From: madvenka@linux.microsoft.com
To: jpoimboe@redhat.com, peterz@infradead.org,
chenzhongjin@huawei.com, mark.rutland@arm.com,
broonie@kernel.org, nobuta.keiya@fujitsu.com,
sjitindarsingh@gmail.com, catalin.marinas@arm.com,
will@kernel.org, jamorris@linux.microsoft.com,
linux-arm-kernel@lists.infradead.org,
live-patching@vger.kernel.org, linux-kernel@vger.kernel.org,
madvenka@linux.microsoft.com
Subject: [RFC PATCH v2 15/20] arm64: Add unwind hints to specific points in code
Date: Mon, 23 May 2022 19:16:32 -0500 [thread overview]
Message-ID: <20220524001637.1707472-16-madvenka@linux.microsoft.com> (raw)
In-Reply-To: <20220524001637.1707472-1-madvenka@linux.microsoft.com>
From: "Madhavan T. Venkataraman" <madvenka@linux.microsoft.com>
Add unwind hints to the following:
- Ftrace entry code
- Interrupt and Exception handlers
- Kretprobe trampoline
Signed-off-by: Madhavan T. Venkataraman <madvenka@linux.microsoft.com>
---
arch/arm64/kernel/entry-ftrace.S | 23 +++++++++++++++++++
arch/arm64/kernel/entry.S | 3 +++
arch/arm64/kernel/probes/kprobes_trampoline.S | 3 +++
3 files changed, 29 insertions(+)
diff --git a/arch/arm64/kernel/entry-ftrace.S b/arch/arm64/kernel/entry-ftrace.S
index e535480a4069..6d68833e8cec 100644
--- a/arch/arm64/kernel/entry-ftrace.S
+++ b/arch/arm64/kernel/entry-ftrace.S
@@ -11,6 +11,7 @@
#include <asm/assembler.h>
#include <asm/ftrace.h>
#include <asm/insn.h>
+#include <asm/unwind_hints.h>
#ifdef CONFIG_DYNAMIC_FTRACE_WITH_REGS
/*
@@ -95,7 +96,14 @@ SYM_CODE_START(ftrace_common)
mov x3, sp // regs
SYM_INNER_LABEL(ftrace_call, SYM_L_GLOBAL)
+ /*
+ * Tracer functions are patched at ftrace_stub. Stack traces
+ * taken from tracer functions will end up here. Place an
+ * unwind hint based on the stackframe setup in ftrace_regs_entry.
+ */
bl ftrace_stub
+SYM_INNER_LABEL(ftrace_call_entry, SYM_L_GLOBAL)
+ UNWIND_HINT_REGS PT_REGS_SIZE
#ifdef CONFIG_FUNCTION_GRAPH_TRACER
SYM_INNER_LABEL(ftrace_graph_call, SYM_L_GLOBAL) // ftrace_graph_caller();
@@ -134,10 +142,25 @@ SYM_CODE_START(ftrace_graph_caller)
add x1, sp, #S_LR // parent_ip (callsite's LR)
ldr x2, [sp, #PT_REGS_SIZE] // parent fp (callsite's FP)
bl prepare_ftrace_return
+SYM_INNER_LABEL(ftrace_graph_caller_entry, SYM_L_GLOBAL)
+ UNWIND_HINT_REGS PT_REGS_SIZE
b ftrace_common_return
SYM_CODE_END(ftrace_graph_caller)
#endif
+/*
+ * ftrace_regs_entry() sets up two stackframes - one for the callsite and
+ * one for the ftrace entry code. Unwind hints have been placed for the
+ * ftrace entry code above. We need an unwind hint for the callsite. Callsites
+ * are numerous. But the unwind hint required for all the callsites is the
+ * same. Define a dummy function here with the callsite unwind hint for the
+ * benefit of the unwinder.
+ */
+SYM_CODE_START(ftrace_callsite)
+ UNWIND_HINT_FTRACE 16 // for the callsite
+ ret
+SYM_CODE_END(ftrace_callsite)
+
#else /* CONFIG_DYNAMIC_FTRACE_WITH_REGS */
/*
diff --git a/arch/arm64/kernel/entry.S b/arch/arm64/kernel/entry.S
index ede028dee81b..95d5f3c08aa1 100644
--- a/arch/arm64/kernel/entry.S
+++ b/arch/arm64/kernel/entry.S
@@ -28,6 +28,7 @@
#include <asm/thread_info.h>
#include <asm/asm-uaccess.h>
#include <asm/unistd.h>
+#include <asm/unwind_hints.h>
.macro clear_gp_regs
.irp n,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29
@@ -560,6 +561,7 @@ SYM_CODE_START_LOCAL(el\el\ht\()_\regsize\()_\label)
.if \el == 0
b ret_to_user
.else
+ UNWIND_HINT_REGS PT_REGS_SIZE
b ret_to_kernel
.endif
SYM_CODE_END(el\el\ht\()_\regsize\()_\label)
@@ -894,6 +896,7 @@ SYM_FUNC_START(call_on_irq_stack)
/* Move to the new stack and call the function there */
mov sp, x16
blr x1
+ UNWIND_HINT_IRQ 16
/*
* Restore the SP from the FP, and restore the FP and LR from the frame
diff --git a/arch/arm64/kernel/probes/kprobes_trampoline.S b/arch/arm64/kernel/probes/kprobes_trampoline.S
index 9a6499bed58b..847cbb81ca33 100644
--- a/arch/arm64/kernel/probes/kprobes_trampoline.S
+++ b/arch/arm64/kernel/probes/kprobes_trampoline.S
@@ -6,6 +6,7 @@
#include <linux/linkage.h>
#include <asm/asm-offsets.h>
#include <asm/assembler.h>
+#include <asm/unwind_hints.h>
.text
@@ -71,6 +72,8 @@ SYM_CODE_START(__kretprobe_trampoline)
mov x0, sp
bl trampoline_probe_handler
+ UNWIND_HINT_REGS PT_REGS_SIZE
+
/*
* Replace trampoline address in lr with actual orig_ret_addr return
* address.
--
2.25.1
next prev parent reply other threads:[~2022-05-24 0:17 UTC|newest]
Thread overview: 40+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <e81e773678f88f7c2ff7480e2eb096973ec198db>
2022-05-24 0:16 ` [RFC PATCH v2 00/20] arm64: livepatch: Use ORC for dynamic frame pointer validation madvenka
2022-05-24 0:16 ` [RFC PATCH v2 01/20] objtool: Reorganize CFI code madvenka
2022-05-24 0:16 ` [RFC PATCH v2 02/20] objtool: Reorganize instruction-related code madvenka
2022-05-24 0:16 ` [RFC PATCH v2 03/20] objtool: Move decode_instructions() to a separate file madvenka
2022-05-24 0:16 ` [RFC PATCH v2 04/20] objtool: Reorganize Unwind hint code madvenka
2022-05-24 0:16 ` [RFC PATCH v2 05/20] objtool: Reorganize ORC types madvenka
2022-05-24 14:27 ` Chen Zhongjin
2022-05-29 15:36 ` Madhavan T. Venkataraman
2022-05-24 0:16 ` [RFC PATCH v2 06/20] objtool: Reorganize ORC code madvenka
2022-05-24 0:16 ` [RFC PATCH v2 07/20] objtool: Reorganize ORC kernel code madvenka
2022-05-24 0:16 ` [RFC PATCH v2 08/20] objtool: arm64: Implement decoder for FP validation madvenka
2022-05-24 0:16 ` [RFC PATCH v2 09/20] objtool: arm64: Implement command to invoke the decoder madvenka
2022-05-24 14:09 ` Mark Brown
2022-05-29 14:49 ` Madhavan T. Venkataraman
2022-05-30 7:51 ` Peter Zijlstra
2022-06-01 22:45 ` Madhavan T. Venkataraman
2022-06-07 18:13 ` Madhavan T. Venkataraman
2022-05-24 0:16 ` [RFC PATCH v2 10/20] objtool: arm64: Compute destinations for call and jump instructions madvenka
2022-05-24 0:16 ` [RFC PATCH v2 11/20] objtool: arm64: Walk instructions and compute CFI for each instruction madvenka
2022-05-24 13:45 ` Chen Zhongjin
2022-05-29 15:18 ` Madhavan T. Venkataraman
2022-05-30 1:44 ` Chen Zhongjin
2022-05-24 0:16 ` [RFC PATCH v2 12/20] objtool: arm64: Generate ORC data from CFI for object files madvenka
2022-05-24 0:16 ` [RFC PATCH v2 13/20] objtool: arm64: Dump ORC data present in " madvenka
2022-05-24 0:16 ` [RFC PATCH v2 14/20] objtool: arm64: Add unwind hint support madvenka
2022-05-24 0:16 ` madvenka [this message]
2022-05-24 0:16 ` [RFC PATCH v2 16/20] arm64: Add kernel and module support for ORC madvenka
2022-05-24 0:16 ` [RFC PATCH v2 17/20] arm64: Build the kernel with ORC information madvenka
2022-05-24 0:16 ` [RFC PATCH v2 18/20] arm64: unwinder: Add a reliability check in the unwinder based on ORC madvenka
2022-05-24 0:16 ` [RFC PATCH v2 19/20] arm64: Miscellaneous changes required for enabling livepatch madvenka
2022-07-01 14:16 ` Miroslav Benes
2022-07-01 19:53 ` Madhavan T. Venkataraman
2022-05-24 0:16 ` [RFC PATCH v2 20/20] arm64: Enable livepatch for ARM64 madvenka
2022-05-24 14:24 ` [RFC PATCH v2 00/20] arm64: livepatch: Use ORC for dynamic frame pointer validation Chen Zhongjin
2022-05-29 15:30 ` Madhavan T. Venkataraman
2022-06-15 12:18 ` Ivan T. Ivanov
2022-06-15 13:37 ` Mark Rutland
2022-06-15 14:18 ` Ivan T. Ivanov
2022-06-15 20:50 ` Madhavan T. Venkataraman
2022-06-15 20:47 ` Madhavan T. Venkataraman
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=20220524001637.1707472-16-madvenka@linux.microsoft.com \
--to=madvenka@linux.microsoft.com \
--cc=broonie@kernel.org \
--cc=catalin.marinas@arm.com \
--cc=chenzhongjin@huawei.com \
--cc=jamorris@linux.microsoft.com \
--cc=jpoimboe@redhat.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=live-patching@vger.kernel.org \
--cc=mark.rutland@arm.com \
--cc=nobuta.keiya@fujitsu.com \
--cc=peterz@infradead.org \
--cc=sjitindarsingh@gmail.com \
--cc=will@kernel.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;
as well as URLs for NNTP newsgroup(s).