* [PATCH 03/15] selftests/seccomp: mips: Define SYSCALL_NUM_SET macro
From: Kees Cook @ 2020-09-12 11:08 UTC (permalink / raw)
To: linux-kernel
Cc: Thadeu Lima de Souza Cascardo, Will Drewry, Kees Cook,
linux-xtensa, linux-mips, Andy Lutomirski, Max Filippov,
linux-arm-kernel, linux-kselftest, linuxppc-dev,
Christian Brauner
In-Reply-To: <20200912110820.597135-1-keescook@chromium.org>
Remove the mips special-case in change_syscall().
Signed-off-by: Kees Cook <keescook@chromium.org>
---
tools/testing/selftests/seccomp/seccomp_bpf.c | 17 +++++++++--------
1 file changed, 9 insertions(+), 8 deletions(-)
diff --git a/tools/testing/selftests/seccomp/seccomp_bpf.c b/tools/testing/selftests/seccomp/seccomp_bpf.c
index 1c83e743bfb1..02a9a6599746 100644
--- a/tools/testing/selftests/seccomp/seccomp_bpf.c
+++ b/tools/testing/selftests/seccomp/seccomp_bpf.c
@@ -1742,6 +1742,13 @@ TEST_F(TRACE_poke, getpid_runs_normally)
# define ARCH_REGS struct pt_regs
# define SYSCALL_NUM(_regs) (_regs).regs[2]
# define SYSCALL_SYSCALL_NUM regs[4]
+# define SYSCALL_NUM_SET(_regs, _nr) \
+ do { \
+ if ((_regs).regs[2] == __NR_O32_Linux) \
+ (_regs).regs[4] = _nr; \
+ else \
+ (_regs).regs[2] = _nr; \
+ } while (0)
# define SYSCALL_RET(_regs) (_regs).regs[2]
# define SYSCALL_NUM_RET_SHARE_REG
#elif defined(__xtensa__)
@@ -1839,17 +1846,11 @@ void change_syscall(struct __test_metadata *_metadata,
#if defined(__x86_64__) || defined(__i386__) || defined(__powerpc__) || \
defined(__s390__) || defined(__hppa__) || defined(__riscv) || \
- defined(__xtensa__) || defined(__csky__) || defined(__sh__)
+ defined(__xtensa__) || defined(__csky__) || defined(__sh__) || \
+ defined(__mips__)
{
SYSCALL_NUM_SET(regs, syscall);
}
-#elif defined(__mips__)
- {
- if (SYSCALL_NUM(regs) == __NR_O32_Linux)
- regs.SYSCALL_SYSCALL_NUM = syscall;
- else
- SYSCALL_NUM_SET(regs, syscall);
- }
#elif defined(__arm__)
# ifndef PTRACE_SET_SYSCALL
--
2.25.1
^ permalink raw reply related
* [PATCH 01/15] selftests/seccomp: Refactor arch register macros to avoid xtensa special case
From: Kees Cook @ 2020-09-12 11:08 UTC (permalink / raw)
To: linux-kernel
Cc: Thadeu Lima de Souza Cascardo, Will Drewry, Kees Cook,
linux-xtensa, linux-mips, Andy Lutomirski, Max Filippov,
linux-arm-kernel, linux-kselftest, linuxppc-dev,
Christian Brauner
In-Reply-To: <20200912110820.597135-1-keescook@chromium.org>
To avoid an xtensa special-case, refactor all arch register macros to
take the register variable instead of depending on the macro expanding
as a struct member name.
Signed-off-by: Kees Cook <keescook@chromium.org>
---
tools/testing/selftests/seccomp/seccomp_bpf.c | 97 +++++++++----------
1 file changed, 47 insertions(+), 50 deletions(-)
diff --git a/tools/testing/selftests/seccomp/seccomp_bpf.c b/tools/testing/selftests/seccomp/seccomp_bpf.c
index c5002fc25b00..fef15080b575 100644
--- a/tools/testing/selftests/seccomp/seccomp_bpf.c
+++ b/tools/testing/selftests/seccomp/seccomp_bpf.c
@@ -1698,64 +1698,64 @@ TEST_F(TRACE_poke, getpid_runs_normally)
}
#if defined(__x86_64__)
-# define ARCH_REGS struct user_regs_struct
-# define SYSCALL_NUM orig_rax
-# define SYSCALL_RET rax
+# define ARCH_REGS struct user_regs_struct
+# define SYSCALL_NUM(_regs) (_regs).orig_rax
+# define SYSCALL_RET(_regs) (_regs).rax
#elif defined(__i386__)
-# define ARCH_REGS struct user_regs_struct
-# define SYSCALL_NUM orig_eax
-# define SYSCALL_RET eax
+# define ARCH_REGS struct user_regs_struct
+# define SYSCALL_NUM(_regs) (_regs).orig_eax
+# define SYSCALL_RET(_regs) (_regs).eax
#elif defined(__arm__)
-# define ARCH_REGS struct pt_regs
-# define SYSCALL_NUM ARM_r7
-# define SYSCALL_RET ARM_r0
+# define ARCH_REGS struct pt_regs
+# define SYSCALL_NUM(_regs) (_regs).ARM_r7
+# define SYSCALL_RET(_regs) (_regs).ARM_r0
#elif defined(__aarch64__)
-# define ARCH_REGS struct user_pt_regs
-# define SYSCALL_NUM regs[8]
-# define SYSCALL_RET regs[0]
+# define ARCH_REGS struct user_pt_regs
+# define SYSCALL_NUM(_regs) (_regs).regs[8]
+# define SYSCALL_RET(_regs) (_regs).regs[0]
#elif defined(__riscv) && __riscv_xlen == 64
-# define ARCH_REGS struct user_regs_struct
-# define SYSCALL_NUM a7
-# define SYSCALL_RET a0
+# define ARCH_REGS struct user_regs_struct
+# define SYSCALL_NUM(_regs) (_regs).a7
+# define SYSCALL_RET(_regs) (_regs).a0
#elif defined(__csky__)
-# define ARCH_REGS struct pt_regs
-#if defined(__CSKYABIV2__)
-# define SYSCALL_NUM regs[3]
-#else
-# define SYSCALL_NUM regs[9]
-#endif
-# define SYSCALL_RET a0
+# define ARCH_REGS struct pt_regs
+# if defined(__CSKYABIV2__)
+# define SYSCALL_NUM(_regs) (_regs).regs[3]
+# else
+# define SYSCALL_NUM(_regs) (_regs).regs[9]
+# endif
+# define SYSCALL_RET(_regs) (_regs).a0
#elif defined(__hppa__)
-# define ARCH_REGS struct user_regs_struct
-# define SYSCALL_NUM gr[20]
-# define SYSCALL_RET gr[28]
+# define ARCH_REGS struct user_regs_struct
+# define SYSCALL_NUM(_regs) (_regs).gr[20]
+# define SYSCALL_RET(_regs) (_regs).gr[28]
#elif defined(__powerpc__)
-# define ARCH_REGS struct pt_regs
-# define SYSCALL_NUM gpr[0]
-# define SYSCALL_RET gpr[3]
+# define ARCH_REGS struct pt_regs
+# define SYSCALL_NUM(_regs) (_regs).gpr[0]
+# define SYSCALL_RET(_regs) (_regs).gpr[3]
#elif defined(__s390__)
-# define ARCH_REGS s390_regs
-# define SYSCALL_NUM gprs[2]
-# define SYSCALL_RET gprs[2]
+# define ARCH_REGS s390_regs
+# define SYSCALL_NUM(_regs) (_regs).gprs[2]
+# define SYSCALL_RET(_regs) (_regs).gprs[2]
# define SYSCALL_NUM_RET_SHARE_REG
#elif defined(__mips__)
-# define ARCH_REGS struct pt_regs
-# define SYSCALL_NUM regs[2]
-# define SYSCALL_SYSCALL_NUM regs[4]
-# define SYSCALL_RET regs[2]
+# define ARCH_REGS struct pt_regs
+# define SYSCALL_NUM(_regs) (_regs).regs[2]
+# define SYSCALL_SYSCALL_NUM regs[4]
+# define SYSCALL_RET(_regs) (_regs).regs[2]
# define SYSCALL_NUM_RET_SHARE_REG
#elif defined(__xtensa__)
-# define ARCH_REGS struct user_pt_regs
-# define SYSCALL_NUM syscall
+# define ARCH_REGS struct user_pt_regs
+# define SYSCALL_NUM(_regs) (_regs).syscall
/*
* On xtensa syscall return value is in the register
* a2 of the current window which is not fixed.
*/
-#define SYSCALL_RET(reg) a[(reg).windowbase * 4 + 2]
+#define SYSCALL_RET(_regs) (_regs).a[(_regs).windowbase * 4 + 2]
#elif defined(__sh__)
-# define ARCH_REGS struct pt_regs
-# define SYSCALL_NUM gpr[3]
-# define SYSCALL_RET gpr[0]
+# define ARCH_REGS struct pt_regs
+# define SYSCALL_NUM(_regs) (_regs).gpr[3]
+# define SYSCALL_RET(_regs) (_regs).gpr[0]
#else
# error "Do not know how to find your architecture's registers and syscalls"
#endif
@@ -1804,10 +1804,10 @@ int get_syscall(struct __test_metadata *_metadata, pid_t tracee)
#endif
#if defined(__mips__)
- if (regs.SYSCALL_NUM == __NR_O32_Linux)
+ if (SYSCALL_NUM(regs) == __NR_O32_Linux)
return regs.SYSCALL_SYSCALL_NUM;
#endif
- return regs.SYSCALL_NUM;
+ return SYSCALL_NUM(regs);
}
/* Architecture-specific syscall changing routine. */
@@ -1830,14 +1830,14 @@ void change_syscall(struct __test_metadata *_metadata,
defined(__s390__) || defined(__hppa__) || defined(__riscv) || \
defined(__xtensa__) || defined(__csky__) || defined(__sh__)
{
- regs.SYSCALL_NUM = syscall;
+ SYSCALL_NUM(regs) = syscall;
}
#elif defined(__mips__)
{
- if (regs.SYSCALL_NUM == __NR_O32_Linux)
+ if (SYSCALL_NUM(regs) == __NR_O32_Linux)
regs.SYSCALL_SYSCALL_NUM = syscall;
else
- regs.SYSCALL_NUM = syscall;
+ SYSCALL_NUM(regs) = syscall;
}
#elif defined(__arm__)
@@ -1871,11 +1871,8 @@ void change_syscall(struct __test_metadata *_metadata,
if (syscall == -1)
#ifdef SYSCALL_NUM_RET_SHARE_REG
TH_LOG("Can't modify syscall return on this architecture");
-
-#elif defined(__xtensa__)
- regs.SYSCALL_RET(regs) = result;
#else
- regs.SYSCALL_RET = result;
+ SYSCALL_RET(regs) = result;
#endif
#ifdef HAVE_GETREGS
--
2.25.1
^ permalink raw reply related
* [PATCH 00/15] selftests/seccomp: Refactor change_syscall()
From: Kees Cook @ 2020-09-12 11:08 UTC (permalink / raw)
To: linux-kernel
Cc: Thadeu Lima de Souza Cascardo, Will Drewry, Kees Cook,
linux-xtensa, linux-mips, Andy Lutomirski, Max Filippov,
linux-arm-kernel, linux-kselftest, linuxppc-dev,
Christian Brauner
Hi,
This refactors the seccomp selftest macros used in change_syscall(),
in an effort to remove special cases for mips, arm, arm64, and xtensa,
which paves the way for powerpc fixes.
I'm not entirely done testing, but all-arch build tests and x86_64
selftests pass. I'll be doing arm, arm64, and i386 selftests shortly,
but I currently don't have an easy way to check xtensa, mips, nor
powerpc. Any help there would be appreciated!
(FWIW, I expect to take these via the seccomp tree.)
Thanks,
-Kees
Kees Cook (15):
selftests/seccomp: Refactor arch register macros to avoid xtensa
special case
selftests/seccomp: Provide generic syscall setting macro
selftests/seccomp: mips: Define SYSCALL_NUM_SET macro
selftests/seccomp: arm: Define SYSCALL_NUM_SET macro
selftests/seccomp: arm64: Define SYSCALL_NUM_SET macro
selftests/seccomp: mips: Remove O32-specific macro
selftests/seccomp: Remove syscall setting #ifdefs
selftests/seccomp: Convert HAVE_GETREG into ARCH_GETREG/ARCH_SETREG
selftests/seccomp: Convert REGSET calls into ARCH_GETREG/ARCH_SETREG
selftests/seccomp: Avoid redundant register flushes
selftests/seccomp: Remove SYSCALL_NUM_RET_SHARE_REG in favor of
SYSCALL_RET_SET
selftests/seccomp: powerpc: Fix seccomp return value testing
selftests/seccomp: powerpc: Set syscall return during ptrace syscall
exit
selftests/clone3: Avoid OS-defined clone_args
selftests/seccomp: Use __NR_mknodat instead of __NR_mknod
.../selftests/clone3/clone3_selftests.h | 16 +-
tools/testing/selftests/seccomp/seccomp_bpf.c | 313 ++++++++++--------
2 files changed, 184 insertions(+), 145 deletions(-)
--
2.25.1
^ permalink raw reply
* [PATCH 02/15] selftests/seccomp: Provide generic syscall setting macro
From: Kees Cook @ 2020-09-12 11:08 UTC (permalink / raw)
To: linux-kernel
Cc: Thadeu Lima de Souza Cascardo, Will Drewry, Kees Cook,
linux-xtensa, linux-mips, Andy Lutomirski, Max Filippov,
linux-arm-kernel, linux-kselftest, linuxppc-dev,
Christian Brauner
In-Reply-To: <20200912110820.597135-1-keescook@chromium.org>
In order to avoid "#ifdef"s in the main function bodies, create a new
macro, SYSCALL_NUM_SET(), where arch-specific logic can live.
Signed-off-by: Kees Cook <keescook@chromium.org>
---
tools/testing/selftests/seccomp/seccomp_bpf.c | 15 +++++++++++++--
1 file changed, 13 insertions(+), 2 deletions(-)
diff --git a/tools/testing/selftests/seccomp/seccomp_bpf.c b/tools/testing/selftests/seccomp/seccomp_bpf.c
index fef15080b575..1c83e743bfb1 100644
--- a/tools/testing/selftests/seccomp/seccomp_bpf.c
+++ b/tools/testing/selftests/seccomp/seccomp_bpf.c
@@ -1760,6 +1760,17 @@ TEST_F(TRACE_poke, getpid_runs_normally)
# error "Do not know how to find your architecture's registers and syscalls"
#endif
+/*
+ * Most architectures can change the syscall by just updating the
+ * associated register. This is the default if not defined above.
+ */
+#ifndef SYSCALL_NUM_SET
+# define SYSCALL_NUM_SET(_regs, _nr) \
+ do { \
+ SYSCALL_NUM(_regs) = (_nr); \
+ } while (0)
+#endif
+
/* When the syscall return can't be changed, stub out the tests for it. */
#ifdef SYSCALL_NUM_RET_SHARE_REG
# define EXPECT_SYSCALL_RETURN(val, action) EXPECT_EQ(-1, action)
@@ -1830,14 +1841,14 @@ void change_syscall(struct __test_metadata *_metadata,
defined(__s390__) || defined(__hppa__) || defined(__riscv) || \
defined(__xtensa__) || defined(__csky__) || defined(__sh__)
{
- SYSCALL_NUM(regs) = syscall;
+ SYSCALL_NUM_SET(regs, syscall);
}
#elif defined(__mips__)
{
if (SYSCALL_NUM(regs) == __NR_O32_Linux)
regs.SYSCALL_SYSCALL_NUM = syscall;
else
- SYSCALL_NUM(regs) = syscall;
+ SYSCALL_NUM_SET(regs, syscall);
}
#elif defined(__arm__)
--
2.25.1
^ permalink raw reply related
* [PATCH 04/15] selftests/seccomp: arm: Define SYSCALL_NUM_SET macro
From: Kees Cook @ 2020-09-12 11:08 UTC (permalink / raw)
To: linux-kernel
Cc: Thadeu Lima de Souza Cascardo, Will Drewry, Kees Cook,
linux-xtensa, linux-mips, Andy Lutomirski, Max Filippov,
linux-arm-kernel, linux-kselftest, linuxppc-dev,
Christian Brauner
In-Reply-To: <20200912110820.597135-1-keescook@chromium.org>
Remove the arm special-case in change_syscall().
Signed-off-by: Kees Cook <keescook@chromium.org>
---
tools/testing/selftests/seccomp/seccomp_bpf.c | 16 ++++++----------
1 file changed, 6 insertions(+), 10 deletions(-)
diff --git a/tools/testing/selftests/seccomp/seccomp_bpf.c b/tools/testing/selftests/seccomp/seccomp_bpf.c
index 02a9a6599746..610fc036e374 100644
--- a/tools/testing/selftests/seccomp/seccomp_bpf.c
+++ b/tools/testing/selftests/seccomp/seccomp_bpf.c
@@ -1708,6 +1708,11 @@ TEST_F(TRACE_poke, getpid_runs_normally)
#elif defined(__arm__)
# define ARCH_REGS struct pt_regs
# define SYSCALL_NUM(_regs) (_regs).ARM_r7
+# ifndef PTRACE_SET_SYSCALL
+# define PTRACE_SET_SYSCALL 23
+# endif
+# define SYSCALL_NUM_SET(_regs, _nr) \
+ EXPECT_EQ(0, ptrace(PTRACE_SET_SYSCALL, tracee, NULL, _nr))
# define SYSCALL_RET(_regs) (_regs).ARM_r0
#elif defined(__aarch64__)
# define ARCH_REGS struct user_pt_regs
@@ -1847,20 +1852,11 @@ void change_syscall(struct __test_metadata *_metadata,
#if defined(__x86_64__) || defined(__i386__) || defined(__powerpc__) || \
defined(__s390__) || defined(__hppa__) || defined(__riscv) || \
defined(__xtensa__) || defined(__csky__) || defined(__sh__) || \
- defined(__mips__)
+ defined(__mips__) || defined(__arm__)
{
SYSCALL_NUM_SET(regs, syscall);
}
-#elif defined(__arm__)
-# ifndef PTRACE_SET_SYSCALL
-# define PTRACE_SET_SYSCALL 23
-# endif
- {
- ret = ptrace(PTRACE_SET_SYSCALL, tracee, NULL, syscall);
- EXPECT_EQ(0, ret);
- }
-
#elif defined(__aarch64__)
# ifndef NT_ARM_SYSTEM_CALL
# define NT_ARM_SYSTEM_CALL 0x404
--
2.25.1
^ permalink raw reply related
* [PATCH 05/15] selftests/seccomp: arm64: Define SYSCALL_NUM_SET macro
From: Kees Cook @ 2020-09-12 11:08 UTC (permalink / raw)
To: linux-kernel
Cc: Thadeu Lima de Souza Cascardo, Will Drewry, Kees Cook,
linux-xtensa, linux-mips, Andy Lutomirski, Max Filippov,
linux-arm-kernel, linux-kselftest, linuxppc-dev,
Christian Brauner
In-Reply-To: <20200912110820.597135-1-keescook@chromium.org>
Remove the arm64 special-case in change_syscall().
Signed-off-by: Kees Cook <keescook@chromium.org>
---
tools/testing/selftests/seccomp/seccomp_bpf.c | 27 +++++++++----------
1 file changed, 13 insertions(+), 14 deletions(-)
diff --git a/tools/testing/selftests/seccomp/seccomp_bpf.c b/tools/testing/selftests/seccomp/seccomp_bpf.c
index 610fc036e374..cfa606d96086 100644
--- a/tools/testing/selftests/seccomp/seccomp_bpf.c
+++ b/tools/testing/selftests/seccomp/seccomp_bpf.c
@@ -1717,6 +1717,18 @@ TEST_F(TRACE_poke, getpid_runs_normally)
#elif defined(__aarch64__)
# define ARCH_REGS struct user_pt_regs
# define SYSCALL_NUM(_regs) (_regs).regs[8]
+# ifndef NT_ARM_SYSTEM_CALL
+# define NT_ARM_SYSTEM_CALL 0x404
+# endif
+# define SYSCALL_NUM_SET(_regs, _nr) \
+ do { \
+ struct iovec __v; \
+ typeof(_nr) __nr = (_nr); \
+ __v.iov_base = &__nr; \
+ __v.iov_len = sizeof(__nr); \
+ EXPECT_EQ(0, ptrace(PTRACE_SETREGSET, tracee, \
+ NT_ARM_SYSTEM_CALL, &__v)); \
+ } while (0)
# define SYSCALL_RET(_regs) (_regs).regs[0]
#elif defined(__riscv) && __riscv_xlen == 64
# define ARCH_REGS struct user_regs_struct
@@ -1852,23 +1864,10 @@ void change_syscall(struct __test_metadata *_metadata,
#if defined(__x86_64__) || defined(__i386__) || defined(__powerpc__) || \
defined(__s390__) || defined(__hppa__) || defined(__riscv) || \
defined(__xtensa__) || defined(__csky__) || defined(__sh__) || \
- defined(__mips__) || defined(__arm__)
+ defined(__mips__) || defined(__arm__) || defined(__aarch64__)
{
SYSCALL_NUM_SET(regs, syscall);
}
-
-#elif defined(__aarch64__)
-# ifndef NT_ARM_SYSTEM_CALL
-# define NT_ARM_SYSTEM_CALL 0x404
-# endif
- {
- iov.iov_base = &syscall;
- iov.iov_len = sizeof(syscall);
- ret = ptrace(PTRACE_SETREGSET, tracee, NT_ARM_SYSTEM_CALL,
- &iov);
- EXPECT_EQ(0, ret);
- }
-
#else
ASSERT_EQ(1, 0) {
TH_LOG("How is the syscall changed on this architecture?");
--
2.25.1
^ permalink raw reply related
* [PATCH 06/15] selftests/seccomp: mips: Remove O32-specific macro
From: Kees Cook @ 2020-09-12 11:08 UTC (permalink / raw)
To: linux-kernel
Cc: Thadeu Lima de Souza Cascardo, Will Drewry, Kees Cook,
linux-xtensa, linux-mips, Andy Lutomirski, Max Filippov,
linux-arm-kernel, linux-kselftest, linuxppc-dev,
Christian Brauner
In-Reply-To: <20200912110820.597135-1-keescook@chromium.org>
Instead of having the mips O32 macro special-cased, pull the logic into
the SYSCALL_NUM() macro. Additionally include the ABI headers, since
these appear to have been missing, leaving __NR_O32_Linux undefined.
Signed-off-by: Kees Cook <keescook@chromium.org>
---
tools/testing/selftests/seccomp/seccomp_bpf.c | 18 ++++++++++++------
1 file changed, 12 insertions(+), 6 deletions(-)
diff --git a/tools/testing/selftests/seccomp/seccomp_bpf.c b/tools/testing/selftests/seccomp/seccomp_bpf.c
index cfa606d96086..aa1c224371d1 100644
--- a/tools/testing/selftests/seccomp/seccomp_bpf.c
+++ b/tools/testing/selftests/seccomp/seccomp_bpf.c
@@ -1756,9 +1756,19 @@ TEST_F(TRACE_poke, getpid_runs_normally)
# define SYSCALL_RET(_regs) (_regs).gprs[2]
# define SYSCALL_NUM_RET_SHARE_REG
#elif defined(__mips__)
+# include <asm/unistd_nr_n32.h>
+# include <asm/unistd_nr_n64.h>
+# include <asm/unistd_nr_o32.h>
# define ARCH_REGS struct pt_regs
-# define SYSCALL_NUM(_regs) (_regs).regs[2]
-# define SYSCALL_SYSCALL_NUM regs[4]
+# define SYSCALL_NUM(_regs) \
+ ({ \
+ typeof((_regs).regs[2]) _nr; \
+ if ((_regs).regs[2] == __NR_O32_Linux) \
+ _nr = (_regs).regs[4]; \
+ else \
+ _nr = (_regs).regs[2]; \
+ _nr; \
+ })
# define SYSCALL_NUM_SET(_regs, _nr) \
do { \
if ((_regs).regs[2] == __NR_O32_Linux) \
@@ -1838,10 +1848,6 @@ int get_syscall(struct __test_metadata *_metadata, pid_t tracee)
}
#endif
-#if defined(__mips__)
- if (SYSCALL_NUM(regs) == __NR_O32_Linux)
- return regs.SYSCALL_SYSCALL_NUM;
-#endif
return SYSCALL_NUM(regs);
}
--
2.25.1
^ permalink raw reply related
* [PATCH 07/15] selftests/seccomp: Remove syscall setting #ifdefs
From: Kees Cook @ 2020-09-12 11:08 UTC (permalink / raw)
To: linux-kernel
Cc: Thadeu Lima de Souza Cascardo, Will Drewry, Kees Cook,
linux-xtensa, linux-mips, Andy Lutomirski, Max Filippov,
linux-arm-kernel, linux-kselftest, linuxppc-dev,
Christian Brauner
In-Reply-To: <20200912110820.597135-1-keescook@chromium.org>
With all architectures now using the common SYSCALL_NUM_SET() macro, the
arch-specific #ifdef can be removed from change_syscall() itself.
Signed-off-by: Kees Cook <keescook@chromium.org>
---
tools/testing/selftests/seccomp/seccomp_bpf.c | 16 +++-------------
1 file changed, 3 insertions(+), 13 deletions(-)
diff --git a/tools/testing/selftests/seccomp/seccomp_bpf.c b/tools/testing/selftests/seccomp/seccomp_bpf.c
index aa1c224371d1..3b77bdbe7125 100644
--- a/tools/testing/selftests/seccomp/seccomp_bpf.c
+++ b/tools/testing/selftests/seccomp/seccomp_bpf.c
@@ -1865,20 +1865,9 @@ void change_syscall(struct __test_metadata *_metadata,
iov.iov_len = sizeof(regs);
ret = ptrace(PTRACE_GETREGSET, tracee, NT_PRSTATUS, &iov);
#endif
- EXPECT_EQ(0, ret) {}
+ EXPECT_EQ(0, ret);
-#if defined(__x86_64__) || defined(__i386__) || defined(__powerpc__) || \
- defined(__s390__) || defined(__hppa__) || defined(__riscv) || \
- defined(__xtensa__) || defined(__csky__) || defined(__sh__) || \
- defined(__mips__) || defined(__arm__) || defined(__aarch64__)
- {
- SYSCALL_NUM_SET(regs, syscall);
- }
-#else
- ASSERT_EQ(1, 0) {
- TH_LOG("How is the syscall changed on this architecture?");
- }
-#endif
+ SYSCALL_NUM_SET(regs, syscall);
/* If syscall is skipped, change return value. */
if (syscall == -1)
@@ -1888,6 +1877,7 @@ void change_syscall(struct __test_metadata *_metadata,
SYSCALL_RET(regs) = result;
#endif
+ /* Flush any register changes made. */
#ifdef HAVE_GETREGS
ret = ptrace(PTRACE_SETREGS, tracee, 0, ®s);
#else
--
2.25.1
^ permalink raw reply related
* [PATCH 08/15] selftests/seccomp: Convert HAVE_GETREG into ARCH_GETREG/ARCH_SETREG
From: Kees Cook @ 2020-09-12 11:08 UTC (permalink / raw)
To: linux-kernel
Cc: Thadeu Lima de Souza Cascardo, Will Drewry, Kees Cook,
linux-xtensa, linux-mips, Andy Lutomirski, Max Filippov,
linux-arm-kernel, linux-kselftest, linuxppc-dev,
Christian Brauner
In-Reply-To: <20200912110820.597135-1-keescook@chromium.org>
Instead of special-casing the get/set-registers routines, move the
HAVE_GETREG logic into the new ARCH_GETREG() and ARCH_SETREG() macros.
Signed-off-by: Kees Cook <keescook@chromium.org>
---
tools/testing/selftests/seccomp/seccomp_bpf.c | 27 ++++++++++---------
1 file changed, 15 insertions(+), 12 deletions(-)
diff --git a/tools/testing/selftests/seccomp/seccomp_bpf.c b/tools/testing/selftests/seccomp/seccomp_bpf.c
index 3b77bdbe7125..a986f2332327 100644
--- a/tools/testing/selftests/seccomp/seccomp_bpf.c
+++ b/tools/testing/selftests/seccomp/seccomp_bpf.c
@@ -1821,20 +1821,21 @@ TEST_F(TRACE_poke, getpid_runs_normally)
} while (0)
#endif
-/* Use PTRACE_GETREGS and PTRACE_SETREGS when available. This is useful for
+/*
+ * Use PTRACE_GETREGS and PTRACE_SETREGS when available. This is useful for
* architectures without HAVE_ARCH_TRACEHOOK (e.g. User-mode Linux).
*/
#if defined(__x86_64__) || defined(__i386__) || defined(__mips__)
-#define HAVE_GETREGS
+# define ARCH_GETREGS(_regs) ptrace(PTRACE_GETREGS, tracee, 0, &(_regs))
+# define ARCH_SETREGS(_regs) ptrace(PTRACE_SETREGS, tracee, 0, &(_regs))
#endif
/* Architecture-specific syscall fetching routine. */
int get_syscall(struct __test_metadata *_metadata, pid_t tracee)
{
ARCH_REGS regs;
-#ifdef HAVE_GETREGS
- EXPECT_EQ(0, ptrace(PTRACE_GETREGS, tracee, 0, ®s)) {
- TH_LOG("PTRACE_GETREGS failed");
+#ifdef ARCH_GETREGS
+ EXPECT_EQ(0, ARCH_GETREGS(regs)) {
return -1;
}
#else
@@ -1855,17 +1856,19 @@ int get_syscall(struct __test_metadata *_metadata, pid_t tracee)
void change_syscall(struct __test_metadata *_metadata,
pid_t tracee, int syscall, int result)
{
- int ret;
ARCH_REGS regs;
-#ifdef HAVE_GETREGS
- ret = ptrace(PTRACE_GETREGS, tracee, 0, ®s);
+#ifdef ARCH_GETREGS
+ EXPECT_EQ(0, ARCH_GETREGS(regs)) {
+ return;
+ }
#else
+ int ret;
struct iovec iov;
iov.iov_base = ®s;
iov.iov_len = sizeof(regs);
ret = ptrace(PTRACE_GETREGSET, tracee, NT_PRSTATUS, &iov);
-#endif
EXPECT_EQ(0, ret);
+#endif
SYSCALL_NUM_SET(regs, syscall);
@@ -1878,14 +1881,14 @@ void change_syscall(struct __test_metadata *_metadata,
#endif
/* Flush any register changes made. */
-#ifdef HAVE_GETREGS
- ret = ptrace(PTRACE_SETREGS, tracee, 0, ®s);
+#ifdef ARCH_SETREGS
+ EXPECT_EQ(0, ARCH_SETREGS(regs));
#else
iov.iov_base = ®s;
iov.iov_len = sizeof(regs);
ret = ptrace(PTRACE_SETREGSET, tracee, NT_PRSTATUS, &iov);
-#endif
EXPECT_EQ(0, ret);
+#endif
}
void tracer_seccomp(struct __test_metadata *_metadata, pid_t tracee,
--
2.25.1
^ permalink raw reply related
* [PATCH 09/15] selftests/seccomp: Convert REGSET calls into ARCH_GETREG/ARCH_SETREG
From: Kees Cook @ 2020-09-12 11:08 UTC (permalink / raw)
To: linux-kernel
Cc: Thadeu Lima de Souza Cascardo, Will Drewry, Kees Cook,
linux-xtensa, linux-mips, Andy Lutomirski, Max Filippov,
linux-arm-kernel, linux-kselftest, linuxppc-dev,
Christian Brauner
In-Reply-To: <20200912110820.597135-1-keescook@chromium.org>
Consolidate the REGSET logic into the new ARCH_GETREG() and
ARCH_SETREG() macros, avoiding more #ifdef code in function bodies.
Signed-off-by: Kees Cook <keescook@chromium.org>
---
tools/testing/selftests/seccomp/seccomp_bpf.c | 42 +++++++------------
1 file changed, 15 insertions(+), 27 deletions(-)
diff --git a/tools/testing/selftests/seccomp/seccomp_bpf.c b/tools/testing/selftests/seccomp/seccomp_bpf.c
index a986f2332327..d9346121b89b 100644
--- a/tools/testing/selftests/seccomp/seccomp_bpf.c
+++ b/tools/testing/selftests/seccomp/seccomp_bpf.c
@@ -1828,26 +1828,29 @@ TEST_F(TRACE_poke, getpid_runs_normally)
#if defined(__x86_64__) || defined(__i386__) || defined(__mips__)
# define ARCH_GETREGS(_regs) ptrace(PTRACE_GETREGS, tracee, 0, &(_regs))
# define ARCH_SETREGS(_regs) ptrace(PTRACE_SETREGS, tracee, 0, &(_regs))
+#else
+# define ARCH_GETREGS(_regs) ({ \
+ struct iovec __v; \
+ __v.iov_base = &(_regs); \
+ __v.iov_len = sizeof(_regs); \
+ ptrace(PTRACE_GETREGSET, tracee, NT_PRSTATUS, &__v); \
+ })
+# define ARCH_SETREGS(_regs) ({ \
+ struct iovec __v; \
+ __v.iov_base = &(_regs); \
+ __v.iov_len = sizeof(_regs); \
+ ptrace(PTRACE_SETREGSET, tracee, NT_PRSTATUS, &__v); \
+ })
#endif
/* Architecture-specific syscall fetching routine. */
int get_syscall(struct __test_metadata *_metadata, pid_t tracee)
{
ARCH_REGS regs;
-#ifdef ARCH_GETREGS
- EXPECT_EQ(0, ARCH_GETREGS(regs)) {
- return -1;
- }
-#else
- struct iovec iov;
- iov.iov_base = ®s;
- iov.iov_len = sizeof(regs);
- EXPECT_EQ(0, ptrace(PTRACE_GETREGSET, tracee, NT_PRSTATUS, &iov)) {
- TH_LOG("PTRACE_GETREGSET failed");
+ EXPECT_EQ(0, ARCH_GETREGS(regs)) {
return -1;
}
-#endif
return SYSCALL_NUM(regs);
}
@@ -1857,18 +1860,10 @@ void change_syscall(struct __test_metadata *_metadata,
pid_t tracee, int syscall, int result)
{
ARCH_REGS regs;
-#ifdef ARCH_GETREGS
+
EXPECT_EQ(0, ARCH_GETREGS(regs)) {
return;
}
-#else
- int ret;
- struct iovec iov;
- iov.iov_base = ®s;
- iov.iov_len = sizeof(regs);
- ret = ptrace(PTRACE_GETREGSET, tracee, NT_PRSTATUS, &iov);
- EXPECT_EQ(0, ret);
-#endif
SYSCALL_NUM_SET(regs, syscall);
@@ -1881,14 +1876,7 @@ void change_syscall(struct __test_metadata *_metadata,
#endif
/* Flush any register changes made. */
-#ifdef ARCH_SETREGS
EXPECT_EQ(0, ARCH_SETREGS(regs));
-#else
- iov.iov_base = ®s;
- iov.iov_len = sizeof(regs);
- ret = ptrace(PTRACE_SETREGSET, tracee, NT_PRSTATUS, &iov);
- EXPECT_EQ(0, ret);
-#endif
}
void tracer_seccomp(struct __test_metadata *_metadata, pid_t tracee,
--
2.25.1
^ permalink raw reply related
* [PATCH 10/15] selftests/seccomp: Avoid redundant register flushes
From: Kees Cook @ 2020-09-12 11:08 UTC (permalink / raw)
To: linux-kernel
Cc: Thadeu Lima de Souza Cascardo, Will Drewry, Kees Cook,
linux-xtensa, linux-mips, Andy Lutomirski, Max Filippov,
linux-arm-kernel, linux-kselftest, linuxppc-dev,
Christian Brauner
In-Reply-To: <20200912110820.597135-1-keescook@chromium.org>
When none of the registers have changed, don't flush them back. This can
happen if the architecture uses a non-register way to change the syscall
(e.g. arm64) , and a return value hasn't been written.
Signed-off-by: Kees Cook <keescook@chromium.org>
---
tools/testing/selftests/seccomp/seccomp_bpf.c | 6 ++++--
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/tools/testing/selftests/seccomp/seccomp_bpf.c b/tools/testing/selftests/seccomp/seccomp_bpf.c
index d9346121b89b..2790d9cd50f4 100644
--- a/tools/testing/selftests/seccomp/seccomp_bpf.c
+++ b/tools/testing/selftests/seccomp/seccomp_bpf.c
@@ -1859,11 +1859,12 @@ int get_syscall(struct __test_metadata *_metadata, pid_t tracee)
void change_syscall(struct __test_metadata *_metadata,
pid_t tracee, int syscall, int result)
{
- ARCH_REGS regs;
+ ARCH_REGS orig, regs;
EXPECT_EQ(0, ARCH_GETREGS(regs)) {
return;
}
+ orig = regs;
SYSCALL_NUM_SET(regs, syscall);
@@ -1876,7 +1877,8 @@ void change_syscall(struct __test_metadata *_metadata,
#endif
/* Flush any register changes made. */
- EXPECT_EQ(0, ARCH_SETREGS(regs));
+ if (memcmp(&orig, ®s, sizeof(orig)) != 0)
+ EXPECT_EQ(0, ARCH_SETREGS(regs));
}
void tracer_seccomp(struct __test_metadata *_metadata, pid_t tracee,
--
2.25.1
^ permalink raw reply related
* [PATCH 12/15] selftests/seccomp: powerpc: Fix seccomp return value testing
From: Kees Cook @ 2020-09-12 11:08 UTC (permalink / raw)
To: linux-kernel
Cc: Thadeu Lima de Souza Cascardo, Will Drewry, Kees Cook,
linux-xtensa, linux-mips, Andy Lutomirski, Max Filippov,
linux-arm-kernel, linux-kselftest, linuxppc-dev,
Christian Brauner
In-Reply-To: <20200912110820.597135-1-keescook@chromium.org>
On powerpc, the errno is not inverted, and depends on ccr.so being
set. Add this to a powerpc definition of SYSCALL_RET_SET().
Co-developed-by: Thadeu Lima de Souza Cascardo <cascardo@canonical.com>
Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@canonical.com>
Link: https://lore.kernel.org/linux-kselftest/20200911181012.171027-1-cascardo@canonical.com/
Fixes: 5d83c2b37d43 ("selftests/seccomp: Add powerpc support")
Signed-off-by: Kees Cook <keescook@chromium.org>
---
tools/testing/selftests/seccomp/seccomp_bpf.c | 15 +++++++++++++++
1 file changed, 15 insertions(+)
diff --git a/tools/testing/selftests/seccomp/seccomp_bpf.c b/tools/testing/selftests/seccomp/seccomp_bpf.c
index 623953a53032..bbab2420d708 100644
--- a/tools/testing/selftests/seccomp/seccomp_bpf.c
+++ b/tools/testing/selftests/seccomp/seccomp_bpf.c
@@ -1750,6 +1750,21 @@ TEST_F(TRACE_poke, getpid_runs_normally)
# define ARCH_REGS struct pt_regs
# define SYSCALL_NUM(_regs) (_regs).gpr[0]
# define SYSCALL_RET(_regs) (_regs).gpr[3]
+# define SYSCALL_RET_SET(_regs, _val) \
+ do { \
+ typeof(_val) _result = (_val); \
+ /* \
+ * A syscall error is signaled by CR0 SO bit \
+ * and the code is stored as a positive value. \
+ */ \
+ if (_result < 0) { \
+ SYSCALL_RET(_regs) = -result; \
+ (_regs).ccr |= 0x10000000; \
+ } else { \
+ SYSCALL_RET(_regs) = result; \
+ (_regs).ccr &= ~0x10000000; \
+ } \
+ } while (0)
#elif defined(__s390__)
# define ARCH_REGS s390_regs
# define SYSCALL_NUM(_regs) (_regs).gprs[2]
--
2.25.1
^ permalink raw reply related
* [PATCH 13/15] selftests/seccomp: powerpc: Set syscall return during ptrace syscall exit
From: Kees Cook @ 2020-09-12 11:08 UTC (permalink / raw)
To: linux-kernel
Cc: Thadeu Lima de Souza Cascardo, Will Drewry, Kees Cook,
linux-xtensa, linux-mips, Andy Lutomirski, Max Filippov,
linux-arm-kernel, linux-kselftest, linuxppc-dev,
Christian Brauner
In-Reply-To: <20200912110820.597135-1-keescook@chromium.org>
Some archs (like ppc) only support changing the return code during
syscall exit when ptrace is used. As the syscall number might not
be available anymore during syscall exit, it needs to be saved
during syscall enter. Adjust the ptrace tests to do this.
Reported-by: Thadeu Lima de Souza Cascardo <cascardo@canonical.com>
Suggested-by: Thadeu Lima de Souza Cascardo <cascardo@canonical.com>
Link: https://lore.kernel.org/linux-kselftest/20200911181012.171027-1-cascardo@canonical.com/
Fixes: 58d0a862f573 ("seccomp: add tests for ptrace hole")
Signed-off-by: Kees Cook <keescook@chromium.org>
---
tools/testing/selftests/seccomp/seccomp_bpf.c | 34 +++++++++++--------
1 file changed, 20 insertions(+), 14 deletions(-)
diff --git a/tools/testing/selftests/seccomp/seccomp_bpf.c b/tools/testing/selftests/seccomp/seccomp_bpf.c
index bbab2420d708..26c712c6a575 100644
--- a/tools/testing/selftests/seccomp/seccomp_bpf.c
+++ b/tools/testing/selftests/seccomp/seccomp_bpf.c
@@ -1949,12 +1949,19 @@ void tracer_seccomp(struct __test_metadata *_metadata, pid_t tracee,
}
+FIXTURE(TRACE_syscall) {
+ struct sock_fprog prog;
+ pid_t tracer, mytid, mypid, parent;
+ long syscall_nr;
+};
+
void tracer_ptrace(struct __test_metadata *_metadata, pid_t tracee,
int status, void *args)
{
- int ret, nr;
+ int ret;
unsigned long msg;
static bool entry;
+ FIXTURE_DATA(TRACE_syscall) *self = args;
/*
* The traditional way to tell PTRACE_SYSCALL entry/exit
@@ -1968,24 +1975,23 @@ void tracer_ptrace(struct __test_metadata *_metadata, pid_t tracee,
EXPECT_EQ(entry ? PTRACE_EVENTMSG_SYSCALL_ENTRY
: PTRACE_EVENTMSG_SYSCALL_EXIT, msg);
- if (!entry)
- return;
-
- nr = get_syscall(_metadata, tracee);
+ /*
+ * Some architectures only support setting return values during
+ * syscall exit under ptrace, and on exit the syscall number may
+ * no longer be available. Therefore, save it here, and call
+ * "change syscall and set return values" on both entry and exit.
+ */
+ if (entry)
+ self->syscall_nr = get_syscall(_metadata, tracee);
- if (nr == __NR_getpid)
+ if (self->syscall_nr == __NR_getpid)
change_syscall(_metadata, tracee, __NR_getppid, 0);
- if (nr == __NR_gettid)
+ if (self->syscall_nr == __NR_gettid)
change_syscall(_metadata, tracee, -1, 45000);
- if (nr == __NR_openat)
+ if (self->syscall_nr == __NR_openat)
change_syscall(_metadata, tracee, -1, -ESRCH);
}
-FIXTURE(TRACE_syscall) {
- struct sock_fprog prog;
- pid_t tracer, mytid, mypid, parent;
-};
-
FIXTURE_VARIANT(TRACE_syscall) {
/*
* All of the SECCOMP_RET_TRACE behaviors can be tested with either
@@ -2044,7 +2050,7 @@ FIXTURE_SETUP(TRACE_syscall)
self->tracer = setup_trace_fixture(_metadata,
variant->use_ptrace ? tracer_ptrace
: tracer_seccomp,
- NULL, variant->use_ptrace);
+ self, variant->use_ptrace);
ret = prctl(PR_SET_NO_NEW_PRIVS, 1, 0, 0, 0);
ASSERT_EQ(0, ret);
--
2.25.1
^ permalink raw reply related
* [PATCH 11/15] selftests/seccomp: Remove SYSCALL_NUM_RET_SHARE_REG in favor of SYSCALL_RET_SET
From: Kees Cook @ 2020-09-12 11:08 UTC (permalink / raw)
To: linux-kernel
Cc: Thadeu Lima de Souza Cascardo, Will Drewry, Kees Cook,
linux-xtensa, linux-mips, Andy Lutomirski, Max Filippov,
linux-arm-kernel, linux-kselftest, linuxppc-dev,
Christian Brauner
In-Reply-To: <20200912110820.597135-1-keescook@chromium.org>
Instead of special-casing the specific case of shared registers, create
a default SYSCALL_RET_SET() macro (mirroring SYSCALL_NUM_SET()), that
writes to the SYSCALL_RET register. For architectures that can't set the
return value (for whatever reason), they can define SYSCALL_RET_SET()
without an associated SYSCALL_RET() macro. This also paves the way for
architectures that need to do special things to set the return value
(e.g. powerpc).
Signed-off-by: Kees Cook <keescook@chromium.org>
---
tools/testing/selftests/seccomp/seccomp_bpf.c | 33 +++++++++++++------
1 file changed, 23 insertions(+), 10 deletions(-)
diff --git a/tools/testing/selftests/seccomp/seccomp_bpf.c b/tools/testing/selftests/seccomp/seccomp_bpf.c
index 2790d9cd50f4..623953a53032 100644
--- a/tools/testing/selftests/seccomp/seccomp_bpf.c
+++ b/tools/testing/selftests/seccomp/seccomp_bpf.c
@@ -1753,8 +1753,8 @@ TEST_F(TRACE_poke, getpid_runs_normally)
#elif defined(__s390__)
# define ARCH_REGS s390_regs
# define SYSCALL_NUM(_regs) (_regs).gprs[2]
-# define SYSCALL_RET(_regs) (_regs).gprs[2]
-# define SYSCALL_NUM_RET_SHARE_REG
+# define SYSCALL_RET_SET(_regs, _val) \
+ TH_LOG("Can't modify syscall return on this architecture")
#elif defined(__mips__)
# include <asm/unistd_nr_n32.h>
# include <asm/unistd_nr_n64.h>
@@ -1776,8 +1776,8 @@ TEST_F(TRACE_poke, getpid_runs_normally)
else \
(_regs).regs[2] = _nr; \
} while (0)
-# define SYSCALL_RET(_regs) (_regs).regs[2]
-# define SYSCALL_NUM_RET_SHARE_REG
+# define SYSCALL_RET_SET(_regs, _val) \
+ TH_LOG("Can't modify syscall return on this architecture")
#elif defined(__xtensa__)
# define ARCH_REGS struct user_pt_regs
# define SYSCALL_NUM(_regs) (_regs).syscall
@@ -1804,9 +1804,26 @@ TEST_F(TRACE_poke, getpid_runs_normally)
SYSCALL_NUM(_regs) = (_nr); \
} while (0)
#endif
+/*
+ * Most architectures can change the syscall return value by just
+ * writing to the SYSCALL_RET register. This is the default if not
+ * defined above. If an architecture cannot set the return value
+ * (for example when the syscall and return value register is
+ * shared), report it with TH_LOG() in an arch-specific definition
+ * of SYSCALL_RET_SET() above, and leave SYSCALL_RET undefined.
+ */
+#if !defined(SYSCALL_RET) && !defined(SYSCALL_RET_SET)
+# error "One of SYSCALL_RET or SYSCALL_RET_SET is needed for this arch"
+#endif
+#ifndef SYSCALL_RET_SET
+# define SYSCALL_RET_SET(_regs, _val) \
+ do { \
+ SYSCALL_RET(_regs) = (_val); \
+ } while (0)
+#endif
/* When the syscall return can't be changed, stub out the tests for it. */
-#ifdef SYSCALL_NUM_RET_SHARE_REG
+#ifndef SYSCALL_RET
# define EXPECT_SYSCALL_RETURN(val, action) EXPECT_EQ(-1, action)
#else
# define EXPECT_SYSCALL_RETURN(val, action) \
@@ -1870,11 +1887,7 @@ void change_syscall(struct __test_metadata *_metadata,
/* If syscall is skipped, change return value. */
if (syscall == -1)
-#ifdef SYSCALL_NUM_RET_SHARE_REG
- TH_LOG("Can't modify syscall return on this architecture");
-#else
- SYSCALL_RET(regs) = result;
-#endif
+ SYSCALL_RET_SET(regs, result);
/* Flush any register changes made. */
if (memcmp(&orig, ®s, sizeof(orig)) != 0)
--
2.25.1
^ permalink raw reply related
* [PATCH 14/15] selftests/clone3: Avoid OS-defined clone_args
From: Kees Cook @ 2020-09-12 11:08 UTC (permalink / raw)
To: linux-kernel
Cc: Thadeu Lima de Souza Cascardo, Will Drewry, Kees Cook,
linux-xtensa, linux-mips, Andy Lutomirski, Max Filippov,
linux-arm-kernel, linux-kselftest, linuxppc-dev,
Christian Brauner
In-Reply-To: <20200912110820.597135-1-keescook@chromium.org>
As the UAPI headers start to appear in distros, we need to avoid
outdated versions of struct clone_args to be able to test modern
features. Additionally pull in the syscall numbers correctly.
Signed-off-by: Kees Cook <keescook@chromium.org>
---
I needed to fix this to get MIPS to build the seccomp selftests.
---
.../testing/selftests/clone3/clone3_selftests.h | 16 ++++++++++++++--
1 file changed, 14 insertions(+), 2 deletions(-)
diff --git a/tools/testing/selftests/clone3/clone3_selftests.h b/tools/testing/selftests/clone3/clone3_selftests.h
index 91c1a78ddb39..bc0f34e37ae1 100644
--- a/tools/testing/selftests/clone3/clone3_selftests.h
+++ b/tools/testing/selftests/clone3/clone3_selftests.h
@@ -4,11 +4,19 @@
#define _CLONE3_SELFTESTS_H
#define _GNU_SOURCE
+
+/* Pull in syscall numbers. */
+#include <unistd.h>
+#include <sys/syscall.h>
+
+/* Avoid old OS versions of "struct clone_args". */
+#define clone_args old_clone_args
#include <sched.h>
#include <linux/sched.h>
+#undef clone_args
+
#include <linux/types.h>
#include <stdint.h>
-#include <syscall.h>
#include <sys/wait.h>
#include "../kselftest.h"
@@ -25,6 +33,7 @@
#ifndef __NR_clone3
#define __NR_clone3 -1
+#endif
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
@@ -34,13 +43,16 @@ struct clone_args {
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
+#ifndef CLONE_ARGS_SIZE_VER1
#define CLONE_ARGS_SIZE_VER1 80
+#endif
__aligned_u64 set_tid;
__aligned_u64 set_tid_size;
+#ifndef CLONE_ARGS_SIZE_VER2
#define CLONE_ARGS_SIZE_VER2 88
+#endif
__aligned_u64 cgroup;
};
-#endif /* __NR_clone3 */
static pid_t sys_clone3(struct clone_args *args, size_t size)
{
--
2.25.1
^ permalink raw reply related
* [PATCH 15/15] selftests/seccomp: Use __NR_mknodat instead of __NR_mknod
From: Kees Cook @ 2020-09-12 11:08 UTC (permalink / raw)
To: linux-kernel
Cc: Thadeu Lima de Souza Cascardo, Will Drewry, Kees Cook,
linux-xtensa, linux-mips, Andy Lutomirski, Max Filippov,
linux-arm-kernel, linux-kselftest, linuxppc-dev,
Christian Brauner
In-Reply-To: <20200912110820.597135-1-keescook@chromium.org>
The __NR_mknod syscall doesn't exist on arm64 (only __NR_mknodat).
Switch to the modern syscall.
Fixes: ad5682184a81 ("selftests/seccomp: Check for EPOLLHUP for user_notif")
Signed-off-by: Kees Cook <keescook@chromium.org>
---
tools/testing/selftests/seccomp/seccomp_bpf.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/testing/selftests/seccomp/seccomp_bpf.c b/tools/testing/selftests/seccomp/seccomp_bpf.c
index 26c712c6a575..b34ede28f314 100644
--- a/tools/testing/selftests/seccomp/seccomp_bpf.c
+++ b/tools/testing/selftests/seccomp/seccomp_bpf.c
@@ -3773,7 +3773,7 @@ TEST(user_notification_filter_empty)
if (pid == 0) {
int listener;
- listener = user_notif_syscall(__NR_mknod, SECCOMP_FILTER_FLAG_NEW_LISTENER);
+ listener = user_notif_syscall(__NR_mknodat, SECCOMP_FILTER_FLAG_NEW_LISTENER);
if (listener < 0)
_exit(EXIT_FAILURE);
--
2.25.1
^ permalink raw reply related
* Re: [PATCH v5 20/23] powerpc/book3s64/hash/kuap: Enable kuap on hash
From: Michael Ellerman @ 2020-09-12 11:54 UTC (permalink / raw)
To: Aneesh Kumar K.V, linuxppc-dev; +Cc: Aneesh Kumar K.V
In-Reply-To: <20200827040931.297759-21-aneesh.kumar@linux.ibm.com>
"Aneesh Kumar K.V" <aneesh.kumar@linux.ibm.com> writes:
> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
> ---
> arch/powerpc/mm/book3s64/pkeys.c | 7 ++++++-
> 1 file changed, 6 insertions(+), 1 deletion(-)
>
> diff --git a/arch/powerpc/mm/book3s64/pkeys.c b/arch/powerpc/mm/book3s64/pkeys.c
> index 391230f93da2..16ea0b2f0ea5 100644
> --- a/arch/powerpc/mm/book3s64/pkeys.c
> +++ b/arch/powerpc/mm/book3s64/pkeys.c
> @@ -257,7 +257,12 @@ void __init setup_kuep(bool disabled)
> #ifdef CONFIG_PPC_KUAP
> void __init setup_kuap(bool disabled)
> {
> - if (disabled || !early_radix_enabled())
> + if (disabled)
> + return;
> + /*
> + * On hash if PKEY feature is not enabled, disable KUAP too.
> + */
> + if (!early_radix_enabled() && !early_mmu_has_feature(MMU_FTR_PKEY))
> return;
>
> if (smp_processor_id() == boot_cpuid) {
I'm seeing userspace crashes, bisect points at this commit. But
obviously this is just the commit that enables the code.
It seems to be TM related, a lot of the TM selftests are failing with
seemingly random segfaults.
eg:
./tm-tmspr
test: tm_tmspr
tags: git_version:v5.9-rc2-146-g928c664a7f4e
!! child died by signal 11
failure: tm_tmspr
And this warning fires:
[ 308.825455] ------------[ cut here ]------------
[ 308.825485] WARNING: CPU: 5 PID: 105478 at arch/powerpc/include/asm/book3s/64/kup.h:287 syscall_exit_prepare+0x350/0x390
[ 308.825495] Modules linked in: iptable_mangle xt_MASQUERADE iptable_nat nf_nat xt_conntrack nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 ipt_REJECT nf_reject_ipv4 xt_tcpudp tun bridge stp llc ip6table_filter ip6_tables iptable_filter fuse kvm_hv kvm binfmt_misc squashfs mlx4_ib ib_uverbs dm_multipath scsi_dh_rdac scsi_dh_alua ib_core mlx4_en bnx2x lpfc mlx4_core crc_t10dif sr_mod crct10dif_generic cdrom vmx_crypto scsi_transport_fc sg mdio gf128mul crct10dif_vpmsum crct10dif_common powernv_rng crc32c_vpmsum rng_core leds_powernv powernv_op_panel led_class sunrpc ip_tables x_tables autofs4
[ 308.825596] CPU: 5 PID: 105478 Comm: tm-tmspr Tainted: P 5.9.0-rc2-00146-g928c664a7f4e #1
[ 308.825607] NIP: c000000000038640 LR: c00000000000ddcc CTR: c0000000003c32d0
[ 308.825616] REGS: c000001e41877b00 TRAP: 0700 Tainted: P (5.9.0-rc2-00146-g928c664a7f4e)
[ 308.825625] MSR: 900000010282b033 <SF,HV,VEC,VSX,EE,FP,ME,IR,DR,RI,LE,TM[E]> CR: 44004484 XER: 20000000
[ 308.825649] CFAR: c000000000038358 IRQMASK: 0
GPR00: c00000000000ddcc c000001e41877da0 c00000000143f000 0000000000000000
GPR04: c000001e41877e80 0000000000000000 00007ffec6ed0000 0000000000000008
GPR08: 0000000000000002 3cffffffffffffff fcffffffffffffff 00000000f5638000
GPR12: 0000000000004400 c0000007ffffa680 0000000000000000 00007ffff7ff0000
GPR16: 00007ffff7f54410 00007ffff7f50320 00007ffec6eff240 00007ffff7f54420
GPR20: 00000000000004ba 00000001000016c0 0000000100034840 0000000000000000
GPR24: 0000000000000000 00007ffec6eff8f0 0000000000000000 00007ffec6efea80
GPR28: 00007ffffffff38f c000001e2e05d600 00007ffec6eff180 c000001e41877e80
[ 308.825724] NIP [c000000000038640] syscall_exit_prepare+0x350/0x390
[ 308.825734] LR [c00000000000ddcc] system_call_common+0xfc/0x27c
[ 308.825741] Call Trace:
[ 308.825751] [c000001e41877da0] [c000001e41877e10] 0xc000001e41877e10 (unreliable)
[ 308.825763] [c000001e41877e10] [c00000000000ddcc] system_call_common+0xfc/0x27c
[ 308.825771] Instruction dump:
[ 308.825778] 713c0800 4082004c f87f0018 395d0080 39001800 7ce050a8 7ce74078 7ce051ad
[ 308.825794] 40c2fff4 4bfffd54 60000000 60000000 <0fe00000> 4bfffd18 60000000 60000000
[ 308.825811] ---[ end trace 57dae589ab37c54c ]---
That's on power8, both bare metal and LPAR.
cheers
^ permalink raw reply
* Re: [PATCH v5 05/10] powerpc/smp: Dont assume l2-cache to be superset of sibling
From: Michael Ellerman @ 2020-09-13 1:46 UTC (permalink / raw)
To: Srikar Dronamraju
Cc: Nathan Lynch, Gautham R Shenoy, Michael Neuling, Peter Zijlstra,
Jordan Niethe, LKML, Nicholas Piggin, Valentin Schneider,
Oliver O'Halloran, linuxppc-dev, Ingo Molnar
In-Reply-To: <20200912044603.GA11808@linux.vnet.ibm.com>
Srikar Dronamraju <srikar@linux.vnet.ibm.com> writes:
> * Michael Ellerman <mpe@ellerman.id.au> [2020-09-11 21:55:23]:
>
>> Srikar Dronamraju <srikar@linux.vnet.ibm.com> writes:
>> > Current code assumes that cpumask of cpus sharing a l2-cache mask will
>> > always be a superset of cpu_sibling_mask.
>> >
>> > Lets stop that assumption. cpu_l2_cache_mask is a superset of
>> > cpu_sibling_mask if and only if shared_caches is set.
>>
>> I'm seeing oopses with this:
>>
>> [ 0.117392][ T1] smp: Bringing up secondary CPUs ...
>> [ 0.156515][ T1] smp: Brought up 2 nodes, 2 CPUs
>> [ 0.158265][ T1] numa: Node 0 CPUs: 0
>> [ 0.158520][ T1] numa: Node 1 CPUs: 1
>> [ 0.167453][ T1] BUG: Unable to handle kernel data access on read at 0x8000000041228298
>> [ 0.167992][ T1] Faulting instruction address: 0xc00000000018c128
>> [ 0.168817][ T1] Oops: Kernel access of bad area, sig: 11 [#1]
>> [ 0.168964][ T1] LE PAGE_SIZE=64K MMU=Hash SMP NR_CPUS=2048 NUMA pSeries
>> [ 0.169417][ T1] Modules linked in:
>> [ 0.170047][ T1] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.9.0-rc2-00095-g7430ad5aa700 #209
>> [ 0.170305][ T1] NIP: c00000000018c128 LR: c00000000018c0cc CTR: c00000000004dce0
>> [ 0.170498][ T1] REGS: c00000007e343880 TRAP: 0380 Not tainted (5.9.0-rc2-00095-g7430ad5aa700)
>> [ 0.170602][ T1] MSR: 8000000002009033 <SF,VEC,EE,ME,IR,DR,RI,LE> CR: 44002222 XER: 00000000
>> [ 0.170985][ T1] CFAR: c00000000018c288 IRQMASK: 0
>> [ 0.170985][ T1] GPR00: 0000000000000000 c00000007e343b10 c00000000173e400 0000000000004000
>> [ 0.170985][ T1] GPR04: 0000000000000000 0000000000000800 0000000000000800 0000000000000000
>> [ 0.170985][ T1] GPR08: 0000000000000000 c00000000122c298 c00000003fffc000 c00000007fd05ce8
>> [ 0.170985][ T1] GPR12: c00000007e0119f8 c000000001930000 00000000ffff8ade 0000000000000000
>> [ 0.170985][ T1] GPR16: c00000007e3c0640 0000000000000917 c00000007e3c0658 0000000000000008
>> [ 0.170985][ T1] GPR20: c0000000015d0bb8 00000000ffff8ade c000000000f57400 c000000001817c28
>> [ 0.170985][ T1] GPR24: c00000000176dc80 c00000007e3c0890 c00000007e3cfe00 0000000000000000
>> [ 0.170985][ T1] GPR28: c000000001772310 c00000007e011900 c00000007e3c0800 0000000000000001
>> [ 0.172750][ T1] NIP [c00000000018c128] build_sched_domains+0x808/0x14b0
>> [ 0.172900][ T1] LR [c00000000018c0cc] build_sched_domains+0x7ac/0x14b0
>> [ 0.173186][ T1] Call Trace:
>> [ 0.173484][ T1] [c00000007e343b10] [c00000000018bfe8] build_sched_domains+0x6c8/0x14b0 (unreliable)
>> [ 0.173821][ T1] [c00000007e343c50] [c00000000018dcdc] sched_init_domains+0xec/0x130
>> [ 0.174037][ T1] [c00000007e343ca0] [c0000000010d59d8] sched_init_smp+0x50/0xc4
>> [ 0.174207][ T1] [c00000007e343cd0] [c0000000010b45c4] kernel_init_freeable+0x1b4/0x378
>> [ 0.174378][ T1] [c00000007e343db0] [c0000000000129fc] kernel_init+0x24/0x158
>> [ 0.174740][ T1] [c00000007e343e20] [c00000000000d9d0] ret_from_kernel_thread+0x5c/0x6c
>> [ 0.175050][ T1] Instruction dump:
>> [ 0.175626][ T1] 554905ee 71480040 7d2907b4 4182016c 2c290000 3920006e 913e002c 41820034
>> [ 0.175841][ T1] 7c6307b4 e9300020 78631f24 7d58182a <7d2a482a> f93e0080 7d404828 314a0001
>> [ 0.178340][ T1] ---[ end trace 6876b88dd1d4b3bb ]---
>> [ 0.178512][ T1]
>> [ 1.180458][ T1] Kernel panic - not syncing: Attempted to kill init! exitcode=0x0000000b
>>
>> That's qemu:
>>
>> qemu-system-ppc64 -nographic -vga none -M pseries -cpu POWER8 \
>> -kernel build~/vmlinux \
>> -m 2G,slots=2,maxmem=4G \
>> -object memory-backend-ram,size=1G,id=m0 \
>> -object memory-backend-ram,size=1G,id=m1 \
>> -numa node,nodeid=0,memdev=m0 \
>> -numa node,nodeid=1,memdev=m1 \
>> -smp 2,sockets=2,maxcpus=2 \
>>
>
> Thanks Michael for the report and also for identifying the patch and also
> giving an easy reproducer. That made my task easy. (My only problem was all
> my PowerKVM hosts had a old compiler that refuse to compile never kernels.)
>
> So in this setup, CPU doesn't have a l2-cache. And in that scenario, we
> miss updating the l2-cache domain. Actually the initial patch had this
> exact code. However it was my mistake. I should have reassessed it before
> making changes suggested by Gautham.
>
> Patch below. Do let me know if you want me to send the patch separately.
>> On mambo I get:
>>
>> [ 0.005069][ T1] smp: Bringing up secondary CPUs ...
>> [ 0.011656][ T1] smp: Brought up 2 nodes, 8 CPUs
>> [ 0.011682][ T1] numa: Node 0 CPUs: 0-3
>> [ 0.011709][ T1] numa: Node 1 CPUs: 4-7
>> [ 0.012015][ T1] BUG: arch topology borken
>> [ 0.012040][ T1] the SMT domain not a subset of the CACHE domain
>> [ 0.012107][ T1] BUG: Unable to handle kernel data access on read at 0x80000001012e7398
>> [ 0.012142][ T1] Faulting instruction address: 0xc0000000001aa4f0
>> [ 0.012174][ T1] Oops: Kernel access of bad area, sig: 11 [#1]
>> [ 0.012206][ T1] LE PAGE_SIZE=64K MMU=Hash SMP NR_CPUS=2048 NUMA PowerNV
>> [ 0.012236][ T1] Modules linked in:
>> [ 0.012264][ T1] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.9.0-rc2-00095-g7430ad5aa700 #1
>> [ 0.012304][ T1] NIP: c0000000001aa4f0 LR: c0000000001aa498 CTR: 0000000000000000
>> [ 0.012341][ T1] REGS: c0000000ef583880 TRAP: 0380 Not tainted (5.9.0-rc2-00095-g7430ad5aa700)
>> [ 0.012379][ T1] MSR: 9000000002009033 <SF,HV,VEC,EE,ME,IR,DR,RI,LE> CR: 44002222 XER: 00040000
>> [ 0.012439][ T1] CFAR: c0000000000101b0 IRQMASK: 0
>> [ 0.012439][ T1] GPR00: 0000000000000000 c0000000ef583b10 c0000000017fd000 0000000000004000
>> [ 0.012439][ T1] GPR04: 0000000000000000 0000000000000800 0000000000000000 0000000000000000
>> [ 0.012439][ T1] GPR08: 0000000000000000 c0000000012eb398 c0000000ffffc000 0000000000000000
>> [ 0.012439][ T1] GPR12: 0000000000000020 c0000000019f0000 00000000ffff8ad1 0000000000000000
>> [ 0.012439][ T1] GPR16: c0000000ef068658 c0000000018d7ba8 0000000000000008 c000000001690bb8
>> [ 0.012439][ T1] GPR20: c00000000182dc80 c0000000ef06be90 00000000ffff8ad1 c000000001014aa8
>> [ 0.012439][ T1] GPR24: 0000000000000917 c0000000ef068e00 0000000000000000 c0000000ef06be00
>> [ 0.012439][ T1] GPR28: 0000000000000001 c0000000ef068640 c0000000ef4a1800 c000000001832310
>> [ 0.012774][ T1] NIP [c0000000001aa4f0] build_sched_domains+0x5c0/0x14f0
>> [ 0.012812][ T1] LR [c0000000001aa498] build_sched_domains+0x568/0x14f0
>> [ 0.012842][ T1] Call Trace:
>> [ 0.012872][ T1] [c0000000ef583b10] [c0000000001aa3b4] build_sched_domains+0x484/0x14f0 (unreliable)
>> [ 0.012922][ T1] [c0000000ef583c50] [c0000000001ac3d8] sched_init_domains+0xd8/0x120
>> [ 0.012966][ T1] [c0000000ef583ca0] [c0000000011962d0] sched_init_smp+0x50/0xc4
>> [ 0.013008][ T1] [c0000000ef583cd0] [c00000000117451c] kernel_init_freeable+0x1b4/0x378
>> [ 0.013051][ T1] [c0000000ef583db0] [c000000000012994] kernel_init+0x2c/0x158
>> [ 0.013092][ T1] [c0000000ef583e20] [c00000000000d9d0] ret_from_kernel_thread+0x5c/0x6c
>> [ 0.013128][ T1] Instruction dump:
>> [ 0.013151][ T1] e93b003a 712a0040 552a05ee 418203c4 2c2a0000 3920006e 913b002c 41820034
>> [ 0.013206][ T1] 7c6307b4 e93d0020 78631f24 7d54182a <7d2a482a> f93b0080 7d404828 314a0001
>> [ 0.013267][ T1] ---[ end trace 1bf5f6f38a9fd096 ]---
>>
>
> I haven't tried Mambo. But the problem report looks similar.
The patch fixes qemu, and I don't see the crash on mambo, but I still
see:
[ 0.010536] smp: Bringing up secondary CPUs ...
[ 0.019189] smp: Brought up 2 nodes, 8 CPUs
[ 0.019210] numa: Node 0 CPUs: 0-3
[ 0.019235] numa: Node 1 CPUs: 4-7
[ 0.023669] BUG: arch topology borken
[ 0.023690] the SMT domain not a subset of the CACHE domain
[ 0.023726] BUG: arch topology borken
[ 0.023747] the CACHE domain not a subset of the MC domain
[ 0.023808] BUG: arch topology borken
[ 0.023829] the SMT domain not a subset of the CACHE domain
[ 0.023865] BUG: arch topology borken
[ 0.023887] the CACHE domain not a subset of the MC domain
[ 0.023948] BUG: arch topology borken
[ 0.023969] the SMT domain not a subset of the CACHE domain
[ 0.024005] BUG: arch topology borken
[ 0.024026] the CACHE domain not a subset of the MC domain
[ 0.024087] BUG: arch topology borken
[ 0.024108] the SMT domain not a subset of the CACHE domain
[ 0.024144] BUG: arch topology borken
[ 0.024165] the CACHE domain not a subset of the MC domain
[ 0.024227] BUG: arch topology borken
[ 0.024248] the SMT domain not a subset of the CACHE domain
[ 0.024284] BUG: arch topology borken
[ 0.024305] the CACHE domain not a subset of the MC domain
[ 0.024366] BUG: arch topology borken
[ 0.024387] the SMT domain not a subset of the CACHE domain
[ 0.024423] BUG: arch topology borken
[ 0.024444] the CACHE domain not a subset of the MC domain
[ 0.024505] BUG: arch topology borken
[ 0.024527] the SMT domain not a subset of the CACHE domain
[ 0.024563] BUG: arch topology borken
[ 0.024584] the CACHE domain not a subset of the MC domain
[ 0.024645] BUG: arch topology borken
[ 0.024666] the SMT domain not a subset of the CACHE domain
[ 0.024702] BUG: arch topology borken
[ 0.024723] the CACHE domain not a subset of the MC domain
That's the p9 mambo model, using skiboot.tcl from skiboot, with CPUS=2,
THREADS=4 and MAMBO_NUMA=1.
Node layout is:
[ 0.000000] Early memory node ranges
[ 0.000000] node 0: [mem 0x0000000000000000-0x00000000ffffffff]
[ 0.000000] node 1: [mem 0x0000200000000000-0x00002000ffffffff]
[ 0.000000] Initmem setup node 0 [mem 0x0000000000000000-0x00000000ffffffff]
[ 0.000000] On node 0 totalpages: 65536
[ 0.000000] Initmem setup node 1 [mem 0x0000200000000000-0x00002000ffffffff]
[ 0.000000] On node 1 totalpages: 65536
There aren't any l2-cache properties in the device-tree under cpus.
I'll try and have a closer look tonight.
cheers
^ permalink raw reply
* Re: [PATCH v2] selftests/seccomp: fix ptrace tests on powerpc
From: Michael Ellerman @ 2020-09-13 7:35 UTC (permalink / raw)
To: Kees Cook, Thadeu Lima de Souza Cascardo
Cc: linuxppc-dev, Shuah Khan, Oleg Nesterov, linux-kselftest
In-Reply-To: <202009111550.07017FE49@keescook>
Kees Cook <keescook@chromium.org> writes:
> On Fri, Sep 11, 2020 at 03:10:12PM -0300, Thadeu Lima de Souza Cascardo wrote:
...
>> diff --git a/tools/testing/selftests/seccomp/seccomp_bpf.c b/tools/testing/selftests/seccomp/seccomp_bpf.c
>> index 7a6d40286a42..0ddc0846e9c0 100644
>> --- a/tools/testing/selftests/seccomp/seccomp_bpf.c
>> +++ b/tools/testing/selftests/seccomp/seccomp_bpf.c
>> @@ -1916,10 +1957,15 @@ void tracer_ptrace(struct __test_metadata *_metadata, pid_t tracee,
>> EXPECT_EQ(entry ? PTRACE_EVENTMSG_SYSCALL_ENTRY
>> : PTRACE_EVENTMSG_SYSCALL_EXIT, msg);
>>
>> - if (!entry)
>> + if (!entry && !variant)
>> return;
>>
>> - nr = get_syscall(_metadata, tracee);
>> + if (entry)
>> + nr = get_syscall(_metadata, tracee);
>> + else if (variant)
>> + nr = variant->syscall_nr;
>> + if (variant)
>> + variant->syscall_nr = nr;
>
> So, to be clear this is _only_ an issue for the ptrace side of things,
> yes? i.e. seccomp's setting of the return value will correct stick?
Yes. There's a comment which (hopefully) explains the difference here:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/arch/powerpc/kernel/ptrace/ptrace.c?commit=ab29a807a7ddaa7c84d2f4cb8d29e74e33759072#n239
Which says:
static int do_seccomp(struct pt_regs *regs)
{
if (!test_thread_flag(TIF_SECCOMP))
return 0;
/*
* The ABI we present to seccomp tracers is that r3 contains
* the syscall return value and orig_gpr3 contains the first
* syscall parameter. This is different to the ptrace ABI where
* both r3 and orig_gpr3 contain the first syscall parameter.
*/
regs->gpr[3] = -ENOSYS;
cheers
^ permalink raw reply
* [PATCH] mm/debug_vm_pgtable: Avoid doing memory allocation with pgtable_t mapped.
From: Aneesh Kumar K.V @ 2020-09-13 11:03 UTC (permalink / raw)
To: linux-mm, akpm
Cc: kernel test robot, linuxppc-dev, Aneesh Kumar K.V,
Anshuman Khandual
In-Reply-To: <20200902114222.181353-1-aneesh.kumar@linux.ibm.com>
With highmem, pte_alloc_map() keep the level4 page table mapped using
kmap_atomic(). Avoid doing new memory allocation with page table
mapped like above.
[ 9.409233] BUG: sleeping function called from invalid context at mm/page_alloc.c:4822
[ 9.410557] in_atomic(): 1, irqs_disabled(): 0, non_block: 0, pid: 1, name: swapper
[ 9.411932] no locks held by swapper/1.
[ 9.412595] CPU: 0 PID: 1 Comm: swapper Not tainted 5.9.0-rc3-00323-gc50eb1ed654b5 #2
[ 9.413824] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.12.0-1 04/01/2014
[ 9.415207] Call Trace:
[ 9.415651] ? ___might_sleep.cold+0xa7/0xcc
[ 9.416367] ? __alloc_pages_nodemask+0x14c/0x5b0
[ 9.417055] ? swap_migration_tests+0x50/0x293
[ 9.417704] ? debug_vm_pgtable+0x4bc/0x708
[ 9.418287] ? swap_migration_tests+0x293/0x293
[ 9.418911] ? do_one_initcall+0x82/0x3cb
[ 9.419465] ? parse_args+0x1bd/0x280
[ 9.419983] ? rcu_read_lock_sched_held+0x36/0x60
[ 9.420673] ? trace_initcall_level+0x1f/0xf3
[ 9.421279] ? trace_initcall_level+0xbd/0xf3
[ 9.421881] ? do_basic_setup+0x9d/0xdd
[ 9.422410] ? do_basic_setup+0xc3/0xdd
[ 9.422938] ? kernel_init_freeable+0x72/0xa3
[ 9.423539] ? rest_init+0x134/0x134
[ 9.424055] ? kernel_init+0x5/0x12c
[ 9.424574] ? ret_from_fork+0x19/0x30
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
---
mm/debug_vm_pgtable.c | 11 ++++++++---
1 file changed, 8 insertions(+), 3 deletions(-)
diff --git a/mm/debug_vm_pgtable.c b/mm/debug_vm_pgtable.c
index d12bde82ae95..612c665a1136 100644
--- a/mm/debug_vm_pgtable.c
+++ b/mm/debug_vm_pgtable.c
@@ -994,7 +994,13 @@ static int __init debug_vm_pgtable(void)
p4dp = p4d_alloc(mm, pgdp, vaddr);
pudp = pud_alloc(mm, p4dp, vaddr);
pmdp = pmd_alloc(mm, pudp, vaddr);
- ptep = pte_alloc_map(mm, pmdp, vaddr);
+ /*
+ * Allocate pgtable_t
+ */
+ if (pte_alloc(mm, pmdp)) {
+ pr_err("pgtable allocation failed\n");
+ return 1;
+ }
/*
* Save all the page table page addresses as the page table
@@ -1048,8 +1054,7 @@ static int __init debug_vm_pgtable(void)
* proper page table lock.
*/
- ptl = pte_lockptr(mm, pmdp);
- spin_lock(ptl);
+ ptep = pte_offset_map_lock(mm, pmdp, vaddr, &ptl);
pte_clear_tests(mm, ptep, pte_aligned, vaddr, prot);
pte_advanced_tests(mm, vma, ptep, pte_aligned, vaddr, prot);
pte_unmap_unlock(ptep, ptl);
--
2.26.2
^ permalink raw reply related
* Re: [PATCH] selftests/seccomp: fix ptrace tests on powerpc
From: Michael Ellerman @ 2020-09-13 12:34 UTC (permalink / raw)
To: Thadeu Lima de Souza Cascardo, Kees Cook
Cc: linuxppc-dev, Shuah Khan, Oleg Nesterov, linux-kselftest
In-Reply-To: <20200911180637.GI4002@mussarela>
Thadeu Lima de Souza Cascardo <cascardo@canonical.com> writes:
> On Tue, Sep 08, 2020 at 04:18:17PM -0700, Kees Cook wrote:
>> On Tue, Jun 30, 2020 at 01:47:39PM -0300, Thadeu Lima de Souza Cascardo wrote:
...
>> > @@ -1809,10 +1818,15 @@ void tracer_ptrace(struct __test_metadata *_metadata, pid_t tracee,
>> > EXPECT_EQ(entry ? PTRACE_EVENTMSG_SYSCALL_ENTRY
>> > : PTRACE_EVENTMSG_SYSCALL_EXIT, msg);
>> >
>> > - if (!entry)
>> > + if (!entry && !syscall_nr)
>> > return;
>> >
>> > - nr = get_syscall(_metadata, tracee);
>> > + if (entry)
>> > + nr = get_syscall(_metadata, tracee);
>> > + else
>> > + nr = *syscall_nr;
>>
>> This is weird? Shouldn't get_syscall() be modified to do the right thing
>> here instead of depending on the extra arg?
>>
>
> R0 might be clobered. Same documentation mentions it as volatile. So, during
> syscall exit, we can't tell for sure that R0 will have the original syscall
> number. So, we need to grab it during syscall enter, save it somewhere and
> reuse it. I used the test context/args for that.
The user r0 (in regs->gpr[0]) shouldn't be clobbered.
But it is modified if the tracer skips the syscall, by setting the
syscall number to -1. Or if the tracer changes the syscall number.
So if you need the original syscall number in the exit path then I think
you do need to save it at entry.
cheers
^ permalink raw reply
* Re: [PATCH] mm/debug_vm_pgtable: Avoid doing memory allocation with pgtable_t mapped.
From: Matthew Wilcox @ 2020-09-13 13:42 UTC (permalink / raw)
To: Aneesh Kumar K.V
Cc: kernel test robot, Anshuman Khandual, linux-mm, akpm,
linuxppc-dev
In-Reply-To: <20200913110327.645310-1-aneesh.kumar@linux.ibm.com>
On Sun, Sep 13, 2020 at 04:33:27PM +0530, Aneesh Kumar K.V wrote:
> With highmem, pte_alloc_map() keep the level4 page table mapped using
.noitcerid etisoppo eht ni selbat egap eht srebmun ygolonimret xuniL
^ permalink raw reply
* RE: [PATCHv7 10/12] arm64: dts: layerscape: Add PCIe EP node for ls1088a
From: Z.q. Hou @ 2020-09-13 16:26 UTC (permalink / raw)
To: Rob Herring
Cc: devicetree@vger.kernel.org, lorenzo.pieralisi@arm.com,
Xiaowei Bao, Roy Zang, linux-pci@vger.kernel.org,
linux-kernel@vger.kernel.org, Leo Li, M.h. Lian,
jingoohan1@gmail.com, andrew.murray@arm.com, Mingkai Hu,
gustavo.pimentel@synopsys.com, bhelgaas@google.com,
shawnguo@kernel.org, kishon@ti.com, linuxppc-dev@lists.ozlabs.org,
linux-arm-kernel@lists.infradead.org
In-Reply-To: <20200910164751.GA501845@bogus>
Hi Rob,
Thanks a lot for your comments!
> -----Original Message-----
> From: Rob Herring <robh@kernel.org>
> Sent: 2020年9月11日 0:48
> To: Z.q. Hou <zhiqiang.hou@nxp.com>
> Cc: linux-pci@vger.kernel.org; devicetree@vger.kernel.org;
> linux-kernel@vger.kernel.org; linux-arm-kernel@lists.infradead.org;
> linuxppc-dev@lists.ozlabs.org; bhelgaas@google.com;
> lorenzo.pieralisi@arm.com; shawnguo@kernel.org; Leo Li
> <leoyang.li@nxp.com>; kishon@ti.com; gustavo.pimentel@synopsys.com;
> Roy Zang <roy.zang@nxp.com>; jingoohan1@gmail.com;
> andrew.murray@arm.com; Mingkai Hu <mingkai.hu@nxp.com>; M.h. Lian
> <minghuan.lian@nxp.com>; Xiaowei Bao <xiaowei.bao@nxp.com>
> Subject: Re: [PATCHv7 10/12] arm64: dts: layerscape: Add PCIe EP node for
> ls1088a
>
> On Tue, Aug 11, 2020 at 05:54:39PM +0800, Zhiqiang Hou wrote:
> > From: Xiaowei Bao <xiaowei.bao@nxp.com>
> >
> > Add PCIe EP node for ls1088a to support EP mode.
> >
> > Signed-off-by: Xiaowei Bao <xiaowei.bao@nxp.com>
> > Reviewed-by: Andrew Murray <andrew.murray@arm.com>
> > Signed-off-by: Hou Zhiqiang <Zhiqiang.Hou@nxp.com>
> > ---
> > V7:
> > - Rebase the patch without functionality change.
> >
> > .../arm64/boot/dts/freescale/fsl-ls1088a.dtsi | 31
> +++++++++++++++++++
> > 1 file changed, 31 insertions(+)
> >
> > diff --git a/arch/arm64/boot/dts/freescale/fsl-ls1088a.dtsi
> b/arch/arm64/boot/dts/freescale/fsl-ls1088a.dtsi
> > index 169f4742ae3b..915592141f1b 100644
> > --- a/arch/arm64/boot/dts/freescale/fsl-ls1088a.dtsi
> > +++ b/arch/arm64/boot/dts/freescale/fsl-ls1088a.dtsi
> > @@ -499,6 +499,17 @@
> > status = "disabled";
> > };
> >
> > + pcie_ep@3400000 {
>
> pci-ep@...
The DT node name must be "xxx-xx" style? If yes, the LS1046A EP node also has the name "pcie_ep", it also need to be fixed.
Thanks,
Zhiqiang
>
> > + compatible = "fsl,ls1088a-pcie-ep","fsl,ls-pcie-ep";
> > + reg = <0x00 0x03400000 0x0 0x00100000
> > + 0x20 0x00000000 0x8 0x00000000>;
> > + reg-names = "regs", "addr_space";
> > + num-ib-windows = <24>;
> > + num-ob-windows = <128>;
> > + max-functions = /bits/ 8 <2>;
> > + status = "disabled";
> > + };
> > +
> > pcie@3500000 {
> > compatible = "fsl,ls1088a-pcie";
> > reg = <0x00 0x03500000 0x0 0x00100000 /* controller
> registers */
> > @@ -525,6 +536,16 @@
> > status = "disabled";
> > };
> >
> > + pcie_ep@3500000 {
> > + compatible = "fsl,ls1088a-pcie-ep","fsl,ls-pcie-ep";
> > + reg = <0x00 0x03500000 0x0 0x00100000
> > + 0x28 0x00000000 0x8 0x00000000>;
> > + reg-names = "regs", "addr_space";
> > + num-ib-windows = <6>;
> > + num-ob-windows = <8>;
> > + status = "disabled";
> > + };
> > +
> > pcie@3600000 {
> > compatible = "fsl,ls1088a-pcie";
> > reg = <0x00 0x03600000 0x0 0x00100000 /* controller
> registers */
> > @@ -551,6 +572,16 @@
> > status = "disabled";
> > };
> >
> > + pcie_ep@3600000 {
> > + compatible = "fsl,ls1088a-pcie-ep","fsl,ls-pcie-ep";
> > + reg = <0x00 0x03600000 0x0 0x00100000
> > + 0x30 0x00000000 0x8 0x00000000>;
> > + reg-names = "regs", "addr_space";
> > + num-ib-windows = <6>;
> > + num-ob-windows = <8>;
> > + status = "disabled";
> > + };
> > +
> > smmu: iommu@5000000 {
> > compatible = "arm,mmu-500";
> > reg = <0 0x5000000 0 0x800000>;
> > --
> > 2.17.1
> >
^ permalink raw reply
* RE: [PATCHv7 02/12] PCI: designware-ep: Add the doorbell mode of MSI-X in EP mode
From: Z.q. Hou @ 2020-09-13 16:28 UTC (permalink / raw)
To: Rob Herring
Cc: devicetree@vger.kernel.org, lorenzo.pieralisi@arm.com,
Xiaowei Bao, Roy Zang, linux-pci@vger.kernel.org,
linux-kernel@vger.kernel.org, Leo Li, M.h. Lian,
jingoohan1@gmail.com, andrew.murray@arm.com, Mingkai Hu,
gustavo.pimentel@synopsys.com, bhelgaas@google.com,
shawnguo@kernel.org, kishon@ti.com, linuxppc-dev@lists.ozlabs.org,
linux-arm-kernel@lists.infradead.org
In-Reply-To: <20200910175804.GA592152@bogus>
Hi Rob,
Thanks a lot for your comments!
> -----Original Message-----
> From: Rob Herring <robh@kernel.org>
> Sent: 2020年9月11日 1:58
> To: Z.q. Hou <zhiqiang.hou@nxp.com>
> Cc: linux-pci@vger.kernel.org; devicetree@vger.kernel.org;
> linux-kernel@vger.kernel.org; linux-arm-kernel@lists.infradead.org;
> linuxppc-dev@lists.ozlabs.org; bhelgaas@google.com;
> lorenzo.pieralisi@arm.com; shawnguo@kernel.org; Leo Li
> <leoyang.li@nxp.com>; kishon@ti.com; gustavo.pimentel@synopsys.com;
> Roy Zang <roy.zang@nxp.com>; jingoohan1@gmail.com;
> andrew.murray@arm.com; Mingkai Hu <mingkai.hu@nxp.com>; M.h. Lian
> <minghuan.lian@nxp.com>; Xiaowei Bao <xiaowei.bao@nxp.com>
> Subject: Re: [PATCHv7 02/12] PCI: designware-ep: Add the doorbell mode of
> MSI-X in EP mode
>
> On Tue, Aug 11, 2020 at 05:54:31PM +0800, Zhiqiang Hou wrote:
> > From: Xiaowei Bao <xiaowei.bao@nxp.com>
> >
> > Add the doorbell mode of MSI-X in DWC EP driver.
> >
> > Signed-off-by: Xiaowei Bao <xiaowei.bao@nxp.com>
> > Reviewed-by: Andrew Murray <andrew.murray@arm.com>
> > Signed-off-by: Hou Zhiqiang <Zhiqiang.Hou@nxp.com>
> > ---
> > V7:
> > - Rebase the patch without functionality change.
> >
> > drivers/pci/controller/dwc/pcie-designware-ep.c | 14 ++++++++++++++
> > drivers/pci/controller/dwc/pcie-designware.h | 12 ++++++++++++
> > 2 files changed, 26 insertions(+)
> >
> > diff --git a/drivers/pci/controller/dwc/pcie-designware-ep.c
> > b/drivers/pci/controller/dwc/pcie-designware-ep.c
> > index e5bd3a5ef380..e76b504ed465 100644
> > --- a/drivers/pci/controller/dwc/pcie-designware-ep.c
> > +++ b/drivers/pci/controller/dwc/pcie-designware-ep.c
> > @@ -471,6 +471,20 @@ int dw_pcie_ep_raise_msi_irq(struct dw_pcie_ep
> *ep, u8 func_no,
> > return 0;
> > }
> >
> > +int dw_pcie_ep_raise_msix_irq_doorbell(struct dw_pcie_ep *ep, u8
> > +func_no,
>
> return void. It never has an error.
>
> It could also just be an inline function.
Yes, make sense and will change in next version.
Thanks,
Zhiqiang
>
> > + u16 interrupt_num)
> > +{
> > + struct dw_pcie *pci = to_dw_pcie_from_ep(ep);
> > + u32 msg_data;
> > +
> > + msg_data = (func_no << PCIE_MSIX_DOORBELL_PF_SHIFT) |
> > + (interrupt_num - 1);
> > +
> > + dw_pcie_writel_dbi(pci, PCIE_MSIX_DOORBELL, msg_data);
> > +
> > + return 0;
> > +}
> > +
> > int dw_pcie_ep_raise_msix_irq(struct dw_pcie_ep *ep, u8 func_no,
> > u16 interrupt_num)
> > {
> > diff --git a/drivers/pci/controller/dwc/pcie-designware.h
> > b/drivers/pci/controller/dwc/pcie-designware.h
> > index 89f8271ec5ee..745b4938225a 100644
> > --- a/drivers/pci/controller/dwc/pcie-designware.h
> > +++ b/drivers/pci/controller/dwc/pcie-designware.h
> > @@ -97,6 +97,9 @@
> > #define PCIE_MISC_CONTROL_1_OFF 0x8BC
> > #define PCIE_DBI_RO_WR_EN BIT(0)
> >
> > +#define PCIE_MSIX_DOORBELL 0x948
> > +#define PCIE_MSIX_DOORBELL_PF_SHIFT 24
> > +
> > #define PCIE_PL_CHK_REG_CONTROL_STATUS 0xB20
> > #define PCIE_PL_CHK_REG_CHK_REG_START BIT(0)
> > #define PCIE_PL_CHK_REG_CHK_REG_CONTINUOUS BIT(1)
> > @@ -434,6 +437,8 @@ int dw_pcie_ep_raise_msi_irq(struct dw_pcie_ep
> *ep, u8 func_no,
> > u8 interrupt_num);
> > int dw_pcie_ep_raise_msix_irq(struct dw_pcie_ep *ep, u8 func_no,
> > u16 interrupt_num);
> > +int dw_pcie_ep_raise_msix_irq_doorbell(struct dw_pcie_ep *ep, u8
> func_no,
> > + u16 interrupt_num);
> > void dw_pcie_ep_reset_bar(struct dw_pcie *pci, enum pci_barno bar);
> > #else static inline void dw_pcie_ep_linkup(struct dw_pcie_ep *ep) @@
> > -475,6 +480,13 @@ static inline int dw_pcie_ep_raise_msix_irq(struct
> dw_pcie_ep *ep, u8 func_no,
> > return 0;
> > }
> >
> > +static inline int dw_pcie_ep_raise_msix_irq_doorbell(struct dw_pcie_ep
> *ep,
> > + u8 func_no,
> > + u16 interrupt_num)
> > +{
> > + return 0;
> > +}
> > +
> > static inline void dw_pcie_ep_reset_bar(struct dw_pcie *pci, enum
> > pci_barno bar) { }
> > --
> > 2.17.1
> >
^ permalink raw reply
* Re: [PATCH v5 05/10] powerpc/smp: Dont assume l2-cache to be superset of sibling
From: Srikar Dronamraju @ 2020-09-13 16:53 UTC (permalink / raw)
To: Michael Ellerman
Cc: Nathan Lynch, Gautham R Shenoy, Michael Neuling, Peter Zijlstra,
Jordan Niethe, LKML, Nicholas Piggin, Ingo Molnar,
Oliver O'Halloran, linuxppc-dev, Valentin Schneider
In-Reply-To: <87imciqwhq.fsf@mpe.ellerman.id.au>
* Michael Ellerman <mpe@ellerman.id.au> [2020-09-13 11:46:41]:
> Srikar Dronamraju <srikar@linux.vnet.ibm.com> writes:
> > * Michael Ellerman <mpe@ellerman.id.au> [2020-09-11 21:55:23]:
> >
> >> Srikar Dronamraju <srikar@linux.vnet.ibm.com> writes:
> >> > Current code assumes that cpumask of cpus sharing a l2-cache mask will
> >> > always be a superset of cpu_sibling_mask.
> >> >
> >> > Lets stop that assumption. cpu_l2_cache_mask is a superset of
> >> > cpu_sibling_mask if and only if shared_caches is set.
> >>
> >> I'm seeing oopses with this:
> >>
>
> The patch fixes qemu, and I don't see the crash on mambo, but I still
> see:
> [ 0.010536] smp: Bringing up secondary CPUs ...
> [ 0.019189] smp: Brought up 2 nodes, 8 CPUs
> [ 0.019210] numa: Node 0 CPUs: 0-3
> [ 0.019235] numa: Node 1 CPUs: 4-7
> [ 0.024444] the CACHE domain not a subset of the MC domain
> [ 0.024505] BUG: arch topology borken
> [ 0.024527] the SMT domain not a subset of the CACHE domain
> [ 0.024563] BUG: arch topology borken
> [ 0.024584] the CACHE domain not a subset of the MC domain
> [ 0.024645] BUG: arch topology borken
> [ 0.024666] the SMT domain not a subset of the CACHE domain
> [ 0.024702] BUG: arch topology borken
> [ 0.024723] the CACHE domain not a subset of the MC domain
>
> That's the p9 mambo model, using skiboot.tcl from skiboot, with CPUS=2,
> THREADS=4 and MAMBO_NUMA=1.
>
I was able to reproduce with
qemu-system-ppc64 -nographic -vga none -M pseries -cpu POWER8 \
-kernel build~/vmlinux \
-m 2G,slots=2,maxmem=4G \
-object memory-backend-ram,size=1G,id=m0 \
-object memory-backend-ram,size=1G,id=m1 \
-numa node,nodeid=0,memdev=m0 \
-numa node,nodeid=1,memdev=m1 \
-smp 8,threads=4,sockets=2,maxcpus=8 \
If the CPU doesn't have a l2-cache element, then CPU not only has to set
itself in the cpu_l2_cache but also the siblings. Otherwise it will so
happen that the Siblings will have 4 Cpus set, and the Cache domain will
have just one cpu set, leading to this BUG message.
Patch follows this mail.
> Node layout is:
>
> [ 0.000000] Early memory node ranges
> [ 0.000000] node 0: [mem 0x0000000000000000-0x00000000ffffffff]
> [ 0.000000] node 1: [mem 0x0000200000000000-0x00002000ffffffff]
> [ 0.000000] Initmem setup node 0 [mem 0x0000000000000000-0x00000000ffffffff]
> [ 0.000000] On node 0 totalpages: 65536
> [ 0.000000] Initmem setup node 1 [mem 0x0000200000000000-0x00002000ffffffff]
> [ 0.000000] On node 1 totalpages: 65536
>
>
> There aren't any l2-cache properties in the device-tree under cpus.
>
> I'll try and have a closer look tonight.
>
> cheers
--
Thanks and Regards
Srikar Dronamraju
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox