Message-ID: <17dfb122-4ae3-477f-93a8-1748b819c59c@coelacanthus.name>
Date: Wed, 15 Jan 2025 17:41:57 +0800
From: Celeste Liu <uwu@...lacanthus.name>
To: Andrew Jones <ajones@...tanamicro.com>
Cc: Oleg Nesterov <oleg@...hat.com>, Paul Walmsley
 <paul.walmsley@...ive.com>, Palmer Dabbelt <palmer@...belt.com>,
 Eric Biederman <ebiederm@...ssion.com>, Kees Cook <kees@...nel.org>,
 Shuah Khan <shuah@...nel.org>, Albert Ou <aou@...s.berkeley.edu>,
 Alexandre Ghiti <alex@...ti.fr>, "Dmitry V. Levin" <ldv@...ace.io>,
 Andrea Bolognani <abologna@...hat.com>, Björn Töpel
 <bjorn@...nel.org>, Thomas Gleixner <tglx@...utronix.de>,
 Ron Economos <re@...z.net>, Charlie Jenkins <charlie@...osinc.com>,
 Quan Zhou <zhouquan@...as.ac.cn>, Felix Yan <felixonmars@...hlinux.org>,
 Ruizhe Pan <c141028@...il.com>, Guo Ren <guoren@...nel.org>,
 Yao Zi <ziyao@...root.org>, linux-riscv@...ts.infradead.org,
 linux-kernel@...r.kernel.org, linux-mm@...ck.org,
 linux-kselftest@...r.kernel.org, Björn Töpel
 <bjorn@...osinc.com>
Subject: Re: [PATCH v5 2/2] riscv: selftests: Add a ptrace test to verify a0
 and orig_a0 access

On 2025-01-15 17:14, Andrew Jones wrote:
> On Wed, Jan 15, 2025 at 04:24:59AM +0800, Celeste Liu wrote:
>> This test checks that orig_a0 and a0 can be modified and accessed
>> independently.
>>
>> Co-developed-by: Quan Zhou <zhouquan@...as.ac.cn>
>> Signed-off-by: Quan Zhou <zhouquan@...as.ac.cn>
>> Co-developed-by: Charlie Jenkins <charlie@...osinc.com>
>> Signed-off-by: Charlie Jenkins <charlie@...osinc.com>
>> Reviewed-by: Björn Töpel <bjorn@...osinc.com>
>> Signed-off-by: Celeste Liu <uwu@...lacanthus.name>
>> ---
>>  tools/testing/selftests/riscv/abi/.gitignore |   1 +
>>  tools/testing/selftests/riscv/abi/Makefile   |   6 +-
>>  tools/testing/selftests/riscv/abi/ptrace.c   | 201 +++++++++++++++++++++++++++
>>  3 files changed, 207 insertions(+), 1 deletion(-)
>>
>> diff --git a/tools/testing/selftests/riscv/abi/.gitignore b/tools/testing/selftests/riscv/abi/.gitignore
>> index b38358f91c4d2240ae64892871d9ca98bda1ae58..378c605919a3b3d58eec2701eb7af430cfe315d6 100644
>> --- a/tools/testing/selftests/riscv/abi/.gitignore
>> +++ b/tools/testing/selftests/riscv/abi/.gitignore
>> @@ -1 +1,2 @@
>>  pointer_masking
>> +ptrace
>> diff --git a/tools/testing/selftests/riscv/abi/Makefile b/tools/testing/selftests/riscv/abi/Makefile
>> index ed82ff9c664e7eb3f760cbab81fb957ff72579c5..359a082c88a401883fb3776b35e4dacf69beaaaa 100644
>> --- a/tools/testing/selftests/riscv/abi/Makefile
>> +++ b/tools/testing/selftests/riscv/abi/Makefile
>> @@ -1,10 +1,14 @@
>>  # SPDX-License-Identifier: GPL-2.0
>>  
>>  CFLAGS += -I$(top_srcdir)/tools/include
>> +CFLAGS += $(KHDR_INCLUDES)
>>  
>> -TEST_GEN_PROGS := pointer_masking
>> +TEST_GEN_PROGS := pointer_masking ptrace
>>  
>>  include ../../lib.mk
>>  
>>  $(OUTPUT)/pointer_masking: pointer_masking.c
>>  	$(CC) -static -o$@ $(CFLAGS) $(LDFLAGS) $^
>> +
>> +$(OUTPUT)/ptrace: ptrace.c
>> +	$(CC) -static -o$@ $(CFLAGS) $(LDFLAGS) $^
>> diff --git a/tools/testing/selftests/riscv/abi/ptrace.c b/tools/testing/selftests/riscv/abi/ptrace.c
>> new file mode 100644
>> index 0000000000000000000000000000000000000000..f1a0458adccdd040bfaa350e2e8d98b1ef34c0ad
>> --- /dev/null
>> +++ b/tools/testing/selftests/riscv/abi/ptrace.c
>> @@ -0,0 +1,201 @@
>> +// SPDX-License-Identifier: GPL-2.0-only
>> +#include <stdio.h>
>> +#include <stdlib.h>
>> +#include <string.h>
>> +#include <unistd.h>
>> +#include <fcntl.h>
>> +#include <signal.h>
>> +#include <errno.h>
>> +#include <sys/types.h>
>> +#include <sys/ptrace.h>
>> +#include <sys/stat.h>
>> +#include <sys/user.h>
>> +#include <sys/wait.h>
>> +#include <sys/uio.h>
>> +#include <linux/elf.h>
>> +#include <linux/unistd.h>
>> +#include <linux/ptrace.h>
>> +#include <asm/ptrace.h>
>> +
>> +#include "../../kselftest_harness.h"
>> +
>> +#ifndef sizeof_field
>> +#define sizeof_field(TYPE, MEMBER) sizeof((((TYPE *)0)->MEMBER))
>> +#endif
>> +#ifndef offsetofend
>> +#define offsetofend(TYPE, MEMBER) \
>> +	(offsetof(TYPE, MEMBER) + sizeof_field(TYPE, MEMBER))
>> +#endif
> 
> I think this is the sixth test to define these. We should copy
> include/linux/stddef.h into tools/include. We already have
> tools/include/uapi/linux/stddef.h with __struct_group and
> __DECLARE_FLEX_ARRAY, so I think it should just work.

Agreed. But it may be better to be a separate patchset
so we can change those definition in different selftests
one pass.

> 
>> +
>> +
>> +#define ORIG_A0_MODIFY      0x01
>> +#define A0_MODIFY           0x02
>> +#define A0_OLD              0xbadbeefbeeff
>> +#define A0_NEW              0xffeebfeebdab
>> +
>> +
>> +struct a0_regs {
>> +	__s64 orig_a0;
>> +	__u64 a0;
>> +};
>> +
>> +#define perr_and_exit(fmt, ...)						\
>> +	({								\
>> +		char buf[256];						\
>> +		snprintf(buf, sizeof(buf), "%s:%d:" fmt ": %m\n",	\
>> +			__func__, __LINE__, ##__VA_ARGS__);		\
>> +		ksft_exit_fail_perror(buf);				\
>> +	})
>> +
>> +static void ptrace_test(int opt, struct a0_regs result[])
>> +{
>> +	int status;
>> +	long rc;
>> +	pid_t pid;
>> +	struct user_regs_struct regs;
>> +	struct iovec iov = {
>> +		.iov_base = &regs,
>> +		.iov_len = sizeof(regs),
>> +	};
>> +
>> +	unsigned long orig_a0;
>> +	struct iovec a0_iov = {
>> +		.iov_base = &orig_a0,
>> +		.iov_len = sizeof(orig_a0),
>> +	};
>> +	struct ptrace_syscall_info syscall_info = {
>> +		.op = 0xff,
>> +	};
>> +	const unsigned int expected_sci_entry_size =
>> +		offsetofend(struct ptrace_syscall_info, entry.args);
>> +	const unsigned int expected_sci_exit_size =
>> +		offsetofend(struct ptrace_syscall_info, exit.is_error);
>> +
>> +	pid = fork();
>> +	if (pid == 0) {
>> +		/* Mark oneself being traced */
>> +		long val = ptrace(PTRACE_TRACEME, 0, 0, 0);
>> +
>> +		if (val < 0)
>> +			perr_and_exit("failed to request for tracer to trace me: %ld", val);
>> +
>> +		kill(getpid(), SIGSTOP);
>> +
>> +		/* Perform chdir syscall that will be intercepted */
>> +		syscall(__NR_chdir, A0_OLD);
>> +
>> +		exit(0);
>> +	}
>> +
>> +	if (pid < 0)
>> +		ksft_exit_fail_perror("failed to fork");
>> +
>> +	for (int i = 0; i < 3; i++) {
>> +		if (waitpid(pid, &status, 0) != pid)
>> +			perr_and_exit("failed to wait for the tracee %d", pid);
>> +		if (WIFSTOPPED(status)) {
>> +			switch (WSTOPSIG(status)) {
>> +			case SIGSTOP:
>> +				if (ptrace(PTRACE_SETOPTIONS, pid, 0, PTRACE_O_TRACESYSGOOD) < 0)
>> +					perr_and_exit("failed to set PTRACE_O_TRACESYSGOOD");
>> +				break;
>> +			case SIGTRAP|0x80:
>> +				/* Modify twice so GET_SYSCALL_INFO get modified a0 and orig_a0 */
>> +				if (ptrace(PTRACE_GETREGSET, pid, NT_PRSTATUS, &iov))
>> +					perr_and_exit("failed to get tracee registers");
>> +				if (ptrace(PTRACE_GETREGSET, pid, NT_RISCV_ORIG_A0, &a0_iov))
>> +					perr_and_exit("failed to get tracee registers");
>> +
>> +				/* Modify a0/orig_a0 for the syscall */
>> +				switch (opt) {
>> +				case A0_MODIFY:
>> +					regs.a0 = A0_NEW;
>> +					break;
>> +				case ORIG_A0_MODIFY:
>> +					orig_a0 = A0_NEW;
>> +					break;
>> +				}
>> +
>> +				if (ptrace(PTRACE_SETREGSET, pid, NT_PRSTATUS, &iov))
>> +					perr_and_exit("failed to set tracee registers");
>> +				if (ptrace(PTRACE_SETREGSET, pid, NT_RISCV_ORIG_A0, &a0_iov))
>> +					perr_and_exit("failed to set tracee registers");
>> +				switch (i) {
>> +				case 1:
>> +					/* Stop at the beginning of syscall */
>> +					rc = ptrace(PTRACE_GET_SYSCALL_INFO, pid,
>> +						sizeof(syscall_info), &syscall_info);
>> +					if (rc < 0)
>> +						perr_and_exit("failed to get syscall info of entry");
>> +					if (rc < expected_sci_entry_size
>> +						|| syscall_info.op != PTRACE_SYSCALL_INFO_ENTRY)
>> +						perr_and_exit("stop position of entry mismatched");
>> +					result[0].orig_a0 = syscall_info.entry.args[0];
>> +					break;
>> +
>> +				case 2:
>> +					/* Stop at the end of syscall */
>> +					rc = ptrace(PTRACE_GET_SYSCALL_INFO, pid,
>> +						sizeof(syscall_info), &syscall_info);
>> +					if (rc < 0)
>> +						perr_and_exit("failed to get syscall info of entry");
>> +					if (rc < expected_sci_exit_size
>> +						|| syscall_info.op != PTRACE_SYSCALL_INFO_EXIT)
>> +						perr_and_exit("stop position of exit mismatched");
>> +					result[0].a0 = syscall_info.exit.rval;
>> +
>> +					if (ptrace(PTRACE_GETREGSET, pid, NT_PRSTATUS, &iov))
>> +						perr_and_exit("failed to get tracee registers");
>> +					result[1].a0 = regs.a0;
>> +					if (ptrace(PTRACE_GETREGSET, pid, NT_RISCV_ORIG_A0,
>> +						   &a0_iov))
>> +						perr_and_exit("failed to get tracee registers");
>> +					result[1].orig_a0 = orig_a0;
>> +				}
>> +			}
>> +			if (ptrace(PTRACE_SYSCALL, pid, 0, 0) < 0)
>> +				perr_and_exit("failed to resume tracee");
>> +		}
>> +	}
>> +
>> +	/* Resume the tracee */
>> +	ptrace(PTRACE_CONT, pid, 0, 0);
>> +	if (waitpid(pid, &status, 0) != pid)
>> +		perr_and_exit("failed to wait for the tracee");
>> +
>> +}
>> +
>> +TEST(ptrace_access_a0)
>> +{
>> +	struct a0_regs result[2];
>> +
>> +	ptrace_test(A0_MODIFY, result);
>> +
>> +	/* Verify PTRACE_SETREGSET */
>> +	/* The modification of a0 cannot affect the first argument of the syscall */
>> +	EXPECT_EQ(A0_OLD, result[0].orig_a0);
>> +	EXPECT_EQ(A0_NEW, result[0].a0);
>> +
>> +	/* Verify PTRACE_GETREGSET */
>> +	EXPECT_EQ(result[1].orig_a0, result[0].orig_a0);
>> +	EXPECT_EQ(result[1].a0, result[0].a0);
>> +}
>> +
>> +TEST(ptrace_access_orig_a0)
>> +{
>> +	struct a0_regs result[2];
>> +
>> +	ptrace_test(ORIG_A0_MODIFY, result);
>> +
>> +	/* Verify PTRACE_SETREGSET */
>> +	/* Only modify orig_a0 to change the first argument of the syscall */
>> +	EXPECT_EQ(A0_NEW, result[0].orig_a0);
>> +	/* a0 will keep default value, orig_a0 or -ENOSYS, depends on internal. */
>> +	EXPECT_NE(A0_NEW, result[0].a0);
> 
> I don't understand this test. Why don't we know what to expect? Also, the
> comment says orig_a0 is an option for the value, but then we don't expect
> it to be A0_NEW, even though we expect orig_a0 to be A0_NEW?

The purpose of the test is to ensure that the ORIG_A0_MODIFY operation
will not affect the a0 register (So it will not be our A0_NEW). But there
is a problem with the comment. It is written for some old wrong code.
I will correct the comment.

> 
>> +
>> +	/* Verify PTRACE_GETREGSET */
>> +	EXPECT_EQ(result[1].orig_a0, result[0].orig_a0);
>> +	EXPECT_EQ(result[1].a0, result[0].a0);
>> +}
>> +
>> +TEST_HARNESS_MAIN
>>
>> -- 
>> 2.48.0
>>
> 
> Other than the two comments/questions, this test looks great.
> 
> Thanks,
> drew

Powered by blists - more mailing lists