From: Masami Hiramatsu (Google) <mhiramat@kernel.org>
To: "Masami Hiramatsu (Google)" <mhiramat@kernel.org>
Cc: Alexei Starovoitov <alexei.starovoitov@gmail.com>,
Steven Rostedt <rostedt@goodmis.org>,
Florent Revest <revest@chromium.org>,
linux-trace-kernel@vger.kernel.org,
LKML <linux-kernel@vger.kernel.org>,
Martin KaFai Lau <martin.lau@linux.dev>,
bpf <bpf@vger.kernel.org>, Alexei Starovoitov <ast@kernel.org>,
Jiri Olsa <jolsa@kernel.org>,
Alan Maguire <alan.maguire@oracle.com>,
Mark Rutland <mark.rutland@arm.com>,
linux-arch@vger.kernel.org
Subject: Re: [PATCH v19 18/19] ftrace: Add ftrace_get_symaddr to convert fentry_ip to symaddr
Date: Mon, 9 Dec 2024 17:05:05 +0900 [thread overview]
Message-ID: <20241209170505.61d4a585c2b0282190f29954@kernel.org> (raw)
In-Reply-To: <173125394102.172790.7669548166614384865.stgit@devnote2>
On Mon, 11 Nov 2024 00:52:21 +0900
"Masami Hiramatsu (Google)" <mhiramat@kernel.org> wrote:
> From: Masami Hiramatsu (Google) <mhiramat@kernel.org>
>
> This introduces ftrace_get_symaddr() which tries to convert fentry_ip
> passed by ftrace or fgraph callback to symaddr without calling
> kallsyms API. It returns the symbol address or 0 if it fails to
> convert it.
>
> Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
> ---
> Changes in v19:
> - Newly added.
> ---
> arch/arm64/include/asm/ftrace.h | 2 +
> arch/arm64/kernel/ftrace.c | 63 +++++++++++++++++++++++++++++++++++++++
> arch/x86/include/asm/ftrace.h | 21 +++++++++++++
> include/linux/ftrace.h | 13 ++++++++
> 4 files changed, 99 insertions(+)
>
> diff --git a/arch/arm64/include/asm/ftrace.h b/arch/arm64/include/asm/ftrace.h
> index 876e88ad4119..f08e70bf09ea 100644
> --- a/arch/arm64/include/asm/ftrace.h
> +++ b/arch/arm64/include/asm/ftrace.h
> @@ -52,6 +52,8 @@ extern unsigned long ftrace_graph_call;
> extern void return_to_handler(void);
>
> unsigned long ftrace_call_adjust(unsigned long addr);
> +unsigned long arch_ftrace_call_adjust(unsigned long fentry_ip);
> +#define ftrace_call_adjust(fentry_ip) arch_ftrace_call_adjust(fentry_ip)
Oops, this is arch_ftrace_get_symaddr()!
It needs to be fixed.
Thanks,
>
> #ifdef CONFIG_DYNAMIC_FTRACE_WITH_ARGS
> #define HAVE_ARCH_FTRACE_REGS
> diff --git a/arch/arm64/kernel/ftrace.c b/arch/arm64/kernel/ftrace.c
> index 606fd6994578..de1223669758 100644
> --- a/arch/arm64/kernel/ftrace.c
> +++ b/arch/arm64/kernel/ftrace.c
> @@ -143,6 +143,69 @@ unsigned long ftrace_call_adjust(unsigned long addr)
> return addr;
> }
>
> +/* Convert fentry_ip to the symbol address without kallsyms */
> +unsigned long arch_ftrace_get_symaddr(unsigned long fentry_ip)
> +{
> + u32 insn;
> +
> + /*
> + * When using patchable-function-entry without pre-function NOPS, ftrace
> + * entry is the address of the first NOP after the function entry point.
> + *
> + * The compiler has either generated:
> + *
> + * func+00: func: NOP // To be patched to MOV X9, LR
> + * func+04: NOP // To be patched to BL <caller>
> + *
> + * Or:
> + *
> + * func-04: BTI C
> + * func+00: func: NOP // To be patched to MOV X9, LR
> + * func+04: NOP // To be patched to BL <caller>
> + *
> + * The fentry_ip is the address of `BL <caller>` which is at `func + 4`
> + * bytes in either case.
> + */
> + if (!IS_ENABLED(CONFIG_DYNAMIC_FTRACE_WITH_CALL_OPS))
> + return fentry_ip - AARCH64_INSN_SIZE;
> +
> + /*
> + * When using patchable-function-entry with pre-function NOPs, BTI is
> + * a bit different.
> + *
> + * func+00: func: NOP // To be patched to MOV X9, LR
> + * func+04: NOP // To be patched to BL <caller>
> + *
> + * Or:
> + *
> + * func+00: func: BTI C
> + * func+04: NOP // To be patched to MOV X9, LR
> + * func+08: NOP // To be patched to BL <caller>
> + *
> + * The fentry_ip is the address of `BL <caller>` which is at either
> + * `func + 4` or `func + 8` depends on whether there is a BTI.
> + */
> +
> + /* If there is no BTI, the func address should be one instruction before. */
> + if (!IS_ENABLED(CONFIG_ARM64_BTI_KERNEL))
> + return fentry_ip - AARCH64_INSN_SIZE;
> +
> + /* We want to be extra safe in case entry ip is on the page edge,
> + * but otherwise we need to avoid get_kernel_nofault()'s overhead.
> + */
> + if ((fentry_ip & ~PAGE_MASK) < AARCH64_INSN_SIZE * 2) {
> + if (get_kernel_nofault(insn, (u32 *)(fentry_ip - AARCH64_INSN_SIZE * 2)))
> + return 0;
> + } else {
> + insn = *(u32 *)(fentry_ip - AARCH64_INSN_SIZE * 2);
> + }
> +
> + if (aarch64_insn_is_bti(le32_to_cpu((__le32)insn)))
> + return fentry_ip - AARCH64_INSN_SIZE * 2;
> +
> + return fentry_ip - AARCH64_INSN_SIZE;
> +}
> +
> /*
> * Replace a single instruction, which may be a branch or NOP.
> * If @validate == true, a replaced instruction is checked against 'old'.
> diff --git a/arch/x86/include/asm/ftrace.h b/arch/x86/include/asm/ftrace.h
> index cc92c99ef276..f9cb4d07df58 100644
> --- a/arch/x86/include/asm/ftrace.h
> +++ b/arch/x86/include/asm/ftrace.h
> @@ -34,6 +34,27 @@ static inline unsigned long ftrace_call_adjust(unsigned long addr)
> return addr;
> }
>
> +static inline unsigned long arch_ftrace_get_symaddr(unsigned long fentry_ip)
> +{
> +#ifdef CONFIG_X86_KERNEL_IBT
> + u32 instr;
> +
> + /* We want to be extra safe in case entry ip is on the page edge,
> + * but otherwise we need to avoid get_kernel_nofault()'s overhead.
> + */
> + if ((fentry_ip & ~PAGE_MASK) < ENDBR_INSN_SIZE) {
> + if (get_kernel_nofault(instr, (u32 *)(fentry_ip - ENDBR_INSN_SIZE)))
> + return fentry_ip;
> + } else {
> + instr = *(u32 *)(fentry_ip - ENDBR_INSN_SIZE);
> + }
> + if (is_endbr(instr))
> + fentry_ip -= ENDBR_INSN_SIZE;
> +#endif
> + return fentry_ip;
> +}
> +#define ftrace_get_symaddr(fentry_ip) arch_ftrace_get_symaddr(fentry_ip)
> +
> #ifdef CONFIG_HAVE_DYNAMIC_FTRACE_WITH_ARGS
>
> #include <linux/ftrace_regs.h>
> diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
> index 4c553fe9c026..9659bb2cd76c 100644
> --- a/include/linux/ftrace.h
> +++ b/include/linux/ftrace.h
> @@ -652,6 +652,19 @@ struct ftrace_ops *ftrace_ops_trampoline(unsigned long addr);
>
> bool is_ftrace_trampoline(unsigned long addr);
>
> +/* Arches can override ftrace_get_symaddr() to convert fentry_ip to symaddr. */
> +#ifndef ftrace_get_symaddr
> +/**
> + * ftrace_get_symaddr - return the symbol address from fentry_ip
> + * @fentry_ip: the address of ftrace location
> + *
> + * Get the symbol address from @fentry_ip (fast path). If there is no fast
> + * search path, this returns 0.
> + * User may need to use kallsyms API to find the symbol address.
> + */
> +#define ftrace_get_symaddr(fentry_ip) (0)
> +#endif
> +
> /*
> * The dyn_ftrace record's flags field is split into two parts.
> * the first part which is '0-FTRACE_REF_MAX' is a counter of
>
>
--
Masami Hiramatsu (Google) <mhiramat@kernel.org>
next prev parent reply other threads:[~2024-12-09 8:05 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-11-10 15:48 [PATCH v19 00/19] tracing: fprobe: function_graph: Multi-function graph and fprobe on fgraph Masami Hiramatsu (Google)
2024-11-10 15:48 ` [PATCH v19 01/19] fgraph: Pass ftrace_regs to entryfunc Masami Hiramatsu (Google)
2024-11-10 15:49 ` [PATCH v19 02/19] fgraph: Replace fgraph_ret_regs with ftrace_regs Masami Hiramatsu (Google)
2024-11-10 15:49 ` [PATCH v19 03/19] fgraph: Pass ftrace_regs to retfunc Masami Hiramatsu (Google)
2024-11-10 15:49 ` [PATCH v19 04/19] fprobe: Use ftrace_regs in fprobe entry handler Masami Hiramatsu (Google)
2024-11-10 15:49 ` [PATCH v19 05/19] fprobe: Use ftrace_regs in fprobe exit handler Masami Hiramatsu (Google)
2024-11-10 17:04 ` Heiko Carstens
2024-11-10 15:49 ` [PATCH v19 06/19] tracing: Add ftrace_partial_regs() for converting ftrace_regs to pt_regs Masami Hiramatsu (Google)
2024-11-10 15:50 ` [PATCH v19 07/19] tracing: Add ftrace_fill_perf_regs() for perf event Masami Hiramatsu (Google)
2024-11-10 17:05 ` Heiko Carstens
2024-11-10 15:50 ` [PATCH v19 08/19] tracing/fprobe: Enable fprobe events with CONFIG_DYNAMIC_FTRACE_WITH_ARGS Masami Hiramatsu (Google)
2024-11-10 15:50 ` [PATCH v19 09/19] bpf: Enable kprobe_multi feature if CONFIG_FPROBE is enabled Masami Hiramatsu (Google)
2024-11-10 15:50 ` [PATCH v19 10/19] ftrace: Add CONFIG_HAVE_FTRACE_GRAPH_FUNC Masami Hiramatsu (Google)
2024-11-10 15:50 ` [PATCH v19 11/19] s390/tracing: Enable HAVE_FTRACE_GRAPH_FUNC Masami Hiramatsu (Google)
2024-11-10 17:05 ` Heiko Carstens
2024-11-10 15:51 ` [PATCH v19 12/19] fprobe: Rewrite fprobe on function-graph tracer Masami Hiramatsu (Google)
2024-11-10 17:07 ` Heiko Carstens
2024-12-05 18:34 ` Steven Rostedt
2024-12-05 23:09 ` Masami Hiramatsu
2024-11-10 15:51 ` [PATCH v19 13/19] fprobe: Add fprobe_header encoding feature Masami Hiramatsu (Google)
2024-11-10 17:16 ` Heiko Carstens
2024-11-10 15:51 ` [PATCH v19 14/19] tracing/fprobe: Remove nr_maxactive from fprobe Masami Hiramatsu (Google)
2024-11-10 15:51 ` [PATCH v19 15/19] selftests: ftrace: Remove obsolate maxactive syntax check Masami Hiramatsu (Google)
2024-11-10 15:52 ` [PATCH v19 16/19] selftests/ftrace: Add a test case for repeating register/unregister fprobe Masami Hiramatsu (Google)
2024-11-10 15:52 ` [PATCH v19 17/19] Documentation: probes: Update fprobe on function-graph tracer Masami Hiramatsu (Google)
2024-11-13 0:42 ` [PATCH 19.1] " Masami Hiramatsu (Google)
2024-11-10 15:52 ` [PATCH v19 18/19] ftrace: Add ftrace_get_symaddr to convert fentry_ip to symaddr Masami Hiramatsu (Google)
2024-12-09 8:05 ` Masami Hiramatsu [this message]
2024-11-10 15:52 ` [PATCH v19 19/19] bpf: Use ftrace_get_symaddr() in get_entry_ip() Masami Hiramatsu (Google)
2024-12-09 9:29 ` Masami Hiramatsu
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20241209170505.61d4a585c2b0282190f29954@kernel.org \
--to=mhiramat@kernel.org \
--cc=alan.maguire@oracle.com \
--cc=alexei.starovoitov@gmail.com \
--cc=ast@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=jolsa@kernel.org \
--cc=linux-arch@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-trace-kernel@vger.kernel.org \
--cc=mark.rutland@arm.com \
--cc=martin.lau@linux.dev \
--cc=revest@chromium.org \
--cc=rostedt@goodmis.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox