Syscall User Dispatch is presently built under CONFIG_GENERIC_SYSCALL and
cannot be disabled independently.
Add CONFIG_SYSCALL_USER_DISPATCH to make it an optional feature.
Signed-off-by: Gregory Price <[email protected]>
Signed-off-by: Thomas Gleixner <[email protected]>
Link: https://patch.msgid.link/[email protected]
Successful syscalls don't change errno, so checking errno is wrong
to ensure that a syscall has failed. For example for the following
sequence:
prctl(PR_SET_SYSCALL_USER_DISPATCH, op, 0x0, 0xff, 0);
EXPECT_EQ(EINVAL, errno);
prctl(PR_SET_SYSCALL_USER_DISPATCH, op, 0x0, 0x0, &sel);
EXPECT_EQ(EINVAL, errno);
only the first syscall may fail and set errno, but the second may succeed
and keep errno intact, and the check will falsely pass.
Or if errno happened to be EINVAL before, even the first check may falsely
pass.
Also use EXPECT/ASSERT consistently. Currently there is an inconsistent mix
without obvious reasons for usage of one or another.
Fixes: 179ef03599 ("selftests: Add kselftest for syscall user dispatch")
Signed-off-by: Dmitry Vyukov <[email protected]>
Signed-off-by: Thomas Gleixner <[email protected]>
Link: https://lore.kernel.org/all/af6a04dbfef9af8570f5bab43e3ef1416b62699a.1747839857.git.dvyukov@google.com
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 <[email protected]>
Reviewed-by: Palmer Dabbelt <[email protected]>
Acked-by: Palmer Dabbelt <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Palmer Dabbelt <[email protected]>
Michael Kerrisk suggested that, from an API perspective, it is a bad
idea to share the PR_SYS_DISPATCH_ defines between the prctl operation
and the selector variable.
Therefore, define two new constants to be used by SUD's selector variable
and update the corresponding documentation and test cases.
While this changes the API syscall user dispatch has never been part of a
Linux release, it will show up for the first time in 5.11.
Suggested-by: Michael Kerrisk (man-pages) <[email protected]>
Signed-off-by: Gabriel Krisman Bertazi <[email protected]>
Signed-off-by: Thomas Gleixner <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
This is the patch I'm using to evaluate the impact syscall user dispatch
has on native syscall (syscalls not redirected to userspace) when
enabled for the process and submiting syscalls though the unblocked
dispatch selector. It works by running a step to define a baseline of
the cost of executing sysinfo, then enabling SUD, and rerunning that
step.
On my test machine, an AMD Ryzen 5 1500X, I have the following results
with the latest version of syscall user dispatch patches.
root@olga:~# syscall_user_dispatch/sud_benchmark
Calibrating test set to last ~5 seconds...
test iterations = 37500000
Avg syscall time 134ns.
Caught sys_ff00
trapped_call_count 1, native_call_count 0.
Avg syscall time 147ns.
Interception overhead: 9.7% (+13ns).
Signed-off-by: Gabriel Krisman Bertazi <[email protected]>
Signed-off-by: Thomas Gleixner <[email protected]>
Reviewed-by: Kees Cook <[email protected]>
Reviewed-by: Andy Lutomirski <[email protected]>
Acked-by: Peter Zijlstra (Intel) <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Gregory Price