Re: [PATCH v2 3/3] arm64: implement KPROBES_ON_FTRACE
From: Jisheng Zhang <hidden>
Date: 2019-08-20 07:19:38
Also in:
linux-doc, lkml
On Tue, 20 Aug 2019 03:54:20 +0000 Jisheng Zhang wrote:
KPROBES_ON_FTRACE avoids much of the overhead with regular kprobes as it eliminates the need for a trap, as well as the need to emulate or single-step instructions. This patch implements KPROBES_ON_FTRACE for arm64. Tested on berlin arm64 platform.
some performance numbers may be interesting.
HW: Berlin arm64 platform, cpufreq is forced to 800MHZ
SW: getppid syscall micro-benchmark, source code is put at the end of this email.
A. Not probed.
B. Probed at __arm64_sys_getppid w/ non-operation probe functions, w/o KPROBES_ON_FTRACE
C. Probed at __arm64_sys_getppid w/ non-operation probe functions, w/ KPROBES_ON_FTRACE
A: 1905 ns/call
B: 5833 ns/call
C: 2169 ns/call
The overhead of kprobes is 5833 - 1905 = 3928 ns/call
The overhead of kprobes w/ KPROBES_ON_FTRACE is 2169 - 1905 = 264 ns/call
As can be seen, KPROBES_ON_FTRACE significantly reduce the overhead of kprobes.
Thanks
<---8---
#include <stdio.h>
#include <stdlib.h>
#include <sys/time.h>
#include <sys/resource.h>
#include <unistd.h>
int main (int argc, char *argv[])
{
struct timeval tv;
unsigned long count;
struct rusage usage;
for (count = 0; count < 10000000; count++)
getppid();
getrusage(RUSAGE_SELF, &usage);
tv = usage.ru_stime;
tv.tv_sec += usage.ru_utime.tv_sec;
tv.tv_usec += usage.ru_utime.tv_usec;
fprintf(stderr, "getppid was called %u times: %d nsec per call\n",
count, (tv.tv_sec*1000*1000 + tv.tv_usec)/(count/1000));
return 0;
}
quoted hunk ↗ jump to hunk
~ # mount -t debugfs debugfs /sys/kernel/debug/ ~ # cd /sys/kernel/debug/ /sys/kernel/debug # echo 'p _do_fork' > tracing/kprobe_events before the patch: /sys/kernel/debug # cat kprobes/list ffffff801009fe28 k _do_fork+0x0 [DISABLED] after the patch: /sys/kernel/debug # cat kprobes/list ffffff801009ff54 k _do_fork+0x4 [DISABLED][FTRACE] Signed-off-by: Jisheng Zhang <redacted> --- .../debug/kprobes-on-ftrace/arch-support.txt | 2 +- arch/arm64/Kconfig | 1 + arch/arm64/kernel/probes/Makefile | 1 + arch/arm64/kernel/probes/ftrace.c | 60 +++++++++++++++++++ 4 files changed, 63 insertions(+), 1 deletion(-) create mode 100644 arch/arm64/kernel/probes/ftrace.cdiff --git a/Documentation/features/debug/kprobes-on-ftrace/arch-support.txt b/Documentation/features/debug/kprobes-on-ftrace/arch-support.txt index 68f266944d5f..e8358a38981c 100644 --- a/Documentation/features/debug/kprobes-on-ftrace/arch-support.txt +++ b/Documentation/features/debug/kprobes-on-ftrace/arch-support.txt@@ -9,7 +9,7 @@ | alpha: | TODO | | arc: | TODO | | arm: | TODO | - | arm64: | TODO | + | arm64: | ok | | c6x: | TODO | | csky: | TODO | | h8300: | TODO |diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig index 663392d1eae2..928700f15e23 100644 --- a/arch/arm64/Kconfig +++ b/arch/arm64/Kconfig@@ -167,6 +167,7 @@ config ARM64 select HAVE_STACKPROTECTOR select HAVE_SYSCALL_TRACEPOINTS select HAVE_KPROBES + select HAVE_KPROBES_ON_FTRACE select HAVE_KRETPROBES select HAVE_GENERIC_VDSO select IOMMU_DMA if IOMMU_SUPPORTdiff --git a/arch/arm64/kernel/probes/Makefile b/arch/arm64/kernel/probes/Makefile index 8e4be92e25b1..4020cfc66564 100644 --- a/arch/arm64/kernel/probes/Makefile +++ b/arch/arm64/kernel/probes/Makefile@@ -4,3 +4,4 @@ obj-$(CONFIG_KPROBES) += kprobes.o decode-insn.o \ simulate-insn.o obj-$(CONFIG_UPROBES) += uprobes.o decode-insn.o \ simulate-insn.o +obj-$(CONFIG_KPROBES_ON_FTRACE) += ftrace.odiff --git a/arch/arm64/kernel/probes/ftrace.c b/arch/arm64/kernel/probes/ftrace.c new file mode 100644 index 000000000000..52901ffff570 --- /dev/null +++ b/arch/arm64/kernel/probes/ftrace.c@@ -0,0 +1,60 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * Dynamic Ftrace based Kprobes Optimization + * + * Copyright (C) Hitachi Ltd., 2012 + * Copyright (C) 2019 Jisheng Zhang <jszhang@kernel.org> + * Synaptics Incorporated + */ + +#include <linux/kprobes.h> + +/* Ftrace callback handler for kprobes -- called under preepmt disabed */ +void kprobe_ftrace_handler(unsigned long ip, unsigned long parent_ip, + struct ftrace_ops *ops, struct pt_regs *regs) +{ + struct kprobe *p; + struct kprobe_ctlblk *kcb; + + /* Preempt is disabled by ftrace */ + p = get_kprobe((kprobe_opcode_t *)ip); + if (unlikely(!p) || kprobe_disabled(p)) + return; + + kcb = get_kprobe_ctlblk(); + if (kprobe_running()) { + kprobes_inc_nmissed_count(p); + } else { + unsigned long orig_ip = instruction_pointer(regs); + /* Kprobe handler expects regs->pc = pc + 1 as breakpoint hit */ + instruction_pointer_set(regs, ip + sizeof(kprobe_opcode_t)); + + __this_cpu_write(current_kprobe, p); + kcb->kprobe_status = KPROBE_HIT_ACTIVE; + if (!p->pre_handler || !p->pre_handler(p, regs)) { + /* + * Emulate singlestep (and also recover regs->pc) + * as if there is a nop + */ + instruction_pointer_set(regs, + (unsigned long)p->addr + MCOUNT_INSN_SIZE); + if (unlikely(p->post_handler)) { + kcb->kprobe_status = KPROBE_HIT_SSDONE; + p->post_handler(p, regs, 0); + } + instruction_pointer_set(regs, orig_ip); + } + /* + * If pre_handler returns !0, it changes regs->pc. We have to + * skip emulating post_handler. + */ + __this_cpu_write(current_kprobe, NULL); + } +} +NOKPROBE_SYMBOL(kprobe_ftrace_handler); + +int arch_prepare_kprobe_ftrace(struct kprobe *p) +{ + p->ainsn.api.insn = NULL; + return 0; +} --2.23.0.rc1 _______________________________________________ linux-arm-kernel mailing list linux-arm-kernel@lists.infradead.org
_______________________________________________ linux-arm-kernel mailing list linux-arm-kernel@lists.infradead.org http://lists.infradead.org/mailman/listinfo/linux-arm-kernel