From: Tiezhu Yang <yangtiezhu@loongson.cn>
To: Thomas Bogendoerfer <tsbogend@alpha.franken.de>,
Peter Zijlstra <peterz@infradead.org>,
Ingo Molnar <mingo@redhat.com>,
Arnaldo Carvalho de Melo <acme@kernel.org>,
Mark Rutland <mark.rutland@arm.com>,
Alexander Shishkin <alexander.shishkin@linux.intel.com>,
Jiri Olsa <jolsa@redhat.com>, Namhyung Kim <namhyung@kernel.org>
Cc: linux-mips@vger.kernel.org, linux-kernel@vger.kernel.org,
Xuefeng Li <lixuefeng@loongson.cn>,
David Daney <david.daney@cavium.com>,
Ralf Baechle <ralf@linux-mips.org>,
Archer Yan <ayan@wavecomp.com>
Subject: [PATCH 1/3] MIPS: kernel: Support extracting off-line stack traces from user-space with perf
Date: Tue, 29 Dec 2020 20:55:59 +0800 [thread overview]
Message-ID: <1609246561-5474-2-git-send-email-yangtiezhu@loongson.cn> (raw)
In-Reply-To: <1609246561-5474-1-git-send-email-yangtiezhu@loongson.cn>
Add perf_event_mips_regs/perf_reg_value/perf_reg_validate to support
features HAVE_PERF_REGS/HAVE_PERF_USER_STACK_DUMP in kernel.
[ayan@wavecomp.com: Repick this patch for unwinding userstack backtrace
by perf and libunwind on MIPS based CPU.]
[ralf@linux-mips.org: Add perf_get_regs_user() which is required after
'commit 88a7c26af8da ("perf: Move task_pt_regs sampling into arch code")'.]
[yangtiezhu@loongson.cn: Fix build error about perf_get_regs_user() after
commit 76a4efa80900 ("perf/arch: Remove perf_sample_data::regs_user_copy"),
and also separate the original patches into two parts (MIPS kernel and perf
tools) to merge easily.]
The original patches:
https://lore.kernel.org/patchwork/patch/1126521/
https://lore.kernel.org/patchwork/patch/1126520/
Signed-off-by: David Daney <david.daney@cavium.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Archer Yan <ayan@wavecomp.com>
Signed-off-by: Tiezhu Yang <yangtiezhu@loongson.cn>
---
arch/mips/Kconfig | 2 +
arch/mips/include/uapi/asm/perf_regs.h | 42 +++++++++++++++++++++
arch/mips/kernel/Makefile | 2 +-
arch/mips/kernel/perf_regs.c | 68 ++++++++++++++++++++++++++++++++++
4 files changed, 113 insertions(+), 1 deletion(-)
create mode 100644 arch/mips/include/uapi/asm/perf_regs.h
create mode 100644 arch/mips/kernel/perf_regs.c
diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index 0a17bed..092c876 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -76,6 +76,8 @@ config MIPS
select HAVE_NMI
select HAVE_OPROFILE
select HAVE_PERF_EVENTS
+ select HAVE_PERF_REGS
+ select HAVE_PERF_USER_STACK_DUMP
select HAVE_REGS_AND_STACK_ACCESS_API
select HAVE_RSEQ
select HAVE_SPARSE_SYSCALL_NR
diff --git a/arch/mips/include/uapi/asm/perf_regs.h b/arch/mips/include/uapi/asm/perf_regs.h
new file mode 100644
index 0000000..f3cef08
--- /dev/null
+++ b/arch/mips/include/uapi/asm/perf_regs.h
@@ -0,0 +1,42 @@
+/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */
+#ifndef _ASM_MIPS_PERF_REGS_H
+#define _ASM_MIPS_PERF_REGS_H
+
+enum perf_event_mips_regs {
+ PERF_REG_MIPS_PC,
+ PERF_REG_MIPS_R1,
+ PERF_REG_MIPS_R2,
+ PERF_REG_MIPS_R3,
+ PERF_REG_MIPS_R4,
+ PERF_REG_MIPS_R5,
+ PERF_REG_MIPS_R6,
+ PERF_REG_MIPS_R7,
+ PERF_REG_MIPS_R8,
+ PERF_REG_MIPS_R9,
+ PERF_REG_MIPS_R10,
+ PERF_REG_MIPS_R11,
+ PERF_REG_MIPS_R12,
+ PERF_REG_MIPS_R13,
+ PERF_REG_MIPS_R14,
+ PERF_REG_MIPS_R15,
+ PERF_REG_MIPS_R16,
+ PERF_REG_MIPS_R17,
+ PERF_REG_MIPS_R18,
+ PERF_REG_MIPS_R19,
+ PERF_REG_MIPS_R20,
+ PERF_REG_MIPS_R21,
+ PERF_REG_MIPS_R22,
+ PERF_REG_MIPS_R23,
+ PERF_REG_MIPS_R24,
+ PERF_REG_MIPS_R25,
+ /*
+ * 26 and 27 are k0 and k1, they are always clobbered thus not
+ * stored.
+ */
+ PERF_REG_MIPS_R28,
+ PERF_REG_MIPS_R29,
+ PERF_REG_MIPS_R30,
+ PERF_REG_MIPS_R31,
+ PERF_REG_MIPS_MAX = PERF_REG_MIPS_R31 + 1,
+};
+#endif /* _ASM_MIPS_PERF_REGS_H */
diff --git a/arch/mips/kernel/Makefile b/arch/mips/kernel/Makefile
index 2a05b92..120075a 100644
--- a/arch/mips/kernel/Makefile
+++ b/arch/mips/kernel/Makefile
@@ -104,7 +104,7 @@ obj-$(CONFIG_MIPSR2_TO_R6_EMULATOR) += mips-r2-to-r6-emul.o
CFLAGS_cpu-bugs64.o = $(shell if $(CC) $(KBUILD_CFLAGS) -Wa,-mdaddi -c -o /dev/null -x c /dev/null >/dev/null 2>&1; then echo "-DHAVE_AS_SET_DADDI"; fi)
-obj-$(CONFIG_PERF_EVENTS) += perf_event.o
+obj-$(CONFIG_PERF_EVENTS) += perf_event.o perf_regs.o
obj-$(CONFIG_HW_PERF_EVENTS) += perf_event_mipsxx.o
obj-$(CONFIG_JUMP_LABEL) += jump_label.o
diff --git a/arch/mips/kernel/perf_regs.c b/arch/mips/kernel/perf_regs.c
new file mode 100644
index 0000000..e686780
--- /dev/null
+++ b/arch/mips/kernel/perf_regs.c
@@ -0,0 +1,68 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Some parts derived from x86 version of this file.
+ *
+ * Copyright (C) 2013 Cavium, Inc.
+ */
+
+#include <linux/perf_event.h>
+
+#include <asm/ptrace.h>
+
+#ifdef CONFIG_32BIT
+u64 perf_reg_abi(struct task_struct *tsk)
+{
+ return PERF_SAMPLE_REGS_ABI_32;
+}
+#else /* Must be CONFIG_64BIT */
+u64 perf_reg_abi(struct task_struct *tsk)
+{
+ if (test_tsk_thread_flag(tsk, TIF_32BIT_REGS))
+ return PERF_SAMPLE_REGS_ABI_32;
+ else
+ return PERF_SAMPLE_REGS_ABI_64;
+}
+#endif /* CONFIG_32BIT */
+
+int perf_reg_validate(u64 mask)
+{
+ if (!mask)
+ return -EINVAL;
+ if (mask & ~((1ull << PERF_REG_MIPS_MAX) - 1))
+ return -EINVAL;
+ return 0;
+}
+
+u64 perf_reg_value(struct pt_regs *regs, int idx)
+{
+ long v;
+
+ switch (idx) {
+ case PERF_REG_MIPS_PC:
+ v = regs->cp0_epc;
+ break;
+ case PERF_REG_MIPS_R1 ... PERF_REG_MIPS_R25:
+ v = regs->regs[idx - PERF_REG_MIPS_R1 + 1];
+ break;
+ case PERF_REG_MIPS_R28 ... PERF_REG_MIPS_R31:
+ v = regs->regs[idx - PERF_REG_MIPS_R28 + 28];
+ break;
+
+ default:
+ WARN_ON_ONCE(1);
+ return 0;
+ }
+
+ return (s64)v; /* Sign extend if 32-bit. */
+}
+
+void perf_get_regs_user(struct perf_regs *regs_user,
+ struct pt_regs *regs)
+{
+ regs_user->regs = task_pt_regs(current);
+ regs_user->abi = perf_reg_abi(current);
+}
--
2.1.0
next prev parent reply other threads:[~2020-12-29 12:57 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-12-29 12:55 [PATCH 0/3] Add some perf support for mips Tiezhu Yang
2020-12-29 12:55 ` Tiezhu Yang [this message]
2021-01-04 10:59 ` [PATCH 1/3] MIPS: kernel: Support extracting off-line stack traces from user-space with perf Peter Zijlstra
2021-01-05 3:45 ` Jiaxun Yang
2021-01-05 10:18 ` Peter Zijlstra
2021-01-27 21:15 ` Thomas Bogendoerfer
2021-01-29 2:48 ` Tiezhu Yang
2021-01-29 17:56 ` Arnaldo Carvalho de Melo
2021-01-29 17:58 ` Arnaldo Carvalho de Melo
2021-02-01 10:43 ` Thomas Bogendoerfer
2021-02-01 12:56 ` Tiezhu Yang
2021-02-03 10:40 ` Thomas Bogendoerfer
2021-02-03 13:12 ` Tiezhu Yang
2021-02-03 13:41 ` Thomas Bogendoerfer
2020-12-29 12:56 ` [PATCH 2/3] perf tools: Support mips unwinding and dwarf-regs Tiezhu Yang
2020-12-29 12:56 ` [PATCH 3/3] perf tools: Generate mips syscalls_n64.c syscall table Tiezhu Yang
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=1609246561-5474-2-git-send-email-yangtiezhu@loongson.cn \
--to=yangtiezhu@loongson.cn \
--cc=acme@kernel.org \
--cc=alexander.shishkin@linux.intel.com \
--cc=ayan@wavecomp.com \
--cc=david.daney@cavium.com \
--cc=jolsa@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mips@vger.kernel.org \
--cc=lixuefeng@loongson.cn \
--cc=mark.rutland@arm.com \
--cc=mingo@redhat.com \
--cc=namhyung@kernel.org \
--cc=peterz@infradead.org \
--cc=ralf@linux-mips.org \
--cc=tsbogend@alpha.franken.de \
/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).