From: Clément Léger <cleger(a)rivosinc.com>
[ Upstream commit 17c67ed752d6a456602b3dbb25c5ae4d3de5deab ]
Currently, the sud_test expects the emulated syscall to return the
emulated syscall number. This assumption only works on architectures
were the syscall calling convention use the same register for syscall
number/syscall return value. This is not the case for RISC-V and thus
the return value must be also emulated using the provided ucontext.
Signed-off-by: Clément Léger <cleger(a)rivosinc.com>
Reviewed-by: Palmer Dabbelt <palmer(a)rivosinc.com>
Acked-by: Palmer Dabbelt <palmer(a)rivosinc.com>
Link: https://lore.kernel.org/r/20231206134438.473166-1-cleger@rivosinc.com
Signed-off-by: Palmer Dabbelt <palmer(a)rivosinc.com>
Signed-off-by: Sasha Levin <sashal(a)kernel.org>
---
.../selftests/syscall_user_dispatch/sud_test.c | 14 ++++++++++++++
1 file changed, 14 insertions(+)
diff --git a/tools/testing/selftests/syscall_user_dispatch/sud_test.c b/tools/testing/selftests/syscall_user_dispatch/sud_test.c
index b5d592d4099e8..d975a67673299 100644
--- a/tools/testing/selftests/syscall_user_dispatch/sud_test.c
+++ b/tools/testing/selftests/syscall_user_dispatch/sud_test.c
@@ -158,6 +158,20 @@ static void handle_sigsys(int sig, siginfo_t *info, void *ucontext)
/* In preparation for sigreturn. */
SYSCALL_DISPATCH_OFF(glob_sel);
+
+ /*
+ * The tests for argument handling assume that `syscall(x) == x`. This
+ * is a NOP on x86 because the syscall number is passed in %rax, which
+ * happens to also be the function ABI return register. Other
+ * architectures may need to swizzle the arguments around.
+ */
+#if defined(__riscv)
+/* REG_A7 is not defined in libc headers */
+# define REG_A7 (REG_A0 + 7)
+
+ ((ucontext_t *)ucontext)->uc_mcontext.__gregs[REG_A0] =
+ ((ucontext_t *)ucontext)->uc_mcontext.__gregs[REG_A7];
+#endif
}
TEST(dispatch_and_return)
--
2.43.0
From: Clément Léger <cleger(a)rivosinc.com>
[ Upstream commit 17c67ed752d6a456602b3dbb25c5ae4d3de5deab ]
Currently, the sud_test expects the emulated syscall to return the
emulated syscall number. This assumption only works on architectures
were the syscall calling convention use the same register for syscall
number/syscall return value. This is not the case for RISC-V and thus
the return value must be also emulated using the provided ucontext.
Signed-off-by: Clément Léger <cleger(a)rivosinc.com>
Reviewed-by: Palmer Dabbelt <palmer(a)rivosinc.com>
Acked-by: Palmer Dabbelt <palmer(a)rivosinc.com>
Link: https://lore.kernel.org/r/20231206134438.473166-1-cleger@rivosinc.com
Signed-off-by: Palmer Dabbelt <palmer(a)rivosinc.com>
Signed-off-by: Sasha Levin <sashal(a)kernel.org>
---
.../selftests/syscall_user_dispatch/sud_test.c | 14 ++++++++++++++
1 file changed, 14 insertions(+)
diff --git a/tools/testing/selftests/syscall_user_dispatch/sud_test.c b/tools/testing/selftests/syscall_user_dispatch/sud_test.c
index b5d592d4099e8..d975a67673299 100644
--- a/tools/testing/selftests/syscall_user_dispatch/sud_test.c
+++ b/tools/testing/selftests/syscall_user_dispatch/sud_test.c
@@ -158,6 +158,20 @@ static void handle_sigsys(int sig, siginfo_t *info, void *ucontext)
/* In preparation for sigreturn. */
SYSCALL_DISPATCH_OFF(glob_sel);
+
+ /*
+ * The tests for argument handling assume that `syscall(x) == x`. This
+ * is a NOP on x86 because the syscall number is passed in %rax, which
+ * happens to also be the function ABI return register. Other
+ * architectures may need to swizzle the arguments around.
+ */
+#if defined(__riscv)
+/* REG_A7 is not defined in libc headers */
+# define REG_A7 (REG_A0 + 7)
+
+ ((ucontext_t *)ucontext)->uc_mcontext.__gregs[REG_A0] =
+ ((ucontext_t *)ucontext)->uc_mcontext.__gregs[REG_A7];
+#endif
}
TEST(dispatch_and_return)
--
2.43.0
Hi, all,
With the latest 6.9-rc7 torvalds tree kernel, the "make kselftest"
always hangs in the following
./pidfd_setns_test.
Symptoms are that all ./pidfd_netne_test processes end up in pause()
syscalls, with
nothing to wake them up.
Please find the config attached. All the options from the
selftests/pidfdconfig are on (verified):
CONFIG_UTS_NS=y
CONFIG_IPC_NS=y
CONFIG_USER_NS=y
CONFIG_PID_NS=y
CONFIG_NET_NS=y
CONFIG_CGROUPS=y
CONFIG_CHECKPOINT_RESTORE=y
root 2090101 5211 0 21:53 pts/1 00:00:00 make
OUTPUT=/home/marvin/linux/kernel/linux_torvalds/tools/testing/selftests/pidfd
-C pidfd run_tests
SRC_PATH=/home/marvin/linux/kernel/linux_torvalds/tools/testing/selftests
OB
root 2090102 2090101 0 21:53 pts/1 00:00:00 /bin/sh -c
BASE_DIR="/home/marvin/linux/kernel/linux_torvalds/tools/testing/selftests";
. /home/marvin/linux/kernel/linux_torvalds/tools/testing/selftests/kselftest/runner.sh;
if
root 2133026 2090102 0 21:54 pts/1 00:00:00 /bin/sh -c
BASE_DIR="/home/marvin/linux/kernel/linux_torvalds/tools/testing/selftests";
. /home/marvin/linux/kernel/linux_torvalds/tools/testing/selftests/kselftest/runner.sh;
if
root 2133027 2133026 0 21:54 pts/1 00:00:00 /bin/sh -c
BASE_DIR="/home/marvin/linux/kernel/linux_torvalds/tools/testing/selftests";
. /home/marvin/linux/kernel/linux_torvalds/tools/testing/selftests/kselftest/runner.sh;
if
root 2133028 2133027 0 21:54 pts/1 00:00:00 /bin/sh -c
BASE_DIR="/home/marvin/linux/kernel/linux_torvalds/tools/testing/selftests";
. /home/marvin/linux/kernel/linux_torvalds/tools/testing/selftests/kselftest/runner.sh;
if
root 2133031 2133028 0 21:54 pts/1 00:00:00 /bin/sh -c
BASE_DIR="/home/marvin/linux/kernel/linux_torvalds/tools/testing/selftests";
. /home/marvin/linux/kernel/linux_torvalds/tools/testing/selftests/kselftest/runner.sh;
if
root 2133033 2133031 0 21:54 pts/1 00:00:00 perl
/home/marvin/linux/kernel/linux_torvalds/tools/testing/selftests/kselftest/prefix.pl
root 2133050 2887 0 21:54 pts/1 00:00:00 ./pidfd_setns_test
root 2133051 2887 0 21:54 pts/1 00:00:00 ./pidfd_setns_test
root 2133056 2887 0 21:54 pts/1 00:00:00 ./pidfd_setns_test
root 2133057 2887 0 21:54 pts/1 00:00:00 ./pidfd_setns_test
root 2133062 2887 0 21:54 pts/1 00:00:00 ./pidfd_setns_test
root 2133063 2887 0 21:54 pts/1 00:00:00 ./pidfd_setns_test
root 2133068 2887 0 21:54 pts/1 00:00:00 ./pidfd_setns_test
root 2133069 2887 0 21:54 pts/1 00:00:00 ./pidfd_setns_test
.
.
.
marvin@defiant:~/linux/kernel/linux_torvalds$ sudo bash
[sudo] password for marvin:
root@defiant:/home/marvin/linux/kernel/linux_torvalds# strace -p 2133050
strace: Process 2133050 attached
pause(^Cstrace: Process 2133050 detached
<detached ...>
root@defiant:/home/marvin/linux/kernel/linux_torvalds# strace -p 2133051
strace: Process 2133051 attached
pause(^Cstrace: Process 2133051 detached
<detached ...>
root@defiant:/home/marvin/linux/kernel/linux_torvalds# strace -p 2133056
strace: Process 2133056 attached
pause(^Cstrace: Process 2133056 detached
<detached ...>
root@defiant:/home/marvin/linux/kernel/linux_torvalds# strace -p 2133057
strace: Process 2133057 attached
pause(^Cstrace: Process 2133057 detached
<detached ...>
root@defiant:/home/marvin/linux/kernel/linux_torvalds# strace -p 2133062
strace: Process 2133062 attached
pause(^Cstrace: Process 2133062 detached
<detached ...>
root@defiant:/home/marvin/linux/kernel/linux_torvalds# strace -p 2133063
strace: Process 2133063 attached
pause(^Cstrace: Process 2133063 detached
<detached ...>
root@defiant:/home/marvin/linux/kernel/linux_torvalds#
root@defiant:/home/marvin/linux/kernel/linux_torvalds#
root@defiant:/home/marvin/linux/kernel/linux_torvalds# strace -p 2133068
strace: Process 2133068 attached
pause(^Cstrace: Process 2133068 detached
<detached ...>
root@defiant:/home/marvin/linux/kernel/linux_torvalds# strace -p 2133069
strace: Process 2133069 attached
pause(^Cstrace: Process 2133069 detached
<detached ...>
root@defiant:/home/marvin/linux/kernel/linux_torvalds#
The output log is:
# selftests: pidfd: pidfd_setns_test
# TAP version 13
# 1..7
# # Starting 7 tests from 2 test cases.
# # RUN global.setns_einval ...
# # OK global.setns_einval
# ok 1 global.setns_einval
# # RUN current_nsset.invalid_flags ...
# # pidfd_setns_test.c:161:invalid_flags:Expected
self->child_pid_exited (0) > 0 (0)
# # OK current_nsset.invalid_flags
# ok 2 current_nsset.invalid_flags
# # RUN current_nsset.pidfd_exited_child ...
# # pidfd_setns_test.c:161:pidfd_exited_child:Expected
self->child_pid_exited (0) > 0 (0)
# # OK current_nsset.pidfd_exited_child
# ok 3 current_nsset.pidfd_exited_child
# # RUN current_nsset.pidfd_incremental_setns ...
# # pidfd_setns_test.c:161:pidfd_incremental_setns:Expected
self->child_pid_exited (0) > 0 (0)
# # pidfd_setns_test.c:408:pidfd_incremental_setns:Managed to
correctly setns to user namespace of 2133050 via pidfd 20
# # pidfd_setns_test.c:408:pidfd_incremental_setns:Managed to
correctly setns to mnt namespace of 2133050 via pidfd 20
# # pidfd_setns_test.c:408:pidfd_incremental_setns:Managed to
correctly setns to pid namespace of 2133050 via pidfd 20
# # pidfd_setns_test.c:408:pidfd_incremental_setns:Managed to
correctly setns to uts namespace of 2133050 via pidfd 20
# # pidfd_setns_test.c:408:pidfd_incremental_setns:Managed to
correctly setns to ipc namespace of 2133050 via pidfd 20
# # pidfd_setns_test.c:408:pidfd_incremental_setns:Managed to
correctly setns to net namespace of 2133050 via pidfd 20
# # pidfd_setns_test.c:408:pidfd_incremental_setns:Managed to
correctly setns to cgroup namespace of 2133050 via pidfd 20
# # pidfd_setns_test.c:408:pidfd_incremental_setns:Managed to
correctly setns to pid_for_children namespace of 2133050 via pidfd 20
# # pidfd_setns_test.c:391:pidfd_incremental_setns:Expected
setns(self->child_pidfd1, info->flag) (-1) == 0 (0)
# # pidfd_setns_test.c:392:pidfd_incremental_setns:Too many users -
Failed to setns to time namespace of 2133050 via pidfd 20
# # pidfd_incremental_setns: Test terminated by assertion
# # FAIL current_nsset.pidfd_incremental_setns
# not ok 4 current_nsset.pidfd_incremental_setns
# # RUN current_nsset.nsfd_incremental_setns ...
# # pidfd_setns_test.c:161:nsfd_incremental_setns:Expected
self->child_pid_exited (0) > 0 (0)
# # pidfd_setns_test.c:444:nsfd_incremental_setns:Managed to correctly
setns to user namespace of 2133056 via nsfd 19
# # pidfd_setns_test.c:444:nsfd_incremental_setns:Managed to correctly
setns to mnt namespace of 2133056 via nsfd 24
# # pidfd_setns_test.c:444:nsfd_incremental_setns:Managed to correctly
setns to pid namespace of 2133056 via nsfd 27
# # pidfd_setns_test.c:444:nsfd_incremental_setns:Managed to correctly
setns to uts namespace of 2133056 via nsfd 30
# # pidfd_setns_test.c:444:nsfd_incremental_setns:Managed to correctly
setns to ipc namespace of 2133056 via nsfd 33
# # pidfd_setns_test.c:444:nsfd_incremental_setns:Managed to correctly
setns to net namespace of 2133056 via nsfd 36
# # pidfd_setns_test.c:444:nsfd_incremental_setns:Managed to correctly
setns to cgroup namespace of 2133056 via nsfd 39
# # pidfd_setns_test.c:444:nsfd_incremental_setns:Managed to correctly
setns to pid_for_children namespace of 2133056 via nsfd 42
# # pidfd_setns_test.c:427:nsfd_incremental_setns:Expected
setns(self->child_nsfds1[i], info->flag) (-1) == 0 (0)
# # pidfd_setns_test.c:428:nsfd_incremental_setns:Too many users -
Failed to setns to time namespace of 2133056 via nsfd 45
# # nsfd_incremental_setns: Test terminated by assertion
# # FAIL current_nsset.nsfd_incremental_setns
# not ok 5 current_nsset.nsfd_incremental_setns
# # RUN current_nsset.pidfd_one_shot_setns ...
# # pidfd_setns_test.c:161:pidfd_one_shot_setns:Expected
self->child_pid_exited (0) > 0 (0)
# # pidfd_setns_test.c:462:pidfd_one_shot_setns:Adding user namespace
of 2133062 to list of namespaces to attach to
# # pidfd_setns_test.c:462:pidfd_one_shot_setns:Adding mnt namespace
of 2133062 to list of namespaces to attach to
# # pidfd_setns_test.c:462:pidfd_one_shot_setns:Adding pid namespace
of 2133062 to list of namespaces to attach to
# # pidfd_setns_test.c:462:pidfd_one_shot_setns:Adding uts namespace
of 2133062 to list of namespaces to attach to
# # pidfd_setns_test.c:462:pidfd_one_shot_setns:Adding ipc namespace
of 2133062 to list of namespaces to attach to
# # pidfd_setns_test.c:462:pidfd_one_shot_setns:Adding net namespace
of 2133062 to list of namespaces to attach to
# # pidfd_setns_test.c:462:pidfd_one_shot_setns:Adding cgroup
namespace of 2133062 to list of namespaces to attach to
# # pidfd_setns_test.c:462:pidfd_one_shot_setns:Adding
pid_for_children namespace of 2133062 to list of namespaces to attach
to
# # pidfd_setns_test.c:462:pidfd_one_shot_setns:Adding time namespace
of 2133062 to list of namespaces to attach to
# # pidfd_setns_test.c:466:pidfd_one_shot_setns:Expected
setns(self->child_pidfd1, flags) (-1) == 0 (0)
# # pidfd_setns_test.c:467:pidfd_one_shot_setns:Too many users -
Failed to setns to namespaces of 2133062
# # pidfd_one_shot_setns: Test terminated by assertion
# # FAIL current_nsset.pidfd_one_shot_setns
# not ok 6 current_nsset.pidfd_one_shot_setns
# # RUN current_nsset.no_foul_play ...
# # pidfd_setns_test.c:161:no_foul_play:Expected
self->child_pid_exited (0) > 0 (0)
# # pidfd_setns_test.c:506:no_foul_play:Adding user namespace of
2133068 to list of namespaces to attach to
# # pidfd_setns_test.c:506:no_foul_play:Adding mnt namespace of
2133068 to list of namespaces to attach to
# # pidfd_setns_test.c:506:no_foul_play:Adding pid namespace of
2133068 to list of namespaces to attach to
# # pidfd_setns_test.c:506:no_foul_play:Adding uts namespace of
2133068 to list of namespaces to attach to
# # pidfd_setns_test.c:506:no_foul_play:Adding ipc namespace of
2133068 to list of namespaces to attach to
# # pidfd_setns_test.c:506:no_foul_play:Adding net namespace of
2133068 to list of namespaces to attach to
# # pidfd_setns_test.c:506:no_foul_play:Adding cgroup namespace of
2133068 to list of namespaces to attach to
# # pidfd_setns_test.c:506:no_foul_play:Adding time namespace of
2133068 to list of namespaces to attach to
# # pidfd_setns_test.c:510:no_foul_play:Expected
setns(self->child_pidfd1, flags) (-1) == 0 (0)
# # pidfd_setns_test.c:511:no_foul_play:Too many users - Failed to
setns to namespaces of 2133068 vid pidfd 20
# # no_foul_play: Test terminated by assertion
# # FAIL current_nsset.no_foul_play
# not ok 7 current_nsset.no_foul_play
# # FAILED: 3 / 7 tests passed.
# # Totals: pass:3 fail:4 xfail:0 xpass:0 skip:0 error:0
Thanks for your time and patience for reviewing this BUG report.
Best regards,
Mirsad Todorovac
P.S.
I have changed the email address because of the uncertainty whether the employer
would continue to support my work on the Linux kernel testing as the
part of my work
research.
Thank you.
809216233555 ("selftests/harness: remove use of LINE_MAX") introduced
asprintf into kselftest_harness.h, which is a GNU extension and needs
_GNU_SOURCE to either be defined prior to including headers or with the
-D_GNU_SOURCE flag passed to the compiler.
Edward Liaw (10):
selftests/sgx: Compile with -D_GNU_SOURCE
selftests/alsa: Compile with -D_GNU_SOURCE
selftests/hid: Compile with -D_GNU_SOURCE
selftests/kvm: Define _GNU_SOURCE
selftests/nci: Compile with -D_GNU_SOURCE
selftests/net: Define _GNU_SOURCE
selftests/prctl: Compile with -D_GNU_SOURCE
selftests/rtc: Compile with -D_GNU_SOURCE
selftests/tdx: Compile with -D_GNU_SOURCE
selftests/user_events: Compiled with -D_GNU_SOURCE
tools/testing/selftests/alsa/Makefile | 1 +
tools/testing/selftests/hid/Makefile | 2 +-
tools/testing/selftests/kvm/x86_64/fix_hypercall_test.c | 2 ++
tools/testing/selftests/nci/Makefile | 2 +-
tools/testing/selftests/net/bind_wildcard.c | 1 +
tools/testing/selftests/net/ip_local_port_range.c | 1 +
tools/testing/selftests/net/reuseaddr_ports_exhausted.c | 2 ++
tools/testing/selftests/prctl/Makefile | 1 +
tools/testing/selftests/rtc/Makefile | 2 +-
tools/testing/selftests/sgx/Makefile | 2 +-
tools/testing/selftests/sgx/sigstruct.c | 2 --
tools/testing/selftests/tdx/Makefile | 2 +-
tools/testing/selftests/user_events/Makefile | 2 +-
tools/testing/selftests/user_events/abi_test.c | 1 -
14 files changed, 14 insertions(+), 9 deletions(-)
--
2.45.0.rc0.197.gbae5840b3b-goog
The cb fields network_offset and inner_network_offset are used instead of
skb->network_header throughout GRO.
These fields are then leveraged in the next commit to remove flush_id state
from napi_gro_cb, and stateful code in {ipv6,inet}_gro_receive which may be
unnecessarily complicated due to encapsulation support in GRO. These fields
are checked in L4 instead.
3rd patch adds tests for different flush_id flows in GRO.
v7 -> v8:
- Remove network_header use in gro
- Re-send commits after the dependent patch to net was applied
- v7:
https://lore.kernel.org/all/20240412155533.115507-1-richardbgobert@gmail.co…
v6 -> v7:
- Moved bug fixes to a separate submission in net
- Added UDP fwd benchmark
- v6:
https://lore.kernel.org/all/20240410153423.107381-1-richardbgobert@gmail.co…
v5 -> v6:
- Write inner_network_offset in vxlan and geneve
- Ignore is_atomic when DF=0
- v5:
https://lore.kernel.org/all/20240408141720.98832-1-richardbgobert@gmail.com/
v4 -> v5:
- Add 1st commit - flush id checks in udp_gro_receive segment which can be
backported by itself
- Add TCP measurements for the 5th commit
- Add flush id tests to ensure flush id logic is preserved in GRO
- Simplify gro_inet_flush by removing a branch
- v4:
https://lore.kernel.org/all/202420325182543.87683-1-richardbgobert@gmail.co…
v3 -> v4:
- Fix code comment and commit message typos
- v3:
https://lore.kernel.org/all/f939c84a-2322-4393-a5b0-9b1e0be8ed8e@gmail.com/
v2 -> v3:
- Use napi_gro_cb instead of skb->{offset}
- v2:
https://lore.kernel.org/all/2ce1600b-e733-448b-91ac-9d0ae2b866a4@gmail.com/
v1 -> v2:
- Pass p_off in *_gro_complete to fix UDP bug
- Remove more conditionals and memory fetches from inet_gro_flush
- v1:
https://lore.kernel.org/netdev/e1d22505-c5f8-4c02-a997-64248480338b@gmail.c…
Richard Gobert (3):
net: gro: use cb instead of skb->network_header
net: gro: move L3 flush checks to tcp_gro_receive and udp_gro_receive_segment
selftests/net: add flush id selftests
include/net/gro.h | 75 +++++++++++++--
net/core/gro.c | 3 -
net/ipv4/af_inet.c | 45 +--------
net/ipv4/tcp_offload.c | 18 +---
net/ipv4/udp_offload.c | 10 +-
net/ipv6/ip6_offload.c | 16 +---
net/ipv6/tcpv6_offload.c | 3 +-
tools/testing/selftests/net/gro.c | 147 ++++++++++++++++++++++++++++++
8 files changed, 225 insertions(+), 92 deletions(-)
--
2.36.1
Hi,
This sixth series just update the last patch description.
Shuah, I think this should be in -next really soon to make sure
everything works fine for the v6.9 release, which is not currently the
case. I cannot test against all kselftests though. I would prefer to
let you handle this, but I guess you're not able to do so and I'll push
it on my branch without reply from you. Even if I push it on my branch,
please push it on yours too as soon as you see this and I'll remove it
from mine.
Mark, Jakub, could you please test this series?
As reported by Kernel Test Robot [1] and Sean Christopherson [2], some
tests fail since v6.9-rc1 . This is due to the use of vfork() which
introduced some side effects. Similarly, while making it more generic,
a previous commit made some Landlock file system tests flaky, and
subject to the host's file system mount configuration.
This series fixes all these side effects by replacing vfork() with
clone3() and CLONE_VFORK, which is cleaner (no arbitrary shared memory)
and makes the Kselftest framework more robust.
I tried different approaches and I found this one to be the cleaner and
less invasive for current test cases.
I successfully ran the following tests (using TEST_F and
fork/clone/clone3, and KVM_ONE_VCPU_TEST) with this series:
- kvm:fix_hypercall_test
- kvm:sync_regs_test
- kvm:userspace_msr_exit_test
- kvm:vmx_pmu_caps_test
- landlock:fs_test
- landlock:net_test
- landlock:ptrace_test
- move_mount_set_group:move_mount_set_group_test
- net/af_unix:scm_pidfd
- perf_events:remove_on_exec
- pidfd:pidfd_getfd_test
- pidfd:pidfd_setns_test
- seccomp:seccomp_bpf
- user_events:abi_test
[1] https://lore.kernel.org/oe-lkp/202403291015.1fcfa957-oliver.sang@intel.com
[2] https://lore.kernel.org/r/ZjPelW6-AbtYvslu@google.com
Previous versions:
v1: https://lore.kernel.org/r/20240426172252.1862930-1-mic@digikod.net
v2: https://lore.kernel.org/r/20240429130931.2394118-1-mic@digikod.net
v3: https://lore.kernel.org/r/20240429191911.2552580-1-mic@digikod.net
v4: https://lore.kernel.org/r/20240502210926.145539-1-mic@digikod.net
v5: https://lore.kernel.org/r/20240503105820.300927-1-mic@digikod.net
Regards,
Mickaël Salaün (10):
selftests/pidfd: Fix config for pidfd_setns_test
selftests/landlock: Fix FS tests when run on a private mount point
selftests/harness: Fix fixture teardown
selftests/harness: Fix interleaved scheduling leading to race
conditions
selftests/landlock: Do not allocate memory in fixture data
selftests/harness: Constify fixture variants
selftests/pidfd: Fix wrong expectation
selftests/harness: Share _metadata between forked processes
selftests/harness: Fix vfork() side effects
selftests/harness: Handle TEST_F()'s explicit exit codes
tools/testing/selftests/kselftest_harness.h | 122 +++++++++++++-----
tools/testing/selftests/landlock/fs_test.c | 83 +++++++-----
tools/testing/selftests/pidfd/config | 2 +
.../selftests/pidfd/pidfd_setns_test.c | 2 +-
4 files changed, 143 insertions(+), 66 deletions(-)
base-commit: e67572cd2204894179d89bd7b984072f19313b03
--
2.45.0
When I introduced HID-BPF, I mentioned that we should ship the HID-BPF
programs in the kernel when they are fixes so that everybody can benefit
from them.
I tried multiple times to do so but I was confronted to a tough problem:
how can I make the kernel load them automatically?
I went over a few solutions, but it always came down to something either
ugly, or either not satisfying (like forcing `bpftool` to be compiled
first, or not being able to insert them as a module).
OTOH, I was working with Peter on `udev-hid-bpf`[0] as a proof of
concept on how a minimal loader should look like. This allowed me to
experiment on the BPF files and how they should look like.
And after further thoughts, I realized that `udev-hid-bpf` could very
well be the `kmod load` that we currently have:
- the kernel handles the device normally
- a udev event is emitted
- a udev rule fires `udev-hid-bpf` and load the appropriate HID-BPF
file(s) based on the modalias
Given that most HID devices are supposed to work to a minimal level when
connected without any driver, this makes the whole HID-BPF programs nice
to have but not critical. We can then postpone the HID-BPF loading when
userspace is ready.
Working with HID-BPF is also a much better user experience for end users
(as I predicted). All they have to do is to go to the `udev-hid-bpf`
project, fetch an artifact from the MR that concerns them, run
`install.sh` (no compilation required), and their devices are fixed
(minus some back and forth when the HID-BPF program needs some changes).
So I already have that loader available, and it works well enough for
our users. But the missing point was still how to "upstream" those BPF
fixes?
That's where this patch series comes in: we simply store the fixes in
the kernel under `drivers/hid/bpf/progs`, provide a way to compile them,
but also add tests for them in the selftests dir.
Once a program is accepted here, for convenience, the same program will
move from a "testing" directory to a "stable" directory on
`udev-hid-bpf`. This way, distributions don't need to follow when there
is a new program added here, they can just ship the "stable" ones from
`udev-hid-bpf`.
Signed-off-by: Benjamin Tissoires <bentiss(a)kernel.org>
[0] https://gitlab.freedesktop.org/libevdev/udev-hid-bpf/
---
Benjamin Tissoires (18):
HID: do not assume HAT Switch logical max < 8
HID: bpf: add first in-tree HID-BPF fix for the XPPen Artist 24
HID: bpf: add in-tree HID-BPF fix for the XPPen Artist 16
HID: bpf: add in-tree HID-BPF fix for the HP Elite Presenter Mouse
HID: bpf: add in-tree HID-BPF fix for the IOGear Kaliber Gaming MMOmentum mouse
HID: bpf: add in-tree HID-BPF fix for the Wacom ArtPen
HID: bpf: add in-tree HID-BPF fix for the XBox Elite 2 over Bluetooth
HID: bpf: add in-tree HID-BPF fix for the Huion Kamvas Pro 19
HID: bpf: add in-tree HID-BPF fix for the Raptor Mach 2
selftests/hid: import base_device.py from hid-tools
selftests/hid: add support for HID-BPF pre-loading before starting a test
selftests/hid: tablets: reduce the number of pen state
selftests/hid: tablets: add a couple of XP-PEN tablets
selftests/hid: tablets: also check for XP-Pen offset correction
selftests/hid: add Huion Kamvas Pro 19 tests
selftests/hid: import base_gamepad.py from hid-tools
selftests/hid: move the gamepads definitions in the test file
selftests/hid: add tests for the Raptor Mach 2 joystick
drivers/hid/bpf/progs/FR-TEC__Raptor-Mach-2.bpf.c | 185 ++++++
drivers/hid/bpf/progs/HP__Elite-Presenter.bpf.c | 58 ++
drivers/hid/bpf/progs/Huion__Kamvas-Pro-19.bpf.c | 290 +++++++++
.../hid/bpf/progs/IOGEAR__Kaliber-MMOmentum.bpf.c | 59 ++
drivers/hid/bpf/progs/Makefile | 91 +++
.../hid/bpf/progs/Microsoft__XBox-Elite-2.bpf.c | 133 ++++
drivers/hid/bpf/progs/README | 102 +++
drivers/hid/bpf/progs/Wacom__ArtPen.bpf.c | 173 +++++
drivers/hid/bpf/progs/XPPen__Artist24.bpf.c | 229 +++++++
drivers/hid/bpf/progs/XPPen__ArtistPro16Gen2.bpf.c | 274 ++++++++
drivers/hid/bpf/progs/hid_bpf.h | 15 +
drivers/hid/bpf/progs/hid_bpf_helpers.h | 170 +++++
include/linux/hid.h | 6 +-
tools/testing/selftests/hid/tests/base.py | 87 ++-
tools/testing/selftests/hid/tests/base_device.py | 421 ++++++++++++
tools/testing/selftests/hid/tests/base_gamepad.py | 238 +++++++
tools/testing/selftests/hid/tests/test_gamepad.py | 457 ++++++++++++-
tools/testing/selftests/hid/tests/test_tablet.py | 723 +++++++++++++++------
18 files changed, 3507 insertions(+), 204 deletions(-)
---
base-commit: 3e78a6c0d3e02e4cf881dc84c5127e9990f939d6
change-id: 20240328-bpf_sources-be1f3c617c5e
Best regards,
--
Benjamin Tissoires <bentiss(a)kernel.org>
Add several new test cases which assert corner cases on the eventfd
mechanism, for example, the supplied buffer is less than 8 bytes,
attempting to write a value that is too large, etc.
./eventfd_test
# Starting 9 tests from 1 test cases.
# RUN global.eventfd01 ...
# OK global.eventfd01
ok 1 global.eventfd01
# RUN global.eventfd02 ...
# OK global.eventfd02
ok 2 global.eventfd02
# RUN global.eventfd03 ...
# OK global.eventfd03
ok 3 global.eventfd03
# RUN global.eventfd04 ...
# OK global.eventfd04
ok 4 global.eventfd04
# RUN global.eventfd05 ...
# OK global.eventfd05
ok 5 global.eventfd05
# RUN global.eventfd06 ...
# OK global.eventfd06
ok 6 global.eventfd06
# RUN global.eventfd07 ...
# OK global.eventfd07
ok 7 global.eventfd07
# RUN global.eventfd08 ...
# OK global.eventfd08
ok 8 global.eventfd08
# RUN global.eventfd09 ...
# OK global.eventfd09
ok 9 global.eventfd09
# PASSED: 9 / 9 tests passed.
# Totals: pass:9 fail:0 xfail:0 xpass:0 skip:0 error:0
Signed-off-by: Wen Yang <wen.yang(a)linux.dev>
Cc: SShuah Khan <shuah(a)kernel.org>
Cc: Christian Brauner <brauner(a)kernel.org>
Cc: Andrei Vagin <avagin(a)google.com>
Cc: Mathieu Desnoyers <mathieu.desnoyers(a)efficios.com>
Cc: Steven Rostedt <rostedt(a)goodmis.org>
Cc: Andrew Morton <akpm(a)linux-foundation.org>
Cc: Dave Young <dyoung(a)redhat.com>
Cc: linux-kselftest(a)vger.kernel.org
Cc: linux-kernel(a)vger.kernel.org
---
.../filesystems/eventfd/eventfd_test.c | 116 ++++++++++++++++++
1 file changed, 116 insertions(+)
diff --git a/tools/testing/selftests/filesystems/eventfd/eventfd_test.c b/tools/testing/selftests/filesystems/eventfd/eventfd_test.c
index f142a137526c..eeab8df5b1b5 100644
--- a/tools/testing/selftests/filesystems/eventfd/eventfd_test.c
+++ b/tools/testing/selftests/filesystems/eventfd/eventfd_test.c
@@ -183,4 +183,120 @@ TEST(eventfd05)
close(fd);
}
+/*
+ * A write(2) fails with the error EINVAL if the size of the supplied buffer
+ * is less than 8 bytes, or if an attempt is made to write the value
+ * 0xffffffffffffffff.
+ */
+TEST(eventfd06)
+{
+ uint64_t value = 1;
+ ssize_t size;
+ int fd;
+
+ fd = sys_eventfd2(0, 0);
+ ASSERT_GE(fd, 0);
+
+ size = write(fd, &value, sizeof(int));
+ EXPECT_EQ(size, -1);
+ EXPECT_EQ(errno, EINVAL);
+
+ value = (uint64_t)-1;
+ size = write(fd, &value, sizeof(value));
+ EXPECT_EQ(size, -1);
+ EXPECT_EQ(errno, EINVAL);
+
+ close(fd);
+}
+
+/*
+ * A read(2) fails with the error EINVAL if the size of the supplied buffer is
+ * less than 8 bytes.
+ */
+TEST(eventfd07)
+{
+ int value = 0;
+ ssize_t size;
+ int fd;
+
+ fd = sys_eventfd2(0, 0);
+ ASSERT_GE(fd, 0);
+
+ size = write(fd, &value, sizeof(value));
+ EXPECT_EQ(size, -1);
+ EXPECT_EQ(errno, EINVAL);
+
+ close(fd);
+}
+
+/*
+ * If EFD_SEMAPHORE was not specified and the eventfd counter has a nonzero
+ * value, then a read(2) returns 8 bytes containing that value, and the
+ * counter's value is reset to zero.
+ * If the eventfd counter is zero at the time of the call to read(2), then the
+ * call fails with the error EAGAIN if the file descriptor has been made nonblocking.
+ */
+TEST(eventfd08)
+{
+ uint64_t value;
+ ssize_t size;
+ int fd;
+ int i;
+
+ fd = sys_eventfd2(0, EFD_NONBLOCK);
+ ASSERT_GE(fd, 0);
+
+ value = 1;
+ for (i = 0; i < 10000000; i++) {
+ size = write(fd, &value, sizeof(value));
+ EXPECT_EQ(size, sizeof(value));
+ }
+
+ size = read(fd, &value, sizeof(value));
+ EXPECT_EQ(size, sizeof(uint64_t));
+ EXPECT_EQ(value, 10000000);
+
+ size = read(fd, &value, sizeof(value));
+ EXPECT_EQ(size, -1);
+ EXPECT_EQ(errno, EAGAIN);
+
+ close(fd);
+}
+
+/*
+ * If EFD_SEMAPHORE was specified and the eventfd counter has a nonzero value,
+ * then a read(2) returns 8 bytes containing the value 1, and the counter's
+ * value is decremented by 1.
+ * If the eventfd counter is zero at the time of the call to read(2), then the
+ * call fails with the error EAGAIN if the file descriptor has been made nonblocking.
+ */
+TEST(eventfd09)
+{
+ uint64_t value;
+ ssize_t size;
+ int fd;
+ int i;
+
+ fd = sys_eventfd2(0, EFD_SEMAPHORE|EFD_NONBLOCK);
+ ASSERT_GE(fd, 0);
+
+ value = 1;
+ for (i = 0; i < 10000000; i++) {
+ size = write(fd, &value, sizeof(value));
+ EXPECT_EQ(size, sizeof(value));
+ }
+
+ for (i = 0; i < 10000000; i++) {
+ size = read(fd, &value, sizeof(value));
+ EXPECT_EQ(size, sizeof(value));
+ EXPECT_EQ(value, 1);
+ }
+
+ size = read(fd, &value, sizeof(value));
+ EXPECT_EQ(size, -1);
+ EXPECT_EQ(errno, EAGAIN);
+
+ close(fd);
+}
+
TEST_HARNESS_MAIN
--
2.25.1
This is a RFC, following[0].
It works, still needs some care but this is mainly to see if this will
have a chance to get upsrteamed or if I should rely on struct_ops
instead.
Cheers,
Benjamin
Signed-off-by: Benjamin Tissoires <bentiss(a)kernel.org>
---
Benjamin Tissoires (8):
bpf: ignore sleepable prog parameter for kfuncs
bpf: add kfunc_meta parameter to push_callback_call()
bpf: implement __async and __s_async kfunc suffixes
bpf: typedef a type for the bpf_wq callbacks
selftests/bpf: rely on wq_callback_fn_t
bpf: remove one special case of is_bpf_wq_set_callback_impl_kfunc
bpf: implement __aux kfunc argument suffix to fetch prog_aux
bpf: rely on __aux suffix for bpf_wq_set_callback_impl
kernel/bpf/helpers.c | 10 +-
kernel/bpf/verifier.c | 333 +++++++++++++++++++-----
tools/testing/selftests/bpf/bpf_experimental.h | 3 +-
tools/testing/selftests/bpf/progs/wq.c | 10 +-
tools/testing/selftests/bpf/progs/wq_failures.c | 4 +-
5 files changed, 280 insertions(+), 80 deletions(-)
---
base-commit: 05cbc217aafbc631a6c2fab4accf95850cb48358
change-id: 20240507-bpf_async-bd2e65847525
Best regards,
--
Benjamin Tissoires <bentiss(a)kernel.org>
Hi,
This fix, originally intended for XFRM/IPsec, has been recommended by
Steffen Klassert to submit to the net tree.
The patch addresses a minor issue related to the IPv4 source address of
ICMP error messages, which originated from an old 2011 commit:
415b3334a21a ("icmp: Fix regression in nexthop resolution during replies.")
The omission of a "Fixes" tag in the following commit is deliberate
to prevent potential test failures and subsequent regression issues
that may arise from backporting this patch all stable kerenels.
This is a minor fix, anot not security fix.
With a seleftest I am submitting this to net-next tree.
v2->v3 : fix testscript. The IFS, space, got mangled.
v1->v2 : add kernel selftest script
Antony Antony (2):
xfrm: fix source address in icmp error generation from IPsec gateway
selftests/net: add ICMP unreachable over IPsec tunnel
net/ipv4/icmp.c | 1 -
tools/testing/selftests/net/Makefile | 1 +
tools/testing/selftests/net/xfrm_state.sh | 624 ++++++++++++++++++++++
3 files changed, 625 insertions(+), 1 deletion(-)
create mode 100755 tools/testing/selftests/net/xfrm_state.sh
--
2.30.2
Cast operation has a higher precedence than addition. The code here
wants to zero the 2nd half of the 64-bit metadata, but due to a pointer
arithmetic mistake, it writes the zero at offset 16 instead.
Just adding parentheses around "data + 4" would fix this, but I think
this will be slightly better readable with array syntax.
I was unable to test this with tools/testing/selftests/bpf/vmtest.sh,
because my glibc is newer than glibc in the provided VM image.
So I just checked the difference in the compiled code.
objdump -S tools/testing/selftests/bpf/xdp_do_redirect.test.o:
- *((__u32 *)data) = 0x42; /* metadata test value */
+ ((__u32 *)data)[0] = 0x42; /* metadata test value */
be7: 48 8d 85 30 fc ff ff lea -0x3d0(%rbp),%rax
bee: c7 00 42 00 00 00 movl $0x42,(%rax)
- *((__u32 *)data + 4) = 0;
+ ((__u32 *)data)[1] = 0;
bf4: 48 8d 85 30 fc ff ff lea -0x3d0(%rbp),%rax
- bfb: 48 83 c0 10 add $0x10,%rax
+ bfb: 48 83 c0 04 add $0x4,%rax
bff: c7 00 00 00 00 00 movl $0x0,(%rax)
Fixes: 5640b6d89434 ("selftests/bpf: fix "metadata marker" getting overwritten by the netstack")
Signed-off-by: Michal Schmidt <mschmidt(a)redhat.com>
---
tools/testing/selftests/bpf/prog_tests/xdp_do_redirect.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/tools/testing/selftests/bpf/prog_tests/xdp_do_redirect.c b/tools/testing/selftests/bpf/prog_tests/xdp_do_redirect.c
index 498d3bdaa4b0..bad0ea167be7 100644
--- a/tools/testing/selftests/bpf/prog_tests/xdp_do_redirect.c
+++ b/tools/testing/selftests/bpf/prog_tests/xdp_do_redirect.c
@@ -107,8 +107,8 @@ void test_xdp_do_redirect(void)
.attach_point = BPF_TC_INGRESS);
memcpy(&data[sizeof(__u64)], &pkt_udp, sizeof(pkt_udp));
- *((__u32 *)data) = 0x42; /* metadata test value */
- *((__u32 *)data + 4) = 0;
+ ((__u32 *)data)[0] = 0x42; /* metadata test value */
+ ((__u32 *)data)[1] = 0;
skel = test_xdp_do_redirect__open();
if (!ASSERT_OK_PTR(skel, "skel"))
--
2.44.0
From: Geliang Tang <tanggeliang(a)kylinos.cn>
v4:
- fix a bug in v3, it should be 'if (err)', not 'if (!err)'.
- move "selftests/bpf: Use log_err in network_helpers" out of this
series.
v3:
- add two more patches.
- use log_err instead of ASSERT in v3.
- let send_recv_data return int as Martin suggested.
v2:
Address Martin's comments for v1 (thanks.)
- drop patch 1, "export send_byte helper".
- drop "WRITE_ONCE(arg.stop, 0)".
- rebased.
send_recv_data will be re-used in MPTCP bpf tests, but not included
in this set because it depends on other patches that have not been
in the bpf-next yet. It will be sent as another set soon.
Geliang Tang (3):
selftests/bpf: Add struct send_recv_arg
selftests/bpf: Export send_recv_data helper
selftests/bpf: Support nonblock for send_recv_data
tools/testing/selftests/bpf/network_helpers.c | 103 ++++++++++++++++++
tools/testing/selftests/bpf/network_helpers.h | 1 +
.../selftests/bpf/prog_tests/bpf_tcp_ca.c | 71 +-----------
3 files changed, 105 insertions(+), 70 deletions(-)
--
2.40.1
Fixes clang compilation warnings by adding explicit unsigned conversion:
parse_vdso.c:206:22: warning: passing 'const char *' to parameter of
type 'const unsigned char *' converts between pointers to integer types
where one is of the unique plain 'char' type and the other is not
[-Wpointer-sign]
ver_hash = elf_hash(version);
^~~~~~~
parse_vdso.c:59:52: note: passing argument to parameter 'name' here
static unsigned long elf_hash(const unsigned char *name)
^
parse_vdso.c:207:46: warning: passing 'const char *' to parameter of
type 'const unsigned char *' converts between pointers to integer types
where one is of the unique plain 'char' type and the other is not
[-Wpointer-sign]
ELF(Word) chain = vdso_info.bucket[elf_hash(name) % vdso_info.nbucket];
^~~~
parse_vdso.c:59:52: note: passing argument to parameter 'name' here
static unsigned long elf_hash(const unsigned char *name)
Fixes: 98eedc3a9dbf ("Document the vDSO and add a reference parser")
Signed-off-by: Edward Liaw <edliaw(a)google.com>
---
v2: update commit message with correct compiler warning
v3: fix checkpatch errors and indentation
---
tools/testing/selftests/vDSO/parse_vdso.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/tools/testing/selftests/vDSO/parse_vdso.c b/tools/testing/selftests/vDSO/parse_vdso.c
index 413f75620a35..9e29ff0657ea 100644
--- a/tools/testing/selftests/vDSO/parse_vdso.c
+++ b/tools/testing/selftests/vDSO/parse_vdso.c
@@ -203,8 +203,9 @@ void *vdso_sym(const char *version, const char *name)
if (!vdso_info.valid)
return 0;
- ver_hash = elf_hash(version);
- ELF(Word) chain = vdso_info.bucket[elf_hash(name) % vdso_info.nbucket];
+ ver_hash = elf_hash((const unsigned char *)version);
+ ELF(Word) chain = vdso_info.bucket[
+ elf_hash((const unsigned char *)name) % vdso_info.nbucket];
for (; chain != STN_UNDEF; chain = vdso_info.chain[chain]) {
ELF(Sym) *sym = &vdso_info.symtab[chain];
--
2.45.0.rc0.197.gbae5840b3b-goog
When building either tools/bpf/bpftool, or tools/testing/selftests/hid,
(the same Makefile is used for these), clang generates many instances of
the following:
"clang: warning: -lLLVM-17: 'linker' input unused"
Quentin points out that the LLVM version is only required in $(LIBS),
not in $(CFLAGS), so the fix is to remove it from CFLAGS.
Suggested-by: Quentin Monnet <qmo(a)kernel.org>
Signed-off-by: John Hubbard <jhubbard(a)nvidia.com>
---
tools/bpf/bpftool/Makefile | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/bpf/bpftool/Makefile b/tools/bpf/bpftool/Makefile
index e9154ace80ff..a5445a422109 100644
--- a/tools/bpf/bpftool/Makefile
+++ b/tools/bpf/bpftool/Makefile
@@ -143,7 +143,7 @@ ifeq ($(feature-llvm),1)
# If LLVM is available, use it for JIT disassembly
CFLAGS += -DHAVE_LLVM_SUPPORT
LLVM_CONFIG_LIB_COMPONENTS := mcdisassembler all-targets
- CFLAGS += $(shell $(LLVM_CONFIG) --cflags --libs $(LLVM_CONFIG_LIB_COMPONENTS))
+ CFLAGS += $(shell $(LLVM_CONFIG) --cflags)
LIBS += $(shell $(LLVM_CONFIG) --libs $(LLVM_CONFIG_LIB_COMPONENTS))
ifeq ($(shell $(LLVM_CONFIG) --shared-mode),static)
LIBS += $(shell $(LLVM_CONFIG) --system-libs $(LLVM_CONFIG_LIB_COMPONENTS))
base-commit: f462ae0edd3703edd6f22fe41d336369c38b884b
prerequisite-patch-id: b901ece2a5b78503e2fb5480f20e304d36a0ea27
--
2.45.0
clang doesn't deal well with "-pie -static": it warns that -pie is an
unused option here. Changing to "-fPIE -static" solves this problem for
clang, while keeping the gcc results identical.
The problem is visible when building via:
make LLVM=1 -C tools/testing/selftests
Again: gcc 13 produces identical binaries for all of these programs,
both before and after this commit (using "-pie"), and after (using
"-fPIE").
Also, the runtime results are the same for both clang and gcc builds.
Signed-off-by: John Hubbard <jhubbard(a)nvidia.com>
---
tools/testing/selftests/exec/Makefile | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/tools/testing/selftests/exec/Makefile b/tools/testing/selftests/exec/Makefile
index fb4472ddffd8..b7b54d442378 100644
--- a/tools/testing/selftests/exec/Makefile
+++ b/tools/testing/selftests/exec/Makefile
@@ -29,8 +29,8 @@ $(OUTPUT)/execveat.denatured: $(OUTPUT)/execveat
cp $< $@
chmod -x $@
$(OUTPUT)/load_address_4096: load_address.c
- $(CC) $(CFLAGS) $(LDFLAGS) -Wl,-z,max-page-size=0x1000 -pie -static $< -o $@
+ $(CC) $(CFLAGS) $(LDFLAGS) -Wl,-z,max-page-size=0x1000 -fPIE -static $< -o $@
$(OUTPUT)/load_address_2097152: load_address.c
- $(CC) $(CFLAGS) $(LDFLAGS) -Wl,-z,max-page-size=0x200000 -pie -static $< -o $@
+ $(CC) $(CFLAGS) $(LDFLAGS) -Wl,-z,max-page-size=0x200000 -fPIE -static $< -o $@
$(OUTPUT)/load_address_16777216: load_address.c
- $(CC) $(CFLAGS) $(LDFLAGS) -Wl,-z,max-page-size=0x1000000 -pie -static $< -o $@
+ $(CC) $(CFLAGS) $(LDFLAGS) -Wl,-z,max-page-size=0x1000000 -fPIE -static $< -o $@
base-commit: ddb4c3f25b7b95df3d6932db0b379d768a6ebdf7
prerequisite-patch-id: b901ece2a5b78503e2fb5480f20e304d36a0ea27
--
2.45.0
When building with clang, via:
make LLVM=1 -C tools/testing/selftest
...clang warns that "a variable sized type not at the end of a struct or
class is a GNU extension".
These cases are not easily changed, because they involve structs that
are part of the API. Fortunately, however, the tests seem to be doing
just fine (specifically, neither affected test runs any differently with
gcc vs. clang builds, on my test system) regardless of the warning. So,
all the warning is doing is preventing a clean build of selftests/net.
Fix this by suppressing this particular clang warning for the
selftests/net suite.
Signed-off-by: John Hubbard <jhubbard(a)nvidia.com>
---
tools/testing/selftests/net/Makefile | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/tools/testing/selftests/net/Makefile b/tools/testing/selftests/net/Makefile
index 7b6918d5f4af..956481174783 100644
--- a/tools/testing/selftests/net/Makefile
+++ b/tools/testing/selftests/net/Makefile
@@ -6,6 +6,10 @@ CFLAGS += -I../../../../usr/include/ $(KHDR_INCLUDES)
# Additional include paths needed by kselftest.h
CFLAGS += -I../
+ifneq ($(LLVM),)
+ CFLAGS += -Wno-gnu-variable-sized-type-not-at-end
+endif
+
TEST_PROGS := run_netsocktests run_afpackettests test_bpf.sh netdevice.sh \
rtnetlink.sh xfrm_policy.sh test_blackhole_dev.sh
TEST_PROGS += fib_tests.sh fib-onlink-tests.sh pmtu.sh udpgso.sh ip_defrag.sh
base-commit: f462ae0edd3703edd6f22fe41d336369c38b884b
prerequisite-patch-id: b901ece2a5b78503e2fb5480f20e304d36a0ea27
--
2.45.0
When building with clang, via:
make LLVM=1 -C tools/testing/selftests
...clang warns, correctly, that several functions declared with an "int"
return type are not always returning values in all cases (or at least,
clang cannot prove that they always return a value).
Fix this by returning an appropriate value for each function. Thanks to
Felix Huettner for recommending MNL_CB_OK (which is non-zero) for the
return value of the count_entries() callback.
Cc: Felix Huettner <felix.huettner(a)mail.schwarz>
Signed-off-by: John Hubbard <jhubbard(a)nvidia.com>
---
tools/testing/selftests/netfilter/conntrack_dump_flush.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/tools/testing/selftests/netfilter/conntrack_dump_flush.c b/tools/testing/selftests/netfilter/conntrack_dump_flush.c
index b11ea8ee6719..e9df4ae14e16 100644
--- a/tools/testing/selftests/netfilter/conntrack_dump_flush.c
+++ b/tools/testing/selftests/netfilter/conntrack_dump_flush.c
@@ -43,6 +43,7 @@ static int build_cta_tuple_v4(struct nlmsghdr *nlh, int type,
mnl_attr_nest_end(nlh, nest_proto);
mnl_attr_nest_end(nlh, nest);
+ return 0;
}
static int build_cta_tuple_v6(struct nlmsghdr *nlh, int type,
@@ -71,6 +72,7 @@ static int build_cta_tuple_v6(struct nlmsghdr *nlh, int type,
mnl_attr_nest_end(nlh, nest_proto);
mnl_attr_nest_end(nlh, nest);
+ return 0;
}
static int build_cta_proto(struct nlmsghdr *nlh)
@@ -90,6 +92,7 @@ static int build_cta_proto(struct nlmsghdr *nlh)
mnl_attr_nest_end(nlh, nest_proto);
mnl_attr_nest_end(nlh, nest);
+ return 0;
}
static int conntrack_data_insert(struct mnl_socket *sock, struct nlmsghdr *nlh,
@@ -207,6 +210,7 @@ static int conntrack_data_generate_v6(struct mnl_socket *sock,
static int count_entries(const struct nlmsghdr *nlh, void *data)
{
reply_counter++;
+ return MNL_CB_OK;
}
static int conntracK_count_zone(struct mnl_socket *sock, uint16_t zone)
base-commit: f462ae0edd3703edd6f22fe41d336369c38b884b
prerequisite-patch-id: b901ece2a5b78503e2fb5480f20e304d36a0ea27
prerequisite-patch-id: 9db2d20be98dc44731d8605a3da64ff118d2546d
--
2.45.0
On 5/6/24 7:41 AM, Felix Huettner wrote:
> On Sun, May 05, 2024 at 02:47:16PM -0700, John Hubbard wrote:
...
> > @@ -207,6 +210,7 @@ static int conntrack_data_generate_v6(struct
> mnl_socket *sock,
> > static int count_entries(const struct nlmsghdr *nlh, void *data)
> > {
> > reply_counter++;
> > + return 0;
>
> Hi John,
>
> This will need to return MNL_CB_OK.
> Otherwise mnl_cb_run below will abort early and the connection count
> will be wrong.
>
Thanks for catching that, I'm sending a v2 with that fix.
I was thinking about it, and expected that the pre-existing code
appeared to work because the return value was some non-zero garbage
value scrounged off of the stack (or %rax, for example on x86).
However, just a quick test showed that *any* value (O, 1==MNL_CB_OK,
or no value at all) allows the test to report success...oh, I see,
it's reporting PASSED when it really ought to say SKIPPED:
$ ./conntrack_dump_flush
TAP version 13
1..3
# Starting 3 tests from 1 test cases.
# RUN conntrack_dump_flush.test_dump_by_zone ...
mnl_socket_open: Protocol not supported
# OK conntrack_dump_flush.test_dump_by_zone
ok 1 conntrack_dump_flush.test_dump_by_zone
# RUN conntrack_dump_flush.test_flush_by_zone ...
mnl_socket_open: Protocol not supported
# OK conntrack_dump_flush.test_flush_by_zone
ok 2 conntrack_dump_flush.test_flush_by_zone
# RUN conntrack_dump_flush.test_flush_by_zone_default ...
mnl_socket_open: Protocol not supported
# OK conntrack_dump_flush.test_flush_by_zone_default
ok 3 conntrack_dump_flush.test_flush_by_zone_default
# PASSED: 3 / 3 tests passed.
# Totals: pass:3 fail:0 xfail:0 xpass:0 skip:0 error:0
As long as we are looking at this, what do you think about
this:
diff --git a/tools/testing/selftests/netfilter/conntrack_dump_flush.c
b/tools/testing/selftests/netfilter/conntrack_dump_flush.c
index e9df4ae14e16..4a73afad4de4 100644
--- a/tools/testing/selftests/netfilter/conntrack_dump_flush.c
+++ b/tools/testing/selftests/netfilter/conntrack_dump_flush.c
@@ -317,12 +317,12 @@ FIXTURE_SETUP(conntrack_dump_flush)
self->sock = mnl_socket_open(NETLINK_NETFILTER);
if (!self->sock) {
perror("mnl_socket_open");
- exit(EXIT_FAILURE);
+ SKIP(exit(EXIT_FAILURE), "mnl_socket_open() failed");
}
if (mnl_socket_bind(self->sock, 0, MNL_SOCKET_AUTOPID) < 0) {
perror("mnl_socket_bind");
- exit(EXIT_FAILURE);
+ SKIP(exit(EXIT_FAILURE), "mnl_socket_bind() failed");
}
ret = conntracK_count_zone(self->sock, TEST_ZONE_ID);
...which changes the above output, to:
$ ./conntrack_dump_flush
TAP version 13
1..3
# Starting 3 tests from 1 test cases.
# RUN conntrack_dump_flush.test_dump_by_zone ...
mnl_socket_open: Protocol not supported
# SKIP mnl_socket_open() failed
# OK conntrack_dump_flush.test_dump_by_zone
ok 1 conntrack_dump_flush.test_dump_by_zone # SKIP mnl_socket_open() failed
# RUN conntrack_dump_flush.test_flush_by_zone ...
mnl_socket_open: Protocol not supported
# SKIP mnl_socket_open() failed
# OK conntrack_dump_flush.test_flush_by_zone
ok 2 conntrack_dump_flush.test_flush_by_zone # SKIP mnl_socket_open() failed
# RUN conntrack_dump_flush.test_flush_by_zone_default ...
mnl_socket_open: Protocol not supported
# SKIP mnl_socket_open() failed
# OK conntrack_dump_flush.test_flush_by_zone_default
ok 3 conntrack_dump_flush.test_flush_by_zone_default # SKIP
mnl_socket_open() failed
# PASSED: 3 / 3 tests passed.
# Totals: pass:0 fail:0 xfail:0 xpass:0 skip:3 error:0
?
thanks,
--
John Hubbard
NVIDIA
Fixes clang compilation warnings by changing elf_hash's parameter type
to char * and casting to unsigned char * inside elf_hash:
parse_vdso.c:206:22: warning: passing 'const char *' to parameter of
type 'const unsigned char *' converts between pointers to integer types
where one is of the unique plain 'char' type and the other is not
[-Wpointer-sign]
ver_hash = elf_hash(version);
^~~~~~~
parse_vdso.c:59:52: note: passing argument to parameter 'name' here
static unsigned long elf_hash(const unsigned char *name)
^
parse_vdso.c:207:46: warning: passing 'const char *' to parameter of
type 'const unsigned char *' converts between pointers to integer types
where one is of the unique plain 'char' type and the other is not
[-Wpointer-sign]
ELF(Word) chain = vdso_info.bucket[elf_hash(name) % vdso_info.nbucket];
^~~~
parse_vdso.c:59:52: note: passing argument to parameter 'name' here
static unsigned long elf_hash(const unsigned char *name)
Fixes: 98eedc3a9dbf ("Document the vDSO and add a reference parser")
Signed-off-by: Edward Liaw <edliaw(a)google.com>
---
v2: updated commit message with correct compiler warning
v3: fixed checkpatch errors and indentation
https://lore.kernel.org/all/20240501180622.1676340-1-edliaw@google.com/
v4: moved the typecast into elf_hash based on libelf
https://sourceforge.net/p/elftoolchain/code/HEAD/tree/trunk/libelf/elf_hash…
---
tools/testing/selftests/vDSO/parse_vdso.c | 9 ++++++---
1 file changed, 6 insertions(+), 3 deletions(-)
diff --git a/tools/testing/selftests/vDSO/parse_vdso.c b/tools/testing/selftests/vDSO/parse_vdso.c
index 413f75620a35..33db8abd7d59 100644
--- a/tools/testing/selftests/vDSO/parse_vdso.c
+++ b/tools/testing/selftests/vDSO/parse_vdso.c
@@ -56,12 +56,15 @@ static struct vdso_info
} vdso_info;
/* Straight from the ELF specification. */
-static unsigned long elf_hash(const unsigned char *name)
+static unsigned long elf_hash(const char *name)
{
unsigned long h = 0, g;
- while (*name)
+ const unsigned char *s;
+
+ s = (const unsigned char *) name;
+ while (*s)
{
- h = (h << 4) + *name++;
+ h = (h << 4) + *s++;
if (g = h & 0xf0000000)
h ^= g >> 24;
h &= ~g;
--
2.45.0.rc1.225.g2a3ae87e7f-goog
dump_config_tree() is declared to return an int, but the compiler cannot
prove that it always returns any value at all. This leads to a clang
warning, when building via:
make LLVM=1 -C tools/testing/selftests
Furthermore, Mark Brown noticed that dump_config_tree() isn't even used
anymore, so just delete the entire function.
Cc: Mark Brown <broonie(a)kernel.org>
Signed-off-by: John Hubbard <jhubbard(a)nvidia.com>
---
tools/testing/selftests/alsa/conf.c | 13 -------------
1 file changed, 13 deletions(-)
diff --git a/tools/testing/selftests/alsa/conf.c b/tools/testing/selftests/alsa/conf.c
index 89e3656a042d..357561c1759b 100644
--- a/tools/testing/selftests/alsa/conf.c
+++ b/tools/testing/selftests/alsa/conf.c
@@ -105,19 +105,6 @@ static struct card_cfg_data *conf_data_by_card(int card, bool msg)
return NULL;
}
-static int dump_config_tree(snd_config_t *top)
-{
- snd_output_t *out;
- int err;
-
- err = snd_output_stdio_attach(&out, stdout, 0);
- if (err < 0)
- ksft_exit_fail_msg("stdout attach\n");
- if (snd_config_save(top, out))
- ksft_exit_fail_msg("config save\n");
- snd_output_close(out);
-}
-
snd_config_t *conf_load_from_file(const char *filename)
{
snd_config_t *dst;
base-commit: f462ae0edd3703edd6f22fe41d336369c38b884b
prerequisite-patch-id: b901ece2a5b78503e2fb5480f20e304d36a0ea27
--
2.45.0
First of all, in order to build with clang at all, one must first apply
Valentin Obst's build fix for LLVM [1]. Furthermore, for this particular
resctrl directory, my pending fix [2] must also be applied. Once those
fixes are in place, then when building with clang, via:
make LLVM=1 -C tools/testing/selftests
...two types of warnings occur:
warning: absolute value function 'abs' given an argument of type
'long' but has parameter of type 'int' which may cause truncation of
value
warning: taking the absolute value of unsigned type 'unsigned long'
has no effect
Fix these by:
a) using labs() in place of abs(), when long integers are involved, and
b) don't call labs() unnecessarily.
[1] https://lore.kernel.org/all/20240329-selftests-libmk-llvm-rfc-v1-1-2f9ed7d1…
[2] https://lore.kernel.org/all/20240503021712.78601-1-jhubbard@nvidia.com/
Signed-off-by: John Hubbard <jhubbard(a)nvidia.com>
---
tools/testing/selftests/resctrl/cmt_test.c | 4 ++--
tools/testing/selftests/resctrl/mba_test.c | 2 +-
tools/testing/selftests/resctrl/mbm_test.c | 2 +-
3 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/tools/testing/selftests/resctrl/cmt_test.c b/tools/testing/selftests/resctrl/cmt_test.c
index a81f91222a89..05a241519ae8 100644
--- a/tools/testing/selftests/resctrl/cmt_test.c
+++ b/tools/testing/selftests/resctrl/cmt_test.c
@@ -40,11 +40,11 @@ static int show_results_info(unsigned long sum_llc_val, int no_of_bits,
int ret;
avg_llc_val = sum_llc_val / num_of_runs;
- avg_diff = (long)abs(cache_span - avg_llc_val);
+ avg_diff = (long)(cache_span - avg_llc_val);
diff_percent = ((float)cache_span - avg_llc_val) / cache_span * 100;
ret = platform && abs((int)diff_percent) > max_diff_percent &&
- abs(avg_diff) > max_diff;
+ labs(avg_diff) > max_diff;
ksft_print_msg("%s Check cache miss rate within %lu%%\n",
ret ? "Fail:" : "Pass:", max_diff_percent);
diff --git a/tools/testing/selftests/resctrl/mba_test.c b/tools/testing/selftests/resctrl/mba_test.c
index 7946e32e85c8..673b2bb800f7 100644
--- a/tools/testing/selftests/resctrl/mba_test.c
+++ b/tools/testing/selftests/resctrl/mba_test.c
@@ -77,7 +77,7 @@ static bool show_mba_info(unsigned long *bw_imc, unsigned long *bw_resc)
avg_bw_imc = sum_bw_imc / (NUM_OF_RUNS - 1);
avg_bw_resc = sum_bw_resc / (NUM_OF_RUNS - 1);
- avg_diff = (float)labs(avg_bw_resc - avg_bw_imc) / avg_bw_imc;
+ avg_diff = (float)(avg_bw_resc - avg_bw_imc) / avg_bw_imc;
avg_diff_per = (int)(avg_diff * 100);
ksft_print_msg("%s Check MBA diff within %d%% for schemata %u\n",
diff --git a/tools/testing/selftests/resctrl/mbm_test.c b/tools/testing/selftests/resctrl/mbm_test.c
index d67ffa3ec63a..c873793d016d 100644
--- a/tools/testing/selftests/resctrl/mbm_test.c
+++ b/tools/testing/selftests/resctrl/mbm_test.c
@@ -33,7 +33,7 @@ show_bw_info(unsigned long *bw_imc, unsigned long *bw_resc, size_t span)
avg_bw_imc = sum_bw_imc / 4;
avg_bw_resc = sum_bw_resc / 4;
- avg_diff = (float)labs(avg_bw_resc - avg_bw_imc) / avg_bw_imc;
+ avg_diff = (float)(avg_bw_resc - avg_bw_imc) / avg_bw_imc;
avg_diff_per = (int)(avg_diff * 100);
ret = avg_diff_per > MAX_DIFF_PERCENT;
base-commit: f03359bca01bf4372cf2c118cd9a987a5951b1c8
prerequisite-patch-id: b901ece2a5b78503e2fb5480f20e304d36a0ea27
prerequisite-patch-id: 8d96c4b8c3ed6d9ea2588ef7f594ae0f9f83c279
--
2.45.0
Hi,
This fifth series fixes _metadata reset and fixes the last patch to
handle code set with direct calls to _exit().
As reported by Kernel Test Robot [1], some pidfd tests fail. This is
due to the use of vfork() which introduced some side effects.
Similarly, while making it more generic, a previous commit made some
Landlock file system tests flaky, and subject to the host's file system
mount configuration.
This series fixes all these side effects by replacing vfork() with
clone3() and CLONE_VFORK, which is cleaner (no arbitrary shared memory)
and makes the Kselftest framework more robust.
I tried different approaches and I found this one to be the cleaner and
less invasive for current test cases.
I successfully ran the following tests (using TEST_F and
fork/clone/clone3, and KVM_ONE_VCPU_TEST) with this series:
- kvm:fix_hypercall_test
- kvm:sync_regs_test
- kvm:userspace_msr_exit_test
- kvm:vmx_pmu_caps_test
- landlock:fs_test
- landlock:net_test
- landlock:ptrace_test
- move_mount_set_group:move_mount_set_group_test
- net/af_unix:scm_pidfd
- perf_events:remove_on_exec
- pidfd:pidfd_getfd_test
- pidfd:pidfd_setns_test
- seccomp:seccomp_bpf
- user_events:abi_test
[1] https://lore.kernel.org/oe-lkp/202403291015.1fcfa957-oliver.sang@intel.com
Previous versions:
v1: https://lore.kernel.org/r/20240426172252.1862930-1-mic@digikod.net
v2: https://lore.kernel.org/r/20240429130931.2394118-1-mic@digikod.net
v3: https://lore.kernel.org/r/20240429191911.2552580-1-mic@digikod.net
v4: https://lore.kernel.org/r/20240502210926.145539-1-mic@digikod.net
Regards,
Mickaël Salaün (10):
selftests/pidfd: Fix config for pidfd_setns_test
selftests/landlock: Fix FS tests when run on a private mount point
selftests/harness: Fix fixture teardown
selftests/harness: Fix interleaved scheduling leading to race
conditions
selftests/landlock: Do not allocate memory in fixture data
selftests/harness: Constify fixture variants
selftests/pidfd: Fix wrong expectation
selftests/harness: Share _metadata between forked processes
selftests/harness: Fix vfork() side effects
selftests/harness: Handle TEST_F()'s explicit exit codes
tools/testing/selftests/kselftest_harness.h | 122 +++++++++++++-----
tools/testing/selftests/landlock/fs_test.c | 83 +++++++-----
tools/testing/selftests/pidfd/config | 2 +
.../selftests/pidfd/pidfd_setns_test.c | 2 +-
4 files changed, 143 insertions(+), 66 deletions(-)
base-commit: e67572cd2204894179d89bd7b984072f19313b03
--
2.45.0
When building either tools/bpf/bpftool, or tools/testing/selftests/hid,
(the same Makefile is used for these), clang generates many instances of
a warning that is useless here:
"clang: warning: -lLLVM-17: 'linker' input unused"
Silence this in both locations, by disabling that warning when building
with clang.
Signed-off-by: John Hubbard <jhubbard(a)nvidia.com>
---
tools/bpf/bpftool/Makefile | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/tools/bpf/bpftool/Makefile b/tools/bpf/bpftool/Makefile
index e9154ace80ff..c7457921d136 100644
--- a/tools/bpf/bpftool/Makefile
+++ b/tools/bpf/bpftool/Makefile
@@ -133,6 +133,10 @@ CFLAGS += -DUSE_LIBCAP
LIBS += -lcap
endif
+ifneq ($(LLVM),)
+ CFLAGS += -Wno-unused-command-line-argument
+endif
+
include $(wildcard $(OUTPUT)*.d)
all: $(OUTPUT)bpftool
base-commit: f462ae0edd3703edd6f22fe41d336369c38b884b
prerequisite-patch-id: b901ece2a5b78503e2fb5480f20e304d36a0ea27
--
2.45.0
When building with clang, via:
make LLVM=1 -C tools/testing/selftest
...clang warns about using "int *" interchangeably with "socklen_t *".
clang is correct, so fix this by declaring len as a socklen_t, instead
of as an int.
Signed-off-by: John Hubbard <jhubbard(a)nvidia.com>
---
tools/testing/selftests/netfilter/sctp_collision.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/tools/testing/selftests/netfilter/sctp_collision.c b/tools/testing/selftests/netfilter/sctp_collision.c
index 21bb1cfd8a85..91df996367e9 100644
--- a/tools/testing/selftests/netfilter/sctp_collision.c
+++ b/tools/testing/selftests/netfilter/sctp_collision.c
@@ -9,7 +9,8 @@
int main(int argc, char *argv[])
{
struct sockaddr_in saddr = {}, daddr = {};
- int sd, ret, len = sizeof(daddr);
+ int sd, ret;
+ socklen_t len = sizeof(daddr);
struct timeval tv = {25, 0};
char buf[] = "hello";
base-commit: f462ae0edd3703edd6f22fe41d336369c38b884b
prerequisite-patch-id: b901ece2a5b78503e2fb5480f20e304d36a0ea27
--
2.45.0
When building with clang, via:
make LLVM=1 -C tools/testing/selftest
...clang warns about "taking address of packed member 'write_index' ".
This is not particularly helpful, because the test code really wants to
write to exactly this location, and if it ends up being unaligned, then
the test won't work (and may segfault, depending on the CPU type).
If that ever comes up, it will be obvious and can be fixed. But all it's
doing now is prevent a clean clang build, so disable the warning.
Signed-off-by: John Hubbard <jhubbard(a)nvidia.com>
---
tools/testing/selftests/user_events/Makefile | 5 +++++
1 file changed, 5 insertions(+)
diff --git a/tools/testing/selftests/user_events/Makefile b/tools/testing/selftests/user_events/Makefile
index 10fcd0066203..617e94344711 100644
--- a/tools/testing/selftests/user_events/Makefile
+++ b/tools/testing/selftests/user_events/Makefile
@@ -1,5 +1,10 @@
# SPDX-License-Identifier: GPL-2.0
CFLAGS += -Wl,-no-as-needed -Wall $(KHDR_INCLUDES)
+
+ifneq ($(LLVM),)
+ CFLAGS += -Wno-address-of-packed-member
+endif
+
LDLIBS += -lrt -lpthread -lm
TEST_GEN_PROGS = ftrace_test dyn_test perf_test abi_test
base-commit: f462ae0edd3703edd6f22fe41d336369c38b884b
prerequisite-patch-id: b901ece2a5b78503e2fb5480f20e304d36a0ea27
--
2.45.0
dump_config_tree() is declared to return an int, but the compiler cannot
prove that it always returns any value at all. This leads to a clang
warning, when building via:
make LLVM=1 -C tools/testing/selftests
Fix this by unconditionally returning the "err" variable if the code
reaches the end of the function.
Signed-off-by: John Hubbard <jhubbard(a)nvidia.com>
---
tools/testing/selftests/alsa/conf.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/tools/testing/selftests/alsa/conf.c b/tools/testing/selftests/alsa/conf.c
index 89e3656a042d..0109fde53e6f 100644
--- a/tools/testing/selftests/alsa/conf.c
+++ b/tools/testing/selftests/alsa/conf.c
@@ -116,6 +116,8 @@ static int dump_config_tree(snd_config_t *top)
if (snd_config_save(top, out))
ksft_exit_fail_msg("config save\n");
snd_output_close(out);
+
+ return err;
}
snd_config_t *conf_load_from_file(const char *filename)
base-commit: ddb4c3f25b7b95df3d6932db0b379d768a6ebdf7
prerequisite-patch-id: b901ece2a5b78503e2fb5480f20e304d36a0ea27
--
2.45.0
Hi,
This fix, originally intended for XFRM/IPsec, has been recommended by
Steffen Klassert to submit to the net tree.
The patch addresses a minor issue related to the IPv4 source address of
ICMP error messages, which originated from an old 2011 commit:
415b3334a21a ("icmp: Fix regression in nexthop resolution during replies.")
The omission of a "Fixes" tag in the following commit is deliberate
to prevent potential test failures and subsequent regression issues
that may arise from backporting this patch all stable kerenels.
This is a minor fix, anot not security fix.
With a seleftest I am submitting this to net-next tree.
v1->v2 : add kernel selftest script
Antony Antony (2):
xfrm: fix source address in icmp error generation from IPsec gateway
selftests: add ICMP unreachable over IPsec tunnel
net/ipv4/icmp.c | 1 -
tools/testing/selftests/net/Makefile | 1 +
tools/testing/selftests/net/xfrm_state.sh | 624 ++++++++++++++++++++++
3 files changed, 625 insertions(+), 1 deletion(-)
create mode 100755 tools/testing/selftests/net/xfrm_state.sh
--
2.30.2
In the function l5_test(), variable $tests is empty when there is no .mk
file in the subsystem to be tested. It causes the following grep operation
stuck.
This fix check the variable $tests, return when it is empty.
Signed-off-by: Lu Dai <dai.lu(a)exordes.com>
---
tools/testing/selftests/kselftest_deps.sh | 1 +
1 file changed, 1 insertion(+)
diff --git a/tools/testing/selftests/kselftest_deps.sh b/tools/testing/selftests/kselftest_deps.sh
index de59cc8f03c3..487e49fdf2a6 100755
--- a/tools/testing/selftests/kselftest_deps.sh
+++ b/tools/testing/selftests/kselftest_deps.sh
@@ -244,6 +244,7 @@ l4_test()
l5_test()
{
tests=$(find $(dirname "$test") -type f -name "*.mk")
+ [[ -z "${tests// }" ]] && return
test_libs=$(grep "^IOURING_EXTRA_LIBS +\?=" $tests | \
cut -d "=" -f 2)
--
2.39.2
From: Masami Hiramatsu (Google) <mhiramat(a)kernel.org>
Since the VFS type argument test case uses fprobe events, it must
check the availablity of dynamic_events file and fprobe events syntax
in README. Without this fix, the test fails if CONFIG_FPROBE_EVENTS=n.
Fixes: ee97e5e135c6 ("selftests/ftrace: add fprobe test cases for VFS type "%pd" and "%pD"")
Signed-off-by: Masami Hiramatsu (Google) <mhiramat(a)kernel.org>
---
.../ftrace/test.d/dynevent/fprobe_args_vfs.tc | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/tools/testing/selftests/ftrace/test.d/dynevent/fprobe_args_vfs.tc b/tools/testing/selftests/ftrace/test.d/dynevent/fprobe_args_vfs.tc
index 49a833bf334c..c6a9d2466a71 100644
--- a/tools/testing/selftests/ftrace/test.d/dynevent/fprobe_args_vfs.tc
+++ b/tools/testing/selftests/ftrace/test.d/dynevent/fprobe_args_vfs.tc
@@ -1,7 +1,8 @@
#!/bin/sh
# SPDX-License-Identifier: GPL-2.0
# description: Fprobe event VFS type argument
-# requires: kprobe_events "%pd/%pD":README
+# requires: dynamic_events "%pd/%pD":README "f[:[<group>/][<event>]] <func-name>[%return] [<args>]":README
+
: "Test argument %pd with name for fprobe"
echo 'f:testprobe dput name=$arg1:%pd' > dynamic_events
Here is a couple of patches for fixing errors on ftracetest.
Shuah, can you pick these to your fixes branch? Or I also can push it.
Thank you,
---
Masami Hiramatsu (Google) (2):
selftests/ftrace: Fix BTFARG testcase to check fprobe is enabled correctly
selftests/ftrace: Fix checkbashisms errors
.../ftrace/test.d/dynevent/add_remove_btfarg.tc | 2 +-
.../ftrace/test.d/dynevent/fprobe_entry_arg.tc | 2 +-
.../ftrace/test.d/kprobe/kretprobe_entry_arg.tc | 2 +-
3 files changed, 3 insertions(+), 3 deletions(-)
--
Masami Hiramatsu (Google) <mhiramat(a)kernel.org>
Use kselftest_harness.h to simplify the code structure by eliminating
conditional logic. Enhance diagnostics by directly printing relevant info,
such as access and modification times, upon test failure. Reflecting
common I/O optimizations, the access time usually remains unchanged, while
the modify time is expected to update. Accordingly, these elements have
been logically separated.
Signed-off-by: Shengyu Li <shengyu.li.evgeny(a)gmail.com>
---
v3: Explain the need for refactoring
v2: Fixed the last Assert
---
.../testing/selftests/tty/tty_tstamp_update.c | 49 +++++++++----------
1 file changed, 22 insertions(+), 27 deletions(-)
diff --git a/tools/testing/selftests/tty/tty_tstamp_update.c b/tools/testing/selftests/tty/tty_tstamp_update.c
index 0ee97943dccc..38de211e0715 100644
--- a/tools/testing/selftests/tty/tty_tstamp_update.c
+++ b/tools/testing/selftests/tty/tty_tstamp_update.c
@@ -9,7 +9,7 @@
#include <unistd.h>
#include <linux/limits.h>
-#include "../kselftest.h"
+#include "../kselftest_harness.h"
#define MIN_TTY_PATH_LEN 8
@@ -42,47 +42,42 @@ static int write_dev_tty(void)
return r;
}
-int main(int argc, char **argv)
+TEST(tty_tstamp_update)
{
int r;
char tty[PATH_MAX] = {};
struct stat st1, st2;
- ksft_print_header();
- ksft_set_plan(1);
+ ASSERT_GE(readlink("/proc/self/fd/0", tty, PATH_MAX), 0)
+ TH_LOG("readlink on /proc/self/fd/0 failed: %m");
- r = readlink("/proc/self/fd/0", tty, PATH_MAX);
- if (r < 0)
- ksft_exit_fail_msg("readlink on /proc/self/fd/0 failed: %m\n");
-
- if (!tty_valid(tty))
- ksft_exit_skip("invalid tty path '%s'\n", tty);
+ ASSERT_TRUE(tty_valid(tty)) {
+ TH_LOG("SKIP: invalid tty path '%s'", tty);
+ _exit(KSFT_SKIP);
+ }
- r = stat(tty, &st1);
- if (r < 0)
- ksft_exit_fail_msg("stat failed on tty path '%s': %m\n", tty);
+ ASSERT_GE(stat(tty, &st1), 0)
+ TH_LOG("stat failed on tty path '%s': %m", tty);
/* We need to wait at least 8 seconds in order to observe timestamp change */
/* https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… */
sleep(10);
r = write_dev_tty();
- if (r < 0)
- ksft_exit_fail_msg("failed to write to /dev/tty: %s\n",
- strerror(-r));
+ ASSERT_GE(r, 0)
+ TH_LOG("failed to write to /dev/tty: %s", strerror(-r));
- r = stat(tty, &st2);
- if (r < 0)
- ksft_exit_fail_msg("stat failed on tty path '%s': %m\n", tty);
+ ASSERT_GE(stat(tty, &st2), 0)
+ TH_LOG("stat failed on tty path '%s': %m", tty);
+
+ /* Validate unchanged atime under 'relatime' to ensure minimal disk I/O */
+ EXPECT_EQ(st1.st_atim.tv_sec, st2.st_atim.tv_sec);
/* We wrote to the terminal so timestamps should have been updated */
- if (st1.st_atim.tv_sec == st2.st_atim.tv_sec &&
- st1.st_mtim.tv_sec == st2.st_mtim.tv_sec) {
- ksft_test_result_fail("tty timestamps not updated\n");
- ksft_exit_fail();
- }
+ ASSERT_NE(st1.st_mtim.tv_sec, st2.st_mtim.tv_sec)
+ TH_LOG("tty timestamps not updated");
- ksft_test_result_pass(
- "timestamps of terminal '%s' updated after write to /dev/tty\n", tty);
- return EXIT_SUCCESS;
+ TH_LOG("timestamps of terminal '%s' updated after write to /dev/tty",
+ tty);
}
+TEST_HARNESS_MAIN
--
2.25.1
Currently, the migration worker delays 1-10 us, assuming that one
KVM_RUN iteration only takes a few microseconds. But if the CPU low
power wakeup latency is large enough, for example, hundreds or even
thousands of microseconds deep C-state exit latencies on x86 server
CPUs, it may happen that it's not able to wakeup the target CPU before
the migration worker starts to migrate the vCPU thread to the next CPU.
If the system workload is light, most CPUs could be at a certain low
power state, which may result in less successful migrations and fail the
migration/KVM_RUN ratio sanity check. But this is not supposed to be
deemed a test failure.
This patch adds a command line option to skip the sanity check in
this case.
Co-developed-by: Dongsheng Zhang <dongsheng.x.zhang(a)intel.com>
Signed-off-by: Dongsheng Zhang <dongsheng.x.zhang(a)intel.com>
Signed-off-by: Zide Chen <zide.chen(a)intel.com>
---
V2:
- removed the busy loop implementation
- add the new "-s" option
V3:
- drop the usleep randomization code
- removed the term C-state for less confusion for non-x86 archetectures
- changed patch subject
v4:
- replaced Signed-off-by with Co-developed-by
- changed command line option from "-s" to "-u"
- Adopted the much clearer assertion error messages provided by Sean.
V5:
- Fixed the missing SoB
---
tools/testing/selftests/kvm/rseq_test.c | 35 +++++++++++++++++++++++--
1 file changed, 33 insertions(+), 2 deletions(-)
diff --git a/tools/testing/selftests/kvm/rseq_test.c b/tools/testing/selftests/kvm/rseq_test.c
index 28f97fb52044..ad418a5c59dd 100644
--- a/tools/testing/selftests/kvm/rseq_test.c
+++ b/tools/testing/selftests/kvm/rseq_test.c
@@ -186,12 +186,35 @@ static void calc_min_max_cpu(void)
"Only one usable CPU, task migration not possible");
}
+static void help(const char *name)
+{
+ puts("");
+ printf("usage: %s [-h] [-u]\n", name);
+ printf(" -u: Don't sanity check the number of successful KVM_RUNs\n");
+ puts("");
+ exit(0);
+}
+
int main(int argc, char *argv[])
{
int r, i, snapshot;
struct kvm_vm *vm;
struct kvm_vcpu *vcpu;
u32 cpu, rseq_cpu;
+ bool skip_sanity_check = false;
+ int opt;
+
+ while ((opt = getopt(argc, argv, "hu")) != -1) {
+ switch (opt) {
+ case 'u':
+ skip_sanity_check = true;
+ break;
+ case 'h':
+ default:
+ help(argv[0]);
+ break;
+ }
+ }
r = sched_getaffinity(0, sizeof(possible_mask), &possible_mask);
TEST_ASSERT(!r, "sched_getaffinity failed, errno = %d (%s)", errno,
@@ -254,9 +277,17 @@ int main(int argc, char *argv[])
* getcpu() to stabilize. A 2:1 migration:KVM_RUN ratio is a fairly
* conservative ratio on x86-64, which can do _more_ KVM_RUNs than
* migrations given the 1us+ delay in the migration task.
+ *
+ * Another reason why it may have small migration:KVM_RUN ratio is that,
+ * on systems with large low power mode wakeup latency, it may happen
+ * quite often that the scheduler is not able to wake up the target CPU
+ * before the vCPU thread is scheduled to another CPU.
*/
- TEST_ASSERT(i > (NR_TASK_MIGRATIONS / 2),
- "Only performed %d KVM_RUNs, task stalled too much?", i);
+ TEST_ASSERT(skip_sanity_check || i > (NR_TASK_MIGRATIONS / 2),
+ "Only performed %d KVM_RUNs, task stalled too much? \n"
+ " Try disabling deep sleep states to reduce CPU wakeup latency,\n"
+ " e.g. via cpuidle.off=1 or setting /dev/cpu_dma_latency to '0',\n"
+ " or run with -u to disable this sanity check.", i);
pthread_join(migration_thread, NULL);
--
2.34.1
Align the behavior for gcc and clang builds by interpreting unset
`ARCH` and `CROSS_COMPILE` variables in `LLVM` builds as a sign that the
user wants to build for the host architecture.
This patch preserves the properties that setting the `ARCH` variable to an
unknown value will trigger an error that complains about insufficient
information, and that a set `CROSS_COMPILE` variable will override the
target triple that is determined based on presence/absence of `ARCH`.
When compiling with clang, i.e., `LLVM` is set, an unset `ARCH` variable in
combination with an unset `CROSS_COMPILE` variable, i.e., compiling for
the host architecture, leads to compilation failures since `lib.mk` can
not determine the clang target triple. In this case, the following error
message is displayed for each subsystem that does not set `ARCH` in its
own Makefile before including `lib.mk` (lines wrapped at 75 chrs):
make[1]: Entering directory '/mnt/build/linux/tools/testing/selftests/
sysctl'
../lib.mk:33: *** Specify CROSS_COMPILE or add '--target=' option to
lib.mk. Stop.
make[1]: Leaving directory '/mnt/build/linux/tools/testing/selftests/
sysctl'
In the same scenario a gcc build would default to the host architecture,
i.e., it would use plain `gcc`.
Fixes: 795285ef2425 ("selftests: Fix clang cross compilation")
Reviewed-by: Mark Brown <broonie(a)kernel.org>
Signed-off-by: Valentin Obst <kernel(a)valentinobst.de>
---
I am not entirely sure whether this behavior is in fact known and intended
and whether the way to obtain the host target triple is sufficiently
general. The flag was introduced in llvm-8 with [1], it will be an error in
older clang versions.
The target triple you get with `-print-target-triple` may not be the
same that you would get when explicitly setting ARCH to you host
architecture. For example on my x86_64 system it get
`x86_64-pc-linux-gnu` instead of `x86_64-linux-gnu`, similar deviations
were observed when testing other clang binaries on compiler-explorer,
e.g., [2].
An alternative could be to simply do:
ARCH ?= $(shell uname -m)
before using it to select the target. Possibly with some post processing,
but at that point we would likely be replicating `scripts/subarch.include`.
This is what some subsystem Makefiles do before including `lib.mk`. This
change might make it possible to remove the explicit setting of `ARCH` from
the few subsystem Makefiles that do it.
[1]: https://reviews.llvm.org/D50755
[2]: https://godbolt.org/z/r7Gn9bvv1
Changes in v1:
- Shortened commit message.
- Link to RFC: https://lore.kernel.org/r/20240303-selftests-libmk-llvm-rfc-v1-1-9ab53e365e…
---
tools/testing/selftests/lib.mk | 12 ++++++++++--
1 file changed, 10 insertions(+), 2 deletions(-)
diff --git a/tools/testing/selftests/lib.mk b/tools/testing/selftests/lib.mk
index da2cade3bab0..8ae203d8ed7f 100644
--- a/tools/testing/selftests/lib.mk
+++ b/tools/testing/selftests/lib.mk
@@ -7,6 +7,8 @@ else ifneq ($(filter -%,$(LLVM)),)
LLVM_SUFFIX := $(LLVM)
endif
+CLANG := $(LLVM_PREFIX)clang$(LLVM_SUFFIX)
+
CLANG_TARGET_FLAGS_arm := arm-linux-gnueabi
CLANG_TARGET_FLAGS_arm64 := aarch64-linux-gnu
CLANG_TARGET_FLAGS_hexagon := hexagon-linux-musl
@@ -18,7 +20,13 @@ CLANG_TARGET_FLAGS_riscv := riscv64-linux-gnu
CLANG_TARGET_FLAGS_s390 := s390x-linux-gnu
CLANG_TARGET_FLAGS_x86 := x86_64-linux-gnu
CLANG_TARGET_FLAGS_x86_64 := x86_64-linux-gnu
-CLANG_TARGET_FLAGS := $(CLANG_TARGET_FLAGS_$(ARCH))
+
+# Default to host architecture if ARCH is not explicitly given.
+ifeq ($(ARCH),)
+CLANG_TARGET_FLAGS := $(shell $(CLANG) -print-target-triple)
+else
+CLANG_TARGET_FLAGS := $(CLANG_TARGET_FLAGS_$(ARCH))
+endif
ifeq ($(CROSS_COMPILE),)
ifeq ($(CLANG_TARGET_FLAGS),)
@@ -30,7 +38,7 @@ else
CLANG_FLAGS += --target=$(notdir $(CROSS_COMPILE:%-=%))
endif # CROSS_COMPILE
-CC := $(LLVM_PREFIX)clang$(LLVM_SUFFIX) $(CLANG_FLAGS) -fintegrated-as
+CC := $(CLANG) $(CLANG_FLAGS) -fintegrated-as
else
CC := $(CROSS_COMPILE)gcc
endif # LLVM
---
base-commit: 4cece764965020c22cff7665b18a012006359095
change-id: 20240303-selftests-libmk-llvm-rfc-5fe3cfa9f094
Best regards,
--
Valentin Obst <kernel(a)valentinobst.de>