the stack and the falling kprobe structure.
Hmm, kprobes already has recursion detection(kp->nmiss), so
maybe, we can check it.
>
> That would avoid such hard lockups and also help to identify
> the dangerous symbols to probe.
>
>
>
>>> The problem is that I don't have any serial line in this
>>> box then I can't catch any crash log.
>>> My K7 testbox also died in my arms this afternoon.
>>>
>>> But I still have two other testboxes (one P2 and one P3),
>>> hopefully I could reproduce the problem in these boxes
>>> in which I can connect a serial line.
>>
>> Thank you for helping me to find it!
>>
>>> I've pushed your patches in the following git tree:
>>>
>>> git://git.kernel.org/pub/scm/linux/kernel/git/fgrederic/random-tracing.git \
>>> tracing/kprobes
>>>
>>> So you can send patches on top of this one.
>>
>> Great! I've found another trivial bugs, so I'll fix those on it.
>
> Cool :)
>
> Btw, here is the result of your stress test in a PIII (attaching the log
> and the config).
Thanks, I'll check that.
Thank you,
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America), Inc.
Software Solutions Division
e-mail: mhiramat@redhat.com
^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH tracing/kprobes 1/4] x86: Fix x86 instruction decoder selftest to check only .text
[not found] ` <20090821000108.GC6078@nowhere>
2009-08-21 0:56 ` Masami Hiramatsu
2009-08-21 19:40 ` [PATCH tracing/kprobes 2/4] x86: Check awk features before generating inat-tables.c Masami Hiramatsu
@ 2009-08-21 19:40 ` Masami Hiramatsu
2009-08-21 19:40 ` [PATCH tracing/kprobes 3/4] tracing/kprobes: Fix format typo in trace_kprobes Masami Hiramatsu
2009-08-21 19:41 ` [PATCH tracing/kprobes 4/4] tracing/kprobes: Change trace_arg to probe_arg Masami Hiramatsu
4 siblings, 0 replies; 29+ messages in thread
From: Masami Hiramatsu @ 2009-08-21 19:40 UTC (permalink / raw)
To: Frederic Weisbecker, Steven Rostedt, lkml
Cc: systemtap, kvm, DLE, Masami Hiramatsu, Jim Keniston,
H. Peter Anvin, Ananth N Mavinakayanahalli, Avi Kivity,
Andi Kleen, Christoph Hellwig, Frank Ch. Eigler,
Frederic Weisbecker, Ingo Molnar, Jason Baron, K.Prasad,
Lai Jiangshan, Li Zefan, PrzemysławPawełczyk,
Roland McGrath, Sam Ravnborg, Srikar Dronamraju, Steven Rostedt,
Tom Zanussi, Vegard Nossum
Fix x86 instruction decoder selftest to check only .text because other
sections (e.g. .notes) will have random bytes which don't need to be checked.
Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Jim Keniston <jkenisto@us.ibm.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Avi Kivity <avi@redhat.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Frank Ch. Eigler <fche@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Jason Baron <jbaron@redhat.com>
Cc: K.Prasad <prasad@linux.vnet.ibm.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Przemysław Pawełczyk <przemyslaw@pawelczyk.it>
Cc: Roland McGrath <roland@redhat.com>
Cc: Sam Ravnborg <sam@ravnborg.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Tom Zanussi <tzanussi@gmail.com>
Cc: Vegard Nossum <vegard.nossum@gmail.com>
---
arch/x86/tools/Makefile | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/arch/x86/tools/Makefile b/arch/x86/tools/Makefile
index 3dd626b..95e9cc4 100644
--- a/arch/x86/tools/Makefile
+++ b/arch/x86/tools/Makefile
@@ -1,6 +1,6 @@
PHONY += posttest
quiet_cmd_posttest = TEST $@
- cmd_posttest = $(OBJDUMP) -d $(objtree)/vmlinux | awk -f $(srctree)/arch/x86/tools/distill.awk | $(obj)/test_get_len
+ cmd_posttest = $(OBJDUMP) -d -j .text $(objtree)/vmlinux | awk -f $(srctree)/arch/x86/tools/distill.awk | $(obj)/test_get_len
posttest: $(obj)/test_get_len vmlinux
$(call cmd,posttest)
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America), Inc.
Software Solutions Division
e-mail: mhiramat@redhat.com
^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH tracing/kprobes 2/4] x86: Check awk features before generating inat-tables.c
[not found] ` <20090821000108.GC6078@nowhere>
2009-08-21 0:56 ` Masami Hiramatsu
@ 2009-08-21 19:40 ` Masami Hiramatsu
2009-08-21 19:40 ` [PATCH tracing/kprobes 1/4] x86: Fix x86 instruction decoder selftest to check only .text Masami Hiramatsu
` (2 subsequent siblings)
4 siblings, 0 replies; 29+ messages in thread
From: Masami Hiramatsu @ 2009-08-21 19:40 UTC (permalink / raw)
To: Frederic Weisbecker, Steven Rostedt, lkml
Cc: systemtap, kvm, DLE, Masami Hiramatsu, Jim Keniston,
H. Peter Anvin, Ananth N Mavinakayanahalli, Avi Kivity,
Andi Kleen, Christoph Hellwig, Frank Ch. Eigler,
Frederic Weisbecker, Ingo Molnar, Jason Baron, K.Prasad,
Lai Jiangshan, Li Zefan, PrzemysławPawełczyk,
Roland McGrath, Sam Ravnborg, Srikar Dronamraju, Steven Rostedt,
Tom Zanussi, Vegard Nossum
Check some awk features which old mawk doesn't support.
Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Jim Keniston <jkenisto@us.ibm.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Avi Kivity <avi@redhat.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Frank Ch. Eigler <fche@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Jason Baron <jbaron@redhat.com>
Cc: K.Prasad <prasad@linux.vnet.ibm.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Przemysław Pawełczyk <przemyslaw@pawelczyk.it>
Cc: Roland McGrath <roland@redhat.com>
Cc: Sam Ravnborg <sam@ravnborg.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Tom Zanussi <tzanussi@gmail.com>
Cc: Vegard Nossum <vegard.nossum@gmail.com>
---
arch/x86/tools/gen-insn-attr-x86.awk | 20 ++++++++++++++++++++
1 files changed, 20 insertions(+), 0 deletions(-)
diff --git a/arch/x86/tools/gen-insn-attr-x86.awk b/arch/x86/tools/gen-insn-attr-x86.awk
index 93b62c9..19ba096 100644
--- a/arch/x86/tools/gen-insn-attr-x86.awk
+++ b/arch/x86/tools/gen-insn-attr-x86.awk
@@ -4,7 +4,25 @@
#
# Usage: awk -f gen-insn-attr-x86.awk x86-opcode-map.txt > inat-tables.c
+# Awk implementation sanity check
+function check_awk_implement() {
+ if (!match("abc", "[[:lower:]]+"))
+ return "Your awk doesn't support charactor-class."
+ if (sprintf("%x", 0) != "0")
+ return "Your awk has a printf-format problem."
+ return ""
+}
+
BEGIN {
+ # Implementation error checking
+ awkchecked = check_awk_implement()
+ if (awkchecked != "") {
+ print "Error: " awkchecked > "/dev/stderr"
+ print "Please try to use gawk." > "/dev/stderr"
+ exit 1
+ }
+
+ # Setup generating tables
print "/* x86 opcode map generated from x86-opcode-map.txt */"
print "/* Do not change this code. */"
ggid = 1
@@ -293,6 +311,8 @@ function convert_operands(opnd, i,imm,mod)
}
END {
+ if (awkchecked != "")
+ exit 1
# print escape opcode map's array
print "/* Escape opcode map array */"
print "const insn_attr_t const *inat_escape_tables[INAT_ESC_MAX + 1]" \
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America), Inc.
Software Solutions Division
e-mail: mhiramat@redhat.com
^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH tracing/kprobes 3/4] tracing/kprobes: Fix format typo in trace_kprobes
[not found] ` <20090821000108.GC6078@nowhere>
` (2 preceding siblings ...)
2009-08-21 19:40 ` [PATCH tracing/kprobes 1/4] x86: Fix x86 instruction decoder selftest to check only .text Masami Hiramatsu
@ 2009-08-21 19:40 ` Masami Hiramatsu
2009-08-21 19:41 ` [PATCH tracing/kprobes 4/4] tracing/kprobes: Change trace_arg to probe_arg Masami Hiramatsu
4 siblings, 0 replies; 29+ messages in thread
From: Masami Hiramatsu @ 2009-08-21 19:40 UTC (permalink / raw)
To: Frederic Weisbecker, Steven Rostedt, lkml
Cc: systemtap, kvm, DLE, Masami Hiramatsu, Jim Keniston,
H. Peter Anvin, Ananth N Mavinakayanahalli, Avi Kivity,
Andi Kleen, Christoph Hellwig, Frank Ch. Eigler,
Frederic Weisbecker, Ingo Molnar, Jason Baron, K.Prasad,
Lai Jiangshan, Li Zefan, PrzemysławPawełczyk,
Roland McGrath, Sam Ravnborg, Srikar Dronamraju, Steven Rostedt,
Tom Zanussi, Vegard Nossum
Fix a format typo in kprobe-tracer.
Currently, it shows 'tsize' in format;
$ cat /debug/tracing/events/kprobes/event/format
...
field: unsigned long ip; offset:16;tsize:8;
field: int nargs; offset:24;tsize:4;
...
This should be '\tsize';
$ cat /debug/tracing/events/kprobes/event/format
...
field: unsigned long ip; offset:16; size:8;
field: int nargs; offset:24; size:4;
...
Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Jim Keniston <jkenisto@us.ibm.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Avi Kivity <avi@redhat.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Frank Ch. Eigler <fche@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Jason Baron <jbaron@redhat.com>
Cc: K.Prasad <prasad@linux.vnet.ibm.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Przemysław Pawełczyk <przemyslaw@pawelczyk.it>
Cc: Roland McGrath <roland@redhat.com>
Cc: Sam Ravnborg <sam@ravnborg.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Tom Zanussi <tzanussi@gmail.com>
Cc: Vegard Nossum <vegard.nossum@gmail.com>
---
kernel/trace/trace_kprobe.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c
index 7cd726e..22e91c0 100644
--- a/kernel/trace/trace_kprobe.c
+++ b/kernel/trace/trace_kprobe.c
@@ -1069,7 +1069,7 @@ static int __probe_event_show_format(struct trace_seq *s,
#define SHOW_FIELD(type, item, name) \
do { \
ret = trace_seq_printf(s, "\tfield: " #type " %s;\t" \
- "offset:%u;tsize:%u;\n", name, \
+ "offset:%u;\tsize:%u;\n", name, \
(unsigned int)offsetof(typeof(field), item),\
(unsigned int)sizeof(type)); \
if (!ret) \
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America), Inc.
Software Solutions Division
e-mail: mhiramat@redhat.com
^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH tracing/kprobes 4/4] tracing/kprobes: Change trace_arg to probe_arg
[not found] ` <20090821000108.GC6078@nowhere>
` (3 preceding siblings ...)
2009-08-21 19:40 ` [PATCH tracing/kprobes 3/4] tracing/kprobes: Fix format typo in trace_kprobes Masami Hiramatsu
@ 2009-08-21 19:41 ` Masami Hiramatsu
4 siblings, 0 replies; 29+ messages in thread
From: Masami Hiramatsu @ 2009-08-21 19:41 UTC (permalink / raw)
To: Frederic Weisbecker, Steven Rostedt, lkml
Cc: systemtap, kvm, DLE, Masami Hiramatsu, Jim Keniston,
H. Peter Anvin, Ananth N Mavinakayanahalli, Avi Kivity,
Andi Kleen, Christoph Hellwig, Frank Ch. Eigler,
Frederic Weisbecker, Ingo Molnar, Jason Baron, K.Prasad,
Lai Jiangshan, Li Zefan, PrzemysławPawełczyk,
Roland McGrath, Sam Ravnborg, Srikar Dronamraju, Steven Rostedt,
Tom Zanussi, Vegard Nossum
Change trace_arg_string() and parse_trace_arg() to probe_arg_string()
and parse_probe_arg(), since those are kprobe-tracer local functions.
Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Jim Keniston <jkenisto@us.ibm.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Avi Kivity <avi@redhat.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Frank Ch. Eigler <fche@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Jason Baron <jbaron@redhat.com>
Cc: K.Prasad <prasad@linux.vnet.ibm.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Przemysław Pawełczyk <przemyslaw@pawelczyk.it>
Cc: Roland McGrath <roland@redhat.com>
Cc: Sam Ravnborg <sam@ravnborg.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Tom Zanussi <tzanussi@gmail.com>
Cc: Vegard Nossum <vegard.nossum@gmail.com>
---
kernel/trace/trace_kprobe.c | 18 +++++++++---------
1 files changed, 9 insertions(+), 9 deletions(-)
diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c
index 22e91c0..783d2db 100644
--- a/kernel/trace/trace_kprobe.c
+++ b/kernel/trace/trace_kprobe.c
@@ -220,7 +220,7 @@ static __kprobes void *probe_address(struct trace_probe *tp)
return (probe_is_return(tp)) ? tp->rp.kp.addr : tp->kp.addr;
}
-static int trace_arg_string(char *buf, size_t n, struct fetch_func *ff)
+static int probe_arg_string(char *buf, size_t n, struct fetch_func *ff)
{
int ret = -EINVAL;
@@ -250,7 +250,7 @@ static int trace_arg_string(char *buf, size_t n, struct fetch_func *ff)
if (ret >= n)
goto end;
l += ret;
- ret = trace_arg_string(buf + l, n - l, &id->orig);
+ ret = probe_arg_string(buf + l, n - l, &id->orig);
if (ret < 0)
goto end;
l += ret;
@@ -408,7 +408,7 @@ static int split_symbol_offset(char *symbol, long *offset)
#define PARAM_MAX_ARGS 16
#define PARAM_MAX_STACK (THREAD_SIZE / sizeof(unsigned long))
-static int parse_trace_arg(char *arg, struct fetch_func *ff, int is_return)
+static int parse_probe_arg(char *arg, struct fetch_func *ff, int is_return)
{
int ret = 0;
unsigned long param;
@@ -499,7 +499,7 @@ static int parse_trace_arg(char *arg, struct fetch_func *ff, int is_return)
if (!id)
return -ENOMEM;
id->offset = offset;
- ret = parse_trace_arg(arg, &id->orig, is_return);
+ ret = parse_probe_arg(arg, &id->orig, is_return);
if (ret)
kfree(id);
else {
@@ -617,7 +617,7 @@ static int create_trace_probe(int argc, char **argv)
ret = -ENOSPC;
goto error;
}
- ret = parse_trace_arg(argv[i], &tp->args[i], is_return);
+ ret = parse_probe_arg(argv[i], &tp->args[i], is_return);
if (ret)
goto error;
}
@@ -680,7 +680,7 @@ static int probes_seq_show(struct seq_file *m, void *v)
seq_printf(m, " 0x%p", probe_address(tp));
for (i = 0; i < tp->nr_args; i++) {
- ret = trace_arg_string(buf, MAX_ARGSTR_LEN, &tp->args[i]);
+ ret = probe_arg_string(buf, MAX_ARGSTR_LEN, &tp->args[i]);
if (ret < 0) {
pr_warning("Argument%d decoding error(%d).\n", i, ret);
return ret;
@@ -996,7 +996,7 @@ static int kprobe_event_define_fields(struct ftrace_event_call *event_call)
sprintf(buf, "arg%d", i);
DEFINE_FIELD(unsigned long, args[i], buf, 0);
/* Set argument string as an alias field */
- ret = trace_arg_string(buf, MAX_ARGSTR_LEN, &tp->args[i]);
+ ret = probe_arg_string(buf, MAX_ARGSTR_LEN, &tp->args[i]);
if (ret < 0)
return ret;
DEFINE_FIELD(unsigned long, args[i], buf, 0);
@@ -1023,7 +1023,7 @@ static int kretprobe_event_define_fields(struct ftrace_event_call *event_call)
sprintf(buf, "arg%d", i);
DEFINE_FIELD(unsigned long, args[i], buf, 0);
/* Set argument string as an alias field */
- ret = trace_arg_string(buf, MAX_ARGSTR_LEN, &tp->args[i]);
+ ret = probe_arg_string(buf, MAX_ARGSTR_LEN, &tp->args[i]);
if (ret < 0)
return ret;
DEFINE_FIELD(unsigned long, args[i], buf, 0);
@@ -1040,7 +1040,7 @@ static int __probe_event_show_format(struct trace_seq *s,
/* Show aliases */
for (i = 0; i < tp->nr_args; i++) {
- ret = trace_arg_string(buf, MAX_ARGSTR_LEN, &tp->args[i]);
+ ret = probe_arg_string(buf, MAX_ARGSTR_LEN, &tp->args[i]);
if (ret < 0)
return ret;
if (!trace_seq_printf(s, "\talias: %s;\toriginal: arg%d;\n",
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America), Inc.
Software Solutions Division
e-mail: mhiramat@redhat.com
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [TOOL] c2kpe: C expression to kprobe event format converter
[not found] ` <20090830195041.GA6189@nowhere>
@ 2009-08-31 4:12 ` Masami Hiramatsu
0 siblings, 0 replies; 29+ messages in thread
From: Masami Hiramatsu @ 2009-08-31 4:12 UTC (permalink / raw)
To: Frederic Weisbecker
Cc: Ingo Molnar, Steven Rostedt, lkml, Ananth N Mavinakayanahalli,
Avi Kivity, Andi Kleen, Christoph Hellwig, Frank Ch. Eigler,
H. Peter Anvin, Jason Baron, Jim Keniston, K.Prasad,
Lai Jiangshan, Li Zefan, PrzemysławPawełczyk,
Roland McGrath, Sam Ravnborg, Srikar Dronamraju, Tom Zanussi,
Vegard Nossum, systemtap, kvm, DLE
Frederic Weisbecker wrote:
> On Thu, Aug 13, 2009 at 04:59:19PM -0400, Masami Hiramatsu wrote:
>> This program converts probe point in C expression to kprobe event
>> format for kprobe-based event tracer. This helps to define kprobes
>> events by C source line number or function name, and local variable
>> name. Currently, this supports only x86(32/64) kernels.
>>
>>
>> Compile
>> --------
>> Before compilation, please install libelf and libdwarf development
>> packages.
>> (e.g. elfutils-libelf-devel and libdwarf-devel on Fedora)
>
>
> This may probably need a specific libdwarf version?
>
> c2kpe.c: In function ‘die_get_entrypc’:
> c2kpe.c:422: erreur: ‘Dwarf_Ranges’ undeclared (first use in this function)
> c2kpe.c:422: erreur: (Each undeclared identifier is reported only once
> c2kpe.c:422: erreur: for each function it appears in.)
> c2kpe.c:422: erreur: ‘ranges’ undeclared (first use in this function)
> c2kpe.c:447: attention : implicit declaration of function ‘dwarf_get_ranges’
> c2kpe.c:451: attention : implicit declaration of function ‘dwarf_ranges_dealloc’
Aah, sure, it should be compiled with libdwarf newer than 20090324.
You can find it in http://reality.sgiweb.org/davea/dwarf.html
BTW, libdwarf and libdw (which is the yet another implementation of
dwarf library) are still under development, e.g. libdwarf doesn't
support gcc-4.4.1(very new) and only the latest libdw(0.142) can
support it. So, perhaps I might better port it on libdw, even that is
less documented...:(
>> TODO
>> ----
>> - Fix bugs.
>> - Support multiple probepoints from stdin.
>> - Better kmodule support.
>> - Use elfutils-libdw?
>> - Merge into trace-cmd or perf-tools?
>
>
> Yeah definetly, that would be a veeery interesting thing to have.
> I've played with kprobe ftrace to debug something this evening.
>
> It's very cool to be able to put dynamic tracepoints in desired places.
>
> But...
> I firstly needed to put random trace_printk() in some places to
> observe some variables values. And then I thought about the kprobes
> tracer and realized I could do that without the need of rebuilding
> my kernel. Then I've played with it and indeed it works well and
> it's useful, but at the cost of reading objdump based assembly
> code to find the places where I could find my variables values.
> And after two or three probes in such conditions, I've become
> tired of that, then I wanted to try this tool.
>
>
> While I cannot yet because of this build error, I can imagine
> the power of such facility from perf.
>
> We could have a perf probe that creates a kprobe event in debugfs
> (default enable = 0) and which then rely on perf record for the actual
> recording.
>
> Then we could analyse it through perf trace.
> Let's imagine a simple example:
>
> int foo(int arg1, int arg2)
> {
> int var1;
>
> var1 = arg1;
> var1 *= arg2;
> var1 -= arg1;
>
> ------> insert a probe here (file bar.c : line 60)
>
> var1 ^= ...
>
> return var1;
> }
>
> ./perf kprobe --file bar.c:60 --action "arg1=%d","arg2=%d","var1=%d" -- ls -R /
I recommend it should be separated from record, like below:
# set new event
./perf kprobe --add kprobe:event1 --file bar.c:60 --action "arg1=%d","arg2=%d","var1=%d"
# record new event
./perf record -e kprobe:event1 -a -R -- ls -R /
This will allow us to focus on one thing -- convert C to kprobe-tracer.
And also, it can be listed as like as tracepoint events.
> ./perf trace
> arg1=1 arg2=1 var1=0
> arg1=2 arg2=2 var1=2
> etc..
>
> You may want to sort by field:
>
> ./perf trace -s arg1 --order desc
> arg1=1
> |
> ------- arg2=1 var=1
> |
> ------- arg2=2 var=1
>
> arg1=2
> |
> ------- arg2=1 var=0
> |
> ------- [...]
>
> ./perf trace -s arg1,arg2 --order asc
> arg1=1
> |
> ------- arg2=1
> |
> --------- var1=0
> |
> --------- var1=....
> arg2=...
> |
>
> Ok the latter is a bad example because var1 will always have only one
> value for a given arg1 and arg2. But I guess you see the point.
>
> You won't have to care about the perf trace part, it's already
> implemented and I'll soon handle the sorting part.
>
> All we need is the perf kprobes that translate a C level
> probing expression to a /debug/tracing/kprobe_events compliant
> thing. And then just call perf record with the new created
> event as an argument.
Indeed, that's what I imagine.
Thank you,
--
Masami Hiramatsu
Software Engineer
Hitachi Computer Products (America), Inc.
Software Solutions Division
e-mail: mhiramat@redhat.com
^ permalink raw reply [flat|nested] 29+ messages in thread
end of thread, other threads:[~2009-08-31 4:12 UTC | newest]
Thread overview: 29+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2009-08-13 20:31 [PATCH -tip v14 00/12] tracing: kprobe-based event tracer and x86 instruction decoder Masami Hiramatsu
2009-08-13 20:31 ` [PATCH -tip v14 03/12] kprobes: checks probe address is instruction boudary on x86 Masami Hiramatsu
[not found] ` <20090818230340.GF5231@nowhere>
2009-08-18 23:16 ` Masami Hiramatsu
[not found] ` <20090818234341.GG5231@nowhere>
2009-08-19 0:18 ` Masami Hiramatsu
2009-08-13 20:31 ` [PATCH -tip v14 01/12] x86: instruction decoder API Masami Hiramatsu
[not found] ` <20090819234227.GJ4972@nowhere>
2009-08-20 14:38 ` Masami Hiramatsu
[not found] ` <20090820002113.GK4972@nowhere>
2009-08-20 15:00 ` Masami Hiramatsu
[not found] ` <20090820152549.GD6389@nowhere>
2009-08-20 16:12 ` Masami Hiramatsu
[not found] ` <20090820180737.GA6229@nowhere>
2009-08-20 18:58 ` Masami Hiramatsu
2009-08-13 20:31 ` [PATCH -tip v14 05/12] x86: add pt_regs register and stack access APIs Masami Hiramatsu
2009-08-13 20:31 ` [PATCH -tip v14 04/12] kprobes: cleanup fix_riprel() using insn decoder on x86 Masami Hiramatsu
2009-08-13 20:31 ` [PATCH -tip v14 02/12] x86: x86 instruction decoder build-time selftest Masami Hiramatsu
2009-08-13 20:32 ` [PATCH -tip v14 08/12] tracing: add kprobe-based event tracer Masami Hiramatsu
2009-08-13 20:32 ` [PATCH -tip v14 10/12] tracing: Generate names for each kprobe event automatically Masami Hiramatsu
2009-08-13 20:32 ` [PATCH -tip v14 09/12] tracing: Kprobe-tracer supports more than 6 arguments Masami Hiramatsu
2009-08-13 20:33 ` [PATCH -tip v14 12/12] tracing: Add kprobes event profiling interface Masami Hiramatsu
2009-08-13 20:33 ` [PATCH -tip v14 11/12] tracing: Kprobe tracer assigns new event ids for each event Masami Hiramatsu
2009-08-13 20:47 ` [PATCH -tip v14 07/12] tracing: Introduce TRACE_FIELD_ZERO() macro Masami Hiramatsu
[not found] ` <20090819010903.GI5231@nowhere>
2009-08-19 2:18 ` Masami Hiramatsu
2009-08-13 20:47 ` [PATCH -tip v14 06/12] tracing: ftrace dynamic ftrace_event_call support Masami Hiramatsu
2009-08-13 20:54 ` [TOOL] kprobestest : Kprobe stress test tool Masami Hiramatsu
[not found] ` <20090820184331.GA6078@nowhere>
2009-08-20 19:42 ` Masami Hiramatsu
[not found] ` <20090821000108.GC6078@nowhere>
2009-08-21 0:56 ` Masami Hiramatsu
2009-08-21 19:40 ` [PATCH tracing/kprobes 2/4] x86: Check awk features before generating inat-tables.c Masami Hiramatsu
2009-08-21 19:40 ` [PATCH tracing/kprobes 1/4] x86: Fix x86 instruction decoder selftest to check only .text Masami Hiramatsu
2009-08-21 19:40 ` [PATCH tracing/kprobes 3/4] tracing/kprobes: Fix format typo in trace_kprobes Masami Hiramatsu
2009-08-21 19:41 ` [PATCH tracing/kprobes 4/4] tracing/kprobes: Change trace_arg to probe_arg Masami Hiramatsu
2009-08-13 20:56 ` [TOOL] c2kpe: C expression to kprobe event format converter Masami Hiramatsu
[not found] ` <20090830195041.GA6189@nowhere>
2009-08-31 4:12 ` Masami Hiramatsu
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).