All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH -next 1/2] riscv: implemented auipc simulate instruction
@ 2021-06-29  2:34 ` Chen Lifu
  0 siblings, 0 replies; 10+ messages in thread
From: Chen Lifu @ 2021-06-29  2:34 UTC (permalink / raw)
  To: paul.walmsley, palmer, aou, guoren, chenlifu, penberg, mhiramat,
	me, linux-riscv, linux-kernel

To test the kprobe-based event tracing, we prepare
a kernel module 'kprobe_test.ko' to add the probes.
The assembly codes (partially) of the module are as follows:
...
0000000000000000 <kprobe_test_branch>:
...
0000000000000038 <.LVL1>:
  38:	00000597          	auipc	a1,0x0
  3c:	00058593          	mv	a1,a1
...

Test the kprobe-based event tracing in qemu-system-riscv64:
First, install the kprobe test module:
insmod /root/kprobe_test.ko

Then, add a probe as a new event at an 'auipc' instruction,
the following error occurs due to the instruction not allowed to probe yet:
echo "p:auipc kprobe_test:kprobe_test_branch+0x38 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
sh: write error: Invalid argument

This patch implemented the 'auipc' simulate instruction and allowed to probe it.
Merge this patch and perform the test again, the test results are as follows:
First, add a probe at the 'auipc' instruction:
echo "p:auipc kprobe_test:kprobe_test_branch+0x38 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
echo 1 > /sys/kernel/debug/tracing/events/kprobes/auipc/enable

Then, do something to run to the probe.
After that, see the traced information:
cat /sys/kernel/debug/tracing/trace
sysctl-58      [001] d...   179.126350: auipc: (kprobe_test_branch+0x38/0x10e [kprobe_test]) epc=0xffffffff016122aa opcode=0x100073

Now we can see the traced information.
The actual address of the symbol 'kprobe_test_branch' is as follows:
cat /proc/kallsyms | grep kprobe_test_branch
ffffffff01612272 t kprobe_test_branch   [kprobe_test]

Based on the traced information and the actual address of the symbol
'kprobe_test_branch', we can also see that the 'auipc' instruction
has been replaced by 'ebreak(0x100073)' instruction.

--------

Signed-off-by: Chen Lifu <chenlifu@huawei.com>
---
 arch/riscv/kernel/probes/decode-insn.c   |  2 +-
 arch/riscv/kernel/probes/simulate-insn.c | 34 ++++++++++++++++++++++++
 2 files changed, 35 insertions(+), 1 deletion(-)

diff --git a/arch/riscv/kernel/probes/decode-insn.c b/arch/riscv/kernel/probes/decode-insn.c
index 0ed043acc882..5eb03fb61450 100644
--- a/arch/riscv/kernel/probes/decode-insn.c
+++ b/arch/riscv/kernel/probes/decode-insn.c
@@ -38,11 +38,11 @@ riscv_probe_decode_insn(probe_opcode_t *addr, struct arch_probe_insn *api)
 	RISCV_INSN_REJECTED(c_ebreak,		insn);
 #endif
 
-	RISCV_INSN_REJECTED(auipc,		insn);
 	RISCV_INSN_REJECTED(branch,		insn);
 
 	RISCV_INSN_SET_SIMULATE(jal,		insn);
 	RISCV_INSN_SET_SIMULATE(jalr,		insn);
+	RISCV_INSN_SET_SIMULATE(auipc,		insn);
 
 	return INSN_GOOD;
 }
diff --git a/arch/riscv/kernel/probes/simulate-insn.c b/arch/riscv/kernel/probes/simulate-insn.c
index 2519ce26377d..b81719522d5c 100644
--- a/arch/riscv/kernel/probes/simulate-insn.c
+++ b/arch/riscv/kernel/probes/simulate-insn.c
@@ -83,3 +83,37 @@ bool __kprobes simulate_jalr(u32 opcode, unsigned long addr, struct pt_regs *reg
 
 	return ret;
 }
+
+#define auipc_rd_idx(opcode) \
+	((opcode >> 7) & 0x1f)
+
+#define auipc_imm(opcode) \
+	((((opcode) >> 12) & 0xfffff) << 12)
+
+#if __riscv_xlen == 64
+#define auipc_offset(opcode)	sign_extend64(auipc_imm(opcode), 31)
+#elif __riscv_xlen == 32
+#define auipc_offset(opcode)	auipc_imm(opcode)
+#else
+#error "Unexpected __riscv_xlen"
+#endif
+
+bool __kprobes simulate_auipc(u32 opcode, unsigned long addr, struct pt_regs *regs)
+{
+	/*
+	 * auipc instruction:
+	 *  31        12 11 7 6      0
+	 * | imm[31:12] | rd | opcode |
+	 *        20       5     7
+	 */
+
+	u32 rd_idx = auipc_rd_idx(opcode);
+	unsigned long rd_val = addr + auipc_offset(opcode);
+
+	if (!rv_insn_reg_set_val(regs, rd_idx, rd_val))
+		return false;
+
+	instruction_pointer_set(regs, addr + 4);
+
+	return true;
+}
-- 
2.32.0


^ permalink raw reply related	[flat|nested] 10+ messages in thread

* [PATCH -next 1/2] riscv: implemented auipc simulate instruction
@ 2021-06-29  2:34 ` Chen Lifu
  0 siblings, 0 replies; 10+ messages in thread
From: Chen Lifu @ 2021-06-29  2:34 UTC (permalink / raw)
  To: paul.walmsley, palmer, aou, guoren, chenlifu, penberg, mhiramat,
	me, linux-riscv, linux-kernel

To test the kprobe-based event tracing, we prepare
a kernel module 'kprobe_test.ko' to add the probes.
The assembly codes (partially) of the module are as follows:
...
0000000000000000 <kprobe_test_branch>:
...
0000000000000038 <.LVL1>:
  38:	00000597          	auipc	a1,0x0
  3c:	00058593          	mv	a1,a1
...

Test the kprobe-based event tracing in qemu-system-riscv64:
First, install the kprobe test module:
insmod /root/kprobe_test.ko

Then, add a probe as a new event at an 'auipc' instruction,
the following error occurs due to the instruction not allowed to probe yet:
echo "p:auipc kprobe_test:kprobe_test_branch+0x38 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
sh: write error: Invalid argument

This patch implemented the 'auipc' simulate instruction and allowed to probe it.
Merge this patch and perform the test again, the test results are as follows:
First, add a probe at the 'auipc' instruction:
echo "p:auipc kprobe_test:kprobe_test_branch+0x38 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
echo 1 > /sys/kernel/debug/tracing/events/kprobes/auipc/enable

Then, do something to run to the probe.
After that, see the traced information:
cat /sys/kernel/debug/tracing/trace
sysctl-58      [001] d...   179.126350: auipc: (kprobe_test_branch+0x38/0x10e [kprobe_test]) epc=0xffffffff016122aa opcode=0x100073

Now we can see the traced information.
The actual address of the symbol 'kprobe_test_branch' is as follows:
cat /proc/kallsyms | grep kprobe_test_branch
ffffffff01612272 t kprobe_test_branch   [kprobe_test]

Based on the traced information and the actual address of the symbol
'kprobe_test_branch', we can also see that the 'auipc' instruction
has been replaced by 'ebreak(0x100073)' instruction.

--------

Signed-off-by: Chen Lifu <chenlifu@huawei.com>
---
 arch/riscv/kernel/probes/decode-insn.c   |  2 +-
 arch/riscv/kernel/probes/simulate-insn.c | 34 ++++++++++++++++++++++++
 2 files changed, 35 insertions(+), 1 deletion(-)

diff --git a/arch/riscv/kernel/probes/decode-insn.c b/arch/riscv/kernel/probes/decode-insn.c
index 0ed043acc882..5eb03fb61450 100644
--- a/arch/riscv/kernel/probes/decode-insn.c
+++ b/arch/riscv/kernel/probes/decode-insn.c
@@ -38,11 +38,11 @@ riscv_probe_decode_insn(probe_opcode_t *addr, struct arch_probe_insn *api)
 	RISCV_INSN_REJECTED(c_ebreak,		insn);
 #endif
 
-	RISCV_INSN_REJECTED(auipc,		insn);
 	RISCV_INSN_REJECTED(branch,		insn);
 
 	RISCV_INSN_SET_SIMULATE(jal,		insn);
 	RISCV_INSN_SET_SIMULATE(jalr,		insn);
+	RISCV_INSN_SET_SIMULATE(auipc,		insn);
 
 	return INSN_GOOD;
 }
diff --git a/arch/riscv/kernel/probes/simulate-insn.c b/arch/riscv/kernel/probes/simulate-insn.c
index 2519ce26377d..b81719522d5c 100644
--- a/arch/riscv/kernel/probes/simulate-insn.c
+++ b/arch/riscv/kernel/probes/simulate-insn.c
@@ -83,3 +83,37 @@ bool __kprobes simulate_jalr(u32 opcode, unsigned long addr, struct pt_regs *reg
 
 	return ret;
 }
+
+#define auipc_rd_idx(opcode) \
+	((opcode >> 7) & 0x1f)
+
+#define auipc_imm(opcode) \
+	((((opcode) >> 12) & 0xfffff) << 12)
+
+#if __riscv_xlen == 64
+#define auipc_offset(opcode)	sign_extend64(auipc_imm(opcode), 31)
+#elif __riscv_xlen == 32
+#define auipc_offset(opcode)	auipc_imm(opcode)
+#else
+#error "Unexpected __riscv_xlen"
+#endif
+
+bool __kprobes simulate_auipc(u32 opcode, unsigned long addr, struct pt_regs *regs)
+{
+	/*
+	 * auipc instruction:
+	 *  31        12 11 7 6      0
+	 * | imm[31:12] | rd | opcode |
+	 *        20       5     7
+	 */
+
+	u32 rd_idx = auipc_rd_idx(opcode);
+	unsigned long rd_val = addr + auipc_offset(opcode);
+
+	if (!rv_insn_reg_set_val(regs, rd_idx, rd_val))
+		return false;
+
+	instruction_pointer_set(regs, addr + 4);
+
+	return true;
+}
-- 
2.32.0


_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

^ permalink raw reply related	[flat|nested] 10+ messages in thread

* [PATCH -next 2/2] riscv: implemented branch simulate instructions
  2021-06-29  2:34 ` Chen Lifu
@ 2021-06-29  2:34   ` Chen Lifu
  -1 siblings, 0 replies; 10+ messages in thread
From: Chen Lifu @ 2021-06-29  2:34 UTC (permalink / raw)
  To: paul.walmsley, palmer, aou, guoren, chenlifu, penberg, mhiramat,
	me, linux-riscv, linux-kernel

To test the kprobe-based event tracing, we prepare
a kernel module 'kprobe_test.ko' to add the probes.
The assembly codes (partially) of the module are as follows:
...
0000000000000000 <kprobe_test_branch>:
...

0000000000000080 <.L7>:
  80:	414986bb          	subw	a3,s3,s4
  84:	4785                	li	a5,1
  86:	00d7c363          	blt	a5,a3,8c <.L8>
  8a:	2485                	addiw	s1,s1,1

000000000000008c <.L8>:
  8c:	012987bb          	addw	a5,s3,s2
  90:	4711                	li	a4,4
  92:	00f75363          	bge	a4,a5,98 <.L9>
  96:	2485                	addiw	s1,s1,1
...

00000000000000aa <.L11>:
  aa:	01299363          	bne	s3,s2,b0 <.L12>
  ae:	2485                	addiw	s1,s1,1

00000000000000b0 <.L12>:
  b0:	012a0363          	beq	s4,s2,b6 <.L13>
  b4:	2485                	addiw	s1,s1,1
...

00000000000000c2 <.L14>:
  c2:	0009861b          	sext.w	a2,s3
  c6:	013a7363          	bgeu	s4,s3,cc <.L15>
  ca:	2485                	addiw	s1,s1,1
...

00000000000000d2 <.L16>:
  d2:	00e7e363          	bltu	a5,a4,d8 <.L17>
  d6:	2485                	addiw	s1,s1,1
...

Test the kprobe-based event tracing in qemu-system-riscv64:
First, install the kprobe test module:
insmod /root/kprobe_test.ko

Then, add probes as new events at the branch instructions,
i.e. blt, bge, bne, beq, bgeu and bltu.
The following errors occur due to the instructions not allowed to probe yet:
echo "p:blt kprobe_test:kprobe_test_branch+0x86 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
sh: write error: Invalid argument
echo "p:bge kprobe_test:kprobe_test_branch+0x92 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
sh: write error: Invalid argument
echo "p:bne kprobe_test:kprobe_test_branch+0xaa epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
sh: write error: Invalid argument
echo "p:beq kprobe_test:kprobe_test_branch+0xb0 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
sh: write error: Invalid argument
echo "p:bgeu kprobe_test:kprobe_test_branch+0xc6 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
sh: write error: Invalid argument
echo "p:bltu kprobe_test:kprobe_test_branch+0xd2 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
sh: write error: Invalid argument

This patch implemented the branch simulate instructions and allowed to probe them.
Merge this patch and perform the test again, the test results are as follows:
First, add probes at the branch instructions:
echo "p:blt kprobe_test:kprobe_test_branch+0x86 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
echo "p:bge kprobe_test:kprobe_test_branch+0x92 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
echo "p:bne kprobe_test:kprobe_test_branch+0xaa epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
echo "p:beq kprobe_test:kprobe_test_branch+0xb0 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
echo "p:bgeu kprobe_test:kprobe_test_branch+0xc6 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
echo "p:bltu kprobe_test:kprobe_test_branch+0xd2 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
echo 1 > /sys/kernel/debug/tracing/events/kprobes/blt/enable
echo 1 > /sys/kernel/debug/tracing/events/kprobes/bge/enable
echo 1 > /sys/kernel/debug/tracing/events/kprobes/bne/enable
echo 1 > /sys/kernel/debug/tracing/events/kprobes/beq/enable
echo 1 > /sys/kernel/debug/tracing/events/kprobes/bgeu/enable
echo 1 > /sys/kernel/debug/tracing/events/kprobes/bltu/enable

Then, do something to run to the probes.
After that, see the traced information:
cat /sys/kernel/debug/tracing/trace
sysctl-63      [001] d...  2505.263969: blt: (kprobe_test_branch+0x86/0x10e [kprobe_test]) epc=0xffffffff016122f8 opcode=0x100073
sysctl-63      [001] d...  2505.263991: bge: (kprobe_test_branch+0x92/0x10e [kprobe_test]) epc=0xffffffff01612304 opcode=0x100073
sysctl-63      [001] d...  2505.264001: bne: (kprobe_test_branch+0xaa/0x10e [kprobe_test]) epc=0xffffffff0161231c opcode=0x100073
sysctl-63      [001] d...  2505.264011: beq: (kprobe_test_branch+0xb0/0x10e [kprobe_test]) epc=0xffffffff01612322 opcode=0x100073
sysctl-63      [001] d...  2505.264019: bgeu: (kprobe_test_branch+0xc6/0x10e [kprobe_test]) epc=0xffffffff01612338 opcode=0x100073
sysctl-63      [001] d...  2505.264027: bltu: (kprobe_test_branch+0xd2/0x10e [kprobe_test]) epc=0xffffffff01612344 opcode=0x100073

Now we can see the traced information.
The actual address of the symbol 'kprobe_test_branch' is as follows:
cat /proc/kallsyms | grep kprobe_test_branch
ffffffff01612272 t kprobe_test_branch   [kprobe_test]

Based on the traced information and the actual address of the symbol
'kprobe_test_branch', we can also see that the branch instructions
have been replaced by 'ebreak(0x100073)' instructions.

The pesudoinstructions of the branch instructions,
i.e. bnez, beqz, blez, bgez, bltz, bgtz, bleu, bgtu and ble
are allowed to probe as well.

--------

Signed-off-by: Chen Lifu <chenlifu@huawei.com>
---
 arch/riscv/kernel/probes/decode-insn.c   |  3 +-
 arch/riscv/kernel/probes/simulate-insn.c | 78 ++++++++++++++++++++++++
 2 files changed, 79 insertions(+), 2 deletions(-)

diff --git a/arch/riscv/kernel/probes/decode-insn.c b/arch/riscv/kernel/probes/decode-insn.c
index 5eb03fb61450..64f6183b4717 100644
--- a/arch/riscv/kernel/probes/decode-insn.c
+++ b/arch/riscv/kernel/probes/decode-insn.c
@@ -38,11 +38,10 @@ riscv_probe_decode_insn(probe_opcode_t *addr, struct arch_probe_insn *api)
 	RISCV_INSN_REJECTED(c_ebreak,		insn);
 #endif
 
-	RISCV_INSN_REJECTED(branch,		insn);
-
 	RISCV_INSN_SET_SIMULATE(jal,		insn);
 	RISCV_INSN_SET_SIMULATE(jalr,		insn);
 	RISCV_INSN_SET_SIMULATE(auipc,		insn);
+	RISCV_INSN_SET_SIMULATE(branch,		insn);
 
 	return INSN_GOOD;
 }
diff --git a/arch/riscv/kernel/probes/simulate-insn.c b/arch/riscv/kernel/probes/simulate-insn.c
index b81719522d5c..efc989b28859 100644
--- a/arch/riscv/kernel/probes/simulate-insn.c
+++ b/arch/riscv/kernel/probes/simulate-insn.c
@@ -117,3 +117,81 @@ bool __kprobes simulate_auipc(u32 opcode, unsigned long addr, struct pt_regs *re
 
 	return true;
 }
+
+#define branch_rs1_idx(opcode) \
+	(((opcode) >> 15) & 0x1f)
+
+#define branch_rs2_idx(opcode) \
+	(((opcode) >> 20) & 0x1f)
+
+#define branch_funct3(opcode) \
+	(((opcode) >> 12) & 0x7)
+
+#define branch_imm(opcode) \
+	(((((opcode) >> 8 ) & 0xf ) << 1 ) | \
+	 ((((opcode) >> 25) & 0x3f) << 5 ) | \
+	 ((((opcode) >> 7 ) & 0x1 ) << 11) | \
+	 ((((opcode) >> 31) & 0x1 ) << 12))
+
+#define branch_offset(opcode) \
+	sign_extend32((branch_imm(opcode)), 12)
+
+#define BRANCH_BEQ	0x0
+#define BRANCH_BNE	0x1
+#define BRANCH_BLT	0x4
+#define BRANCH_BGE	0x5
+#define BRANCH_BLTU	0x6
+#define BRANCH_BGEU	0x7
+
+bool __kprobes simulate_branch(u32 opcode, unsigned long addr, struct pt_regs *regs)
+{
+	/*
+	 * branch instructions:
+	 *      31    30       25 24 20 19 15 14    12 11       8    7      6      0
+	 * | imm[12] | imm[10:5] | rs2 | rs1 | funct3 | imm[4:1] | imm[11] | opcode |
+	 *     1           6        5     5      3         4         1         7
+	 *     imm[12|10:5]        rs2   rs1    000       imm[4:1|11]       1100011  BEQ
+	 *     imm[12|10:5]        rs2   rs1    001       imm[4:1|11]       1100011  BNE
+	 *     imm[12|10:5]        rs2   rs1    100       imm[4:1|11]       1100011  BLT
+	 *     imm[12|10:5]        rs2   rs1    101       imm[4:1|11]       1100011  BGE
+	 *     imm[12|10:5]        rs2   rs1    110       imm[4:1|11]       1100011  BLTU
+	 *     imm[12|10:5]        rs2   rs1    111       imm[4:1|11]       1100011  BGEU
+	 */
+
+	s32 offset;
+	s32 offset_tmp;
+	unsigned long rs1_val;
+	unsigned long rs2_val;
+
+	if (!rv_insn_reg_get_val(regs, branch_rs1_idx(opcode), &rs1_val) ||
+	    !rv_insn_reg_get_val(regs, branch_rs2_idx(opcode), &rs2_val))
+		return false;
+
+	offset_tmp = branch_offset(opcode);
+	switch (branch_funct3(opcode)) {
+	case BRANCH_BEQ:
+		offset = (rs1_val == rs2_val) ? offset_tmp : 4;
+		break;
+	case BRANCH_BNE:
+		offset = (rs1_val != rs2_val) ? offset_tmp : 4;
+		break;
+	case BRANCH_BLT:
+		offset = ((long)rs1_val < (long)rs2_val) ? offset_tmp : 4;
+		break;
+	case BRANCH_BGE:
+		offset = ((long)rs1_val >= (long)rs2_val) ? offset_tmp : 4;
+		break;
+	case BRANCH_BLTU:
+		offset = (rs1_val < rs2_val) ? offset_tmp : 4;
+		break;
+	case BRANCH_BGEU:
+		offset = (rs1_val >= rs2_val) ? offset_tmp : 4;
+		break;
+	default:
+		return false;
+	}
+
+	instruction_pointer_set(regs, addr + offset);
+
+	return true;
+}
-- 
2.32.0


^ permalink raw reply related	[flat|nested] 10+ messages in thread

* [PATCH -next 2/2] riscv: implemented branch simulate instructions
@ 2021-06-29  2:34   ` Chen Lifu
  0 siblings, 0 replies; 10+ messages in thread
From: Chen Lifu @ 2021-06-29  2:34 UTC (permalink / raw)
  To: paul.walmsley, palmer, aou, guoren, chenlifu, penberg, mhiramat,
	me, linux-riscv, linux-kernel

To test the kprobe-based event tracing, we prepare
a kernel module 'kprobe_test.ko' to add the probes.
The assembly codes (partially) of the module are as follows:
...
0000000000000000 <kprobe_test_branch>:
...

0000000000000080 <.L7>:
  80:	414986bb          	subw	a3,s3,s4
  84:	4785                	li	a5,1
  86:	00d7c363          	blt	a5,a3,8c <.L8>
  8a:	2485                	addiw	s1,s1,1

000000000000008c <.L8>:
  8c:	012987bb          	addw	a5,s3,s2
  90:	4711                	li	a4,4
  92:	00f75363          	bge	a4,a5,98 <.L9>
  96:	2485                	addiw	s1,s1,1
...

00000000000000aa <.L11>:
  aa:	01299363          	bne	s3,s2,b0 <.L12>
  ae:	2485                	addiw	s1,s1,1

00000000000000b0 <.L12>:
  b0:	012a0363          	beq	s4,s2,b6 <.L13>
  b4:	2485                	addiw	s1,s1,1
...

00000000000000c2 <.L14>:
  c2:	0009861b          	sext.w	a2,s3
  c6:	013a7363          	bgeu	s4,s3,cc <.L15>
  ca:	2485                	addiw	s1,s1,1
...

00000000000000d2 <.L16>:
  d2:	00e7e363          	bltu	a5,a4,d8 <.L17>
  d6:	2485                	addiw	s1,s1,1
...

Test the kprobe-based event tracing in qemu-system-riscv64:
First, install the kprobe test module:
insmod /root/kprobe_test.ko

Then, add probes as new events at the branch instructions,
i.e. blt, bge, bne, beq, bgeu and bltu.
The following errors occur due to the instructions not allowed to probe yet:
echo "p:blt kprobe_test:kprobe_test_branch+0x86 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
sh: write error: Invalid argument
echo "p:bge kprobe_test:kprobe_test_branch+0x92 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
sh: write error: Invalid argument
echo "p:bne kprobe_test:kprobe_test_branch+0xaa epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
sh: write error: Invalid argument
echo "p:beq kprobe_test:kprobe_test_branch+0xb0 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
sh: write error: Invalid argument
echo "p:bgeu kprobe_test:kprobe_test_branch+0xc6 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
sh: write error: Invalid argument
echo "p:bltu kprobe_test:kprobe_test_branch+0xd2 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
sh: write error: Invalid argument

This patch implemented the branch simulate instructions and allowed to probe them.
Merge this patch and perform the test again, the test results are as follows:
First, add probes at the branch instructions:
echo "p:blt kprobe_test:kprobe_test_branch+0x86 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
echo "p:bge kprobe_test:kprobe_test_branch+0x92 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
echo "p:bne kprobe_test:kprobe_test_branch+0xaa epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
echo "p:beq kprobe_test:kprobe_test_branch+0xb0 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
echo "p:bgeu kprobe_test:kprobe_test_branch+0xc6 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
echo "p:bltu kprobe_test:kprobe_test_branch+0xd2 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
echo 1 > /sys/kernel/debug/tracing/events/kprobes/blt/enable
echo 1 > /sys/kernel/debug/tracing/events/kprobes/bge/enable
echo 1 > /sys/kernel/debug/tracing/events/kprobes/bne/enable
echo 1 > /sys/kernel/debug/tracing/events/kprobes/beq/enable
echo 1 > /sys/kernel/debug/tracing/events/kprobes/bgeu/enable
echo 1 > /sys/kernel/debug/tracing/events/kprobes/bltu/enable

Then, do something to run to the probes.
After that, see the traced information:
cat /sys/kernel/debug/tracing/trace
sysctl-63      [001] d...  2505.263969: blt: (kprobe_test_branch+0x86/0x10e [kprobe_test]) epc=0xffffffff016122f8 opcode=0x100073
sysctl-63      [001] d...  2505.263991: bge: (kprobe_test_branch+0x92/0x10e [kprobe_test]) epc=0xffffffff01612304 opcode=0x100073
sysctl-63      [001] d...  2505.264001: bne: (kprobe_test_branch+0xaa/0x10e [kprobe_test]) epc=0xffffffff0161231c opcode=0x100073
sysctl-63      [001] d...  2505.264011: beq: (kprobe_test_branch+0xb0/0x10e [kprobe_test]) epc=0xffffffff01612322 opcode=0x100073
sysctl-63      [001] d...  2505.264019: bgeu: (kprobe_test_branch+0xc6/0x10e [kprobe_test]) epc=0xffffffff01612338 opcode=0x100073
sysctl-63      [001] d...  2505.264027: bltu: (kprobe_test_branch+0xd2/0x10e [kprobe_test]) epc=0xffffffff01612344 opcode=0x100073

Now we can see the traced information.
The actual address of the symbol 'kprobe_test_branch' is as follows:
cat /proc/kallsyms | grep kprobe_test_branch
ffffffff01612272 t kprobe_test_branch   [kprobe_test]

Based on the traced information and the actual address of the symbol
'kprobe_test_branch', we can also see that the branch instructions
have been replaced by 'ebreak(0x100073)' instructions.

The pesudoinstructions of the branch instructions,
i.e. bnez, beqz, blez, bgez, bltz, bgtz, bleu, bgtu and ble
are allowed to probe as well.

--------

Signed-off-by: Chen Lifu <chenlifu@huawei.com>
---
 arch/riscv/kernel/probes/decode-insn.c   |  3 +-
 arch/riscv/kernel/probes/simulate-insn.c | 78 ++++++++++++++++++++++++
 2 files changed, 79 insertions(+), 2 deletions(-)

diff --git a/arch/riscv/kernel/probes/decode-insn.c b/arch/riscv/kernel/probes/decode-insn.c
index 5eb03fb61450..64f6183b4717 100644
--- a/arch/riscv/kernel/probes/decode-insn.c
+++ b/arch/riscv/kernel/probes/decode-insn.c
@@ -38,11 +38,10 @@ riscv_probe_decode_insn(probe_opcode_t *addr, struct arch_probe_insn *api)
 	RISCV_INSN_REJECTED(c_ebreak,		insn);
 #endif
 
-	RISCV_INSN_REJECTED(branch,		insn);
-
 	RISCV_INSN_SET_SIMULATE(jal,		insn);
 	RISCV_INSN_SET_SIMULATE(jalr,		insn);
 	RISCV_INSN_SET_SIMULATE(auipc,		insn);
+	RISCV_INSN_SET_SIMULATE(branch,		insn);
 
 	return INSN_GOOD;
 }
diff --git a/arch/riscv/kernel/probes/simulate-insn.c b/arch/riscv/kernel/probes/simulate-insn.c
index b81719522d5c..efc989b28859 100644
--- a/arch/riscv/kernel/probes/simulate-insn.c
+++ b/arch/riscv/kernel/probes/simulate-insn.c
@@ -117,3 +117,81 @@ bool __kprobes simulate_auipc(u32 opcode, unsigned long addr, struct pt_regs *re
 
 	return true;
 }
+
+#define branch_rs1_idx(opcode) \
+	(((opcode) >> 15) & 0x1f)
+
+#define branch_rs2_idx(opcode) \
+	(((opcode) >> 20) & 0x1f)
+
+#define branch_funct3(opcode) \
+	(((opcode) >> 12) & 0x7)
+
+#define branch_imm(opcode) \
+	(((((opcode) >> 8 ) & 0xf ) << 1 ) | \
+	 ((((opcode) >> 25) & 0x3f) << 5 ) | \
+	 ((((opcode) >> 7 ) & 0x1 ) << 11) | \
+	 ((((opcode) >> 31) & 0x1 ) << 12))
+
+#define branch_offset(opcode) \
+	sign_extend32((branch_imm(opcode)), 12)
+
+#define BRANCH_BEQ	0x0
+#define BRANCH_BNE	0x1
+#define BRANCH_BLT	0x4
+#define BRANCH_BGE	0x5
+#define BRANCH_BLTU	0x6
+#define BRANCH_BGEU	0x7
+
+bool __kprobes simulate_branch(u32 opcode, unsigned long addr, struct pt_regs *regs)
+{
+	/*
+	 * branch instructions:
+	 *      31    30       25 24 20 19 15 14    12 11       8    7      6      0
+	 * | imm[12] | imm[10:5] | rs2 | rs1 | funct3 | imm[4:1] | imm[11] | opcode |
+	 *     1           6        5     5      3         4         1         7
+	 *     imm[12|10:5]        rs2   rs1    000       imm[4:1|11]       1100011  BEQ
+	 *     imm[12|10:5]        rs2   rs1    001       imm[4:1|11]       1100011  BNE
+	 *     imm[12|10:5]        rs2   rs1    100       imm[4:1|11]       1100011  BLT
+	 *     imm[12|10:5]        rs2   rs1    101       imm[4:1|11]       1100011  BGE
+	 *     imm[12|10:5]        rs2   rs1    110       imm[4:1|11]       1100011  BLTU
+	 *     imm[12|10:5]        rs2   rs1    111       imm[4:1|11]       1100011  BGEU
+	 */
+
+	s32 offset;
+	s32 offset_tmp;
+	unsigned long rs1_val;
+	unsigned long rs2_val;
+
+	if (!rv_insn_reg_get_val(regs, branch_rs1_idx(opcode), &rs1_val) ||
+	    !rv_insn_reg_get_val(regs, branch_rs2_idx(opcode), &rs2_val))
+		return false;
+
+	offset_tmp = branch_offset(opcode);
+	switch (branch_funct3(opcode)) {
+	case BRANCH_BEQ:
+		offset = (rs1_val == rs2_val) ? offset_tmp : 4;
+		break;
+	case BRANCH_BNE:
+		offset = (rs1_val != rs2_val) ? offset_tmp : 4;
+		break;
+	case BRANCH_BLT:
+		offset = ((long)rs1_val < (long)rs2_val) ? offset_tmp : 4;
+		break;
+	case BRANCH_BGE:
+		offset = ((long)rs1_val >= (long)rs2_val) ? offset_tmp : 4;
+		break;
+	case BRANCH_BLTU:
+		offset = (rs1_val < rs2_val) ? offset_tmp : 4;
+		break;
+	case BRANCH_BGEU:
+		offset = (rs1_val >= rs2_val) ? offset_tmp : 4;
+		break;
+	default:
+		return false;
+	}
+
+	instruction_pointer_set(regs, addr + offset);
+
+	return true;
+}
-- 
2.32.0


_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

^ permalink raw reply related	[flat|nested] 10+ messages in thread

* ping//Re: [PATCH -next 2/2] riscv: implemented branch simulate instructions
  2021-06-29  2:34   ` Chen Lifu
@ 2021-07-05  2:09     ` chenlifu
  -1 siblings, 0 replies; 10+ messages in thread
From: chenlifu @ 2021-07-05  2:09 UTC (permalink / raw)
  To: paul.walmsley, palmer, aou, guoren, penberg, mhiramat, me,
	linux-riscv, linux-kernel



在 2021/6/29 10:34, Chen Lifu 写道:
> To test the kprobe-based event tracing, we prepare
> a kernel module 'kprobe_test.ko' to add the probes.
> The assembly codes (partially) of the module are as follows:
> ...
> 0000000000000000 <kprobe_test_branch>:
> ...
> 
> 0000000000000080 <.L7>:
>    80:	414986bb          	subw	a3,s3,s4
>    84:	4785                	li	a5,1
>    86:	00d7c363          	blt	a5,a3,8c <.L8>
>    8a:	2485                	addiw	s1,s1,1
> 
> 000000000000008c <.L8>:
>    8c:	012987bb          	addw	a5,s3,s2
>    90:	4711                	li	a4,4
>    92:	00f75363          	bge	a4,a5,98 <.L9>
>    96:	2485                	addiw	s1,s1,1
> ...
> 
> 00000000000000aa <.L11>:
>    aa:	01299363          	bne	s3,s2,b0 <.L12>
>    ae:	2485                	addiw	s1,s1,1
> 
> 00000000000000b0 <.L12>:
>    b0:	012a0363          	beq	s4,s2,b6 <.L13>
>    b4:	2485                	addiw	s1,s1,1
> ...
> 
> 00000000000000c2 <.L14>:
>    c2:	0009861b          	sext.w	a2,s3
>    c6:	013a7363          	bgeu	s4,s3,cc <.L15>
>    ca:	2485                	addiw	s1,s1,1
> ...
> 
> 00000000000000d2 <.L16>:
>    d2:	00e7e363          	bltu	a5,a4,d8 <.L17>
>    d6:	2485                	addiw	s1,s1,1
> ...
> 
> Test the kprobe-based event tracing in qemu-system-riscv64:
> First, install the kprobe test module:
> insmod /root/kprobe_test.ko
> 
> Then, add probes as new events at the branch instructions,
> i.e. blt, bge, bne, beq, bgeu and bltu.
> The following errors occur due to the instructions not allowed to probe yet:
> echo "p:blt kprobe_test:kprobe_test_branch+0x86 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
> sh: write error: Invalid argument
> echo "p:bge kprobe_test:kprobe_test_branch+0x92 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
> sh: write error: Invalid argument
> echo "p:bne kprobe_test:kprobe_test_branch+0xaa epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
> sh: write error: Invalid argument
> echo "p:beq kprobe_test:kprobe_test_branch+0xb0 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
> sh: write error: Invalid argument
> echo "p:bgeu kprobe_test:kprobe_test_branch+0xc6 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
> sh: write error: Invalid argument
> echo "p:bltu kprobe_test:kprobe_test_branch+0xd2 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
> sh: write error: Invalid argument
> 
> This patch implemented the branch simulate instructions and allowed to probe them.
> Merge this patch and perform the test again, the test results are as follows:
> First, add probes at the branch instructions:
> echo "p:blt kprobe_test:kprobe_test_branch+0x86 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
> echo "p:bge kprobe_test:kprobe_test_branch+0x92 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
> echo "p:bne kprobe_test:kprobe_test_branch+0xaa epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
> echo "p:beq kprobe_test:kprobe_test_branch+0xb0 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
> echo "p:bgeu kprobe_test:kprobe_test_branch+0xc6 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
> echo "p:bltu kprobe_test:kprobe_test_branch+0xd2 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
> echo 1 > /sys/kernel/debug/tracing/events/kprobes/blt/enable
> echo 1 > /sys/kernel/debug/tracing/events/kprobes/bge/enable
> echo 1 > /sys/kernel/debug/tracing/events/kprobes/bne/enable
> echo 1 > /sys/kernel/debug/tracing/events/kprobes/beq/enable
> echo 1 > /sys/kernel/debug/tracing/events/kprobes/bgeu/enable
> echo 1 > /sys/kernel/debug/tracing/events/kprobes/bltu/enable
> 
> Then, do something to run to the probes.
> After that, see the traced information:
> cat /sys/kernel/debug/tracing/trace
> sysctl-63      [001] d...  2505.263969: blt: (kprobe_test_branch+0x86/0x10e [kprobe_test]) epc=0xffffffff016122f8 opcode=0x100073
> sysctl-63      [001] d...  2505.263991: bge: (kprobe_test_branch+0x92/0x10e [kprobe_test]) epc=0xffffffff01612304 opcode=0x100073
> sysctl-63      [001] d...  2505.264001: bne: (kprobe_test_branch+0xaa/0x10e [kprobe_test]) epc=0xffffffff0161231c opcode=0x100073
> sysctl-63      [001] d...  2505.264011: beq: (kprobe_test_branch+0xb0/0x10e [kprobe_test]) epc=0xffffffff01612322 opcode=0x100073
> sysctl-63      [001] d...  2505.264019: bgeu: (kprobe_test_branch+0xc6/0x10e [kprobe_test]) epc=0xffffffff01612338 opcode=0x100073
> sysctl-63      [001] d...  2505.264027: bltu: (kprobe_test_branch+0xd2/0x10e [kprobe_test]) epc=0xffffffff01612344 opcode=0x100073
> 
> Now we can see the traced information.
> The actual address of the symbol 'kprobe_test_branch' is as follows:
> cat /proc/kallsyms | grep kprobe_test_branch
> ffffffff01612272 t kprobe_test_branch   [kprobe_test]
> 
> Based on the traced information and the actual address of the symbol
> 'kprobe_test_branch', we can also see that the branch instructions
> have been replaced by 'ebreak(0x100073)' instructions.
> 
> The pesudoinstructions of the branch instructions,
> i.e. bnez, beqz, blez, bgez, bltz, bgtz, bleu, bgtu and ble
> are allowed to probe as well.
> 
> --------
> 
> Signed-off-by: Chen Lifu <chenlifu@huawei.com>
> ---
>   arch/riscv/kernel/probes/decode-insn.c   |  3 +-
>   arch/riscv/kernel/probes/simulate-insn.c | 78 ++++++++++++++++++++++++
>   2 files changed, 79 insertions(+), 2 deletions(-)
> 
> diff --git a/arch/riscv/kernel/probes/decode-insn.c b/arch/riscv/kernel/probes/decode-insn.c
> index 5eb03fb61450..64f6183b4717 100644
> --- a/arch/riscv/kernel/probes/decode-insn.c
> +++ b/arch/riscv/kernel/probes/decode-insn.c
> @@ -38,11 +38,10 @@ riscv_probe_decode_insn(probe_opcode_t *addr, struct arch_probe_insn *api)
>   	RISCV_INSN_REJECTED(c_ebreak,		insn);
>   #endif
>   
> -	RISCV_INSN_REJECTED(branch,		insn);
> -
>   	RISCV_INSN_SET_SIMULATE(jal,		insn);
>   	RISCV_INSN_SET_SIMULATE(jalr,		insn);
>   	RISCV_INSN_SET_SIMULATE(auipc,		insn);
> +	RISCV_INSN_SET_SIMULATE(branch,		insn);
>   
>   	return INSN_GOOD;
>   }
> diff --git a/arch/riscv/kernel/probes/simulate-insn.c b/arch/riscv/kernel/probes/simulate-insn.c
> index b81719522d5c..efc989b28859 100644
> --- a/arch/riscv/kernel/probes/simulate-insn.c
> +++ b/arch/riscv/kernel/probes/simulate-insn.c
> @@ -117,3 +117,81 @@ bool __kprobes simulate_auipc(u32 opcode, unsigned long addr, struct pt_regs *re
>   
>   	return true;
>   }
> +
> +#define branch_rs1_idx(opcode) \
> +	(((opcode) >> 15) & 0x1f)
> +
> +#define branch_rs2_idx(opcode) \
> +	(((opcode) >> 20) & 0x1f)
> +
> +#define branch_funct3(opcode) \
> +	(((opcode) >> 12) & 0x7)
> +
> +#define branch_imm(opcode) \
> +	(((((opcode) >> 8 ) & 0xf ) << 1 ) | \
> +	 ((((opcode) >> 25) & 0x3f) << 5 ) | \
> +	 ((((opcode) >> 7 ) & 0x1 ) << 11) | \
> +	 ((((opcode) >> 31) & 0x1 ) << 12))
> +
> +#define branch_offset(opcode) \
> +	sign_extend32((branch_imm(opcode)), 12)
> +
> +#define BRANCH_BEQ	0x0
> +#define BRANCH_BNE	0x1
> +#define BRANCH_BLT	0x4
> +#define BRANCH_BGE	0x5
> +#define BRANCH_BLTU	0x6
> +#define BRANCH_BGEU	0x7
> +
> +bool __kprobes simulate_branch(u32 opcode, unsigned long addr, struct pt_regs *regs)
> +{
> +	/*
> +	 * branch instructions:
> +	 *      31    30       25 24 20 19 15 14    12 11       8    7      6      0
> +	 * | imm[12] | imm[10:5] | rs2 | rs1 | funct3 | imm[4:1] | imm[11] | opcode |
> +	 *     1           6        5     5      3         4         1         7
> +	 *     imm[12|10:5]        rs2   rs1    000       imm[4:1|11]       1100011  BEQ
> +	 *     imm[12|10:5]        rs2   rs1    001       imm[4:1|11]       1100011  BNE
> +	 *     imm[12|10:5]        rs2   rs1    100       imm[4:1|11]       1100011  BLT
> +	 *     imm[12|10:5]        rs2   rs1    101       imm[4:1|11]       1100011  BGE
> +	 *     imm[12|10:5]        rs2   rs1    110       imm[4:1|11]       1100011  BLTU
> +	 *     imm[12|10:5]        rs2   rs1    111       imm[4:1|11]       1100011  BGEU
> +	 */
> +
> +	s32 offset;
> +	s32 offset_tmp;
> +	unsigned long rs1_val;
> +	unsigned long rs2_val;
> +
> +	if (!rv_insn_reg_get_val(regs, branch_rs1_idx(opcode), &rs1_val) ||
> +	    !rv_insn_reg_get_val(regs, branch_rs2_idx(opcode), &rs2_val))
> +		return false;
> +
> +	offset_tmp = branch_offset(opcode);
> +	switch (branch_funct3(opcode)) {
> +	case BRANCH_BEQ:
> +		offset = (rs1_val == rs2_val) ? offset_tmp : 4;
> +		break;
> +	case BRANCH_BNE:
> +		offset = (rs1_val != rs2_val) ? offset_tmp : 4;
> +		break;
> +	case BRANCH_BLT:
> +		offset = ((long)rs1_val < (long)rs2_val) ? offset_tmp : 4;
> +		break;
> +	case BRANCH_BGE:
> +		offset = ((long)rs1_val >= (long)rs2_val) ? offset_tmp : 4;
> +		break;
> +	case BRANCH_BLTU:
> +		offset = (rs1_val < rs2_val) ? offset_tmp : 4;
> +		break;
> +	case BRANCH_BGEU:
> +		offset = (rs1_val >= rs2_val) ? offset_tmp : 4;
> +		break;
> +	default:
> +		return false;
> +	}
> +
> +	instruction_pointer_set(regs, addr + offset);
> +
> +	return true;
> +}
> 

^ permalink raw reply	[flat|nested] 10+ messages in thread

* ping//Re: [PATCH -next 2/2] riscv: implemented branch simulate instructions
@ 2021-07-05  2:09     ` chenlifu
  0 siblings, 0 replies; 10+ messages in thread
From: chenlifu @ 2021-07-05  2:09 UTC (permalink / raw)
  To: paul.walmsley, palmer, aou, guoren, penberg, mhiramat, me,
	linux-riscv, linux-kernel



在 2021/6/29 10:34, Chen Lifu 写道:
> To test the kprobe-based event tracing, we prepare
> a kernel module 'kprobe_test.ko' to add the probes.
> The assembly codes (partially) of the module are as follows:
> ...
> 0000000000000000 <kprobe_test_branch>:
> ...
> 
> 0000000000000080 <.L7>:
>    80:	414986bb          	subw	a3,s3,s4
>    84:	4785                	li	a5,1
>    86:	00d7c363          	blt	a5,a3,8c <.L8>
>    8a:	2485                	addiw	s1,s1,1
> 
> 000000000000008c <.L8>:
>    8c:	012987bb          	addw	a5,s3,s2
>    90:	4711                	li	a4,4
>    92:	00f75363          	bge	a4,a5,98 <.L9>
>    96:	2485                	addiw	s1,s1,1
> ...
> 
> 00000000000000aa <.L11>:
>    aa:	01299363          	bne	s3,s2,b0 <.L12>
>    ae:	2485                	addiw	s1,s1,1
> 
> 00000000000000b0 <.L12>:
>    b0:	012a0363          	beq	s4,s2,b6 <.L13>
>    b4:	2485                	addiw	s1,s1,1
> ...
> 
> 00000000000000c2 <.L14>:
>    c2:	0009861b          	sext.w	a2,s3
>    c6:	013a7363          	bgeu	s4,s3,cc <.L15>
>    ca:	2485                	addiw	s1,s1,1
> ...
> 
> 00000000000000d2 <.L16>:
>    d2:	00e7e363          	bltu	a5,a4,d8 <.L17>
>    d6:	2485                	addiw	s1,s1,1
> ...
> 
> Test the kprobe-based event tracing in qemu-system-riscv64:
> First, install the kprobe test module:
> insmod /root/kprobe_test.ko
> 
> Then, add probes as new events at the branch instructions,
> i.e. blt, bge, bne, beq, bgeu and bltu.
> The following errors occur due to the instructions not allowed to probe yet:
> echo "p:blt kprobe_test:kprobe_test_branch+0x86 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
> sh: write error: Invalid argument
> echo "p:bge kprobe_test:kprobe_test_branch+0x92 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
> sh: write error: Invalid argument
> echo "p:bne kprobe_test:kprobe_test_branch+0xaa epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
> sh: write error: Invalid argument
> echo "p:beq kprobe_test:kprobe_test_branch+0xb0 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
> sh: write error: Invalid argument
> echo "p:bgeu kprobe_test:kprobe_test_branch+0xc6 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
> sh: write error: Invalid argument
> echo "p:bltu kprobe_test:kprobe_test_branch+0xd2 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
> sh: write error: Invalid argument
> 
> This patch implemented the branch simulate instructions and allowed to probe them.
> Merge this patch and perform the test again, the test results are as follows:
> First, add probes at the branch instructions:
> echo "p:blt kprobe_test:kprobe_test_branch+0x86 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
> echo "p:bge kprobe_test:kprobe_test_branch+0x92 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
> echo "p:bne kprobe_test:kprobe_test_branch+0xaa epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
> echo "p:beq kprobe_test:kprobe_test_branch+0xb0 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
> echo "p:bgeu kprobe_test:kprobe_test_branch+0xc6 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
> echo "p:bltu kprobe_test:kprobe_test_branch+0xd2 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
> echo 1 > /sys/kernel/debug/tracing/events/kprobes/blt/enable
> echo 1 > /sys/kernel/debug/tracing/events/kprobes/bge/enable
> echo 1 > /sys/kernel/debug/tracing/events/kprobes/bne/enable
> echo 1 > /sys/kernel/debug/tracing/events/kprobes/beq/enable
> echo 1 > /sys/kernel/debug/tracing/events/kprobes/bgeu/enable
> echo 1 > /sys/kernel/debug/tracing/events/kprobes/bltu/enable
> 
> Then, do something to run to the probes.
> After that, see the traced information:
> cat /sys/kernel/debug/tracing/trace
> sysctl-63      [001] d...  2505.263969: blt: (kprobe_test_branch+0x86/0x10e [kprobe_test]) epc=0xffffffff016122f8 opcode=0x100073
> sysctl-63      [001] d...  2505.263991: bge: (kprobe_test_branch+0x92/0x10e [kprobe_test]) epc=0xffffffff01612304 opcode=0x100073
> sysctl-63      [001] d...  2505.264001: bne: (kprobe_test_branch+0xaa/0x10e [kprobe_test]) epc=0xffffffff0161231c opcode=0x100073
> sysctl-63      [001] d...  2505.264011: beq: (kprobe_test_branch+0xb0/0x10e [kprobe_test]) epc=0xffffffff01612322 opcode=0x100073
> sysctl-63      [001] d...  2505.264019: bgeu: (kprobe_test_branch+0xc6/0x10e [kprobe_test]) epc=0xffffffff01612338 opcode=0x100073
> sysctl-63      [001] d...  2505.264027: bltu: (kprobe_test_branch+0xd2/0x10e [kprobe_test]) epc=0xffffffff01612344 opcode=0x100073
> 
> Now we can see the traced information.
> The actual address of the symbol 'kprobe_test_branch' is as follows:
> cat /proc/kallsyms | grep kprobe_test_branch
> ffffffff01612272 t kprobe_test_branch   [kprobe_test]
> 
> Based on the traced information and the actual address of the symbol
> 'kprobe_test_branch', we can also see that the branch instructions
> have been replaced by 'ebreak(0x100073)' instructions.
> 
> The pesudoinstructions of the branch instructions,
> i.e. bnez, beqz, blez, bgez, bltz, bgtz, bleu, bgtu and ble
> are allowed to probe as well.
> 
> --------
> 
> Signed-off-by: Chen Lifu <chenlifu@huawei.com>
> ---
>   arch/riscv/kernel/probes/decode-insn.c   |  3 +-
>   arch/riscv/kernel/probes/simulate-insn.c | 78 ++++++++++++++++++++++++
>   2 files changed, 79 insertions(+), 2 deletions(-)
> 
> diff --git a/arch/riscv/kernel/probes/decode-insn.c b/arch/riscv/kernel/probes/decode-insn.c
> index 5eb03fb61450..64f6183b4717 100644
> --- a/arch/riscv/kernel/probes/decode-insn.c
> +++ b/arch/riscv/kernel/probes/decode-insn.c
> @@ -38,11 +38,10 @@ riscv_probe_decode_insn(probe_opcode_t *addr, struct arch_probe_insn *api)
>   	RISCV_INSN_REJECTED(c_ebreak,		insn);
>   #endif
>   
> -	RISCV_INSN_REJECTED(branch,		insn);
> -
>   	RISCV_INSN_SET_SIMULATE(jal,		insn);
>   	RISCV_INSN_SET_SIMULATE(jalr,		insn);
>   	RISCV_INSN_SET_SIMULATE(auipc,		insn);
> +	RISCV_INSN_SET_SIMULATE(branch,		insn);
>   
>   	return INSN_GOOD;
>   }
> diff --git a/arch/riscv/kernel/probes/simulate-insn.c b/arch/riscv/kernel/probes/simulate-insn.c
> index b81719522d5c..efc989b28859 100644
> --- a/arch/riscv/kernel/probes/simulate-insn.c
> +++ b/arch/riscv/kernel/probes/simulate-insn.c
> @@ -117,3 +117,81 @@ bool __kprobes simulate_auipc(u32 opcode, unsigned long addr, struct pt_regs *re
>   
>   	return true;
>   }
> +
> +#define branch_rs1_idx(opcode) \
> +	(((opcode) >> 15) & 0x1f)
> +
> +#define branch_rs2_idx(opcode) \
> +	(((opcode) >> 20) & 0x1f)
> +
> +#define branch_funct3(opcode) \
> +	(((opcode) >> 12) & 0x7)
> +
> +#define branch_imm(opcode) \
> +	(((((opcode) >> 8 ) & 0xf ) << 1 ) | \
> +	 ((((opcode) >> 25) & 0x3f) << 5 ) | \
> +	 ((((opcode) >> 7 ) & 0x1 ) << 11) | \
> +	 ((((opcode) >> 31) & 0x1 ) << 12))
> +
> +#define branch_offset(opcode) \
> +	sign_extend32((branch_imm(opcode)), 12)
> +
> +#define BRANCH_BEQ	0x0
> +#define BRANCH_BNE	0x1
> +#define BRANCH_BLT	0x4
> +#define BRANCH_BGE	0x5
> +#define BRANCH_BLTU	0x6
> +#define BRANCH_BGEU	0x7
> +
> +bool __kprobes simulate_branch(u32 opcode, unsigned long addr, struct pt_regs *regs)
> +{
> +	/*
> +	 * branch instructions:
> +	 *      31    30       25 24 20 19 15 14    12 11       8    7      6      0
> +	 * | imm[12] | imm[10:5] | rs2 | rs1 | funct3 | imm[4:1] | imm[11] | opcode |
> +	 *     1           6        5     5      3         4         1         7
> +	 *     imm[12|10:5]        rs2   rs1    000       imm[4:1|11]       1100011  BEQ
> +	 *     imm[12|10:5]        rs2   rs1    001       imm[4:1|11]       1100011  BNE
> +	 *     imm[12|10:5]        rs2   rs1    100       imm[4:1|11]       1100011  BLT
> +	 *     imm[12|10:5]        rs2   rs1    101       imm[4:1|11]       1100011  BGE
> +	 *     imm[12|10:5]        rs2   rs1    110       imm[4:1|11]       1100011  BLTU
> +	 *     imm[12|10:5]        rs2   rs1    111       imm[4:1|11]       1100011  BGEU
> +	 */
> +
> +	s32 offset;
> +	s32 offset_tmp;
> +	unsigned long rs1_val;
> +	unsigned long rs2_val;
> +
> +	if (!rv_insn_reg_get_val(regs, branch_rs1_idx(opcode), &rs1_val) ||
> +	    !rv_insn_reg_get_val(regs, branch_rs2_idx(opcode), &rs2_val))
> +		return false;
> +
> +	offset_tmp = branch_offset(opcode);
> +	switch (branch_funct3(opcode)) {
> +	case BRANCH_BEQ:
> +		offset = (rs1_val == rs2_val) ? offset_tmp : 4;
> +		break;
> +	case BRANCH_BNE:
> +		offset = (rs1_val != rs2_val) ? offset_tmp : 4;
> +		break;
> +	case BRANCH_BLT:
> +		offset = ((long)rs1_val < (long)rs2_val) ? offset_tmp : 4;
> +		break;
> +	case BRANCH_BGE:
> +		offset = ((long)rs1_val >= (long)rs2_val) ? offset_tmp : 4;
> +		break;
> +	case BRANCH_BLTU:
> +		offset = (rs1_val < rs2_val) ? offset_tmp : 4;
> +		break;
> +	case BRANCH_BGEU:
> +		offset = (rs1_val >= rs2_val) ? offset_tmp : 4;
> +		break;
> +	default:
> +		return false;
> +	}
> +
> +	instruction_pointer_set(regs, addr + offset);
> +
> +	return true;
> +}
> 

_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

^ permalink raw reply	[flat|nested] 10+ messages in thread

* ping//Re: ping//Re: [PATCH -next 2/2] riscv: implemented branch simulate instructions
  2021-07-05  2:09     ` chenlifu
@ 2021-07-22 10:15       ` chenlifu
  -1 siblings, 0 replies; 10+ messages in thread
From: chenlifu @ 2021-07-22 10:15 UTC (permalink / raw)
  To: paul.walmsley, palmer, aou, guoren, penberg, mhiramat, me,
	linux-riscv, linux-kernel



在 2021/7/5 10:09, chenlifu 写道:
> 
> 
> 在 2021/6/29 10:34, Chen Lifu 写道:
>> To test the kprobe-based event tracing, we prepare
>> a kernel module 'kprobe_test.ko' to add the probes.
>> The assembly codes (partially) of the module are as follows:
>> ...
>> 0000000000000000 <kprobe_test_branch>:
>> ...
>>
>> 0000000000000080 <.L7>:
>>    80:    414986bb              subw    a3,s3,s4
>>    84:    4785                    li    a5,1
>>    86:    00d7c363              blt    a5,a3,8c <.L8>
>>    8a:    2485                    addiw    s1,s1,1
>>
>> 000000000000008c <.L8>:
>>    8c:    012987bb              addw    a5,s3,s2
>>    90:    4711                    li    a4,4
>>    92:    00f75363              bge    a4,a5,98 <.L9>
>>    96:    2485                    addiw    s1,s1,1
>> ...
>>
>> 00000000000000aa <.L11>:
>>    aa:    01299363              bne    s3,s2,b0 <.L12>
>>    ae:    2485                    addiw    s1,s1,1
>>
>> 00000000000000b0 <.L12>:
>>    b0:    012a0363              beq    s4,s2,b6 <.L13>
>>    b4:    2485                    addiw    s1,s1,1
>> ...
>>
>> 00000000000000c2 <.L14>:
>>    c2:    0009861b              sext.w    a2,s3
>>    c6:    013a7363              bgeu    s4,s3,cc <.L15>
>>    ca:    2485                    addiw    s1,s1,1
>> ...
>>
>> 00000000000000d2 <.L16>:
>>    d2:    00e7e363              bltu    a5,a4,d8 <.L17>
>>    d6:    2485                    addiw    s1,s1,1
>> ...
>>
>> Test the kprobe-based event tracing in qemu-system-riscv64:
>> First, install the kprobe test module:
>> insmod /root/kprobe_test.ko
>>
>> Then, add probes as new events at the branch instructions,
>> i.e. blt, bge, bne, beq, bgeu and bltu.
>> The following errors occur due to the instructions not allowed to 
>> probe yet:
>> echo "p:blt kprobe_test:kprobe_test_branch+0x86 epc=%epc 
>> opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
>> sh: write error: Invalid argument
>> echo "p:bge kprobe_test:kprobe_test_branch+0x92 epc=%epc 
>> opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
>> sh: write error: Invalid argument
>> echo "p:bne kprobe_test:kprobe_test_branch+0xaa epc=%epc 
>> opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
>> sh: write error: Invalid argument
>> echo "p:beq kprobe_test:kprobe_test_branch+0xb0 epc=%epc 
>> opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
>> sh: write error: Invalid argument
>> echo "p:bgeu kprobe_test:kprobe_test_branch+0xc6 epc=%epc 
>> opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
>> sh: write error: Invalid argument
>> echo "p:bltu kprobe_test:kprobe_test_branch+0xd2 epc=%epc 
>> opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
>> sh: write error: Invalid argument
>>
>> This patch implemented the branch simulate instructions and allowed to 
>> probe them.
>> Merge this patch and perform the test again, the test results are as 
>> follows:
>> First, add probes at the branch instructions:
>> echo "p:blt kprobe_test:kprobe_test_branch+0x86 epc=%epc 
>> opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
>> echo "p:bge kprobe_test:kprobe_test_branch+0x92 epc=%epc 
>> opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
>> echo "p:bne kprobe_test:kprobe_test_branch+0xaa epc=%epc 
>> opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
>> echo "p:beq kprobe_test:kprobe_test_branch+0xb0 epc=%epc 
>> opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
>> echo "p:bgeu kprobe_test:kprobe_test_branch+0xc6 epc=%epc 
>> opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
>> echo "p:bltu kprobe_test:kprobe_test_branch+0xd2 epc=%epc 
>> opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
>> echo 1 > /sys/kernel/debug/tracing/events/kprobes/blt/enable
>> echo 1 > /sys/kernel/debug/tracing/events/kprobes/bge/enable
>> echo 1 > /sys/kernel/debug/tracing/events/kprobes/bne/enable
>> echo 1 > /sys/kernel/debug/tracing/events/kprobes/beq/enable
>> echo 1 > /sys/kernel/debug/tracing/events/kprobes/bgeu/enable
>> echo 1 > /sys/kernel/debug/tracing/events/kprobes/bltu/enable
>>
>> Then, do something to run to the probes.
>> After that, see the traced information:
>> cat /sys/kernel/debug/tracing/trace
>> sysctl-63      [001] d...  2505.263969: blt: 
>> (kprobe_test_branch+0x86/0x10e [kprobe_test]) epc=0xffffffff016122f8 
>> opcode=0x100073
>> sysctl-63      [001] d...  2505.263991: bge: 
>> (kprobe_test_branch+0x92/0x10e [kprobe_test]) epc=0xffffffff01612304 
>> opcode=0x100073
>> sysctl-63      [001] d...  2505.264001: bne: 
>> (kprobe_test_branch+0xaa/0x10e [kprobe_test]) epc=0xffffffff0161231c 
>> opcode=0x100073
>> sysctl-63      [001] d...  2505.264011: beq: 
>> (kprobe_test_branch+0xb0/0x10e [kprobe_test]) epc=0xffffffff01612322 
>> opcode=0x100073
>> sysctl-63      [001] d...  2505.264019: bgeu: 
>> (kprobe_test_branch+0xc6/0x10e [kprobe_test]) epc=0xffffffff01612338 
>> opcode=0x100073
>> sysctl-63      [001] d...  2505.264027: bltu: 
>> (kprobe_test_branch+0xd2/0x10e [kprobe_test]) epc=0xffffffff01612344 
>> opcode=0x100073
>>
>> Now we can see the traced information.
>> The actual address of the symbol 'kprobe_test_branch' is as follows:
>> cat /proc/kallsyms | grep kprobe_test_branch
>> ffffffff01612272 t kprobe_test_branch   [kprobe_test]
>>
>> Based on the traced information and the actual address of the symbol
>> 'kprobe_test_branch', we can also see that the branch instructions
>> have been replaced by 'ebreak(0x100073)' instructions.
>>
>> The pesudoinstructions of the branch instructions,
>> i.e. bnez, beqz, blez, bgez, bltz, bgtz, bleu, bgtu and ble
>> are allowed to probe as well.
>>
>> --------
>>
>> Signed-off-by: Chen Lifu <chenlifu@huawei.com>
>> ---
>>   arch/riscv/kernel/probes/decode-insn.c   |  3 +-
>>   arch/riscv/kernel/probes/simulate-insn.c | 78 ++++++++++++++++++++++++
>>   2 files changed, 79 insertions(+), 2 deletions(-)
>>
>> diff --git a/arch/riscv/kernel/probes/decode-insn.c 
>> b/arch/riscv/kernel/probes/decode-insn.c
>> index 5eb03fb61450..64f6183b4717 100644
>> --- a/arch/riscv/kernel/probes/decode-insn.c
>> +++ b/arch/riscv/kernel/probes/decode-insn.c
>> @@ -38,11 +38,10 @@ riscv_probe_decode_insn(probe_opcode_t *addr, 
>> struct arch_probe_insn *api)
>>       RISCV_INSN_REJECTED(c_ebreak,        insn);
>>   #endif
>> -    RISCV_INSN_REJECTED(branch,        insn);
>> -
>>       RISCV_INSN_SET_SIMULATE(jal,        insn);
>>       RISCV_INSN_SET_SIMULATE(jalr,        insn);
>>       RISCV_INSN_SET_SIMULATE(auipc,        insn);
>> +    RISCV_INSN_SET_SIMULATE(branch,        insn);
>>       return INSN_GOOD;
>>   }
>> diff --git a/arch/riscv/kernel/probes/simulate-insn.c 
>> b/arch/riscv/kernel/probes/simulate-insn.c
>> index b81719522d5c..efc989b28859 100644
>> --- a/arch/riscv/kernel/probes/simulate-insn.c
>> +++ b/arch/riscv/kernel/probes/simulate-insn.c
>> @@ -117,3 +117,81 @@ bool __kprobes simulate_auipc(u32 opcode, 
>> unsigned long addr, struct pt_regs *re
>>       return true;
>>   }
>> +
>> +#define branch_rs1_idx(opcode) \
>> +    (((opcode) >> 15) & 0x1f)
>> +
>> +#define branch_rs2_idx(opcode) \
>> +    (((opcode) >> 20) & 0x1f)
>> +
>> +#define branch_funct3(opcode) \
>> +    (((opcode) >> 12) & 0x7)
>> +
>> +#define branch_imm(opcode) \
>> +    (((((opcode) >> 8 ) & 0xf ) << 1 ) | \
>> +     ((((opcode) >> 25) & 0x3f) << 5 ) | \
>> +     ((((opcode) >> 7 ) & 0x1 ) << 11) | \
>> +     ((((opcode) >> 31) & 0x1 ) << 12))
>> +
>> +#define branch_offset(opcode) \
>> +    sign_extend32((branch_imm(opcode)), 12)
>> +
>> +#define BRANCH_BEQ    0x0
>> +#define BRANCH_BNE    0x1
>> +#define BRANCH_BLT    0x4
>> +#define BRANCH_BGE    0x5
>> +#define BRANCH_BLTU    0x6
>> +#define BRANCH_BGEU    0x7
>> +
>> +bool __kprobes simulate_branch(u32 opcode, unsigned long addr, struct 
>> pt_regs *regs)
>> +{
>> +    /*
>> +     * branch instructions:
>> +     *      31    30       25 24 20 19 15 14    12 11       8    
>> 7      6      0
>> +     * | imm[12] | imm[10:5] | rs2 | rs1 | funct3 | imm[4:1] | 
>> imm[11] | opcode |
>> +     *     1           6        5     5      3         4         
>> 1         7
>> +     *     imm[12|10:5]        rs2   rs1    000       
>> imm[4:1|11]       1100011  BEQ
>> +     *     imm[12|10:5]        rs2   rs1    001       
>> imm[4:1|11]       1100011  BNE
>> +     *     imm[12|10:5]        rs2   rs1    100       
>> imm[4:1|11]       1100011  BLT
>> +     *     imm[12|10:5]        rs2   rs1    101       
>> imm[4:1|11]       1100011  BGE
>> +     *     imm[12|10:5]        rs2   rs1    110       
>> imm[4:1|11]       1100011  BLTU
>> +     *     imm[12|10:5]        rs2   rs1    111       
>> imm[4:1|11]       1100011  BGEU
>> +     */
>> +
>> +    s32 offset;
>> +    s32 offset_tmp;
>> +    unsigned long rs1_val;
>> +    unsigned long rs2_val;
>> +
>> +    if (!rv_insn_reg_get_val(regs, branch_rs1_idx(opcode), &rs1_val) ||
>> +        !rv_insn_reg_get_val(regs, branch_rs2_idx(opcode), &rs2_val))
>> +        return false;
>> +
>> +    offset_tmp = branch_offset(opcode);
>> +    switch (branch_funct3(opcode)) {
>> +    case BRANCH_BEQ:
>> +        offset = (rs1_val == rs2_val) ? offset_tmp : 4;
>> +        break;
>> +    case BRANCH_BNE:
>> +        offset = (rs1_val != rs2_val) ? offset_tmp : 4;
>> +        break;
>> +    case BRANCH_BLT:
>> +        offset = ((long)rs1_val < (long)rs2_val) ? offset_tmp : 4;
>> +        break;
>> +    case BRANCH_BGE:
>> +        offset = ((long)rs1_val >= (long)rs2_val) ? offset_tmp : 4;
>> +        break;
>> +    case BRANCH_BLTU:
>> +        offset = (rs1_val < rs2_val) ? offset_tmp : 4;
>> +        break;
>> +    case BRANCH_BGEU:
>> +        offset = (rs1_val >= rs2_val) ? offset_tmp : 4;
>> +        break;
>> +    default:
>> +        return false;
>> +    }
>> +
>> +    instruction_pointer_set(regs, addr + offset);
>> +
>> +    return true;
>> +}
>>
> 
> _______________________________________________
> linux-riscv mailing list
> linux-riscv@lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-riscv

^ permalink raw reply	[flat|nested] 10+ messages in thread

* ping//Re: ping//Re: [PATCH -next 2/2] riscv: implemented branch simulate instructions
@ 2021-07-22 10:15       ` chenlifu
  0 siblings, 0 replies; 10+ messages in thread
From: chenlifu @ 2021-07-22 10:15 UTC (permalink / raw)
  To: paul.walmsley, palmer, aou, guoren, penberg, mhiramat, me,
	linux-riscv, linux-kernel



在 2021/7/5 10:09, chenlifu 写道:
> 
> 
> 在 2021/6/29 10:34, Chen Lifu 写道:
>> To test the kprobe-based event tracing, we prepare
>> a kernel module 'kprobe_test.ko' to add the probes.
>> The assembly codes (partially) of the module are as follows:
>> ...
>> 0000000000000000 <kprobe_test_branch>:
>> ...
>>
>> 0000000000000080 <.L7>:
>>    80:    414986bb              subw    a3,s3,s4
>>    84:    4785                    li    a5,1
>>    86:    00d7c363              blt    a5,a3,8c <.L8>
>>    8a:    2485                    addiw    s1,s1,1
>>
>> 000000000000008c <.L8>:
>>    8c:    012987bb              addw    a5,s3,s2
>>    90:    4711                    li    a4,4
>>    92:    00f75363              bge    a4,a5,98 <.L9>
>>    96:    2485                    addiw    s1,s1,1
>> ...
>>
>> 00000000000000aa <.L11>:
>>    aa:    01299363              bne    s3,s2,b0 <.L12>
>>    ae:    2485                    addiw    s1,s1,1
>>
>> 00000000000000b0 <.L12>:
>>    b0:    012a0363              beq    s4,s2,b6 <.L13>
>>    b4:    2485                    addiw    s1,s1,1
>> ...
>>
>> 00000000000000c2 <.L14>:
>>    c2:    0009861b              sext.w    a2,s3
>>    c6:    013a7363              bgeu    s4,s3,cc <.L15>
>>    ca:    2485                    addiw    s1,s1,1
>> ...
>>
>> 00000000000000d2 <.L16>:
>>    d2:    00e7e363              bltu    a5,a4,d8 <.L17>
>>    d6:    2485                    addiw    s1,s1,1
>> ...
>>
>> Test the kprobe-based event tracing in qemu-system-riscv64:
>> First, install the kprobe test module:
>> insmod /root/kprobe_test.ko
>>
>> Then, add probes as new events at the branch instructions,
>> i.e. blt, bge, bne, beq, bgeu and bltu.
>> The following errors occur due to the instructions not allowed to 
>> probe yet:
>> echo "p:blt kprobe_test:kprobe_test_branch+0x86 epc=%epc 
>> opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
>> sh: write error: Invalid argument
>> echo "p:bge kprobe_test:kprobe_test_branch+0x92 epc=%epc 
>> opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
>> sh: write error: Invalid argument
>> echo "p:bne kprobe_test:kprobe_test_branch+0xaa epc=%epc 
>> opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
>> sh: write error: Invalid argument
>> echo "p:beq kprobe_test:kprobe_test_branch+0xb0 epc=%epc 
>> opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
>> sh: write error: Invalid argument
>> echo "p:bgeu kprobe_test:kprobe_test_branch+0xc6 epc=%epc 
>> opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
>> sh: write error: Invalid argument
>> echo "p:bltu kprobe_test:kprobe_test_branch+0xd2 epc=%epc 
>> opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
>> sh: write error: Invalid argument
>>
>> This patch implemented the branch simulate instructions and allowed to 
>> probe them.
>> Merge this patch and perform the test again, the test results are as 
>> follows:
>> First, add probes at the branch instructions:
>> echo "p:blt kprobe_test:kprobe_test_branch+0x86 epc=%epc 
>> opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
>> echo "p:bge kprobe_test:kprobe_test_branch+0x92 epc=%epc 
>> opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
>> echo "p:bne kprobe_test:kprobe_test_branch+0xaa epc=%epc 
>> opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
>> echo "p:beq kprobe_test:kprobe_test_branch+0xb0 epc=%epc 
>> opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
>> echo "p:bgeu kprobe_test:kprobe_test_branch+0xc6 epc=%epc 
>> opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
>> echo "p:bltu kprobe_test:kprobe_test_branch+0xd2 epc=%epc 
>> opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
>> echo 1 > /sys/kernel/debug/tracing/events/kprobes/blt/enable
>> echo 1 > /sys/kernel/debug/tracing/events/kprobes/bge/enable
>> echo 1 > /sys/kernel/debug/tracing/events/kprobes/bne/enable
>> echo 1 > /sys/kernel/debug/tracing/events/kprobes/beq/enable
>> echo 1 > /sys/kernel/debug/tracing/events/kprobes/bgeu/enable
>> echo 1 > /sys/kernel/debug/tracing/events/kprobes/bltu/enable
>>
>> Then, do something to run to the probes.
>> After that, see the traced information:
>> cat /sys/kernel/debug/tracing/trace
>> sysctl-63      [001] d...  2505.263969: blt: 
>> (kprobe_test_branch+0x86/0x10e [kprobe_test]) epc=0xffffffff016122f8 
>> opcode=0x100073
>> sysctl-63      [001] d...  2505.263991: bge: 
>> (kprobe_test_branch+0x92/0x10e [kprobe_test]) epc=0xffffffff01612304 
>> opcode=0x100073
>> sysctl-63      [001] d...  2505.264001: bne: 
>> (kprobe_test_branch+0xaa/0x10e [kprobe_test]) epc=0xffffffff0161231c 
>> opcode=0x100073
>> sysctl-63      [001] d...  2505.264011: beq: 
>> (kprobe_test_branch+0xb0/0x10e [kprobe_test]) epc=0xffffffff01612322 
>> opcode=0x100073
>> sysctl-63      [001] d...  2505.264019: bgeu: 
>> (kprobe_test_branch+0xc6/0x10e [kprobe_test]) epc=0xffffffff01612338 
>> opcode=0x100073
>> sysctl-63      [001] d...  2505.264027: bltu: 
>> (kprobe_test_branch+0xd2/0x10e [kprobe_test]) epc=0xffffffff01612344 
>> opcode=0x100073
>>
>> Now we can see the traced information.
>> The actual address of the symbol 'kprobe_test_branch' is as follows:
>> cat /proc/kallsyms | grep kprobe_test_branch
>> ffffffff01612272 t kprobe_test_branch   [kprobe_test]
>>
>> Based on the traced information and the actual address of the symbol
>> 'kprobe_test_branch', we can also see that the branch instructions
>> have been replaced by 'ebreak(0x100073)' instructions.
>>
>> The pesudoinstructions of the branch instructions,
>> i.e. bnez, beqz, blez, bgez, bltz, bgtz, bleu, bgtu and ble
>> are allowed to probe as well.
>>
>> --------
>>
>> Signed-off-by: Chen Lifu <chenlifu@huawei.com>
>> ---
>>   arch/riscv/kernel/probes/decode-insn.c   |  3 +-
>>   arch/riscv/kernel/probes/simulate-insn.c | 78 ++++++++++++++++++++++++
>>   2 files changed, 79 insertions(+), 2 deletions(-)
>>
>> diff --git a/arch/riscv/kernel/probes/decode-insn.c 
>> b/arch/riscv/kernel/probes/decode-insn.c
>> index 5eb03fb61450..64f6183b4717 100644
>> --- a/arch/riscv/kernel/probes/decode-insn.c
>> +++ b/arch/riscv/kernel/probes/decode-insn.c
>> @@ -38,11 +38,10 @@ riscv_probe_decode_insn(probe_opcode_t *addr, 
>> struct arch_probe_insn *api)
>>       RISCV_INSN_REJECTED(c_ebreak,        insn);
>>   #endif
>> -    RISCV_INSN_REJECTED(branch,        insn);
>> -
>>       RISCV_INSN_SET_SIMULATE(jal,        insn);
>>       RISCV_INSN_SET_SIMULATE(jalr,        insn);
>>       RISCV_INSN_SET_SIMULATE(auipc,        insn);
>> +    RISCV_INSN_SET_SIMULATE(branch,        insn);
>>       return INSN_GOOD;
>>   }
>> diff --git a/arch/riscv/kernel/probes/simulate-insn.c 
>> b/arch/riscv/kernel/probes/simulate-insn.c
>> index b81719522d5c..efc989b28859 100644
>> --- a/arch/riscv/kernel/probes/simulate-insn.c
>> +++ b/arch/riscv/kernel/probes/simulate-insn.c
>> @@ -117,3 +117,81 @@ bool __kprobes simulate_auipc(u32 opcode, 
>> unsigned long addr, struct pt_regs *re
>>       return true;
>>   }
>> +
>> +#define branch_rs1_idx(opcode) \
>> +    (((opcode) >> 15) & 0x1f)
>> +
>> +#define branch_rs2_idx(opcode) \
>> +    (((opcode) >> 20) & 0x1f)
>> +
>> +#define branch_funct3(opcode) \
>> +    (((opcode) >> 12) & 0x7)
>> +
>> +#define branch_imm(opcode) \
>> +    (((((opcode) >> 8 ) & 0xf ) << 1 ) | \
>> +     ((((opcode) >> 25) & 0x3f) << 5 ) | \
>> +     ((((opcode) >> 7 ) & 0x1 ) << 11) | \
>> +     ((((opcode) >> 31) & 0x1 ) << 12))
>> +
>> +#define branch_offset(opcode) \
>> +    sign_extend32((branch_imm(opcode)), 12)
>> +
>> +#define BRANCH_BEQ    0x0
>> +#define BRANCH_BNE    0x1
>> +#define BRANCH_BLT    0x4
>> +#define BRANCH_BGE    0x5
>> +#define BRANCH_BLTU    0x6
>> +#define BRANCH_BGEU    0x7
>> +
>> +bool __kprobes simulate_branch(u32 opcode, unsigned long addr, struct 
>> pt_regs *regs)
>> +{
>> +    /*
>> +     * branch instructions:
>> +     *      31    30       25 24 20 19 15 14    12 11       8    
>> 7      6      0
>> +     * | imm[12] | imm[10:5] | rs2 | rs1 | funct3 | imm[4:1] | 
>> imm[11] | opcode |
>> +     *     1           6        5     5      3         4         
>> 1         7
>> +     *     imm[12|10:5]        rs2   rs1    000       
>> imm[4:1|11]       1100011  BEQ
>> +     *     imm[12|10:5]        rs2   rs1    001       
>> imm[4:1|11]       1100011  BNE
>> +     *     imm[12|10:5]        rs2   rs1    100       
>> imm[4:1|11]       1100011  BLT
>> +     *     imm[12|10:5]        rs2   rs1    101       
>> imm[4:1|11]       1100011  BGE
>> +     *     imm[12|10:5]        rs2   rs1    110       
>> imm[4:1|11]       1100011  BLTU
>> +     *     imm[12|10:5]        rs2   rs1    111       
>> imm[4:1|11]       1100011  BGEU
>> +     */
>> +
>> +    s32 offset;
>> +    s32 offset_tmp;
>> +    unsigned long rs1_val;
>> +    unsigned long rs2_val;
>> +
>> +    if (!rv_insn_reg_get_val(regs, branch_rs1_idx(opcode), &rs1_val) ||
>> +        !rv_insn_reg_get_val(regs, branch_rs2_idx(opcode), &rs2_val))
>> +        return false;
>> +
>> +    offset_tmp = branch_offset(opcode);
>> +    switch (branch_funct3(opcode)) {
>> +    case BRANCH_BEQ:
>> +        offset = (rs1_val == rs2_val) ? offset_tmp : 4;
>> +        break;
>> +    case BRANCH_BNE:
>> +        offset = (rs1_val != rs2_val) ? offset_tmp : 4;
>> +        break;
>> +    case BRANCH_BLT:
>> +        offset = ((long)rs1_val < (long)rs2_val) ? offset_tmp : 4;
>> +        break;
>> +    case BRANCH_BGE:
>> +        offset = ((long)rs1_val >= (long)rs2_val) ? offset_tmp : 4;
>> +        break;
>> +    case BRANCH_BLTU:
>> +        offset = (rs1_val < rs2_val) ? offset_tmp : 4;
>> +        break;
>> +    case BRANCH_BGEU:
>> +        offset = (rs1_val >= rs2_val) ? offset_tmp : 4;
>> +        break;
>> +    default:
>> +        return false;
>> +    }
>> +
>> +    instruction_pointer_set(regs, addr + offset);
>> +
>> +    return true;
>> +}
>>
> 
> _______________________________________________
> linux-riscv mailing list
> linux-riscv@lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-riscv

_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [PATCH -next 1/2] riscv: implemented auipc simulate instruction
  2021-06-29  2:34 ` Chen Lifu
@ 2021-07-22 17:26   ` Palmer Dabbelt
  -1 siblings, 0 replies; 10+ messages in thread
From: Palmer Dabbelt @ 2021-07-22 17:26 UTC (permalink / raw)
  To: chenlifu
  Cc: Paul Walmsley, aou, guoren, chenlifu, penberg, mhiramat, me,
	linux-riscv, linux-kernel

On Mon, 28 Jun 2021 19:34:54 PDT (-0700), chenlifu@huawei.com wrote:
> To test the kprobe-based event tracing, we prepare
> a kernel module 'kprobe_test.ko' to add the probes.
> The assembly codes (partially) of the module are as follows:
> ...
> 0000000000000000 <kprobe_test_branch>:
> ...
> 0000000000000038 <.LVL1>:
>   38:	00000597          	auipc	a1,0x0
>   3c:	00058593          	mv	a1,a1
> ...
>
> Test the kprobe-based event tracing in qemu-system-riscv64:
> First, install the kprobe test module:
> insmod /root/kprobe_test.ko
>
> Then, add a probe as a new event at an 'auipc' instruction,
> the following error occurs due to the instruction not allowed to probe yet:
> echo "p:auipc kprobe_test:kprobe_test_branch+0x38 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
> sh: write error: Invalid argument
>
> This patch implemented the 'auipc' simulate instruction and allowed to probe it.
> Merge this patch and perform the test again, the test results are as follows:
> First, add a probe at the 'auipc' instruction:
> echo "p:auipc kprobe_test:kprobe_test_branch+0x38 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
> echo 1 > /sys/kernel/debug/tracing/events/kprobes/auipc/enable
>
> Then, do something to run to the probe.
> After that, see the traced information:
> cat /sys/kernel/debug/tracing/trace
> sysctl-58      [001] d...   179.126350: auipc: (kprobe_test_branch+0x38/0x10e [kprobe_test]) epc=0xffffffff016122aa opcode=0x100073
>
> Now we can see the traced information.
> The actual address of the symbol 'kprobe_test_branch' is as follows:
> cat /proc/kallsyms | grep kprobe_test_branch
> ffffffff01612272 t kprobe_test_branch   [kprobe_test]
>
> Based on the traced information and the actual address of the symbol
> 'kprobe_test_branch', we can also see that the 'auipc' instruction
> has been replaced by 'ebreak(0x100073)' instruction.
>
> --------
>
> Signed-off-by: Chen Lifu <chenlifu@huawei.com>
> ---
>  arch/riscv/kernel/probes/decode-insn.c   |  2 +-
>  arch/riscv/kernel/probes/simulate-insn.c | 34 ++++++++++++++++++++++++
>  2 files changed, 35 insertions(+), 1 deletion(-)
>
> diff --git a/arch/riscv/kernel/probes/decode-insn.c b/arch/riscv/kernel/probes/decode-insn.c
> index 0ed043acc882..5eb03fb61450 100644
> --- a/arch/riscv/kernel/probes/decode-insn.c
> +++ b/arch/riscv/kernel/probes/decode-insn.c
> @@ -38,11 +38,11 @@ riscv_probe_decode_insn(probe_opcode_t *addr, struct arch_probe_insn *api)
>  	RISCV_INSN_REJECTED(c_ebreak,		insn);
>  #endif
>
> -	RISCV_INSN_REJECTED(auipc,		insn);
>  	RISCV_INSN_REJECTED(branch,		insn);
>
>  	RISCV_INSN_SET_SIMULATE(jal,		insn);
>  	RISCV_INSN_SET_SIMULATE(jalr,		insn);
> +	RISCV_INSN_SET_SIMULATE(auipc,		insn);
>
>  	return INSN_GOOD;
>  }
> diff --git a/arch/riscv/kernel/probes/simulate-insn.c b/arch/riscv/kernel/probes/simulate-insn.c
> index 2519ce26377d..b81719522d5c 100644
> --- a/arch/riscv/kernel/probes/simulate-insn.c
> +++ b/arch/riscv/kernel/probes/simulate-insn.c
> @@ -83,3 +83,37 @@ bool __kprobes simulate_jalr(u32 opcode, unsigned long addr, struct pt_regs *reg
>
>  	return ret;
>  }
> +
> +#define auipc_rd_idx(opcode) \
> +	((opcode >> 7) & 0x1f)
> +
> +#define auipc_imm(opcode) \
> +	((((opcode) >> 12) & 0xfffff) << 12)
> +
> +#if __riscv_xlen == 64
> +#define auipc_offset(opcode)	sign_extend64(auipc_imm(opcode), 31)
> +#elif __riscv_xlen == 32
> +#define auipc_offset(opcode)	auipc_imm(opcode)
> +#else
> +#error "Unexpected __riscv_xlen"
> +#endif
> +
> +bool __kprobes simulate_auipc(u32 opcode, unsigned long addr, struct pt_regs *regs)
> +{
> +	/*
> +	 * auipc instruction:
> +	 *  31        12 11 7 6      0
> +	 * | imm[31:12] | rd | opcode |
> +	 *        20       5     7
> +	 */
> +
> +	u32 rd_idx = auipc_rd_idx(opcode);
> +	unsigned long rd_val = addr + auipc_offset(opcode);
> +
> +	if (!rv_insn_reg_set_val(regs, rd_idx, rd_val))
> +		return false;
> +
> +	instruction_pointer_set(regs, addr + 4);
> +
> +	return true;
> +}

Thanks.  These are on for-next, with the checkpatch errors fixed and the 
commit messages mostly removed -- it was all pretty awkwardly phrased so 
I didn't want to clean it up.

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [PATCH -next 1/2] riscv: implemented auipc simulate instruction
@ 2021-07-22 17:26   ` Palmer Dabbelt
  0 siblings, 0 replies; 10+ messages in thread
From: Palmer Dabbelt @ 2021-07-22 17:26 UTC (permalink / raw)
  To: chenlifu
  Cc: Paul Walmsley, aou, guoren, chenlifu, penberg, mhiramat, me,
	linux-riscv, linux-kernel

On Mon, 28 Jun 2021 19:34:54 PDT (-0700), chenlifu@huawei.com wrote:
> To test the kprobe-based event tracing, we prepare
> a kernel module 'kprobe_test.ko' to add the probes.
> The assembly codes (partially) of the module are as follows:
> ...
> 0000000000000000 <kprobe_test_branch>:
> ...
> 0000000000000038 <.LVL1>:
>   38:	00000597          	auipc	a1,0x0
>   3c:	00058593          	mv	a1,a1
> ...
>
> Test the kprobe-based event tracing in qemu-system-riscv64:
> First, install the kprobe test module:
> insmod /root/kprobe_test.ko
>
> Then, add a probe as a new event at an 'auipc' instruction,
> the following error occurs due to the instruction not allowed to probe yet:
> echo "p:auipc kprobe_test:kprobe_test_branch+0x38 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
> sh: write error: Invalid argument
>
> This patch implemented the 'auipc' simulate instruction and allowed to probe it.
> Merge this patch and perform the test again, the test results are as follows:
> First, add a probe at the 'auipc' instruction:
> echo "p:auipc kprobe_test:kprobe_test_branch+0x38 epc=%epc opcode=+0(%epc):x32" >> /sys/kernel/debug/tracing/kprobe_events
> echo 1 > /sys/kernel/debug/tracing/events/kprobes/auipc/enable
>
> Then, do something to run to the probe.
> After that, see the traced information:
> cat /sys/kernel/debug/tracing/trace
> sysctl-58      [001] d...   179.126350: auipc: (kprobe_test_branch+0x38/0x10e [kprobe_test]) epc=0xffffffff016122aa opcode=0x100073
>
> Now we can see the traced information.
> The actual address of the symbol 'kprobe_test_branch' is as follows:
> cat /proc/kallsyms | grep kprobe_test_branch
> ffffffff01612272 t kprobe_test_branch   [kprobe_test]
>
> Based on the traced information and the actual address of the symbol
> 'kprobe_test_branch', we can also see that the 'auipc' instruction
> has been replaced by 'ebreak(0x100073)' instruction.
>
> --------
>
> Signed-off-by: Chen Lifu <chenlifu@huawei.com>
> ---
>  arch/riscv/kernel/probes/decode-insn.c   |  2 +-
>  arch/riscv/kernel/probes/simulate-insn.c | 34 ++++++++++++++++++++++++
>  2 files changed, 35 insertions(+), 1 deletion(-)
>
> diff --git a/arch/riscv/kernel/probes/decode-insn.c b/arch/riscv/kernel/probes/decode-insn.c
> index 0ed043acc882..5eb03fb61450 100644
> --- a/arch/riscv/kernel/probes/decode-insn.c
> +++ b/arch/riscv/kernel/probes/decode-insn.c
> @@ -38,11 +38,11 @@ riscv_probe_decode_insn(probe_opcode_t *addr, struct arch_probe_insn *api)
>  	RISCV_INSN_REJECTED(c_ebreak,		insn);
>  #endif
>
> -	RISCV_INSN_REJECTED(auipc,		insn);
>  	RISCV_INSN_REJECTED(branch,		insn);
>
>  	RISCV_INSN_SET_SIMULATE(jal,		insn);
>  	RISCV_INSN_SET_SIMULATE(jalr,		insn);
> +	RISCV_INSN_SET_SIMULATE(auipc,		insn);
>
>  	return INSN_GOOD;
>  }
> diff --git a/arch/riscv/kernel/probes/simulate-insn.c b/arch/riscv/kernel/probes/simulate-insn.c
> index 2519ce26377d..b81719522d5c 100644
> --- a/arch/riscv/kernel/probes/simulate-insn.c
> +++ b/arch/riscv/kernel/probes/simulate-insn.c
> @@ -83,3 +83,37 @@ bool __kprobes simulate_jalr(u32 opcode, unsigned long addr, struct pt_regs *reg
>
>  	return ret;
>  }
> +
> +#define auipc_rd_idx(opcode) \
> +	((opcode >> 7) & 0x1f)
> +
> +#define auipc_imm(opcode) \
> +	((((opcode) >> 12) & 0xfffff) << 12)
> +
> +#if __riscv_xlen == 64
> +#define auipc_offset(opcode)	sign_extend64(auipc_imm(opcode), 31)
> +#elif __riscv_xlen == 32
> +#define auipc_offset(opcode)	auipc_imm(opcode)
> +#else
> +#error "Unexpected __riscv_xlen"
> +#endif
> +
> +bool __kprobes simulate_auipc(u32 opcode, unsigned long addr, struct pt_regs *regs)
> +{
> +	/*
> +	 * auipc instruction:
> +	 *  31        12 11 7 6      0
> +	 * | imm[31:12] | rd | opcode |
> +	 *        20       5     7
> +	 */
> +
> +	u32 rd_idx = auipc_rd_idx(opcode);
> +	unsigned long rd_val = addr + auipc_offset(opcode);
> +
> +	if (!rv_insn_reg_set_val(regs, rd_idx, rd_val))
> +		return false;
> +
> +	instruction_pointer_set(regs, addr + 4);
> +
> +	return true;
> +}

Thanks.  These are on for-next, with the checkpatch errors fixed and the 
commit messages mostly removed -- it was all pretty awkwardly phrased so 
I didn't want to clean it up.

_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

^ permalink raw reply	[flat|nested] 10+ messages in thread

end of thread, other threads:[~2021-07-22 17:27 UTC | newest]

Thread overview: 10+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-06-29  2:34 [PATCH -next 1/2] riscv: implemented auipc simulate instruction Chen Lifu
2021-06-29  2:34 ` Chen Lifu
2021-06-29  2:34 ` [PATCH -next 2/2] riscv: implemented branch simulate instructions Chen Lifu
2021-06-29  2:34   ` Chen Lifu
2021-07-05  2:09   ` ping//Re: " chenlifu
2021-07-05  2:09     ` chenlifu
2021-07-22 10:15     ` ping//Re: " chenlifu
2021-07-22 10:15       ` chenlifu
2021-07-22 17:26 ` [PATCH -next 1/2] riscv: implemented auipc simulate instruction Palmer Dabbelt
2021-07-22 17:26   ` Palmer Dabbelt

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.