diff options
author | Rohan McLure <rmclure@linux.ibm.com> | 2022-09-21 16:56:01 +1000 |
---|---|---|
committer | Michael Ellerman <mpe@ellerman.id.au> | 2022-09-28 19:22:09 +1000 |
commit | 7e92e01b724526b98cbc7f03dd4afa0295780d56 (patch) | |
tree | d2c5648812a88af011dceb991917c820b8d80482 /arch/powerpc/kernel | |
parent | f8971c627b14040e533768985a99f4fd6ffa420f (diff) | |
download | linux-7e92e01b724526b98cbc7f03dd4afa0295780d56.tar.gz linux-7e92e01b724526b98cbc7f03dd4afa0295780d56.tar.bz2 linux-7e92e01b724526b98cbc7f03dd4afa0295780d56.zip |
powerpc: Provide syscall wrapper
Implement syscall wrapper as per s390, x86, arm64. When enabled
cause handlers to accept parameters from a stack frame rather than
from user scratch register state. This allows for user registers to be
safely cleared in order to reduce caller influence on speculation
within syscall routine. The wrapper is a macro that emits syscall
handler symbols that call into the target handler, obtaining its
parameters from a struct pt_regs on the stack.
As registers are already saved to the stack prior to calling
system_call_exception, it appears that this function is executed more
efficiently with the new stack-pointer convention than with parameters
passed by registers, avoiding the allocation of a stack frame for this
method. On a 32-bit system, we see >20% performance increases on the
null_syscall microbenchmark, and on a Power 8 the performance gains
amortise the cost of clearing and restoring registers which is
implemented at the end of this series, seeing final result of ~5.6%
performance improvement on null_syscall.
Syscalls are wrapped in this fashion on all platforms except for the
Cell processor as this commit does not provide SPU support. This can be
quickly fixed in a successive patch, but requires spu_sys_callback to
allocate a pt_regs structure to satisfy the wrapped calling convention.
Co-developed-by: Andrew Donnellan <ajd@linux.ibm.com>
Signed-off-by: Andrew Donnellan <ajd@linux.ibm.com>
Signed-off-by: Rohan McLure <rmclure@linux.ibm.com>
Reviewed-by: Nicholas Piggin <npiggin@gmai.com>
[mpe: Make incompatible with COMPAT to retain clearing of high bits of args]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20220921065605.1051927-22-rmclure@linux.ibm.com
Diffstat (limited to 'arch/powerpc/kernel')
-rw-r--r-- | arch/powerpc/kernel/syscall.c | 34 | ||||
-rw-r--r-- | arch/powerpc/kernel/systbl.c | 7 | ||||
-rw-r--r-- | arch/powerpc/kernel/vdso.c | 2 |
3 files changed, 27 insertions, 16 deletions
diff --git a/arch/powerpc/kernel/syscall.c b/arch/powerpc/kernel/syscall.c index 2c002cbbc676..18b9d325395f 100644 --- a/arch/powerpc/kernel/syscall.c +++ b/arch/powerpc/kernel/syscall.c @@ -15,7 +15,6 @@ /* Has to run notrace because it is entered not completely "reconciled" */ notrace long system_call_exception(struct pt_regs *regs, unsigned long r0) { - unsigned long r3, r4, r5, r6, r7, r8; long ret; syscall_fn f; @@ -145,31 +144,34 @@ notrace long system_call_exception(struct pt_regs *regs, unsigned long r0) return -ENOSYS; } - r3 = regs->gpr[3]; - r4 = regs->gpr[4]; - r5 = regs->gpr[5]; - r6 = regs->gpr[6]; - r7 = regs->gpr[7]; - r8 = regs->gpr[8]; - /* May be faster to do array_index_nospec? */ barrier_nospec(); +#ifdef CONFIG_ARCH_HAS_SYSCALL_WRAPPER + // No COMPAT if we have SYSCALL_WRAPPER, see Kconfig + f = (void *)sys_call_table[r0]; + ret = f(regs); +#else if (unlikely(is_compat_task())) { + unsigned long r3, r4, r5, r6, r7, r8; + f = (void *)compat_sys_call_table[r0]; - r3 &= 0x00000000ffffffffULL; - r4 &= 0x00000000ffffffffULL; - r5 &= 0x00000000ffffffffULL; - r6 &= 0x00000000ffffffffULL; - r7 &= 0x00000000ffffffffULL; - r8 &= 0x00000000ffffffffULL; + r3 = regs->gpr[3] & 0x00000000ffffffffULL; + r4 = regs->gpr[4] & 0x00000000ffffffffULL; + r5 = regs->gpr[5] & 0x00000000ffffffffULL; + r6 = regs->gpr[6] & 0x00000000ffffffffULL; + r7 = regs->gpr[7] & 0x00000000ffffffffULL; + r8 = regs->gpr[8] & 0x00000000ffffffffULL; + ret = f(r3, r4, r5, r6, r7, r8); } else { f = (void *)sys_call_table[r0]; - } - ret = f(r3, r4, r5, r6, r7, r8); + ret = f(regs->gpr[3], regs->gpr[4], regs->gpr[5], + regs->gpr[6], regs->gpr[7], regs->gpr[8]); + } +#endif /* * Ultimately, this value will get limited by KSTACK_OFFSET_MAX(), diff --git a/arch/powerpc/kernel/systbl.c b/arch/powerpc/kernel/systbl.c index d64dfafc2e5e..9d7c5a596171 100644 --- a/arch/powerpc/kernel/systbl.c +++ b/arch/powerpc/kernel/systbl.c @@ -15,13 +15,20 @@ #include <asm/unistd.h> #include <asm/syscalls.h> +#undef __SYSCALL_WITH_COMPAT #define __SYSCALL_WITH_COMPAT(nr, entry, compat) __SYSCALL(nr, entry) +#undef __SYSCALL +#ifdef CONFIG_ARCH_HAS_SYSCALL_WRAPPER +#define __SYSCALL(nr, entry) [nr] = __powerpc_##entry, +#define __powerpc_sys_ni_syscall sys_ni_syscall +#else /* * Coerce syscall handlers with arbitrary parameters to common type * requires cast to void* to avoid -Wcast-function-type. */ #define __SYSCALL(nr, entry) [nr] = (void *) entry, +#endif const syscall_fn sys_call_table[] = { #ifdef CONFIG_PPC64 diff --git a/arch/powerpc/kernel/vdso.c b/arch/powerpc/kernel/vdso.c index fcca06d200d3..e1f36fd61db3 100644 --- a/arch/powerpc/kernel/vdso.c +++ b/arch/powerpc/kernel/vdso.c @@ -39,6 +39,8 @@ extern char vdso32_start, vdso32_end; extern char vdso64_start, vdso64_end; +long sys_ni_syscall(void); + /* * The vdso data page (aka. systemcfg for old ppc64 fans) is here. * Once the early boot kernel code no longer needs to muck around |